CoAST Full Paper With Acknowledgement PDF

Proceedings
of
National Conference
On
Advances in Structural Technologies

(CoAST-2019)
Feb 1 to 3, 2019
Organized by
Department of Civil Engineering
National Institute of Technology Silchar
Under TEQIP-III
Proceedings
of
National Conference
on

(CoAST-2019)
Feb 1 to 3, 2019
Organized by
Under TEQIP-III
Proceedings of
National Conference
on

(CoAST-2019)
Feb 1 to 3, 2019
Organized by
Under TEQIP-III
Advisory Committee
Prof. T.K. Datta, IIT Delhi, India
Prof. T. Kant, IIT Bombay, India
Prof. S. Adhikari, Swansea University, UK
Prof. C.S. Manohar, IISc Bangalore, India
Prof. J.N. Bandyopadhyay, IIT Kharagpur, India
Prof. S. Majumder, IIT Kharagpur, India
Prof. V.K. Gupta, IIT Kanpur, India
Prof. D.K. Paul, IIT Roorkee, India
Prof. S.K. Bhattacharya, IIT Kharagpur, India
Prof. A. Dutta, IIT Guwahati, India
Prof. B. Bhattacharjee, IIT Delhi, India
Prof. Y. Singh, IIT Roorkee, India
Prof. S.K. Deb, IIT Guwahati, India
Prof. S. Talukdar, IIT Guwahati, India
Prof. A.K. Singh IIT Guwahati, India
Prof. S. Chakraborty, IIEST Shibpur, India
Prof. C.B. Kameshwar Rao, NIT Warangal, India
Organizing Committee
Patron-in-Chief: Director, NIT Silchar
Patron: HoD, CE, NIT Silchar
Chairman: Prof. S. Choudhury, CE, NIT Silchar
Prof. A.I. Laskar, CE, NIT Silchar
Secretaries: Dr. N. Debnath, CE, NIT Silchar
Dr. A. Sil, CE, NIT Silchar
Dr. M.L.V. Prasad, CE, NIT Silchar
Dr. S. Dey, ME, NIT Silchar
Dr. D. Bhowmik, CE, NIT Silchar
Dr. B.K. Roy, CE, NIT Silchar
Dr. Nitesh A., CE, NIT Silchar
Mr. P. Das, CE, NIT Silchar
Dr. Subhrajit Dutta, CE, NIT Silchar
Proceedings
of
National Conference
on

(CoAST-2019)
Feb 1 to 3, 2019
Organized by
Under TEQIP-III
Keynote Speakers
Prof. Biswajit Bhattacharjee

IIT Delhi
Prof. Nirjhar Dhang

IIT Kraragpur
Prof. Siddhartha Ghosh

IIT Bombay
Dr. Balaji Raghavan

INSA Rennes, France
Dr. Debiprasad Ghosh

DGM (Civil), L&T Construction
Message from the Director

It is a matter of great pleasure for me to know that Department of Civil Engineering
is organizing a National Conference on Advances in Structural Technologies
(CoAST-2019) under TEQIP-III from 1st to 3rd February, 2019. The
Department of Civil Engineering started its journey in 1977. Ever since the inception,
it has been imparting quality education to undergraduate & postgraduate students. The
Department presently fosters five PG programs viz. Structural Engineering,
Structural Dynamics & Earthquake Engineering, Geotechnical Engineering,
Transportation Engineering and Water Resources Engineering.
Modern structural technologies not only help to build safe and survivable structures but
also make the construction processes faster. National Conference on Advances in
Structural Technologies is intended towards a national level forum for profound
discussions in the field of structural engineering for innovative ideas, research and
development. I wish the state-of-the-art thoughts evolved from this conference help the
modern design and construction activities.
I hope that this conference would certainly induce innovative ideas among the
participants paving way for new inventions and new technologies in the domain of
Structural Engineering.
I wish the CoAST-2019 to be a grand success.
Prof. Sivaji Bandyopadhyay

Director, NIT Silchar
i
Message from the HoD, Civil Engineering
I am very happy to know that Structural Engineering Division is organizing a

National Conference on Advances in Structural technologies (CoAST-2019) under
TEQIP-III from 1st to 3rd February. The Department presently has five PG
programs viz. Structural Engineering, Structural Dynamics & Earthquake
Engineering, Geotechnical Engineering, Transportation Engineering and Water
Resources Engineering. A good number of research scholars are carrying out research
in various fields. There are about a dozen laboratories with various research and
testing facilities. The faculty consists of eminent specialists from diverse fields. The
Structural Engineering specialization commenced in the year 2013. Structural
Engineering specialization has applied for NBA accreditation process. All other
specializations and UG program are already with NBA accreditation.
First time ever of this scale the National Conference is organized by the Structural
Engineering division. I expect that this conference would cultivate state-of-the-art
ideas among the participants towards modern structural technologies. I strongly believe
the conference will be highly successful one.
Prof. Parthasarathi Choudhury

Head, Department of Civil Engineering
NIT Silchar
ii
Message from the Chairpersons, CoAST-2019
It gives us immense pleasure to note that the Department of Civil Engineering, NIT
Silchar, is going to organize a National level Conference on Advances in Structural
Technologies (CoAST-2019) from 1st February to 3rd February 2019. Structural
Engineering is the core of Civil Engineering. It advances fast and every time it is
renewed by fresh technologies and techniques. It is necessary that we take stock of such
advancement so that the same can find applications for general users. Taking of such
stock is possible through a conference like this. A conference is also a meeting place of
experts and learners. We earnestly believe that the conference will serve its intended
purpose.
We wish all success of the Conference.
A.I. Laskar S. Choudhury

Chairman, CoAST-2019 Chairman, CoAST-2019
iii
Brief Bio-Sketch of Prof. Biswajit Bhattacharjee, Keynote speaker
Biswajit Bhattachrjee is a Professor at the Department of Civil Engineering, Indian

Institute of Technology Delhi-India.After obtaining B. Tech (Hons.) degree from
I.I.T Kharagpur in 1978. He worked for M/s Gammon Ind ia Limited for a short
period of two years. Subsequently he obtained M. Tech. and Ph.D. degree from
I.I.T. Delhi in the year 1982 and 1990 respectively.
His areas of active research interest includes corrosion o f rebar in concrete, high-
performance concrete, microstructure modeling of concrete, chloride ingress,
service life pred iction and life cycle costing of co ncrete structures besides
condition evaluation and health monitoring of structures .
Minor area of research and teaching interests are Build ing Physics and
Build ingEngineering, Construction Technology and Co nstructio n management.
He is nominated Member of: 1) Academic Advisory Board in Engineering Sciences
at TERI; 2) Governing Cou ncil of Dr. Fixit Institute o f Structural Protectio n
&Rehabilitation;3) Research Council of CBRI.
He is EDITORIAL BOARD MEMBER of: 1) Magazine of Concrete
Research,Technical Journal Pub lished by Thomas Telford Ltd., U.K; 2)
International Journal of 3R's, Dr. FIXIT Laboratory for Repair & Rehabilitation,
INDIA.
He has authored more than 150 journal and co nference papers in leading
International journals such as ACI materials Journal, Journal of Materials in Civil
Engg. ASCE, Cement and Concrete research etc. and Indian Journals such as
ICJand ICI journals, besides authoring state o f the art report, lecture notes and
articles workshop proceed ings etc.
He has handled more than 200 consultancy and research projects and is actively
involved in various national committees including BIS committee.
Dr. Bhattacharjee has supervised around 130 M. Tech. and more than 20 Ph. D.
dissertations till date and currently gu iding few M. Tech. theses work besides few
ongoing Ph. D. research projects.
iv
Brief Bio-Sketch of Prof. Nirjhar Dhang, Keynote speaker
Nirjhar Dhang (born 1962) is a structural engineer and an educator. He is known

for his research on the structural engineering, in the area of structural dynamics &
control and impact stud y on concrete. Presently, he is working on soil structure
interaction, railway bridges applicable for high speed rail.
He is currently Pro fessor of the Department o f Civil Engineering, Indian Institute
of Technolo gy, Kharagpur, where he teaches Bridge Engineering, Design of
Reinforced Co ncrete Structures, Construction Planning & Management and
Highrise Structures.
He was also engaged as a Chairman, Civil Co nstruction and Maintenance for
theinfrastructure development o f the Institute. Different constructio n projects such
as,
laboratory complex for under graduate students, 2000 capacity boys’ ho stel,
students’ activity centre, resid ential accommodation for faculty and staff members
etc. are completed under his chairmanship.
He received B.E. degree in Civil Engineering from the University of Calcutta
(Bengal Engineering College, Sibpur, presently known as IIEST, Sibpur) in 1983,
M.Tech. degree in Structural Engineering from the Indian Institute of Technology,
Kharagpur and Ph.D. degree in Civil Engineering from the Indian Institute of
Technology, Kharagpur.
He has coordinated short term courses on Training o f Faculty Members of State
Resource Institutes under the National Programme of Cap acity Building of
Engineers in Earthquake Risk Management (NPCBEERM) and has coordinated
short term courses, such as, “Seismic Reliability and Life Assessment of
Structures” under National Pro gramme on Earthquake Engineering Educatio n
(NPEEE).
He has done many consultancy and research project works. He has published
numerous research papers in International/National journals and conferences.
v
Brief Bio-Sketch of Siddhartha Ghosh, Keynote speaker
Siddhartha Ghosh is currently a Pro fessor in the Department of Civil Engineering,

IIT Bombay. After comp leting his Ph.D. from the University of Michigan in 2003,
he joined IIT Bombay, where he has held various acad emic positions. His research
interests are primarily in the application of probab ilistic methodologies in
earthquake risk reduction. He has supervised/is supervising several doctoral and
masters research students working in the areas o f performance-based seismic
design, structural reliability, design of structural and co ld-formed steel, analysis of
masonry domes and arches, etc. Besides academic and sponsored research,
heteaches graduate and undergraduate courses in structural engineering and
probabilistic methods. He has conducted industrial research/consultancy works in
the areas related to structural dynamics, structural reliability, design o f steel and
RC structures, and retrofitting of structures. He is a member of the Bureau of
Indian Standards' committee CED 7 on steel structures.
For the academic year 2014-15, he was on sabbatical as a guest professor in the
Chair of Structural Dynamics and Earthquake Engineering in IBK, D-BAUG, ETH
Zurich.
Currently, he is the Professor-in-Charge of the 'Structural Safety, Risk and
Reliability (SSRR) Lab' and the 'Heavy Structures Lab'.
vi
Brief Bio-Sketch of Prof. Balaji Raghavan, Keynote speaker
Balaji Raghavan is a tenured Associate Professor of Civil and Mechanical

Engineering at the Institut National des Sciences Appliquees (INSA) at Rennes in
France since 2013. He has obtained his Bachelor of Techno logy degree in
Mechanical Engineering from IIT Bombay, India, his Master o f Sciences in
Mechanical Engineering with a minor in High Performance Computing from the
Pennsylvania State University (PSU), PA, USA and hisPhD in Advanced
Mechanics from the Universite de Technolo gie de Compiegne (UTC) in
Compiegne, France.
He conducts his research in the field of multi-scale modeling and simu lation for the
multi-physics characterization of materials in the Géomatériauxet Ouvrages Sous
Actio ns compleXes (GEOSAX) research group of the Laboratoire de Genie Civil et
Genie Mecanique (LGCGM EA 3913). He has published 25 papers in international
journals and over 50 papers in variou s international conferences, with anhindex of
12. He is actively invo lved in research co llaborations with the Narodowe Centrum
Bad ańJądrowych (NCBJ) in Świerk Poland, UTC, France as well as the reputed
Northwestern Polytechnical University (NPU) Xi'an, China, where he has also
served as Visiting Professor of nonlinear mechanics. In addition, he serves as
scientific reviewerand/or guest editor for various international journals.
vii
Brief bio-sketch of Dr. Debiprasad Ghosh, Keynote speaker
Debiprasad Ghosh is Deputy General Manager at Larsen & Toubro Constru ction,
where he works in Finite Element Analysis and Artificial Intelligence for the
design of different Metallurgical & Material Handling units. He lives in Kolkata.
He graduated from IIEST, Shibpur (formerly known as Bengal Engineering

College) with b achelor’s degrees in Civil Engineering and subsequently, he earned
his Master of Civil Engineering from Jadavpur University. Over the past two
decades, he has worked as Stru ctural Engineer in a wide variety of professional
capacities.
Debiprasad Ghosh did his Ph.D. from Indian Institute o f Science, Bangalore, and
subsequently, he has worked in the Solver team of STAAD/Pro. He has worked in
McNally Bharat Engineering Company Limited as Structural Design Engineer and
worked for Automation of Structural Enginee ring.
He has published many papers on subjects ranging from smart materials to

structural health monitoring. His research Interest is Finite Element Analysis,
Artificial Neural Network, Genetic Algorithm, and Structural Health Mo nitoring.
More recently due to upsurge of Deep Learning he is rethinking about its
utilization in Construction Industries.
Presently, he is exploring quantum structural health mo nitoring, and studying

quantum mechanics and computatio n on his own for a couple of years now, besides
his other works.
viii
National Conference
on
(CoAST-2019)
Feb 1 to 3, 2019
Organized by
Under TEQIP-III
FOREWORD
There has been constant and intense research & development in the arena of Structural
Engineering. As a result, new technologies and techniques are ever pouring down in design
philosophies, design procedures, construction techniques, optimization methods etc. To take a
stock on these, we need to halt for a moment and look back at what has been accumulated.
This helps in digesting the accumulated knowledge and its final applications. Such
conference gives us an opportunity to look back and take stock of new findings. Suggestions
also come as how to utilize the new knowledge. In this context the CoAST-2019 is a need of
the day.
A National Conference in Structural Engineering gives us an opportunity to exchange ideas

amongst the peer groups. Such interaction not only leads to a feeling of fraternity amongst the
Structural engineers, but also may open the avenue for inter-institute collaboration in
research, pedagogy and consultancy.
It was a long cherished dream for the Department of Civil Engineering to have a National
level conference in Structural Engineering.
In this volume the soft copies of papers registered in CoAST-2019 have been incorporated.
The Conference has been funded by TEQIP-III and this is freely acknowledged.
We have a term of reference with Springer Publications in the sense that select papers will be
published in Springer Proceedings with Scopus indexing. This shall certainly benefit the
authors.
At the last we sincerely hope that the Conference will benefit the participants in particular
and Structural Engineering Community in general.
A.I. Laskar S. Choudhury

Chairman CoAST-2019 Chairman CoAST-2019
ix
Index of Papers
Paper Id Title Page No.

10 Finite Element Simulation of Low Velocity Impact of a Conical Impactor 1-10
on a Honeycomb Sandwich Structure
12 Effect of Wind and Seismic forces on different Components of Cable 11-19
Suspension Bridge: An Overview
14 Replacement of Sand with Stone Dust as a Fine Aggregate in Concrete 20-27
15 A Brief Review of the Effect of Mineral Admixtures on the Properties of 28-35
Alkali Activated Composites
17 Use of Brick Kiln Dust as a Supplementary Material for Fine Aggregate 36-43
in Concrete
18 The Aerodynamic and Seismic Behaviour of Cable-Stayed Bridge 44-52
20 Basalt Fiber and Its Composites: An Overview 53-62
22 Mechanical and Durability Properties of Concrete Using Recycled 63-75
Aggregate
31 Seismic Response Control of a Building Using Passive Hybrid Damper 76-87
under Near Field Earthquakes
32 Structural Modification of a 65 Degree Delta Wing using a Step-Wise 88-100
Discontinuity
34 Socio-Economic Effect of Earthquake: A Case Study of Bhuj Earthquake 101-109
26th Jan 2001
35 Effect of Triaxiality and Lode Angle on Failure Behaviour of Riveted Lap 110-121
joints
36 Effect of Superplasticizer and Mineral Admixtures on Mechanical and 122-131
Durability Properties of Geopolymer Concrete: A Review
37 Innovative Optimization Techniques of Time and Cost Trade Off 132-141
39 Oil Contaminated Sand: Towards Cleaner Future 142-151
43 Wall Effects on Terminal Velocity of Test Fuel Bundle in the Fuel Test 152-163
Loop of High Flux Research Reactor
44 Risk Assessment and Management in Construction Projects 164-175
45 Influence of Material Damping on Dynamic Response of Foundation 176-183
48 Comparison of Seismic Performance of Composite (RCS) Frame with RC 184-195
Frame using Pushover Analysis
50 A Review on Geo-Polymer Pervious Concrete by Using Recycled Coarse 196-208
Aggregate
51 Investigating Load Withstand By L-Shape Concrete Cube, RCC Slab and 209-216
Safeguarded Reinforcement of RCC Slab In Salt-Water Environment
Using Cathodic Protection
52 Performance Evaluation of Two-Way RC Slab Subjected to Blast 217-228
Loading Using Finite Element Analysis
55 Organic and Inorganic Elements Used for Co2 Absorption in Concrete 229-236
57 Development of Fragility Curves for Different Types of RC Frame 237-248
Structures
63 Chances of fall from Height: Identification of Root Cause in Building 249-255
Construction Projects
64 Smart Light Weight MR Damper for Enhancement of Seismic Mitigation 256-264
66 Rheological Behavior of Geopolymer Mortar with Fly ash, Slag and their 265-282
blending
67 Effect of Pile Spacing and Raft Thickness on Behaviour of Piled-Raft 283-294
Foundation – A Parametric Study Using FEM
69 Removal of VOCs and Improvement of Indoor Air Quality Using 295-302
Activated Carbon Air Filter
73 A Comparative Study of Normal and Self Compacting Concrete 303-308
74 Effect of Acidic Environment on Durability of RHA Concrete 309-312
76 Alkali-Activated Concrete – A Review 313-319
80 Stability of Different shapes of Tunnels in Weathering Stages of Basalt 320-327
83 Evaluating Toughness as a Parameter to Determine Fatigue Life of 328-348
Wollastonite Micro Fiber Reinforced High Flow Pavement Quality
Concrete
85 High Flow Pavement Quality Concrete Reinforced with Wollastonite 349-371
Microfiber for Rigid Pavements
86 Wet State Analysis of WMF Reinforced & Flyash Admixed Cement 372-394
88 Comparison of the Behaviour of Deep Beams Designed Using IS Code 395-401
and STM Method
89 Efficient Static Analysis of Laminated Composite and Sandwich Shell 402-413
Structures Using ABAQUS
93 Dynamic Response of Weathered Jointed Rock Slope Having the 414-422
Transmission Tower
98 Response of Single and Multilayered Flexible base for Static and 423-433
Earthquake loading under Framed RC Structure
100 An Approach for Studying Dynamics of a Rotor Bearing System using 434-439
ANSYS
102 Construction of Overhead Water Tank - An Unconventional Approach 440-451
104 Purview of Nonlinear Static and Dynamic Analysis Procedures on 452-463
Seismic Fragility Estimation of RC Buildings with Masonry Infills
107 Effect on Curing Techniques on Durability Properties of Conventional 464-475
and HVGGBFS Concrete
108 Comparative Study of Mixture Design Methods used for Self-Compacting 476-486
Concrete
109 Management of Sustainable Development of Territories 487-498
110 Effect of Different Pozzolanic Materials on Partially Replacing Natural 499-510
Sand by M-Sand
111 Finite Element Analysis of Unconfined Brick Masonry with Experimental 511-524
Verification
112 Analysis of Moment and Torsion in skew plates using ABAQUS 525-540
113 Experimental Study of the Use of Alccofine as a Corrosion Inhibiter in 541-552
Steel Fiber Reinforced Concrete
114 Utilization of Lime in Cement-Based Materials: Review of Fresh, 553-573
Mechanical and Durability Properties
116 Finite Element Analysis of a Beam with Corrugated Web of a Jib Crane 574-591
118 Behaviour of Liquid Storage Tank under Multi-Directional Excitation 592-605
120 Effect of Nitric Acid on Metakaolin Concrete 606-613
121 Prospective Construction Products Using Sugarcane Bagasse Ash- A 614-621
Review
124 Osdag: A Software for Structural Steel Design using IS 800:2007 622-634
128 Experimental and Analytical Investigations on Two-way Flexural 635-648
Capacity of Biaxial Voided Slab
130 Review on Limestone Calcined Clay Cements (LC3) 649-662
131 Performance Issues and Durability Criteria of Concrete Structures under 663-668
Marine Exposed Condition: A Global Trend of Research
135 Sustainable Mortar Production by Utilizing Post Consumed PET Waste 669-679
and Copper Slag
136 Finite Element Analysis of Profiled Deck Composite Slab Using ANSYS 680-687
137 A comparative study of Seismic Response of Structure isolated with 688-699
Triple Friction Pendulum Bearing and Single Friction Pendulum Bearing
under Different Hazard Level of Earthquake
138 Assessment of Important Parameters for Seismic Analysis and Design of 700-711
Confined Masonry Buildings
139 Design and Performance Criteria for Fire Resistant Design of Structures – 712-726
An Overview
140 Performance of Metakaolin Concrete under the Acidic Environment 727-735
141 Static Analysis of Skew Laminated Composite Plates using Isogeometric 736-746
Finite Element Method
143 Fabrication and Mechanical properties of Marble Dust Filled Aluminum 747-751
Metal Matrix structural Composite
144 Properties of Sustainable Concrete Using Copper Slag at Varying Cement 752-761
Content
145 Wear Behavior of Marble Dust Filled Aluminium Metal Matrix Structural 762-767
Composite
148 Seismic Response of Overhead Water Tank with different Staging 768-779
systems
152 A Relational Approach to Quantify Sustainable Performance of Concrete 780-789
using Preference Selection Index (PSI) Method
154 Quantifying Uncertainty in Structural Responses of Polymer Sandwich 790-797
Composites: A Comparative Analysis of Neural Networks
155 Optimization of Cantilever Retaining Wall using Particle Swarm 798-806
Optimization
156 Impact Analysis of Ultra High Performance Fiber Reinforced Concrete 807-812
slabs
157 Development of Low Carbon Binder for Sustainable Construction as an 813-823
Alternative to Conventional Cement
161 Structural Health Monitoring Using Wavelet Transform 824-833
162 Numerical Simulation of Combined Pile-Raft Foundation under 834-841
Horizontal Loading
163 Fluid-Structure Interaction Problem in Elevated water tank 842-849
164 Buckling Analysis of braced frames under axial and lateral loadings 850-869
165 Structural Evaluation of Existing RCC Bridges for Durability 870-879
166 Experimental Study of Concrete Prepared as a Partial Replacement of 880-885
Cement by Various Admixtures
168 Dampness Silent Killer in Structural Buildings- A Case study 886-895
170 Prediction of Compressive Strength of High Strength Fiber Reinforced 896-906
Concrete Using Artificial Neural Networks
171 Effective Public Private Partnership Model for Redevelopment of 907-924
Government Buildings
172 Key Performance Indicators for Public Private Partnership Projects in 925-936
India-a case study
173 Performance of Reinforced Concrete Beams with M-Sand as Alternate 937-948
Fine Aggregates
177 Load-Moment Interaction Behaviour of Axially Loaded Short 949-957
Compression Members
178 A Study on Moment– Curvature Relationships for Reinforced Concrete 958-964
Beams with Varying Fire Loading Conditions
179 Effect of Slab Thickness on Period of Vibration of Reinforced Concrete 965-972
Building
180 Coir Fibre Reinforced Concrete mixed with GGBS 973-981
181 Axial Stress - Strain Model for Plain Concrete and Concrete Filled Steel 982-990
Tube Columns
186 Bamboo Reinforced Concrete Beam - Design Approach for Low Cost 991-996
Housing
188 Sampling Based Techniques for Finite Element Model Updating in 997-1010
Bayesian Framework using Commercial Software
192 Stochastic Structural Optimization of Multiple Tuned Mass Damper 1011-1020
(MTMD) System
193 Behaviour of UHPFRC under Explosive Loading: An Overview 1021-1026
195 Experimental Study on Liquid Sloshing Dynamics in a Sway Excited 1027-1031
Rectangular Tank
196 Hearth Monitoring of Blast Furnace using Finite Element Analysis and 1032-1040
Artificial Intelligence
201 Coconut Fibre Reinforced Concrete 1041-1048
202 Vibration of Simply Supported Laminated Composite Plates 1049-1056
204 Effect of Temperature Differential on Fatigue Performance of Marble 1057-1069
Dust added Concrete Pavement
205 Fatigue Resistance of Recycled Steel Fibres (Discarded Vehicle Tyre 1070-1088
Steel Fibres) Concrete Pavement
206 Innovative Techniques to Improve Strength and Sustainability of 1089-1097
Concrete by Shock Wave Applications
Proceedings of National Conference on Advances in Structural Technologies (CoAST-2019), 1-3 Feb, 2019
Paper ID- 10
Finite Element Simulation of Low Velocity Impact of a

Conical Impactor on a Honeycomb Sandwich Structure
Hrishikesh Dev Sarma1

1
Research Fellow, Indian Institute of Technology, Gandhinagar Gujarat 382355, India
hsarma.ce@gmail.com
Abstract. Honeycomb sandwich structures are widely used in the aerospace

industry. They are preferred for their high flexural rigidity, high crushing
strength and light weight. These structures especially when used in airplanes are
susceptible to low velocity impacts due to tool drops during maintenance work,
bird hits, runway debris etc. Therefore it is important to understand the response
of the structure to such low velocity impacts. This paper focuses on predicting
the response of a honeycomb sandwich structure to a low velocity impact by a
conical impactor. The conical impactor is used in this case as most objects that
hit the structure are expected to have rough or pointed edges. It considers an
aluminum honeycomb structure sandwiched in between two face sheets made
of aluminum. Initially, experiments are performed to study the response of the
structure to quasi static crushing load at different rates of loading. The
specimen is then modelled using a commercially available finite element
analysis package (ABAQUS) and the model is validated by comparing its
response to that obtained from the experiments. The same model is used to
simulate the low velocity impact of a conical impactor on it. The deflection at
the point of contact and the contact force thus obtained are studied by varying
various material and geometric parameters. The results obtained from the finite
element analysis are then compared with analytical formulations.
Keywords: Sandwich Structures, Honeycomb Structures, Impact Mechanics,

Lightweight Structures, Finite Element Analysis, Composite Structures.
1 Introduction
The aluminium honeycomb sandwich structure used in this study consists of a top and bottom
sheet along with a hexagonal honeycomb core all made of aluminium. The honeycomb core is
manufactured by expanding aluminium laminates which are periodically bonded. Upon
expansion the final product is that of a structure that consists of an array of hollow hexagonal
shells. A honeycomb sandwich structure has minimum density and a relative high out-of-plane
compressive strength and out-of-plane shear strength. This makes it a very efficient energy
absorbing structure. These properties make them very desirable for a large range of applications
especially in the fields of transportation and aerospace.
The crushing properties of honeycomb structures have been studied in details by early
researchers like Kelsey [1], McFarland [2], Wierzbicki [3] etc. Wierzbicki [1] has suggested
1
various simple formulas to predict the crushing strength, folding length and deflection of
honeycombs. Recent studies by Mohr and Doyoyo [4] examine the mechanical behaviour of
aluminium honeycomb subjected to combined normal and transverse loads. Wilbert et al [5]
give a detailed finite element analysis of the honeycomb and also have studied the effect of
various imperfections on the out of plane properties of the honeycomb. Indentation of a
sandwich panel with elastic core by a spherical impactor has been investigated by Abrate [6].
Zhou & Stonge [7] have developed an analytical model of deformation of plates, to calculate
the impact force and deflection during low speed impact on circular sandwich panels when
impacted on by a spherical impactor. This model based on an energy balance approach has
been utilised in this study to predict and verify the obtained results.
In this study, the honeycomb sandwich structure is subjected to a quasi-static crushing test.
The response is then used to validate a numerical simulation which is done using the finite
element analysis package ABAQUS [8]. Further a conical impactor is modelled numerically
and a low velocity impact is simulated using on same model of the sandwich structure. The
deflection at the point of impact is obtained from this simulation. The results are compared to
analytical prediction from formulations obtained.
2 Quasi-static crushing of an aluminum honeycomb sandwich

structure
2.1 Experimental investigations and observations

The aluminium honeycomb that forms the core of the sandwich structure is subject to a series
of crushing tests. A displacement controlled crushing load is applied on the honeycomb placed
between the two platens as shown in Fig. 1. The honeycomb sections, Fig. 1 (c), used for this
test typically consist of a 5 x 10 array of cells extracted from a larger sample. The dimensions
of the specimens are shown in Table 1. In the table, s is the distance between two opposite
faces of any hexagonal cell in the direction of extrusion, d is the length of a side of the hexagon
and h is the thickness of the cell wall. hc, LL and LW height (out of plane) length and width of
the specimen respectively.
Fig. 1 (a) honeycomb specimen before loading (b) honeycomb after complete crushing (c)
diagram of typical honeycomb specimen (top view) (d) loading setup
2
Table 1 geometric specifications of the honeycomb used in the crushing test.
s (mm) d (mm) h (μm) hc (mm) LL (mm) LW (mm)

8.0 4.5 60.0 36.5 80.0 40.0
Although the loading rates are varied from 2 mm/minute to 60 mm/minute (Table 2), but the
response of the specimens does not show any significant variations. Fig. 2 shows the
experimental stress strain curve obtained for a rate of loading of 60 mm/min. The stress-strain
curves for all the specimens are almost linear up to the peak crushing stress (σp). The peak
crushing stress obtained show minor variations with an average of 1.76 MPa whereas one
specimen (loaded at 60 mm/minute) resisting a crushing stress of 2.04 MPa. These minor
variations can be attributed to various factors like geometric imperfections, minor platen
misalignments, etc. (Wilbert et al [5]) No direct correlation is established between the rates of
loading and the induced stresses. Prior to the achievement of the peak the honeycomb remains
in an elastic state with the cell walls showing a global buckling behaviour. Consequent to the
peak crushing stress the stress-strain curve drops abruptly and reaches a local minimum. This
minimum is concomitant to the first local fold in the honeycomb specimen. Further loading of
the specimen results in an undulating curve as seen in Fig. 2 which hovers about a plateau or
mean stress. Each of these undulations represents formation of local folds in the honeycomb
cells. The average crushing stress, , is considered to be the mean value of the stresses
corresponding to the crushing strain of 0.1 to 0.8. The average crushing stress is observed to
be around 0.96 MPa. As the crushing (δ) increases, i.e. as the specimen gets progressively
crushed it reaches a point where the whole specimen has failed by the formation of progressive
folds. At this point no more crushing can occur and densification starts. For the specimen the
densification starts at a crushing strain of around 0.8 (not shown).
Table 2 peak and plateau stresses at various rates of loading
Rate of loading
2 5 10 20 30 40 50 60 Average
(mm/min)
Peak Stress (MPa) 1.66 1.80 1.64 1.79 1.70 1.72 1.72 2.04 1.76
Plateau Stress (MPa) 1.10 1.07 1.02 0.88 0.90 0.91 0.95 0.88 0.96
Experiment
2 FEM
Nominal Strain (MPa)
1.5
0.5
0
0 0.1 0.2 0.3 0.4 0.5 0.6
Strain
Fig. 2 nominal stress vs. nominal strain curve of a honeycomb specimen subjected to crushing
3
2.2 Numerical modeling and simulation

The quasi-static crushing of the aluminium honeycomb as described in the previous section is
replicated in a finite element setup using ABAQUS[8]. The honeycomb having the same
geometry as that given in Table 1 is modelled. The material properties of aluminium are
tabulated in table 3. The ABAQUS dynamic-explicit solver was used to perform the
simulations. The honeycomb core was sandwiched in between two platens modelled as rigid
shells. Fixed boundary condition was applied to the bottom platen. The top platen was allowed
translation in the direction corresponding to the loading and fixed against translation and
rotation in all the other directions. Velocity boundary conditions were applied to the top platen.
In order to reduce the computation time the velocity was given as 1 m/s. Strain rate effects were
disabled and in order to prevent any effect of inertia, a smooth step function was used such that
there is no sudden change in velocity. The honeycomb core was modelled using 4-node shell
elements S4R with five-integration points through thickness.
Fig. 2 shows the comparison of the stress strain curve obtained from the experiment and
simulation. It is seen that the results obtained through simulation agree very well with those
obtained from the experiments. The peak stress obtained in the simulation is 2 MPa which is
almost equal to that obtained in the experiment i.e. 2.04 MPa. However it is apparent that the
initial stiffness of the experiment and the simulation do not show good agreement. This occurs
because the simulation assumes that out model is exactly vertical with no imperfection. In
reality the transfer of load to the complete specimen is gradual with some parts of the core
coming in contact with the loading plated before the rest. This coupled with any minor
misalignment of the platens usually results in a lower stiffness. Wilbert et al. [5] shows how
various imperfections in the honeycomb such as irregularity of the hexagonal structure, platen
misalignment and imperfect bond length could lead to changes in response of an axially loaded
honeycomb structure. This model was also able to give a somewhat accurate prediction of the
mean crushing force of the honeycomb which is around 1.02 MPa.
3 Modeling impact of the conical impactor
3.1 Numerical modeling of the impactor, sheets and core

The model (Fig. 3) consists of a conical impactor, the face sheet on which the impact occurs
and the honeycomb core. The interaction between the impactor and face sheets is given by
surface to surface hard contact interaction. The boundary condition given to the top plate is
such that it is free to rotate but restrained against translation in the in plane directions. The
impactor can only move in the out of plane direction corresponding to the plate and the
honeycomb. The plates and the core are constrained to each other using a tie constraint. A face
plate of diameter 65 mm (constant for all simulations) has been modelled as a solid deformable
plate created by extrusion. The aluminium material is modelled as elastic perfectly plastic. The
ABAQUS dynamic-explicit solver was used to perform the simulations. The meshing was done
using 8-noded C3D8R elements. The mesh becomes finer as we approach the point of contact
of the impactor. The impactor is modelled as a solid deformable body conical in shape. The
discretization of impactor is done with different degrees of mesh refinement .The finest mesh is
provided at the point of contact. It was meshed with tetrahedral elements of second order (using
the modified 10-node tetrahedral elements C3D10M). The opening angle of the cone, i.e. the
angle between the axis and the lateral face of the cone is kept constant at 55 degrees. In order to
prevent concentration of stresses at the vertex of the cone, a fillet of 0.5 mm radius has been
used. This helps to prevent singularities. The core is a honeycomb structure made of shell
elements. The structure is uniformly meshed using a 4 noded conventional shell element (S4R).
4
Fig. 3 finite element model of impactor and sandwich panel with meshing
Table 3 material Properties used for the finite element model
Material Part Young’s Yield stress Density Poisson’s

Modulus (GPa) (MPa) (kg/m3) ratio
Steel Impactor 200 600 7850 0.3
Aluminum Core and faceplates 69 250 2700 0.3
3.2 Results of numerical simulations and observations

Fig. 4 shows a typical Von Mises stress contour contour for the sandwich panel (cut at the mid
section) impacted upon by a conical impactor. The indentation shown here is caused by a 2 kg
mass impacting on a sandwich structure at a velocity of 5 m/s.
Fig. 4 deformation of the sandwich panel seen after impact at the mid-section
5
As the impactor hits the sandwich panel, the face sheet gets deformed first and simultaeously
the core also gets deformed. The stresses near the point of impact are maximum. The typical
displacement-time, force-time and force-displacement plots of finite element simulations have
been shown in Fig. 5. In the displacement-time curve it is seen that there is a smooth increase in
displacement up to a maximum, after which the elastic bending and membrane stresses in the
plate are released. Consequently, the impactor rebounds away from the surface of the face
sheet. This is characterized by a smooth increase in the force-time curve as long as the impactor
is in contact with the plate. As soon as the impactor leaves the plate, there is a steep decline in
the contact force. The residual defection seen in the deflection-time curve is due to the plastic
deformation which has caused the indentation. The honeycomb core, however, cannot regain
any of the deformation that has occurred in the area in contact with the face plate. In this region
delamination generally occurs. Observing the force-displacement curve it can be said that the
initial stiffness of the panel remains high as it remains in the elastic region. As the material
starts yielding and the area of contact between the impactor and the facesheet increases the
slope of the force-displacement plot decreases
Deflection vs TIme Reaction Force vs TIme Reaction Force vs Deflection
6
Central Deflection (mm)
Reaction Force (kN)
Reaction Force (kN)

2 2
4 1.5 1.5
1 1
2
0.5 0.5
0 0 0
0 5 10 0 5 10 0 2 4 6
Time (msec) Time (msec) Deflection (mm)
Fig 5 plots showing deflection-time, force-time and force-deflection relationships as the

impactor hits and rebounds off the sandwich panel
4 Analytical method to find deflection and reaction force

The problem of impact on a sandwich structure may be divided into two parts: the local
indentation and the global panel deflection. Zhou and Stronge [7] developed a model to
calculate the total deformation due to the impact and reactive force during low speed impact on
circular sandwich panels. A similar approach is adopted here to find the local indentation and
panel deflection.
4.1 Contact law for local indentation
A rigid conical impactor with mass , has been considered that is striking the center of a
circular sandwich panel at velocity at normally. The panel has radius and is composed of
two facesheets and a core with thicknesses and , respectively. The transverse shear
modulus of the core is represented by and Young’s modulus of the facesheet material is .
The indentation of the contact region at the centre of the upper face sheet is denoted by α0 while
the global transverse deflection at the centre of the panel is represented by ω0 measured at the
mid-surface. Rigid boundary condition has been assumed at the bottom of the sandwich panel
for formulating the indentation law. The boundary condition at the periphery of the indentation
area is assumed to be clamped. The core of the sandwich panel is assumed to be rigid-perfectly
plastic. The indentation is assumed to be larger than the thickness of the face sheet.
6
After running a number of simulations for different impactors, face sheets and cores, an
equation has been chosen to trace the profile of the deformed surface after the indentation has
occurred. This has been illustrated in fig. 6(b).
0
Normalised Deflection
-0.2
-0.4
-0.6
-0.8 actual profile

predictive equation
-1
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
Normalised Radius
Fig. 6 (a) parameters defining the indentation (b) deformation profile obtained numerically
against that suggested theoretically
Using the notations shown in fig. 6(a) the profile of local indentation of the upper face sheet
can be to be represented by
⁄
( ( ) )
where α and a are the central transverse deflection and the radius of region of local indentation
on the upper face sheet, respectively. The equation of the profile should be chosen carefully as
it greatly affects the final results.
The flexible strain bending energy in the face sheet as expressed by Raju et al [9]
∫ ∫ [( ) ( ) ]
This can be simplified by substituting for Df, a and α to obtain, ,
where Ef , hf , and ν are the Young’s modulus, thickness and Poisson ratio of the faceplate,
respectively. The corresponding strain energy due to membrane stretching is
∫ ( ) ∫ ( )
7
where ϵr is the radial strain in the plate and ϵθ is the circumferential strain. This is solved to
obtain the expression for strain energy due to membrane stretching as
The work done by crushing of a perfectly plastic core is ∫

where q is the yield stress of the core. The work done by the contact force P is given by
The total potential energy, . Minimizing the total potential energy with
respect to the central deflection, i.e., yields a contact force
* +
Minimizing the load P with respect to the radius of the region of local indentation a (note: this
is not the contact area) i.e., ∂P/∂a=0 and substituting into the expression for P gives a force-
indentation relationship
√ ( )
Specifically, for a Poisson ratio of ν = 0.3, the above relationship becomes
√ ( ) (1)
Denoting √ and ⁄ , √
4.2 Global panel deflection
Assuming that a mass of M1 hits the panel, the initial kinetic energy of the mass is equal to the
sum of the energies due to localized indentation in the contact area together with global
bending, shear, and membrane deformations in the panel, i.e.,
where and are the mass and initial velocity of the colliding projectile, respectively, is
the bending and shear strain energy, is the indentation energy in the contact region, and
is the global membrane stretching energy.
Substituting the expressions for Ebs, Ec and Em in the expression we have [7]
∫ √ (2)
8
Here Kbs is the effective stiffness due to the combined bending and shear, w0 is the central
deflection and Km is the membrane stiffness. The expression has two unknowns w0 and α0m.
Hence Newton-Rhapson method is employed to obtain the final value of deflection from
equations (1) and (2) by assuming an initial value of indentation radius a.
5 Comparison of FEA with analytical results

The Fig. 7 shows the comparison of the analytical and theoretical prediction of the central
transverse deflection (δ) of the face sheet while varying the mass of impactor, velocity of
impact and plate thickness.
Fig. 7 central Transverse Deflection (δ) for (a) varying velocities of impact; (b) varying masses
of impactor; (c) varying plate thicknesses
For the first case the mass of the impactor is varied from 0.5 kg to 5.5 kg. The simulation is
carried out for 3.0 m/s impact velocity. The other parameters of the sandwich structure and the
impactor remain same. As expected deflection increases as we increase the mass of the
impactor. Larger mass resulted in a longer period of contact and larger deflection as well as a
larger contact force. However, it was observed that the theoretical prediction of the deflection
was better for greater masses. In the second plot, the velocity of impact is varied from 1.0 m/s
to 6.0 m/s. The simulation is carried out for a 2 kg impactor hitting the sandwich panel at this
velocity range. The other parameters of the sandwich structure and the impactor remain same.
Here too we get an expected increase in deflection with an increased velocity of impact. FEA
results seem to agree well with the theory. In the third plot, the thickness of the face sheet is
varied. With constant impact energy, the thicker plate results in larger contact force and smaller
central deflection. The impact duration also tends to decrease with thicker plates. An impactor
of mass 3 kg hitting the sandwich panel at a velocity of 2 m/s has been simulated by varying the
thickness of the plate from 1.0 mm to 4.0 mm. The theory and FEA results do not exactly
match in any of the cases here with the theory over predicting the deflection for thinner plates
and slightly under predicting for thinner plates. The reason for this can be attributed to the fact
9
that, the analytical solution is based on Mindlin-Reissner plate theory which is more suited for
thicker plates.
6 Conclusions
The following conclusions can be gathered from the study. The finite element analysis gives a
robust representation of the impact of the conical impactor on the sandwich structure. The
results obtained show good agreement with the theoretical predictions. The predicted results
using the energy balance model developed by Zhou & Stronge [7] for a spherical impactor are
close to those obtained from the finite element analysis. However care must be taken in order to
choose an accurate deformation profile using which the stored strain energies are computed. In
this case a few other profiles were also tried but with inferior results. The deflection predicted
analytically depends on the initial assumption of the contact radius also. The assumed radius
must be chosen suitably as using Newton-Rhapson method may result in convergence issues.
Experimental validation of the derived expressions and finite element model could lead to a
better understanding of the problem.
Acknowledgements The work herein was completed as part of a master’s degree thesis at the
Indian Institute of Technology Guwahati. The author acknowledges the financial assistance
provided by the Ministry of Human Resource Development (MHRD), and the guidance of Dr.
Amit Shelke, Assistant Professor, IIT Guwahati, for his invaluable insight and expertise.
References
1. Kelsey S, Gellatly RA, & Clark BW (1958) The Shear Modulus of Foil Honeycomb
Cores. Aircraft Engineering and Aerospace Technology 30(10): 294–302.
doi:10.1108/eb033026
2. McFarland R K (1963) Hexagonal Cell Structures Under Post-Buckling Axial Load.
AIAA Journal 1(6): 1380–1385. doi:10.2514/3.1798
3. Wierzbicki T (1983) Crushing analysis of metal honeycombs. International Journal
of Impact Engineering 1(2): 157–174. doi:10.1016/0734-743X(83)90004-0
4. Mohr D, Doyoyo M (2004) Deformation-induced folding systems in thin-walled
monolithic hexagonal metallic honeycomb. International Journal of Solids and
Structures 41(11–12): 3353–3377. doi:10.1016/j.ijsolstr.2004.01.014
5. Wilbert A, Jang WY, Kyriakides S, & Floccari JF (2011) Buckling and progressive
crushing of laterally loaded honeycomb. International Journal of Solids and
Structures 48(5): 803–816. doi:10.1016/J.IJSOLSTR.2010.11.014
6. Abrate S (1997) Localized Impact on Sandwich Structures With Laminated Facings.
Applied Mechanics Reviews 50(2): 69. doi:10.1115/1.3101689
7. Zhou DW, Stronge WJ (2006) Low velocity impact denting of HSSA lightweight
sandwich panel. International Journal of Mechanical Sciences 48(10): 1031–1045.
doi:10.1016/J.IJMECSCI.2006.05.011
8. Hibbitt, Karlsson & Sorensen, Inc. ABAQUS. ABAQUS/Standard user's manual
volumes I-III and ABAQUS CAE manual. Version 6.12. Pawtucket (USA); 2012.
9. Raju KS, Smith BL, Tomblin JS, Liew KH, Guarddon JC (2008) Impact Damage
Resistance and Tolerance of Honeycomb Core Sandwich Panels. Journal of
Composite Materials 42(4): 385–412. doi:10.1177/0021998307088596
10
Paper ID- 12
Effect of Wind and Seismic forces on different Components of Cable

Suspension Bridge: An Overview
Nayan Gupta1, Tanmoy Barik1, Sayan Dey1, Dr. Purnachandra Saha2*(ORCID-0000-0002-2722-1669)
1
Students, School of Civil Engineering, KIIT Deemed to be University, Bhubaneswar
(Mail id: nayanshub@gmail.com, barik.tanmoy.1997@gmail.com, sayandey004@gmail.com)
2
Faculty, School of Civil Engineering, KIIT Deemed to be University, Bhubaneswar, India
(Mail id: dr.purnasaha@gmail.com) *Corresponding Author
ABSTRACT
A Cable Suspension Bridge is a class of bridge in which the deck/stiffening girder is hung by the support of
suspension cables suspenders. Cable suspension bridge can be constructed over large spans and works on the
mechanism of the simply supported beam. Components of the cable suspension bridges are deck, stiffening
girder, suspenders, main cable, pylon and anchorage block. The objective of the present study is to present an
overview of the structural behaviour of different components of cable suspension bridge with respect to
aerodynamic and seismic effect. Effects such as Motion Induced flutter, Vortex Induced Vibration instability
and Turbulence Induced Buffeting are experienced by Long Cable Suspension Bridges. The earth's longest
Cable Suspension Bridge, at present Akashi Kaikyo Bridge connect with a span range of 1991m is in Japan.
Super long-range spans require a propelled comprehension of the wind-bridge communication to fulfil the
expanding wellbeing and financial needs.The seismic examination of long-range spans subjected to various
ground excitations is an imperative issue. The conventional response spectrum technique disregards the spatial
impacts of ground movement, and thusly may result in faulty ends. The irregular vibration approach has been
viewed as more dependable. Lamentably, up until this point, computational troubles have not yet been
satisfactorily resolved.
Keywords: cable-supported bridge, wind, seismic, dynamic behaviour, cable, suspenders, pylons
1. Introduction
In the bridge development history, Suspension bridge has a unique place in establishing the social and
financial needs of the society for effective and helpful transportation frameworks. These bridges have various
specialized and stylish favourable circumstances, allowing overlapping average and large spans. This sort of
structures can be built over long ranges, and because of the higher precision, execution, figuring and control
framework usage, they are safe to use [1-4].
The present-day Suspension Bridge build was in the nineteenth century. In those bridges, the cables are
conveyed by utilizing vertical suspenders. The two closures of the cable are suspended on the towers. The
weight of the bridge is shifted to the anchorages by cables. The anchorages are settled immovably on solid
squares or strong rocks. With the end goal to dispense the weight equally and furthermore to secure the cables
shape breaking, the cables are spread over a large area inside the anchorage. Design of a suspension bridge
involves developing and designing such a model which can satisfy a set of pre-defined criteria within the
framework of a precise idea. These bridges have the fundamental basic parts which include Deck, Stiffening
Girders, Main Cables, Hangers, Pylons, Foundation and Anchorages [5-7].
11
Fig: 1 Components of Suspension Bridge

Since the lamentable fall of the first Tacoma Narrows in 1940, the wind induced instability or
aerodynamic behaviour of different components of these bridges have been a noteworthy issue in the design and
analysis. In the recent times many world records of long and stable Cable suspension bridges have been broken
by many countries around the globe taking the engineering of Bridge design to a whole new standard, managing
to acquire many distinctive ways to deal with aerodynamic instabilities. The main span of Akashi Kaikyo Bridge
in Japan has a length of 1991m, where an opened deck is bolstered on a support brace/truss girder, the spaces
enhance the streamlined strength. With a twin box support and an intermediate hole of 6m the Xihoumen Bridge
in China traverses 1650m [8-10], the intermediate hole was acquainted to decrease the critical flutter wind
speed. The likelihood of longer spans is identified with the usage of more grounded and light materials. [11-13]
Cable supported bridges perform better amid quakes contrasted with different sorts of scaffolds. There are
seismic frail focuses in cable supported bridges which are probably going to end up a wellspring of harms when
the structure encounters a tremor [14]. Tremor harm to an extension can have serious results. Obviously, the fall
of a bridge places individuals on or underneath it in danger, and it must be supplanted after the seismic tremor
except if different possible transportation ways are identified. The outcomes of less serious harm are subtler and
more emotional, yet they are regardless essential. A bridge conclusion, regardless of whether it is brief, can have
tremendous consequences, since bridge provides crucial connections in a transportation framework. In the quick
result of a seismic tremor, shutdown of a bridge can impairs emergency response actions. [15-17].
In this paper we have discussed about the aerodynamic and seismic loading in different components of the
bridge. Cable suspension bridges are highly vulnerable to wind load. This is because of their size, span length,
height of tower and location. These bridges are mostly located on rivers have huge width which makes cable
suspension bridge vulnerable to wind load. Seismic load is also a major concern for a structural engineer when
designing the Bridge. Due to its long span and hanging deck and sky touching towers, Cable Suspension Bridge
is likely to take huge impact in case of seismic excitation [18-21].
2. Structural Behaviour of Different Components of Cable Suspension Bridge under

Aerodynamic and seismic loading
Suspension bridges are the lightest bridge system among the long span bridges around the world. Bending
moment produced due to support condition over the bridge span is less than the other bridge systems as bridge
span acts as a simply supported beam. Structural behaviour including wind and seismic effects on different
components (Cable and hanger, Deck and Tower (Pylon)) of the bridge are studied in detail in this paper.
2.1 Cable and Hanger
Cables are the basic element in all cable suspension bridge. The cables help the deck to transmits the traffic load
to the pylon and anchorages by tension forces. The basic element of the cable in suspension bridge is steel wire
[22]. Generally, the wires are in cylindrical shape of certain diameter and of high strength, high quality steel.
The steel wire has tensile strength in the range of 1550-1800 N/mm2 [23,24]. These steel rods have high carbon
content. After the final drawing the wire is stimulated for corrosion protection. To reduce the sag effect of
cables and to make the structure lighter, Fibre Reinforced Polymer (FRP) has been developed as the substitution
of steel cable. Strength to weight ratio, resistance against corrosion and fatigue strength are higher in FRP
cables. Wind and seismic effects on cable and hangers are discussed below [25].
Cables and hangers show the most critical behaviour in presence of high wind speed [26]. Basically, fatigue
damage is a large problem for the hangers or suspenders. Sometimes due to random wind load, the hangers get
dislocated from its actual position and also possess a higher amount of stresses, which is more than the
allowable the stress or fatigue limit [27]. Hangers get broken one by one due to the rubbing among steel strands
themselves or among strands and the anchorage block, thus lead to the fatigue damage. Main reasons behind the
fatigue damage are- excessive frequency of live load, vortex induced oscillation, buffeting due to random wind
motion (Which is also known as gust response) etc. This damage result in a considerable repair cost and also
make an impact on bridge performance [28]. Hangers are also subjected to bending deformations due to the
relative displacement of the girder and cable. Results from the FE analyses show that the lives of short
12
suspenders without central clamps are insufficient (being only approximately 19–47 years) compared with the
design life of suspenders (i.e., 30 years). To reduce this fatigue damage, we generally increase the damping
capacity of the hangers. Also, several damping devices are used to control the large amplitude of vibration of
hangers due to wind load. Also due to galloping effect, large amount of vibration is observed between the
hangers. Oscillations induced by wind flow can be either in the direction of the main flow or perpendicular to
the main wind direction because of shed vortices on the cable surface is known as galloping [29]. In galloping
effect, both windward and leeward steel ropes have to experience the vibration.
Also due to these wind-induced effects, corrosion phenomenon is occurred in main cables along with the
hangers. With the increase in age of the bridge, corrosion of cables is initiated. Deterioration of strength has also
been occurring in almost all major suspension bridges due to the corrosion in steel cables. At the same time a
serious issue of the safety of the entire bridge arrives. The shared action of compaction and friction in the wire
strands, guarantees that even if any individual wire-strand gets weaken by the action of corrosion and somehow
broken within the cable, then these cables can still transfer some of the tensile load by distributing it to the
remaining unbroken wires. This property of suspension bridge is known as internal redundancy. Domino effect
is triggered when the broken wires reaches a limit, after which, leading to a complete collapse. Corrosion
mechanisms are generally classified into the following types- even corrosion, pitting corrosion, stress corrosion,
hydrogen cracking, and fatigue corrosion [30-32].
Several corrosion protection systems are adopted to enhance the life span of the bridge. Among them three
systems are widely popular. They are the wire painting and wrapping, covers, and S-wire wrapping and cable
Dehumidifiers systems. Painted-wire wrapping system is widely accepted from all over the world. Also, to keep
the cables safe from the effect of fatigue, cable oscillation amplitudes must be reduced by increasing the
damping ratio of the cables. Damping ratio will be increased by means of installing several damping devices
[33].
Ground distortion can destroy a structure. Since an earthquake excitation comprises of the superposition of
countless with various attributes, seismic ground movements at the backings can change fundamentally along a
suspension connect range. The varieties in the help movements can fundamentally impact the inside forces
created in the structure. Different types of earthquake effects on cable and hanger are- Hanger failure due to
uneven load transfer during seismic excitation, Cable failure due to inadequate ductility, damage Cable Saddle
[34,35]. The most important piece of suspension bridge segments as far as struggle against seismic excitation is
the suspension framework which comprises of saddles, suspenders, and main cable. This intriguing ability to
endure seismic tremor shock might be because of substantial wellbeing factor arrangement while cable was
designed. The major disadvantage of suspension framework amid seismic tremor is the slip of saddles which sits
on the pylon. This seismic powerlessness comes up because of tremendous diversion and extraordinary link
point between main span and side spans. [36-38]
2.2 Deck
Primary function of the deck is to convey the live load (i.e. vehicle and movement stack). It is suspended from
the main cables utilizing vertical holders. The self-weight of the deck should be less because the deck is carried
by the cables. Stiffening girder is also provided under the deck to transfer the load from the deck to the hangers
and also to reduce the impact of wind load on the bridge acting. Buckling-controlled propped outlines are an
extraordinary kind of concentrically supported casing with supports extraordinarily and intended to withstand
yield level compressive powers without buckling. The deck and stiffening girder usually resist axial force and
bending moment induced by the stay cables. For central suspension bridge, high torsional stiffness i.e. high
stiffness in bending is essential [39].
The deck along with stiffening girder is always under tension due to live load and dead load acting on the deck.
Deck is highly susceptible to aerodynamic load acting due to heavy wind flow in the region. To cancel out the
aerodynamic effects the stiffening girder plays a major role in stabilizing the stability the deck. Aerodynamic
force can have a disasters impact on the deck which was seen in Tacoma Narrows Bridge, leads to the collapse
of the whole bridge. Different types of decks are- Decks made up of concrete, Decks made up of steel, and deck
made with different elements combined together known as the composite deck.
13
Basically, aerodynamic actions mostly depending upon the different wind directions. Wind flows in the
direction normal to the bridge axis and parallel to bridge axis is named as crosswind. As along wind direction
happens most of the time, it is considered as the standard wind flow direction [28, 29]. The most critical wind
action for deck is buffeting, which is produced due to random wind turbulence. This significantly affects the
torsional response of the deck. Three different features significantly affect the Vortex Induced Vibrations of a
bridge deck. First, the deck size, due to the increment in deck size, the means of the bridge become more and
more compressed, and the equivalent damping turns to be less. Second, air flow along the deck can be non-
uniform. Finally, the most important, is the flow field around the stream-like box girder [40]. Torsional
responses are also observed in the bridge deck due to buffeting and flutter resulting in a total collapse of bridge
deck. Flutter is an event of self-energized vibration, which may make a scaffold shake persistently with
cumulative amplitude until the point when the whole structure fails. Streamlined unsteadiness is likely the
breeze impact that has gotten the most consideration for Cable Suspension Bridges, somewhat as a result of the
disastrous outcomes in the event that it is permitted to happen and halfway due to the fantastic demolition of the
deck structure of the Tacoma Narrows Bridge [41]. Single-mode responses in the vertical and torsional motions
of the deck are less important than Multimode responses. The lateral vibration of bridge deck and tower is
mainly influenced by buffeting of towers and main cables. To mitigated or avoid the aerodynamic instability or
phenomenon certain counter-measures are to be adopted. Installation of Tuned Mass Dampers (TMD) is among
many such countermeasure, which results in the reduction of vertical motion of deck, disadvantage here is the
requirement of large space with the deck, which may or may not be always available [42].
Structural integrity of a deck is mainly affected by the seismic excitation in the built-up zone of bridge. This
factor effects the force of deck vibration and deck bends, also their effects varies along the length of the bridge.
Site situations have significant impact on amplifying deck movement, and subsequent increased weakness of
bridges to fall site due seismic excitation. Importance of this observation increases because mainly bridges and
elevated roadways cross bodies of water where soft soil deposits, high wind speeds, huge traffic and corrosion
are always show their presence. In the event of seismic excitation, the deck is highly vulnerable to fall. The
cause of fall is failure of cable, deck segment detaching from the main line etc. Phenomena such as seismic
excitation and high horizontal and vertical ground accelerations as well as large velocity pulses are result of the
sites directly above the fault rupture. Vertical displacement of the bridge deck is more impacted by the
vulnerabilities considered, contrasted and transverse relocations [15,16]. liquefaction and lateral spreading on
site/ground is the main cause for permanent substructure deformations and loss of superstructure. The
vulnerabilities have less effect on the longitudinal displacements of the bridge deck with respect to vertical and
transverse directions. Most essential questionable parameters are the area of the epicentre, the seismic power
and the constriction law, the speed of seismic waves through the ground, the frequency of the seismic waves, the
local impacts of the site, and so on [43,44]. Ambient vibration observing is regularly led to acquiring the
frequencies and mode shapes.
2.3 Pylon/Tower
Tower or pylon is a slender member that transfers the live and dead load to the foundation. In suspension
bridges, practically there is no direct attachment between the tower and the deck. The most conclusive load on
the pylon will often be the axial compressive force begin from the support of the cable system [8]. Basically,
there are two major materials used in construction of such huge towers, i.e. Concrete and Steel. For Example,
world recognized bridges like the Golden Gate Bridge in the United States and the Akashi Kaikyo Bridge in
Japan comprise of huge Steel towers. Examples for concrete pylon bridge includes the Humber and Great Belt
East Bridges in Europe and the Tsing Ma Bridge in China [9].
Due to longer spans and taller towers these bridges become more flexible and tend to vibrate more. Bridge
towers are widely constructed of steel and they are lighter in weight than the concrete towers. That is why steel
towers are susceptible to vibrate at a large amount. Bridge towers show various instability actions like VIV,
buffeting, flutter and galloping oscillations due to the influence of high-speed wind [45]. During construction
period towers remain in free-standing condition and prone to show resonance oscillation along the pylon weak
axis when pylon local natural frequency accords with the vortex shedding frequency. Depending upon the wind
directions, two types of motions are observed. 1) Motion off the level, in which the Pylon sway along its weak
axis, similar to the bridge axis due to crosswind, and 2) Motion in-plane, where the pylon sway along its strong
axis, perpendicular to the bridge axis due to a long- wind [46].
14
Wind effects are more complicated for the towers/pylons as the height of the towers are too much. Due to higher
wind speed, they are subjected to various wind forces and start to show critical aerodynamic behaviours. Bridge
towers are largely affected by high wind turbulence in the along bridge direction. The response in the along
wind direction is larger than the cross-wind direction. The responses are largest at the top of the bridge tower.
Also vortex induced vibrations create a large influence on the tower during high wind motion. Unsteady flow
over bluff bodies may generate a regular pattern of vortices which is called a vortex street. The vortices interact
with the bluff bodies and provide the source of the effect called VIV [47]. Even if VIV of pylon may generate
stability problem, extreme shaking of the pylon, especially the steel one, surrounding low damping may
influence bridge structural properties and eventually result in fatigue damage [48]. Galloping effect occurs in
bridge tower at a comparatively high wind speed and entire bridge is subjected to a large vibration as well as
oscillation. Resonant vibration of bridge tower in torsional mode is also produced due to the wind turbulence. It
depends upon the torsional stiffness of the bridge tower [45]. In case of low torsional stiffness, large static
displacements occur in bridge tower. Also, critical flutter wind speed creates hindrance during the construction
stage of bridge tower. To avoid this, wind instability, counter solutions, such as cutting off the corners or
making slits and attaching the aerodynamic appendages, such as deflectors and side plates are commonly
implemented.
Pylon in the suspension bridges is the first and beginning part on which compressive loads acts and every now
and again these pylons are built of steel plates. The feeble purpose of the towers amid seismic tremor is the
pylon plate cells. The plate cell of towers isn't solid that is the reason it may buckle [49]. The buckling of the
pylon shaft, which basically bolsters all the load coming through the saddle, loads, will endanger the integrity of
the entire structure. The reason for the failure is because of substantial pylon displacement as well as in light of
the fact that P-delta impact is significantly vast also [50].
Amid seismic excitation, footing raise may arise and therefore, pylon shifts back and forth. On the off chance
that one side of the base pole disengages from the pylon, the load change from its unique position. In this way,
pressure stack on the pylon shaft will considerably increment and in the long run, the pylon shaft will buckle
[51]. A functional method to keep this sort of disappointment is to place, vertical stiffeners to the non-compact
plates of the pylon cells. The steel pylon channel is conveyed by solid platforms. Solid platforms are subjected
to nearby high flexure and very huge quivering. This high flexure and very huge trembling may begin tension
cracks. The most common method to tackle this problem pre-stressed wire strands are used [52,53]. Semi-active
control devices potentially offer the reliability of passive devices, yet keep up the flexibility and versatility of
completely dynamic frameworks. As indicated by and by acknowledged definitions, a semi-active control
gadget is one which can't enter vitality into the framework being controlled. Such gadgets ordinarily have to a
great degree low power prerequisite, or, in other words amid seismic occasions when the main power source to
the structure may fail. Magnetorheological (MR) dampers are new semi-dynamic control gadgets that utilization
MR liquids to give controllable dampers that are very encouraging for structural building applications
[26,42,54]. They offer exceptionally dependable task at a humble expense and can be seen as the fail-safe in that
they become passive dampers should the control hardware malfunction [55].
3. Comparison/Discussion:
The bridge length has very minute effect on the lateral displacement, but noteworthy effect on both the vertical
and torsional displacements. As the suspension of the bridge get bigger, the Cable Suspension bridge behaves as
a simply supported beam, with reduced structural stiffness [56]. Due to increase in the length vertical and
torsional displacements are therefore increased greatly. As shown in Fig:2. Contrarily, as the suspension
partition decreases, the lateral and torsional displacements are both increased. In this manner, any suspension
partition whether it is too short or too long is good for Cable Suspension Bridges [57].
15
Figure. 2. Effect of different wind speeds in lateral and vertical displacement

of deck.
The members of the truss system are: (1) members over the supports at the primary piers; and (2) members at
hanger hinges. The suspension bridge is a flexible structure, the relationship between the load and deformation
of the bridge is non-linear when a normal load is given [58].
Under the seismic waves, the performance of the structure is mainly the longitudinal and vertical. The internal
force and displacement time-history of the control-section in the bridge. From displacement time-history
response, we can see that the peak response is in the middle. As shown in figure 3. The horizontal and vertical
displacement of tower top is far less than the longitudinal displacement [59].
Figure. 3. Variation of moment of deck under seismic excitation. the relative longitudinal displacement
between the top and bottom of Pier
4. Conclusion
In this review study, the transfer of load through different component and component behaviour under various
circumstances like high wind flow and seismic excitation in the tectonic plates have been studied for the
components of a Cable Suspension Bridge. VIV of bridge cables proved to considerably modify due to changes
in flow turbulence and surface roughness of bridge cables.
In particular, increasing the turbulence intensity of the incoming flow causes a reduction in the amplitude of
VIV of ice-accreted bridge cables.
1) These long span suspension bridges at subcritical wind speed range is mainly Gust Induced Buffeting.
2) Most unfavourable force coefficient is the lift force, which occurs at a high wind attack.
16
3) Countless effect on the response of the bridge tower is also due to turbulence, and it may increase the
response significantly. It may also decrease the response of the bridge tower.
4) The galloping of the bridge tower does not occur even at a high wind speed and turbulent intensity
tends to enlarge the VIVs of the bridge tower.
5) The longitudinal element of ground shaking contributes essentially to the vertical reaction of the
bridge.
6) Uniform ground movement over the whole span is destructive for these long structures.
7) Earthquake-induced cable tension might be of significant impact; stresses in different areas may
likewise be unequivocally influenced.
5. Reference
[1] Tanvir Manzur And Alamgir Habib (2006) Optimum Proportions For The Design Of Suspension Bridge.
Journal Of Civil Engineering (IEB), Researchgate
[2] Y L Cheng (2007) A Study Of The Brooklyn Bridge. Proceedings Of Bridge Engineering 2 Conference,
University Of Bath, Bath, UK
[3] C. Neeladharan, G. Gayathri, J. Rasigha, V. Thulasi Priya (2017) Analysis And Design Of Suspension Cable
Bridge. IJIRSET, Vol. 6, Issue 4
[4] Julian Staden (2007) Critical Analysis Of The Manhattan Bridge. Proceedings Of Bridge Engineering 2
Conference, University Of Bath, Bath, UK
[5] Masahiko Yasuda, Makoto Kitagawa, Toshimi Moritni And Susumu Fukunaga (2000) Seismic Design And
Behavior During The Hyogo-Ken Nanbu Earthquake Of The Akashi Kaikyo Bridge. 12WCEE
[6] Franco Bontempi, Luciano Catallo, Luca Sgambi (2014) Structural Analysis And Design Of Long Span
Suspension Bridges With Regards To Nonlinearities, Uncertainties, Interactions And Sustainability.
Researchgate
[7] Peter Sluszka, Christopher Gagnon, Stuart Rankin (2006) Inspection, Evaluation And Monitoring Of
Suspension Bridge Cables. Structures 2006, ASCE
[8] List Of Longest Suspension Bridge Spans:
Https://En.Wikipedia.Org/Wiki/List_Of_Longest_Suspension_Bridge_Spans
[9] Longest Suspension Bridges In The World: Https://Www.Thebalancesmb.Com/Longest-Suspension-
Bridges-In-The-World-4061820
[10] Wai Tak Yim (2007) The Bridge Engineering 2 Conference Akashi Bridge. Proceedings Of Bridge
Engineering 2 Conference, University Of Bath, Bath, UK
[11] Salem S. Faza (2004) A New Generation Of Fiber Reinforced Plastic Rebars For Bridge Deck
Reinforcement. Structures 2000, Advanced Technology In Structural Engineering, ASCE
[12] Taehoon Hong And Makarand Hastak (2006) Construction, Inspection, And Maintenance Of FRP Deck
Panels. Journal Of Composites For Construction, ASCE
[13] Valbona Mara, Reza Haghani, Peter Harryson (2013) Bridge Decks Of Fibre Reinforced Polymer (FRP): A
Sustainable Solution. Construction And Building Materials, ELSEVIER
[14] Hartanto Wibowo, Danielle M. Smith, Ian G. Buckle, David H. Sanders (2012) Live Load Effects On
Response Of Bridges During Earthquakes. Structures Congress 2012, ASCE
[15] Ahmed M. Abdel-Ghaffar And Richard G. Stringfellow (1984) Response Of Suspension Bridges To
Travelling Earthquake Excitations: Part I. Vertical Response. Soil Dynamics And Earthquake Engineering,
1984, Vol. 3, CML Publications
[16] Shrikhande. M., And Gupta. V., (1999) Dynamic Soil Structure Interaction Effects On The Seismic
Response Of Suspension Bridges, Earthquake Engineering And Structural Dynamics, John Wiley & Sons, Ltd.
[17] D. B. Mccallen, A. Astaneh-Asi (1996) Seismic Response Of Steel Suspension Bridge. Seismic Design,
Evaluation, And Retrofit Of Steel Bridges, San Francisco, CA
[18] Theodore Hopwood, James H. Havensy (1984) Introduction To Cable Suspension Bridges. Kentucky
Transportation Center Research, University Of Kentucky
17
[19] Colin O'Connor And Tommy Hung Tin Chan (1988) Dynamic Wheel Loads From Bridge Strains. Journal
Of Structural Engineering, Vol. 114, ASCE
[20] S.S.Law, X.Q.Zhu (2004) Dynamic Behaviour Of Damaged Concrete Bridge Structures Under Moving
Vehicular Loads”, Engineering Structures, Volume 26, Issue 9, ELSEVIER
[21] Patrick Paultre, Omar Chaallal, And Jean Proulx (1992) Bridge Dynamics And Dynamic Amplification
Factors — A Review Of Analytical And Experimental Findings. Canadian Journal Of Civil Engineering, 1992,
Vol. 19
[22] Yuwei Shi, George Deodatis, Raimondo Betti (2007) Random Field-Based Approach For Strength
Evaluation Of Suspension Bridge Cables. Journal Of Structural Engineering, Vol. 133, ASCE
[23] W.Y. Liao1, Y.Q. Ni1 And G. Zheng (2012) Tension Force And Structural Parameter Identification Of
Bridge Cables. Advances In Structural Engineering Vol. 15
[24] Arsenii Trush, Stanislav Pospíšil, Sergey Kuznetsov And Hrvoje Kozmar (2017) Wind-Tunnel
Experiments On Vortex-Induced Vibration Of Rough Bridge Cables. Journal Of Bridge Engineering, ASCE
[25] Llu Ming-Hu, XU Guo-Ping, QIANG Shi-Zhong, REN Wei-Ping (2014) Study And Prototype Design Of
A Suspension Bridge With Ultra-Long Span And CFRP Main Cables. Journal Of Highway And Transportation
Research And Development, ASCE
[26] Lamine Dieng, Guillaume Helbert, Shabnam Arbab Chirani, Thibaut Lecompte, Philippe Pilvin (2013) Use
Of Shape Memory Alloys Damper Device To Mitigate Vibration Amplitudes Of Bridge Cables. Engineering
Structures, ELSEVIER
[27] Munenobu Kondoh, Motoi Okuda, Kouji Kawaguchi And Takefumi Yamazaki (2015) Design Method Of
A Hanger System For Long-Span Suspension Bridge. Journal Of Bridge Engineering, ASCE
[28] Zhongxiang Liu, Tong Guo, Lingyu Huang And Zhihong Pan (2017) Fatigue Life Evaluation On Short
Suspenders Of Long-Span Suspension Bridge With Central Clamps. Journal Of Bridge Engineering, ASCE
[29] Milos Novak, Hiroshi Tanaka (1974) Effect Of Turbulence On Galloping Instability. Researchgate
[30] Efe Karanci And Raimondo Betti (2017) Modeling Corrosion In Suspension Bridge Main Cables. I: Annual
Corrosion Rate. Journal Of Bridge Engineering, ASCE
[31] R. Betti, A. C. West, G. Vermaas And Y. Cao (2005) Corrosion And Embrittlement In High-Strength
Wires Of Suspension Bridge Cables. Journal Of Bridge Engineering, ASCE
[32] Ronald M. Mayrbaur (2000) Corrosion In Suspension Bridge Cables. 16th Congress Of IABSE, Lucerne
[33] Tony S. Vader, Cole C. Mcdaniel (2007) Influence Of Dampers On Seismic Response Of Cable-Supported
Bridge Towers. Journal Of Bridge Engineering, Vol. 12, ASCE
[34] Konishi I., Yamada Y; Studies On The Earthquake Resistance Design Of Suspension Bridge Tower And
Pier System.
[35] Ahmed M. Abdel-Ghaffar And Lawrence I. Rubin (1983) Lateral Earthquake Response Of Suspension
Bridges. Journal Of Structural Engineering, Vol. 109, ASCE
[36] Koel Takena, Michio Sasaki, Kouichi Hata And Kazuo Hasegawa (1992) Slip Behavior Of Cable Against
Saddle In Suspension Bridges. Journal Of Structural Engineering, Vol. 118, ASCE
[37] A. M. Abdel-Ghaffar And L. I. Rubin (1983) Vertical Seismic Behaviour Of Suspension Bridges.
Earthquake Engineering And Structural Dynamic's. Vol. 11, John Wiley & Sons, Ltd.
[38] Q. Zhu, Y.L. Xu, L.D. Zhu, H. Li D (2017) Vortex-Induced Vibration Analysis Of Long-Span Bridges
With Twin-Box Decks Under Non-Uniformly Distributed Turbulent Winds. Journal Of Wind Engineering &
Industrial Aerodynamics, ELSEVIER
[39] Daniel C.Vaz, Raquel A. B. Almeida, Eric Didier, António P.V. Urgueira, A.R. Janeiro Borges (2016)
Improving The Aerodynamic Performance Of Vila-Real Bridge Deck-Section. Journal Of Wind Engineering
And Industrial Aerodynamics, ELSEVIER
[40] Hui Li, Shujin Laima, Qiangqiang Zhang, Nali, Zhiqiang Liu (2013) Field Monitoring And Validation Of
Vortex-Induced Vibrations Of Along-Span Suspension Bridge. Journal Of Wind And Industrial Aerodynamics
Engineering, ELSEVIER
[41] Xinzhong Chen (2006) Analysis Of Long Span Bridge Response To Winds: Building Nexus Between
Flutter And Buffeting. Journal Of Structural Engineering, Vol. 132, ASCE
[42] M. Gu, C.C. Chang, W. Wu, H.F. Xiang (1998) Increase Of Critical Flutter Wind Speed Of Long-Span
Bridges Using Tuned Mass Dampers. Journal Of Wind Engineering And Industrial Aerodynamic, ELSEVIER
[43] M. Gu, C.C. Chang, W. Wu, H.F. Xiang (2014) Longitudinal Restraint Of A Double-Cable Suspension
Bridge. Journal Of Bridge Engineering, ASCE
18
[44] A. A. Dumanoglu, R. T. Severn (1989) Seismic Response Of Modern Suspension Bridges To

Asynchronous Longitudinal And Lateral Ground Motion. Structural Engineering Group, Proc. Instn Civ. Engrs,
Part 2
[45] Arsenii Trush, Stanislav Pospíšil, Sergey Kuznetsov And Hrvoje Kozmar (2017) Wind-Tunnel
Experiments On Vortex-Induced Vibration Of Rough Bridge Cables. Journal Of Bridge Engineering, ASCE
[46] Dionysius M. Siringoringon, Yozo Fujino (2012) Observed Along-Wind Vibration Of A Suspension
Bridge Tower. Journal Of Wind Engineering And Industrial Aerodynamics, ELSEVIER
[47] Bin-Hua Xia, Zeng-Shun Chen, Tian-Long Li, Yong-Da Yang And Lei Liu (2017) Aerodynamic And
Aeroelastic Performance Of An Irregular Bridge Tower: A Case Study. International Journal Of Robotics And
Automation, Vol. 32
[48] D. M. Siringoringo And Y. Fujino (2011) Observed Alongwind Vibration Of A Suspension Bridge Tower
And Girder. The Twelfth East Asia-Pacific Conference On Structural Engineering And Construction, Procedia
Engineering, ELSEVIER
[49] Chong Wu, Qiu Zhao (2014) A Simplified Method For In-Plane Buckling Analysis Of Pylons Of Multi-
Span Suspension Bridges. Fifth International Conference On Intelligent Systems Design And Engineering
Applications, IEEE
[50] Ji Lin (2012) Research On Key Technologies In Design Of Middle Pylon Of Taizhou Bridge. Engineering
Sciences, Vol. 10
[51] Chang-Ke Jiao, Xin Dong, Ai-Qun Li, Guang-Dong Zhou And Xiao-Ping Wu (2017) Seismic Response Of
Long-Span Triple-Tower Suspension Bridge Under Random Ground Motion. Mathematical Problems In
Engineering, Hindawi Publishing Corporation
[52] Ahmed M. Abdel-Ghaffar And Richard G. Stringfellow (1984) Response Of Suspension Bridges To
Travelling Earthquake Excitations: Part II-Lateral Response. Soil Dynamics And Earthquake Engineering, 1984,
Vol. 3, CML Publications
[53] Yan Xu, Shijie Zeng, Xinzhi Duan, Dongbing Ji (2018) Seismic Experimental Study On A Concrete Pylon
From A Typical Medium Span Cable-Stayed Bridge. Frontiers Of Structural And Civil Engineering, Volume 12
[54] Kawashima, K., Unjoh, S., (1994) “SEISMIC RESPONSE CONTROL OF BRIDGES BY
VARIABLE DAMPERS” Journal Of Structural Engineering, ASCE
[55] S.J. Dyke, B.F. Spencer Jr., M.K. Sain And J.D. Carlson (1996) Modeling And Control Of
Magnetorheological Dampers For Seismic Response Reduction. Smart Materials And Structures, Volume 5,
Iopscience
[56] Y. Fujino, M. Iwamoto, M. Ito And Y. Hikami (1992) Wind Tunnel Experiments Using 3D Models And
Response Prediction For A Long.Span Suspension Bridge. Journal Of Wind Engineering And Industrial
Aerodynamics, ELSEVIER
[57] Yongle Lia, Haojun Tanga, Qiaoman Lina, Xinzhong Chena (2017) Vortex-Induced Vibration Of
Suspenders In The Wake Of Bridge Tower By Numerical Simulation And Wind Tunnel Test. Journal Of Wind
Engineering And Industrial Aerodynamics, ELSEVIER
[58] J.D. Yaua, L. FrÝbab, (2007) Response Of Suspended Beams Due To Moving Loads And Vertical
Seismic
Ground Excitations, Engineering Structures, ELSEVIER
[59] I. E. Harik, D. L. Allen, R. L. Street, M. GUO R. C. Graves, J. Harison And M. J. Gawry (1997) Seismic
Evaluation Of Brent-Spence Bridge. Journal of Structural Engineering, ASCE
19
Paper ID- 14
REPLACEMENT OF SAND WITH STONE DUST AS A FINE

AGGREGATE IN CONCRETE
Bikram Paul1, Kushal Ghosh2 and Dr. Partha ghosh3
1
PG Student, 2Ph.D Student and 3Associate Professor
Department of Construction Engineering, Jadavpur University, Kolkata-98
Abstract- Concrete industry is a rapidly growing industry in India consuming lots of natural resources
during the production of concrete. Nowadays, sand has become a rare commodity due to several restrictions
imposed regarding dredging of sand from river beds by the Govt. authorities. As sand has been collected
from the river bed, it will affect river by lowering ground water table and reducing the quality of water to
the downstream users. Now stone dust shows a new ray of hope in the construction industry as an
alternative fine aggregate which can be used in concrete. Here study has been done M35 grade of concrete
and partial replacement of sand has been done by stone dust up to 20% by volume. Different mechanical
properties of concrete like compressive strength, Split tensile, flexural strength etc. and non-destructive
testing like UPV have been studied during this research work.
Keywords: - Sustainable Concrete, Stone Dust, Compressive Strength, Flexural Strength and UPV Test.
1. Introduction:
A civilization means a society with large permanent settlements featuring urban development,
social stratification and written or other formal means of communication. Its’ flowing water
provide us water for drinking, inhabitants and foods of which all are indirect the products of
water. Not only these things but also it is a habitat of many important micro-organisms for the
fertility of soils which are living in the layer of sand. The first great civilizations, such as those
in Mesopotamia, Egypt and even our Indus valley civilization all were started from the big and
important river banks. Today the condition of river is getting worse due to the illegal dredging
of sand from the river bed. Several reports have been published on illegal dredging of sand and
its bad effect. Sand mafias are often involved with this illegal quarrying of sand from the river
bed which leads to the increase of cost of sand. Many deadly diseases like silicosis are often
noticed who are involved in sand mining. [10] Two bridges are also affected by the Illegal sand
dredging in the area across Sutlej, Beas, and Ghaggar River in Punjab. Due to the increasing
development the need of production of concrete has been increased and simultaneously natural
resources. Increased needs of sand have indicated the alarming situation of rivers. To
rejuvenate the condition of rivers, we need to stop mining otherwise the day is not so far when
we does not get water for drinking, agricultural work etc. and balance of ecology will
completely destroy. Time has come to think for an alternate source of fine aggregate in
concrete.
Stone dust have also many bad impacts on our environment but when we used the stone dust
in the production of concrete, it acts as a blessing in our society because it gives us an
opportunity to reduce the use of sand as a fine aggregate in concrete. Mechanical properties of
concrete gives us positive results in terms of compressive strength, flexural strength, split
20
tensile strength and non-destructive test like UPV test etc. Many researchers were studied in
this field those are given below. Celik and Marar [1] investigated the properties of hardened
concrete and mixing stone dust as a partial replacement of sand. They suggested 10%
replacement as an optimum replacement for compressive strength and flexural strength and
15% for water absorption. MD. Nuruzzaman et al. [2] replaced stone dust as a replacement of
sand from 0 to 50% and studied the compressive strength and tensile strength of that mix. They
suggested the optimum percentage of stone dust as 30% for compressive strength whereas
minimum tensile strength for the mix would be 50%. Gurpreet Singh et al. [9] carried out an
examination on partial replacement of sand by Waste foundry sand and put forward for a
consideration of WFS. They used waste foundry sand with concrete as a replacement of sand
by 0%,10%,15%,20% and 25% and testing the mechanical properties (compressive Strength,
Split Tensile) and durability (RCPT) were done at the ages 7 days,28 days and 91 days. They
noticed that there was an increment in compressive and split tensile strength as the age
progressed from 7 to 28, 28 to 91 days. 17% and 19.5% compressive strength increment were
noticed for 28 days and 91 days respectively than that of controlled concrete which was for
20% replacement of sand by waste foundry sand. In case of split tension maximum split value
observed at 15% WFS replacement both for 28 days and 91 days. Maximum values were
observed at 15% for both UPV and compressive strength test. Malkit Singh et al. [11] were
studied the effects of partial replacement of sand using bottom ash as a replacement. They
concluded that the mixing of bottom ash in concrete increased the probability of bleeding of
concrete and this bleeding water gets trapped by aggregate and formed pores in the concrete.
This lead to porous and low-density concrete and followed by lower compressive strength
though it can be ignored by using high dose of admixture with 50% replacement of sand. Zainab
Z. Ismail et al. [14] revealed the use of waste glass as partial replacement of sand in concrete
and also tested the fresh and hardened concrete properties. Waste glass was mixed with
concrete 5%, 10%, 15%, and 20% of sand by volume. Slump value decreased with the increase
in the ratio of waste glass due to poor shape and texture though the mix was described as
workable mix. Compressive Strength of 20% waste glass ratio mix showed the maximum
strength and regarded as optimum percentage. It was also observed that the increment ratio of
strength at 28 days was higher than the 14 days or 7 days due to the pozzolanic reaction. 3 days,
7 days, 14 days and 28 days flexural strength also increased significantly for waste glass mixed
concrete than the normal concrete where 20% waste glass mix had achieved the highest flexural
strength. Aditya Rana et al. [16] investigated the effects on partial replacement of sand using
stone slurry, manufactured sand and crushed sand. They suggested that 15% slurry and 85%
Manufactured sand can be used as full substitution of river sand.
21
1.1 Research Significance
Stone Dust is a by-product found from the stone quarry. Disposal of stone dust found from the
crusher is a big problem. It has also negative impact to the environment. Our aim is to use it in
our concrete industry and to study the properties of concrete with stone dust.
2. Experimental Details
2.1 Materials Used

Cement. Portland Pozzolana cement has been used conforming to IS 1489: part 1
Coarse Aggregate. Coarse aggregates were used of nominal max. Size 20 mm and requirements
met as per IS 383:1970. Sieve Analysis was done to meet the requirements of grading of Coarse
aggregate as per IS 383:1970. Fig.1 shows the grading curve of Coarse Aggregate. Different
test results are given in below table:
Grading Curve of CA
100
Table :2 Physical Properties of Coarse Aggregate
Percentage Passing
Type Crushed 75
Shape Angular 50
Specific Gravity 2.82
25
Fineness Modulus 5.83
0
Water Absorption 2.4%
4.75 10 20 40
Max. Nominal size 20 mm Particle Size, in mm
Coarse Aggregate 20 mm
Fig 1
Fine Aggregate. Locally available sands are used for this research work and sands are tested as
per IS -383:1970. Several physical properties of sand were tested such as moisture content,
water absorption, sieve analysis, specific gravity etc. Grading of sand has been given as per IS
383:1970 in Fig.2.
Grading Curve of Sand

100
Percentage Passing
80
60
40
20
0
1 2 3 4 5 6 7 8 910
Particle Size, in mm
Sand Zone III Min IS 383 Max IS 383
Fig 2
22
Stone dust. Stone Dust was collected from Nalhati, West Bengal. It is a by-product of rock
quarry. We were using it as partial replacement of sand and it was replaced by volume 10%,
15% and 20% Tests results are given below. Sieve Analysis was done to find the zone of the
Stone Dust as per IS 383. Fig.3 shows the grading curve.
Table 4: Design Mix Proportions
Cement 430 Kg/m3

Table :3 Physical Properties of Sand and Stone
Dust Sand 610 Kg/m3
Physical Properties Sand Stone Dust Coarse Aggregate, 20mm 756 Kg/m3
Specific Gravity 2.65 2.68 Coarse Aggregate, 10mm 504 Kg/m3
Fineness Modulus 1.93 1.82 Water 172 Liters
Water Absorption 1.0% 1.16% 0.8% wt of

Admixture
Cement
Grading Curve of Stone Dust

100
Fig.3
80
Percentage Passing
60
40
20
0
1 2 3 4 5 6 7 8 910
Particle Size, mm
new stone Dust mixing stone dust
Min IS 383 Max IS 383
Fig 3
Admixture. Sikament 2004 Ns was mixed with concrete to improve the workability of the mix.
Here 0.8% to 1% weight of cement dose of the admixture was used.
Concrete Mix Proportion. A M35 grade of concrete was designed as per IS 10262:2009 which has
mean target strength of 43.25 MPa. Sand of this mix has been replaced by volume of 10%, 15%
and 20% to make this sustainable and to study and compare its mechanical properties with each
other. Mix proportions have been given in Table 4.
Casting of Specimen. 150mm x 150mm x 150 mm Cube samples were casted for compressive test
and flexural Tests were conducted on 150mm x 150mm x 700mm Beam samples. Split Tensile
Tests also carried out on cylinder of 150 mm diameter samples.
3. Result and Discussion:
Compressive Strength. Concrete Mix was designed for M35 grade which had a target mean
strength of 43.25 MPa. Normal concrete, i.e., 0% replacement of stone dust showed 26.07 MPa
23
and 44.88 MPa compressive strength for 7days and 28 days respectively. When sand was
replaced up to 10%, compressive strength was reduced by 24.08% than that of normal concrete.
This can be attributed to the substantial decrease of workability leading to an extremely stiff
mix. The compressive strength increased for 15% substitution with respect to 10% replacement
both for 7 days and 28 days. Basically, the quantity of fine particles in the matrix was less for
10% and it increased for 15% and 20% replacement by stone dust which led to an increase in
packing density of the specimens leading to an increase in compressive strength. The relative
increase in density could not however offset the damage caused by improper compaction
resulting from a stiff mix.
Compressive Strength Test
Compressive Strength, MPa

48 44.88
Table 5: Compressive Test Result 40.00 40.59
36 34.07
26.07
% of Stone 7 Days, 28 Days, 25.37
24 20.59 25.63
Dust MPa MPa
0% 26.07 44.88
12
10% 20.59 34.07
15% 25.37 40.00 0
20% 25.63 40.59 0%
10%
15%
20%
Percentage Replacement
7 days Comp. Strength 28 days Comp. Strength
Fig 4
Flexural Strength. Flexural Strength of concrete decreased for 10% and 15% replacement of
sand by stone dust than that of normal concrete. Beyond 15% replacement, flexural strength
increased for 20% replacement and increased up to 35% though only results up to 20% have
been included in this study. The above mentioned trend was noticed for both 7 days and 28
days. Increasing flexural strength with increasing replacement of sand by stone dust that may
be due to the rough texture of stone dust that leads to improve the aggregate interlock in the
concrete. Results of flexural strength are tabulated in Table 6 for 7 days and 28 days
respectively. Fig 5 show the graph of the same.
Flexural Strength
5.893
Flexural Strength, Mpa
6.0 5.693
5.267 5.733
Table 6: Flexural Strength 4.467
4.093 4.24 4.467
% of Stone 7 days, 28 days, 4.0
dust MPa MPa
0% 4.467 5.893 2.0
10% 4.093 5.693
0.0
15% 4.240 5.267
0%
5.733 10%
20% 4.467 15%
20%
Fig 5
24
Split Tensile Strength. Reduction in split tensile strength was noticed for the mix with 10%
substitution of sand by stone dust and the trend was almost similar to that of compressive and
flexural strength results. The split tensile strength increased up to 20% replacement for both 7
days and 28 days results respectively. The value of split tensile results are within 5% to 8% of
compressive strength result values.
Split Tensile Strength
Split Tensile Strength, MPa

Table 7: Split Tensile Strength 4.0
% of Stone 7 days, 28 days, 3.0

dust MPa MPa
2.0
0% 2.286 3.158
10% 1.579 2.428 1.0
15% 1.650 2.569 0.0
1.876 2.696 0%
20% 10%
15%
20%
7 days Split Tensile 28 days Split Tensile
Fig 6
UPV Tests. UPV results are within 4.5 to 5 KM/sec which means that the quality of concrete is
very good. A relationship among compressive strength, Upv value and percentage replacement
of sand was shown in Fig. 10. Here UPV value has increased for the specimens with
replacement of sand up to 15%, beyond that the UPV values started decreasing. Results are
given below in the Table 7.
UPV Test
TABLE 8: UPV Result 4.9 4.83
% of Stone Dust 28 days, KM/S 4.8
UPV, Km/S
4.65
0% 4.54 4.7 4.6
4.54
4.6
10% 4.65
4.5
15% 4.83
4.4
20% 4.60 4.3
O% 10% 15% 20%
Fig 7
4. Conclusions
 Due to higher water absorption values of stone dust the mix becomes stiffer due to the
addition of stone dust leading to a decrease in compressive, flexural and split tensile
strength. But addition of stone dust beyond a point also leads to an increase in packing
density which leads to a relative increase in strength.
25
 Stone dust can be used as an alternative for sand provided that the concrete mix is
designed for a higher target compressive strength than what is required. The lowering
of workability due to the addition of stone dust is a major problem which leads to a
lesser strength. This can be solved by carrying out more studies on the addition of
admixture to concrete using stone dust as a fine aggregate.
5. References
1. Celik and Marar (1996).-EFFECTS OF CRUSHED STONE DUST ON SOME PROPERTIES

OF CONCRETE. Cement and Concrete Research, Vol. 26, No. 7, pp. 1121-1130, 1996.
2. MD. NURUZZAMAN et al. STRENGTH ASPECT OF CONCRETE USING STONE DUST AS A
PARTIAL REPLACEMENT OF SAND, Vol. 04, No. 04, October 2015, ISSN 2319-5347,
International Journal of Advanced Structures and Geotechnical Engineering ,237-242.
3. Sanjay Mundra (2016). - CRUSHED ROCK SAND – AN ECONOMICAL AND ECOLOGICAL
ALTERNATIVE TO NATURAL SAND TO OPTIMIZE CONCRETE MIX. 2213-0209/© 2016 Published
by Elsevier GmbH, (http://creativecommons.org/licenses/by-nc-nd/4.0/), 345-347.
4. R. Ilangovana. -STRENGTH AND DURABILITY PROPERTIES OF CONCRETE

CONTAINING QUARRY ROCK DUST AS FINE AGGREGATE, VOL. 3, NO. 5,
OCTOBER 2008 ISSN 1819-6608 ARPN Journal of Engineering and Applied Science, 20-26.
5. Lohani T.K et al. -OPTIMUM UTILIZATION OF QUARRY DUST AS PARTIAL REPLACEMENT OF
SAND IN CONCRETE, Vol. 1, No. 2, 2012 ,Int. Journal of Applied Sciences and Engineering
Research, 392-404
6. Md. Safiuddin et al. - Utilization of Quarry Waste Fine Aggregate in Concrete Mixtures, Journal
of Applied Sciences Research, 3(3): 202-208, 2007 © 2007, INS Inet Publication.
7. Priyanka A. Jadhav, Dilip K. Kulkarni-EFFECT OF REPLACEMENT OF NATURAL SAND BY
MANUFACTURED SAND ON THE PROPERTIES OF CEMENT MORTAR, International Journal of Civil
and Structural Engineering, Volume 3 Issue 3 2013, 622-628.

8. T. Shanmugapriya et al. - OPTIMIZATION OF PARTIAL REPLACEMENT OF M-SAND BY
NATURAL SAND IN HIGH PERFORMANCE CONCRETE WITH SILICA FUME,
International Journal of Engineering Sciences & Emerging Technologies, June 2012.ISSN:
2231 – 6604 Volume 2, Issue 2, pp: 73-80 ©IJESET.
9. Gurpreet Singh et al. - Effect of waste foundry sand (WFS) as partial replacement of sand on
the strength, ultrasonic pulse velocity and permeability of concrete, Construction and Building
Materials 26 (2012) 416–422.
10. https://www.newsclick.in/indias-rivers-ruined-condition-due-unchecked-illegal-sand-mining
11. Malkit Singh et al. - Effect of coal bottom ash as partial replacement of sand on properties of
concrete, Resources, Conservation and Recycling 72 (2013) 20– 32.
12. S.N. Raman et al. - High-strength rice husk ash concrete incorporating quarry dust as a partial
substitute for sand, Construction and Building Materials 25 (2011) 3123–3130.
26
13. Sreekrishnaperumal Thanga Ramesh et al. - Use of furnace slag and welding slag as
replacement for sand in concrete, International Journal of Energy and Environmental
Engineering 2013, 2-6
14. M. Bederina et al. - Effect of partial and total replacement of siliceous river sand with limestone
crushed sand on the durability of mortars exposed to chemical solutions, Construction and
Building Materials 47 (2013) 146–158.
15. Omar M. Omar et al. - Influence of limestone waste as partial replacement material for sand
and marble powder in concrete properties, HBRC Journal (2012) 8, 193–203.
16. Li Beixing et al.- Effect of limestone fines content in manufactured sand on durability of low-
and high-strength concretes, Construction and Building Materials 23 (2009) 2846–2850.
17. Özgür Eren et al. - Effects of limestone crusher dust and steel fibers on concrete, Construction
and Building Materials 23 (2009) 981–988.
18. M. Shahul Hameed et al. - PROPERTIES OF GREEN CONCRETE CONTAINING QUARRY

ROCK DUST AND MARBLE SLUDGE POWDER AS FINE AGGREGATE, ARPN Journal
of Engineering and Applied Sciences, ISSN 1819-6608, VOL. 4, NO. 4, JUNE 2009, 83-89.
19. Abd Elmoaty Mohamed Abd Elmoaty - Mechanical properties and corrosion resistance of
concrete modified with granite dust, Construction and Building Materials 47 (2013) 743–752.
20. Zeki Karaca et al. - Classification of dimension stone wastes, Environ Sci Pollut Res (2012)
19:2354–2362.
21. Jinwoo An et al. - Effect of Aggregate Mineralogy and Concrete Microstructure on Thermal
Expansion and Strength Properties of Concrete, Appl. Sci. 2017, 7, 1307
22. A.S. Adithya Saran et al.- SEM Analysis on Sustainable High-Performance Concrete,
International Journal of Innovative Research in Science, Engineering and Technology, Vol. 6,
Issue 6, June 2017, 10237-10246.
23. MS Shetty – Concrete Technology (Theory and Practice), ISBN: 81-219-0003-4, Reprint 2013
24. P. Kumar Mehta and Paulo J. M. Monteiro – Concrete (Microstructure, Properties and
Materials), Third Edition, 0-07-146289-9.
25. A. M. Neville - Properties of concrete, ISBN: 978-0-273-75580-7, 2011.
26. IS 10262:2009 – Indian Standard, Concrete Mix Proportioning – Guidelines (Fifth Revision).
27. SP 23:1982 – Handbook on Concrete Mixes (Based on Indian Standard), ISBN: 81-7061-012-
5.
28. IS 456:2000 - Indian Standard, Plain and Reinforced Concrete - Code of Practice (Fourth
Revision).
27
Paper ID- 15
A Brief Review of the Effect of Mineral Admixtures on the Properties of

Alkali Activated Composites.
Saswata Chattopadhyay1*, Subhajit Mondal1, Kushal Ghosh2 and Partha Ghosh3

1
2nd Year Undergraduate Student, Department of Construction Engineering, Jadavpur University, Kolkata, India
saswatachatterjee12@gmail.com
1 nd
2 Year Undergraduate Student, Department of Construction Engineering, Jadavpur University, Kolkata, India
subhajitm146@gmail.com
2
Ph.D Student and Junior Research Fellow, Department of Construction Engineering, Jadavpur University, Kolkata,
India
3
Assistant Professor, Department of Construction Engineering, Jadavpur University, Kolkata, India
Abstract. Several leading researchers are putting their effort in finding the most eco-friendly materials which can be used as
sustainable building and construction materials. Ordinary Portland Cement is a vital necessity in the production of concrete
which is the second most consumed material on earth. The production of cement releases huge amounts of greenhouse gases
which plays a vital role in climate change. Therefore to lower the carbon footprint, an alternative to this OPC is Geopolymer
which is an alkali aluminosilicate binder which when activated by alkali silicate solution has the same binding properties to
that of OPC. This paper investigates the effect of addition of mineral admixtures to alkali activated composites and their
subsequent modifications of the properties of the same, carried on by different researchers. Different researchers have aimed
to modify and change the properties of the alkali activated composites with mineral admixtures in recent years, and the data
are summarised to yield a comparative study. Mineral admixtures such as silica fume (SF), Ground granulated blast furnace
slag (GGBS), Metakaolin (MK), Oil Palm Fiber content etc. were studied for the impacts they have on the fresh and the
hardening properties of alkali activated composites. This paper focuses on finding the specific influences of the mineral
admixtures on the fresh and hardened properties of alkali activated composites and attempts to chart out a path to correlate
them in order to help future researchers find an optimal Geopolymeric mix design which balances strength and durability
parameters with sustainability and economic viability.
.
Keywords: Geopolymer, Workability, Flowability.
1. Introduction
With the growth of population and economy the growth of infrastructure is a necessity. The rise of infrastructure
demands has seen a spike of concrete use over the last decade. Concrete has become one of the largest
production materials with an annual global production of about one meter cube for each individual.[8] Concrete
is a mix of the cement paste and aggregates as its main components. The binder, in most of the cases, is Portland
cement, which has a very energy intensive manufacturing process. The raw materials in this case are limestone
and clay burned at very high temperatures. The conventional cement industry uses conventional sources of
energy to obtain this high temperatures. The manufacture of Cement is increasing around 4% annually and with
the production of 1 tonne of cement, nearly 0.93 tonnes of CO2 are produced into the environment. According to
the International Energy Authority, the cement industry is solely responsible for approximately 6% of all the
CO2 emissions globally. CO2 emissions contribute nearly 65% of global warming among all greenhouse gases.
The Cement industry is also responsible for producing SO3 and NOx, other than CO2 , which causes greenhouse
effect and acid rain. Therefore a suitable alternative to this is Geopolymer, an alkali aluminosilicate binder
which when activated by alkali silicate solution has the same binding properties to that of OPC. This new binder
materials which can replace OPC, by alkali activation, can reduce the carbon dioxide emissions to a much lesser
extent when compared to OPC. In addition to its CO2 emission reducing property, the fresh and hardened
28
properties of alkali activated composites can be modified using mineral admixtures such as silica fume (SF),
Ground granulated blast furnace slag (GGBS), Metakaolin (MK). In recent times, Researchers have been
continuously trying to chart out the correlation between the fresh as well as hardened properties of the alkali
activated composites by using different mineral admixtures. In this context, a review of the effect of the mineral
admixtures on the properties of Alkali Activated Composites is presented.
2. METAKAOLIN
2.1 Workability
Zhang et al[4] prepared Metakaolin (MK) blended with FA to produce FA/MK based Alkali Activated
Composites and reported that the fluidity of the mortar was found to increase with the increased percentage of
MK. The increased percentage of MK resulted in prolonging the respective setting times.
Fig 1a: Impact of increasing MK substituted GBFS Fig 1b: Impact of increasing MK substituted GBFS
on flow of GPMs on setting time of GPMs
With increasing the activator solution:binder ratio while keeping all the factors constant the flowability
increased. This was natural since the water content increased which increased the flowability. The mortar
samples activated with high solution took more time to set and the workability also increased. Figure 2
illustrates flow and setting time with increasing S:B ratio.
Figure 2: Effect of activator solution on workability
Cheng et al.[2] also reported that the effect of more metakaolinite in the system resulted in slower setting time.
Zhang et al.[1] also mentions the effects of alkaline solution on the workability of the mortars. Since the alkaline
solution was a mixture of NaOH and SiO2 , he found that increasing the NaOH molarity diminished the flow. As
the SiO2 to Na2O ratio was increased the degree of dissolution and hydrolysis were accelerated, thereby
inhibited the polycondensation. The effect of Na+ negatively affected the setting time and the workability but
29
an increase in heat of hydration. Hence MK doped with sodium could be used as heat evolutionary in AAS
systems.
Fig 3a: Effect of solution modulus in workability Figure 3b: Effect of Solution Modulus on setting time
keeping other factors constant keeping other factors constant
2.2 Strength
Zhang et al.[4] had reported the effect of compressive strength by varying FA/MK based Alkali Activated
Composites. It was reported that the Alkali Activated Composites containing 66.7% MK, after steam curing for
6 days, the compressive strength increased by 35.5%. Fig 4a. illustrates the above.
Fig 4a: Effect of MK substituted samples on compressive Fig 4b: Effect of MK substituted Slag on strength with
strength with time keeping other factors constant varying activator concentration
Yunsheng et al. charted out some properties of FA/MK mortars. FA/MK were taken in the ratios of 70/30,
50/50, 30/70, 10/90 and 0/100, by weight. NaOH and sodium silicate solution having molar ratio of 3.2 was
used as alkaline activator solution. It was ultimately noted that the Alkali Activated Composites containing 30%
FA and 70% MK which was steam cured for 8 hours at 80 C, resulted in higher compressive strength. The
compressive and flexural strengths noted were 32.2 and 7.15 MPa, respectively. Li et al. prepared FA/MK-based
Alkali Activated Composites with the ratios of FA/MK being 50/50, 30/70, and 10/90, by weight. NaOH and
sodium silicate solution was used as alkaline activator solution and polyvinyl alcohol (PVA) fibers (2%) was
used as reinforcements. It was seen that the addition of high content of MK in the Alkali Activated Composites
showed good fiber–matrix bond. It resulted in a higher flexural strength. Similarly, the addition of lower MK
content in the Alkali Activated Composites, led to lower flexural strength.
3. GGBS (GROUND GRANULATED BLAST FURNACE SLAG)

3.1 Workability
30
Pradip Nath et al.[6] reported the effect of varying proportions of GGBS included in fly ash based mortars and
concrete. It was seen that the workability reduced with the the slump and flow values with the inclusion of slag.
All the mixtures were mixed with the same amount of alkali activator, the one with no slag content showed the
highest slump and flow value. Fig 6. Discusses the flow value of GGBS replaced with fly ash mortars. Similarly
it was seen that the binder fly ash took more than 24 hours to set while the same binder replaced with slag
showed considerable amount of decrease in the setting time. Addition of slag increased the reactivity of the
source material leading to an increase in the rate of geopolymerisation. Addition of a calcium based source
material can lead to the formation of a cross-linked hybrid sodium-calcium-aluminate gel system which leads a
decrease in the setting time results in an increase of the stiffness of the geopolymer matrix [12] Sanjay Kumar et
al[14] reported the setting time of partially replaced fly ash with GBFS at 270C and reported that with increasing
GBFS content, the setting time decreased. Neat fly ash samples took more than 295 minutes to set whereas even
an addition of 5% slag reduced the setting time to 105 minutes. The optimum value was 25% slag content which
gave minimum value of 45 minutes. It reports that the distinction between the initial and final setting time was
negligible.
Figure 6a: Effect of GGBS substituted Fly Ash Figure 6b: Effect of various factors combined
in workability[6] with activator solution fixed[6]
The effect of Sodium Silicate to Sodium hydroxide ratio, also known as the solution modulus was investigated
in the Alkali Activated Composites, by keeping the slag and the alkaline activator content fixed at 10% and
40%, and changing the amount of sodium silicate or the sodium hydroxide content in the alkaline activator. The
workability of the Alkali Activated Composites decreased with the increase of the sodium silicate content.
Increasing the Sodium Silicate:Sodium hydroxide ratio, keeping the other variables constant, decreased setting
time to an extent. When the ratio of SiO2:Na2O was increased it amount of silica content increased, which
enhanced the geopolymerisation process.
3.2 Strength
Pradip Nath[6] et al reports that the compressive strength of Alkali Activated Composites having fly ash only,
showed the least strength. Keeping the alkaline activator content fixed (40%) and the solution modulus fixed at
2.5, it was seen that the concrete gained 10 N/mm2 at 28 days for every 10% increment of the slag content. At
the end of 56 days however each mixture shows similar rate of development of compressive strength. Fig7
illustrates the following.
31
Figure 7: Effect of Fly Ash Substituted GGBS in

compressive strength of mortar[6]
Fig 8 also mentions the effect of alkaline activator and the solution modulus on the compressive strength of
concrete and mortar, keeping the amount of GGBS:Fly ash ratio fixed at 10%. The increase of alkaline activator
from 35% to 45% reduced the strength of the concrete samples. For concrete samples the strength decreased by
24% for 40% of alkaline activator compared to that of 35% alkaline activator. Mortar samples showed less
prominent decrease, although the 45% activator solution samples showed 24% reduction in compressive
strength with respect to 35% activator solution samples.
Figure 8: Effect of activator concentration on the

compressive strength of concrete with 10% GGBS[6]
According to Partha Sarathi Deb et al.[7], the effect of solution modulus on the 28 days compressive strength
which was less prominent than the action of GGBS or the activator solution amount. In case of mortar samples,
at the end of 28 days, solution modulus of 1.5 showed higher strength than those with solution modulus of 2.0 or
2.5 and so on. The strength development was prominent up to 56 days with decrease of solution modulus, as
shown in Fig 9.
Figure 9: Effect of solution modulus on varying

proportions of Slag:Fly ash ratio[7]
Sanjay Kumar et al.[14] casted samples of fly ash replaced with GBFS, which were cured at 27 0C. It was seen
that with increasing GBFS content, the compressive strength increased. The rate of development of compressive
strength was more prominent for 15% GBFS addition. The increase in strength is due to the formation of C-S-H
gel.[15] K.Parthiban[16] reported the same results with replacing fly ash with GBFS, which were cured at ambient
temperature. When the solution modulus and mix ratio was held constant, the replacement of GGFS added to
the strength of concrete. 100% replacement of fly ash with solution modulus 1.5 showed the highest strength of
53.87 MPa while neat fly ash gave result of compressive strength of 21.35 MPa.
4. SILICA FUME
4.1 Workability
32
J. Durgaprasad et al.[9] reported that the workability of Geopolymer concrete decreased with the addition of
silica fume. Slump of normal OPC cement was significantly higher than the Geopolymer concrete of fly ash and
silica fume. The viscous nature of the Geopolymer paste with respect to the normal OPC paste was due to the
interaction of cohesion forces acting in the system. Figure 10 shows the slump values for different proportions
of SF added. Collins et al[13] reported that alkali activated concrete, activated by hydrated lime and sodium
metasilicate, which has been replaced by 10% condensed silica fume (CSF) showed less workability than pure
AAS concrete which is only activated with powdered sodium metasilicate and hydrated lime.
Figure 10: Effect of SF on slump value of AAS[9]
3.2 Strength
Debabrata Dutta et al. [10] prepared a Alkali Activated Composites mix containing fly ash with 0%, 2.5%, 5% SF
by weight of fly ash and cured for 850C for 48 hrs. After 28 days from casting the Alkali Activated Composites
mortar was tested for compressive strength. Mortar samples prepares with neat fly ash showed 26 MPa, while
the 2.5% SF replaced mortar showed 31 MPa and the 5% showed 36 MPa. This results clearly showed that
increasing SF content leads to higher strength gains due improvement in porosity.
Figure 11a. Effect of SF on Compressive Strength

of Concrete[9]
J. Durgaprasad et al.[9] reports that with increasing SF content the compressive strength steadily increases. The
concrete mixes were cured at 1000C for 72 hours and the strength was measured at the end of 28 days which
gave results as shown in Fig 11a. The results showed that the compressive strength increased stradily up to 40%
SF, this is evident from the microstructural analysis of the concrete samples tested. It showed that the concrete
has a porous structure and the SF occupies the pores without any hindrance and adds to the compactness of
concrete. Hence the strength of concrete steadily increased. Other authors like Escalante-Garcia et al.[11] found
out that SF levels of 5-10% was optimum amount, any further increase of SF amount decreased the compressive
strength.
33
Figure 11b: Effect of SF on Tensile Strength and Flexural

Strength of Concrete[9]
Flexural Strength and tensile strength characteristics of hardened concrete showed the same type of
improvement like compressive strength. As is evident from Fig 11b. the rate of improvement of tensile strength
rapidly in case of 30 to 40 % replacement value.
5. CONCLUSION
The Mineral Admixtures used to modify the properties of Alkali Activated Composites has been briefly
investigated recently. The conclusion of the above literature review can be summarized as following:
(1) The addition of Metakaolin into Alkali Activated Composite prolonged the setting time and increased
the flowability of the composite.
(2) Researchers have investigated that addition of FA/MK in Alkali Activated increased the compressive
strength of the composite.
(3) The addition of GGBS into the composite retarded the workability and reduced the setting time of the
Alkali Activated Composite. It was also noted that the inclusion of GGBS increased the rate of
development of compressive strength of the composite.
(4) It was observed that silica fume has a negative effect on the workability of Geopolymer Composite,
though the increased SF content had a positive effect on the compressive strength of the composite.
(5) Researchers investigated that increase in SF beyond its optimum content had a negative effect on the
compressive strength of the composite, thus reducing its value.
References
[1] Ghasan F. Huseien, Jahangir Mirza, Mohammad Ismail, S.K. Ghoshal, Mohd Azreen Mohd Ariffin, “Effect
of metakaolin replaced granulated blast furnace slag on fresh and early strength properties of geopolymer
mortar”, Ain Shams Engineering Journal 2016.
[2] Cheng TW, Chiu JP, “Fire-resistant geopolymer produced by granulated blast furnace slag”. Miner Eng
2003;16:205–10.
[3] Buchwald A, Hilbig H, Kaps Ch. “Alkali-activated metakaolin-slag blends performance and structure in
dependence of their composition”. J Mater Sci 2007;42:3024–32.
[4] Alaa M. Rashad, “ A Comprehensive overview about the influence of different activities on the properties of
alkali-activated slag”. Construction and Building Materials 47 (2013) 29-55.
[5] Yip Christina K, Lukey Grant C, Provis John L, van Deventer Jannie SJ. “Effect of calcium silicate sources
on geopolymerisation”. Cem Concr Res 2008;38:554–64.
[6] Pradip Nath, Prabir Kumar Sarker, “Effect of GGBFS on setting, workability and early strength properties of
fly ash geopolymer concrete cured in ambient condition”, Construction and Building Materials 66 (2014) 163–
171.
[7] Partha Sarathi Deb, Pradip Nath, Prabir Kumar Sarker, “The effects of ground granulated blast-furnace slag
blending with fly ash and activator content on the workability and strength properties of geopolymer concrete
cured at ambient temperature”, Materials and Design 62 (2014) 32–39.
[8] David JM Flower, Jay G Sanjayan, “Greenhouse gas emissions due to concrete manufacture”, The
International Journal of Life Cycle Assessment,July 2007, 12:282.
[9] F.N. Okoye, J.Durgaprasad, N.B.Singh, “Effect of silica fume on the mechanical properties of fly ash based-
geopolymer concrete”, Ceramics International 42(2016)3000–3006.
[10] Debabrata Dutta, “Effect of silica fume additions on porosity of fly ash geopolymers”.
[11] Escalante-Garcia Jose Ivan, Palacios-Villanueva Victor M, Gorokhovsky Alexander V, Mendoza-Suarez
Guillermo, Fuentes Antonio F. “Characteristics of a NaOH-activated blast furnace slag blended with a fine
particle silica waste”. J Am Ceram Soc 2002;7(85):1788–92.
[12] Ismail, I., Bernal, S. A., Provis, J. L., San Nicolas, R., Hamdan, S., & van Deventer, J. S. (2014).
“Modification of phase evolution in alkali-activated blast furnace slag by the incorporation of fly ash. Cement
and Concrete Composites”, 45, 125-135.
[13] Collins Frank, Sanjayan JG. “Effects of ultra-fine materials on workability and strength of concrete
containing alkali-activated slag as the binder”. Cem Concr Res 1999;29:459–62.
34
[14] Sanjay Kumar ,Rakesh Kumar, S. P. Mehrotra, “Influence of granulated blast furnace slag on the reaction,
structure and properties of fly ash based geopolymer”, J Mater Sci (2010) 45:607–615 DOI 10.1007/s10853-
009-3934-5.
[15] Buchwald A, Hilbig H, Kaps Ch (2007) J Mater Sci 42:3024 doi:10.1007/s10853-006-0525-6.
[16] K.Parthiban , K.Saravanarajamohan , S.Shobana , A.Anchal Bhaskar, “Effect of Replacement of Slag on
the Mechanical Properties of Fly ash Based Geopolymer Concrete”, International Journal of Engineering and
Technology (IJET).
35
Paper ID- 17
USE OF BRICK KILN DUST AS A SUPPLEMENTARY MATERIAL

FOR FINE AGGREGATE IN CONCRETE
Dipak Hajari1, Bikram Paul2, Kushal Ghosh3 and Dr. Partha Ghosh4
1, 2
ME Student, 3Ph.D Student and 4Associate Professor,
Department of Construction Engineering, Jadavpur University, Kolkata, India
E-mail: dipak.hajari@gmail.com1, bikram.bubai@gmail.com2 and kushalghosh100@gmail.com3
Abstract: At present, environmental pollution is one of the biggest threats facing mankind. And with the
increasing population and demand and the advent of various modern technologies to fulfill those demands,
the threat is becoming more and more uncontrollable day by day. One of such demand is the need for
houses or buildings to accommodate people, offices and factories. And for that, tons of concrete (a mixture
of water, aggregates and cement) are needed in daily basis for the construction of those buildings. Now,
when we extract sand, a key ingredient of concrete, from river-bed, we irreversibly damage our
environment in many ways; bank erosion, loss of valuable timber resources and wildlife habitats in the
riparian areas, lowering of ground water level to name a few. Here, a study is done by replacing sand up
to a certain proportion in the concrete making process by Brick Kiln Dust, a waste product in the brick
and tile factories and an environmental hazard itself, to show that we can tackle the above-mentioned
problems in two ways; by reducing sand mining and by reducing the ill effects of brick dust on
environment due to its storing and disposing problem. M20 grade of concrete is selected for the study and
sand is replaced by brick dust as 0%, 15%, 20%, 25% and 30%. The basic strength and workability
parameter like compressive strength, UPV and slump test etc of concrete were investigated.
Keywords- Brick Kiln Dust (BKD), Compressive strength, Mechanical properties, Water curing, PPC,
Cement, Aggregate.
1. INTRODUCTION
Nature is our mother and she has to sacrifice everything for the so-called development and
economic growth. Today’s increasing demand and growth led us in an alarming situation. It is
time to think for certain models which are essential both for us. The only way to achieve this
goal is to be serious on our nature which should be taken care of by our self to maintain
sustainability.
The construction industry is a huge consumer of natural resources. And as we grow, the
demand for these resources is increasing in an alarming rate. The industry uses concrete in
every structure it builds be it a very tiny one like a lamp post or a gigantic one like a multi-
storied building, bridges etc. This concrete is a mixture of a binding material (Cement), coarse
aggregates (Stones), fine aggregates (Sand) and water. Now, cement is made from naturally
available limestone, stones are picked up from quarries and sand is excavated from riverbeds.
With passage of time it has been proven that sand mining has an enormous negative impact on
the environment. In-stream mining lowers the stream bottom, causing the deepening of rivers
and near the estuaries. Sand extraction from the riverbeds led to large and deep pits on the
bed; as a result of it the water table drops down in that locality. Sand mining from the riverbed
changes the morphology of the bed and for that reason many bridges were damaged across
river Sutlej, Beyas etc. Apart from it, sand mining destabilizes bed and banks and result in
dramatic channel readjustments.
36
In India, brick kiln industries are the third largest industry and there is a big problem
associated with storing and disposing of Brick Kiln Dust (BKD), a hugely generated waste
product in the brick and tile factories and an environmental hazard itself. Now, we have found
that BKD can be a good supplementary material for fine aggregate in concrete in India. In
Holland, BKD was used for the first time in construction Industries. So, if we choose BKD for
that purpose then we can tackle those environmental complications in two ways; by reducing
sand mining and by reducing the ill effects of brick dust on environment due to its storing and
disposing problem.
In economical aspect, the cost of the concrete becomes less, if we use BKD as a
supplementary material in concrete. The benefits for using this type of concrete are,
i) The textured surface of this type of concrete is smoother than the other
conventional concrete.
One more suitable engineering property for use of this type of concrete is that it is
comparatively less permeable to normal concrete.
The aim of this article is to assess the properties of brick Kiln Dust [1] As a natural fine
aggregate partial substitute material, to achieve this fresh and hardened state properties of
concrete which are measured to ensure the fundamental parameter needed for design which
are evaluated. The study in this field shows some problems when using [2] Brick Kiln Dust
with timber in concrete. Some research work done by using [3] well burn bricks as coarse
aggregate in concrete and found that it was possible to achieve concrete of high strength. [4] It
is reported that the strength of concrete is gained within 22 % to 45% by using 33%Brick
Kiln Dust (BKD) and using a natural aggregate replaced [5] by Brick Kiln Dust (BKD) as
both coarse and fine aggregate the strength decreases within 22 % to 30 % , when Brick dust
used as a fine aggregate [6] the strength decreases.
2. EXPERIMENTAL DETAILS
2.1. Materials.
Brick Kiln Dust. In this experiment work we used Brick Kiln Dust (BKD) (it is the by-product
after burning the green brick, it storage and disposal is the biggest problem) from different
quarries after verifying of grading of Brick Kiln Dust (BKD) in different Brick Madding
Industry i.e. i) JADUPUR BRICK FACTORY, KALIACHAK, MALDA, WB, INDIA ii) BBI
ANULIA SATIGHACHA, RANAGHAT, NADIA-741201, WB, INDIA. Here the Brick Kiln
Dust (BKD) used as a partial replacement of sand in M-20 grade of concrete. The Brick Kiln
Dust (BKD) are taking from source and passing through a IS Sieve Size: 2.36 mm and
retained on IS Sieve Size: 75 Micron. Table-01 represent the physical properties of Brick Kiln
Dust (BKD) and Fig. 01 shows the grading of Brick Kiln Dust (BKD). Apart from this a water
absorption and specific gravity is done and found 1.23 % and 2.57 respectively.
37
Physical Properties of Brick kiln dust (BKD)
Property Test Result

Specific Gravity 2.57
Fineness Modulus
2.6
(FM)
Water absorption (%) 1.66
Fig. 1
Cement. PPC (Fly-ash based) cement has been taken confirming to IS 1489: Part-I.
Coarse Aggregate. 20mm down coarse aggregate collected from the Pakur stone quarry,
Jharkhand, India conforming to IS 383:1970. The test result are given in table and the grading
shows in fig. 02
Physical Properties of Coarse Aggregate
SL Test
Lab Experiment
NO Result
01 Specific Gravity 2.85
02 Water Absorption (%) 1.97
03 Bulk Density(Kg/M³) 2450
Fig.2
Fine aggregate. Sand taken from Moyurakhi River, Tilpara Barrage, Birbhum, West Bengal,
India. conforming to IS:383:1970.The test result are given in table no-05 and the grading
shows in fig. 04
Physical Properties of Sand (FA)
SL Test
Lab Experiment
NO Result
01. Fineness Modulus (FM) 2.62

02. Specific Gravity 2.81
Fig.3
3.0 MIX PROPORTION: In this experimental work it was performed a normal M-20 Grade
of concrete. Then incorporation of gel ( i.e. 0%, 15%,20% ,25% and 30%). The cement is
used PPC and Water Binder Ratio 0.50. The water used in this study was a portable drinking
38
water. Table-06 represent the mix composition of concrete.ng the Brick Kiln Dust (BKD) in
concrete as a partial replacement of sand at different percentage.
Table-06
M-20/Cement Qty-349 Kg/PPC & W/C Ratio-0.5
Materials (in Kg) 0 % BKD 15 % BKD 20 % BKD 25 % BKD 30 % BKD
20 MM CA 16.9 16.9 16.9 16.9 16.9

10 MM CA 11.27 11.27 11.27 11.27 11.27
Sand FA 18.56 15.78 14.84 13.92 12.99
BKD Nil 2.78 3.71 4.64 5.56
Cement (PPC) 8.48 8.48 8.48 8.48 8.48
4.0Water
Result and Discussion:4.67 4.67 4.67 4.67 4.67
Compressive Strength. Here M20 grade of concrete was considered for the experiments and
normal concrete met the target strength requirements both for 7 days and 28 days respectively.
As we replaced the sand by 15% BKD abrupt increase in strength was noticed with 7 days and
28 days compressive strength being 19.99 MPa and 26.99 MPa respectively .The values for
20%,25% and 30% did not follow any linear trend but the 7 and 28 day values satisfied the
codal provisions. Generally the increase in percentage of BKD in concrete led to the decrease
in workability and formation of a stiff mix with the mix having 30% BKD showing zero
slump. Higher stiffness of the mix led to improper compaction resulting in the formation of
voids. Also the quantity of fine particles in the matrix adsorbed the initial free water from the
mix and this may have prevented proper hydration for the mix’s with relatively higher
percentage of BKD.
Fig.5
39
Fig.4
Ultra-Sonic Pulse Velocity Test. As the UPV values depends on the several factors, variation in
the UPV between 7 days and 28 days may indicate the variation in moisture level present in
the cube. Though there is no direct relationship present in between compressive strength and
UPV, it’s only represent the localized voids presents and indirectly the density of the concrete.
Here, we can say that the increase or decrease in UPV values indicate the improper
compaction of concrete which can be supported by the results of workability test given in
Fig.5
Fig.6
Split Tensile Strength. All results are showed in the graphical representation in Fig.8. As soon as
the replacement by BKD increased, split tensile strength gets reduced due to weak bond
between brick dust particles and cement paste. Though the increment of split strength noticed
up to 20% replacement, these were lesser than concrete specimens with 0% BKD. Difference
between the concrete with 0% BKD and sustainable concrete with 20% BKD is very small
and 20% BKD can be used in manufacturing concrete.
40
Fig.7
Flexural Strength. For Normal concrete, flexural strength found for 28 days is 5.15 N/mm2. But
when sand was replaced by Brick Kiln Dust (BKD) amounting up to (15 %, 20 %, 25 % and
30 %) in same size of beam, the 28 days flexural strength was found 4.29 N/mm2, 3.48
N/mm2, 3. 46 N/mm2 and 3. 46 N/mm2. From the results, we can say that same trend can be
noticed for flexural strength as that of split tensile strength.
Fig.8
5. Conclusion:
• As we increase replacement of BKD in concrete fines content in the concrete also increase
and the mix become stiff, this leads to decrease in the compressive, split and flexural
strength.
• Based upon experimental result it may be said that the Brick Kiln Dust (BKD) can be used
in place of river sand up to an optimum percentage of 20%.If water reducing admixtures
are used judiciously higher amount of BKD may be used.
• Presence of silica in brick dust indicates that later age strength gain is a possibility and
further study is needed regarding this issue.
Reference:-
1. Sharma, Sharda, Ritesh Mall, and Khalid Raza. "Effect of waste brick kiln dust with
partial replacement of cement with adding superplasticizer in construction of Paver
41
Blocks." International Journal of Science, Engineering and TechnologyResearch 3.9

(2014): 2261-2266.
2. Mr. AnuragVerma, Ms. Disha Srivastava, Mr. Indresh Kumar “Use of Flyash & Lime as
Partial Replacement of Cement and Partial Replacement of Fine Aggregate with Brick
Kiln Dust” International Journal for Scientific Research & Development| Vol. 5, (2017), 1081-
1084
3. S.N. Raman, T.NGO, P. Mendis, H.B.Mahmud, “Replacement of Cement in Concrete”,
International Journal of Environmental Research and Development, ISSN 2249-3131
Volume 4, Number 1 (2014).
4. Nataraja M.C., Nagaraj T.S. and Reddy. A. 2001. Proportioning Concrete Mix with
Quarry Wastes Cement. Concrete and Aggregate Journal, ASTM. 23(2): 81-87
5. Tan KS, Zain MFM, Yus of KM3, Safiuddin M4,Chang TY5, Lee KS “Bilodeau J-
P.Optimisation de la granulométrie des matériaux granulaires de fondation des chaussées
[Dissertation PhD]. Québec: Université Laval; 200.
6. Rjesh Kumar Bharti Mr. R.D. Patel, “A Study on Low Performance Concrete using
MineralAdmixtures (Brick Kiln Dust and Silica Fume)”,International Journal for
Scientific Research &Development, Vol. 2, Issue 10, 2014, ISSN (online):2321-061.
7. Naceri, Abdelghani, and Makhloufi Chikouche Hamina. "Use of waste brick as a partial
replacement of cement in mortar." Waste management 29.8 (2009): 2378-2384.
8. Bektas, F., K. Wang, and H. Ceylan. "Effects of crushed clay brick aggregate on mortar
durability." Construction and Building Materials 23.5 (2009): 1909-1914.
9. Schackow, A., et al. "Influence of fired clay brick waste additions on the durability of
mortars." Cement and Concrete Composites 62 (2015): 82-89.
10. Vieira, T., et al. "Durability-related performance of concrete containing fine recycled
aggregates from crushed bricks and sanitary ware." Materials & Design 90 (2016):767-
776.
11. Tuyan, Murat, Özge Andiç-Çakir, and Kambiz Ramyar. "Effect of alkali activator concentration
and curing condition on strength and microstructure of waste clay brick powder-based
geopolymer." Composites Part B: Engineering 135 (2018): 242-252.
12. Aliabdo, Ali A., Abd-Elmoaty M. Abd-Elmoaty, and Hani H. Hassan. "Utilization of
crushed clay brick in concrete industry." Alexandria Engineering Journal 53.1 (2014):
151-168.
42
13. Elhakam, Ali Abd, Abd Elmoaty Mohamed, and Eslam Awad. "Influence of self-healing,
mixing method and adding silica fume on mechanical properties of recycled aggregates
concrete." Construction and Building Materials 35 (2012): 421-427
14. Cachim, Paulo B. "Mechanical properties of brick aggregate concrete." Construction and
Building Materials 23.3 (2009): 1292-1297.
15. P.K. Mehta, P.J.M. Monteiro, Concrete, Microstructure, Properties and Materials, third
ed., McGraw Hill, New York, 2006.
16. D.M. Sadek, Physico-mechanical properties of solid cement bricks containing recycled
aggregates, J. Adv. Res. 3 (2012) 253–260.
17. Ge, Zhi, et al. "Mix design of concrete with recycled clay-brick-powder using the
orthogonal design method." Construction and Building Materials 31 (2012): 289-293.
18. Ioannou I, Ilia A, Philokprou M. Use of crushed fired clay ceramics in the production of
mortars. Sustain Dev Plan IV 2009;120:257–64.
19. Ptacek P, Opravil T, Soukal F, Wasserbauer J, Masilko J, Baracek J. The influence of
structure order on the kinetics of dehydroxylation of kaolinite. J Eur Ceram Soc
2013;33:2793–9.
20. Aliabdo, Ali A., Abd-Elmoaty M. Abd-Elmoaty, and Hani H. Hassan. "Utilization of
crushed clay brick in cellular concrete production." Alexandria Engineering Journal 53.1
(2014): 119-130.
43
Paper ID- 18
The Aerodynamic and Seismic Behaviour of Cable-Stayed Bridge

Pratik Sen1, Mantu Kumar1, Parth Shukla1, Dr. Purnachandra Saha2* (ORCID-0000-0002-2722-1669)
1
(Mail id: sen.pratik709@gmail.com, mantu7090@gmail.com, parth.shukla099@gmail.com)
2
Senior Associate Professor, School of Civil Engineering, KIIT Deemed to be University, Bhubaneswar, India
(Mail id: dr.purnasaha@gmail.com)
* Corresponding Author
ABSTRACT
The cable stayed bridge is a bridge in which the burden of the deck is upheld by various cables running straight
from the towers by cantilever approach. The cable stayed bridge does not require firm anchorages to oppose the
horizontal force of the fundamental cables as of the suspension bridge. The structural mechanism is simple in
concept: the cables convey the deck loads to the towers and from that point to the foundation. The objective of
this study is to review the aerodynamic and seismic behavior of cable stayed bridge. The expanding range length
of the cable stayed bridges makes the aerodynamic strength a noteworthy concern for this kind of structure.
Vortex, Buffeting and Flutter are the aerodynamic instability process, have turned out to be the decisive criteria
in the construction of the long-span bridges. Long span structure is subjected to seismic excitation, contingent
on the mass distribution of the deck, cable and pylon have a tendency to induce both torsion and flexural
motions in the bridge deck. It is observed from the study that on increasing span length of the cable stayed
bridge up to certain length the vertical displacement is increased and then displacement becomes almost
constant on increasing more span length. Critical flutter speed gradually increases with the increase in main span
of the bridge. Near fault ground motions can introduce more destructive reaction than an equivalent or bigger
and higher vertical acceleration.
Keywords: Cable-Stayed Bridge, Flutter, Seismic, Dynamic Behavior, Stay Cable, Pylons
1. Introduction
The cable stayed bridge consists of one or more supports (pylons) through which the cables run over the bridge.
The bridge consists of deck, girder, cables and pylons. In this type of bridge, the cables are directly connected to
the deck forming fan, harp and semi harp like patterns as shown in Fig 1. The base of the pylons or towers is
firmly held into the bed of the river mostly by caisson footing. The loading mechanism on the bridge is as
follows firstly the live vehicle load is held by the deck then the deck transfers the load to the cables attached
with it. Now this load is further transferred to the tower or pylons of the bridge and then finally to the bridge
foundation. Here the cables are in tension and the towers are in compression. Russky Bridge in Russia is the
longest cable stayed bridge having the main span of length 1104m which was completed in the year 2012.
Different types of loads that act on the Cable stayed bridge are Dead Load, Super-imposed Load, Live Traffic
Load, Seismic Load and Aerodynamic Load. Dead Load is the weight of the bridge deck (pre-cast deck, steel
deck), hangers, and the reinforcing steel, which is used in the Reinforced concrete deck, and the cables.
Super-imposed load will be taken as mainly the components of the road surface. Live traffic load consist of the
ongoing vehicle load over the bridge. Aerodynamic loads are those forces which are exerted on the structure due
to the wind effects acting upon it. Several researchers performed analysis of behavior of cable stayed bridge
under aerodynamic and seismic load. The objective of this study is to review the aerodynamic and seismic
behavior of cable stayed bridge.
44
Fig.1 Cable stayed bridge
2. Aerodynamic Behavior on Cable-stayed Bridge

Cable stayed bridges having long spans are fragile and flexible to wind effects. When the bridge is under
construction or say development stage it is generally observed that the bridge stiffness is lesser as compared to
the operational conditions afterwards, hence these bridges get more prone to the wind impacts during
development stage. Hence expanding span length during the development phase is the issue of major concern in
determining the aerodynamic nature of the bridge. There are basically two categorized into two parts i.e.
limited-amplitude and divergent-amplitude wind induced vibrations as observed from Fig 2 [1]. Limited
amplitude responses include buffeting, vortex induced vibrations, wake induced vibrations and rain wind
induced vibrations which affect the serviceable discomfort and dynamic fatigue in the structure. While divergent
amplitude responses comprises of flutter, galloping and wake instability which affects the structural instability
of the structure [2].
Fig.2 Distribution of Bridge Aerodynamics[2]
2.1 Vortex Induced Effect
The aerodynamic execution is varied because of the adjustment in wind speed of structure. When wind
streams past a thin and tall body at particular speeds, an oscillation is experienced. This act is called vortex
shedding. The shape and size of the pylon plays a vital role in the process of vortex shedding (Fig. 3) [3].
Exactly when wind flow through the pylon, there are low pressure vortices which are confined on its
downstream side. This vortex power will most likely make the pylon to have motion from side to side at its
position. If the frequency of vortex shedding matches with the frequency of the structure then whole structure
will experience some harmonic oscillations. There are various sorts of vortex vibrations, for instance
45
vortex-induced, galloping, vibration, Karman vortex excitation , rain and wind induced at high wind speed
etc.[4] Vortices are the elective shed event in the wake of deck indicating danger of vortex shedding excitation
which impacts the streamlined coefficient of a structure. This is in charge of variance of body surface pressure
which happens at low diminished speeds. At higher wind speeds vortex excitations can be observe due to the
axial flow along the cable. The role of water stream on the cables and disturbance caused by winds is a crucial
factor in the execution of the vortex-induced vibration at high reduced velocity [4]. The combined effect of the
water stream on the cable, the axial flow and the turbulence stream may increase the phenomenon of
aerodynamic instability. The several vibrations can be organized as per their reduced speeds; inclined cables can
be generally sorted as galloping instability and vortex-induced oscillation of which vortex-induced vibration at
high reduced speed (Fig 4) [5-6]. The rain– wind-induced vibration may be clarified as the vortex-induced
vibration at high decreased speed. The vortex induced vibrations have been seen to occur at wind speeds in the
middle of 4m/s to 10m/s. To diminish the impact of vortex induced aerodynamic issues, baffles are provided at
quarter the chord length location, in this manner lessening drag coefficients and slope of lift coefficients.
Various dampeners can also be used in the dynamic system to have the control over the vibrations.
Fig.3 vortex shedding [3] Fig.4 Downstream axial flow at inclined cable [4]
2.2 Buffeting Effect
Buffeting is the sudden instability happened because of shock wave motions or wind current separation made
when objects strike one another. Buffeting excitation is caused by fluctuating forces induced by inflow
turbulence in the wind field. Sudden effect of a seismic load or dynamic load drives cables to strike each other.
This sudden shock prompts vibration in whole structure. Buffeting is additionally not a genuine fluid– structure
interaction process as in the basic approximation to a first estimate does not impact the Properties of the wind
stream. Or maybe it is seen as constrained basic vibrations energized by a random external force, i.e. turbulent
wind [7]. It is a high frequency phenomenon. Buffeting dependably exists when a bridge is presented to
turbulent flow, or in other words in the Atmospheric Boundary Layer. Sudden effect of a seismic load or
dynamic load drives cables to strike each other. This sudden shock instigates vibration over whole structure
[3,8].
Another kind of wind phenomenon on cable stayed or suspension bridges is the buffeting disturbance caused by
atmospherically interference. At the time of development when cables do not provide any support, large
amplitude stabi movements have been seen because of buffeting wind induced reactions, particularly buffeting
reactions because of turbulent winds, are one of the real sources that prompt exhaustion harm in long-range
cable stayed bridges [9-10].
2.2.1 Resonant buffeting

The occurrence was depicted by davenport and can happen for bridges with two symmetrical planes of cables.
Likewise, the wind strike the upwind and down-wind planes of cables with a period postponement of B/U,
Being the separation between the two planes of cables and U the mean wind speed as in Fig.5 in case that this
46
delay corresponds with a large portion of the period Tt related with torsional deck mode, at that point resonant
impacts can be accomplished. [11]
Fig.5 Resonant Buffeting

The critical velocity Ucr for the resonant buffeting is defined as,
2B 1
Ucr =
Tt
2.3 Fluttering Effect
Flutter is defined as a threshold wind velocity beyond which the bridge oscillates in a divergent way. Fluttering
happens because of the consolidated impact of bending and torsion. It is a unsteady vibratory movement of the
structure because of the coupling between elastic deformation of the structure and the aerodynamic force
followed up on it [3]. The bridge with the closed box girder deck has a importantly higher critical flutter wind
velocity than the plate girder deck. Due to which a closed box girder attains a better aerodynamic stability [12].
Long range spans like suspension bridges and cable stayed bridges are more inclined to fluttering due to their
huge d/t value, d being the depth of structure parallel to wind and t being the minimum lateral measurement.
Fluttering is a important vibratory motion and has even lead the fall of Tacoma Narrows cable stayed bridge in
1940 [3].
For the better behavior of cable stayed bridge in aspects of its aerodynamic behavior, the wind induced flutter
instability is a major concern that has to be kept in mind [13].
3. Seismic Behavior of Cable-stayed Bridge

Bridges are exceptionally vulnerable structures, and basic as life-lines, therefore the understanding of their
seismic nature is basic requirement [14]. Cable stayed bridge are getting more popular these days because of
good stability and elegant for long span bridges compared to other types of bridges. It is generally observed that
increasing the length and stiffness of the bridge, the amount of seismic vibrations presented by these impact
increments [15]. At the moment of the seismic activity, most of the initial vibration modes occur in the order of
a couple of moments that exhibit for a critical period, and they are basically deck modes. The cable then trails
this process and combines with the deck modes assuring more vibration for the structure. The role of towers
comes to play later because those need vibrations of higher order to combine with the deck modes [16]. The
phenomenon of torsional and flexural tension are caused due to both deck and pylon of the bridge at a point
when thin Deck of generally flexible Long range structure observes seismic excitation and this process is
governed by the mass distribution of the deck. The deck relationship among tower and deck affect the seismic
execution of the cable stayed bridges to a great extent. The stiff association of deck and tower confine the
displacement of deck horizontally under seismic tremor vibration and prompt transfer of powers from
superstructure to substructure, and consequently increment the maximum lateral force expected i.e base shear of
the tower. It is also observed for a greater value of deck flexibility, the movable or floating design is approached
and it also develops the horizontal displacement of deck under service loading. Furthermore, shape and height of
the tower, generously influence the dynamic reaction of the cable stayed bridge [13]. Cable stayed bridges
present great relationship between the cables, the structure and the surrounding soil that provide a complex
seismic excitation. The typical coupling between vibration modes and the minimized damping and weight of
these structures may prompt huge three directional motions of the deck and the cables during the earthquake.
The relation between the cables and the structure more often than not decreases the seismic reaction of the
47
towers under a broadband seismic excitation; however it might cause huge issues if the earthquake is dominated
by certain frequency bands that resonate with the dominant modes of the structure [15].
Pylon is a column which is connected with all the cables and is responsible for transmission of cable’s forces to
the foundation. By changing the height of column, it will also change the inclination of cable [17]. Shapes of
pylons such as Single pylon, a type pylon and Inverted Y type pylon influence the seismic response of cable
stayed bridge [18]. Single pylon and Inverted Y type pylon are better than A type pylon for parameters cable
forces, pylon and deck deformation. Single pylon is stronger in longitudinal direction in resisting earthquake
force while Inverted Y type pylon is stronger in lateral direction in resisting earthquake force.
Fig. 6 Different mode shapes of Bill Emerson Memorial Bridge (Benchmark CSB) [8]
3.1 Near field effect
Near field vibrations are normally observed in the regions which are in the range of 10-15 km from the epicenter
of the earthquake. In the very beginning of ground motions high pulse qualities are exhibited by near fault
ground motions with high power and the duration of the pulse is generally more than 2 seconds [18]. The
phenomenon of progressive direction is predominant in near field vibrations i.e. if each part of fault’s portion
gets ruptured in some specific period of time, there are some waves which gets shifted to the site in a fault to
fault manner and this effect is much harmful when the site is in same direction to the fault Because at that time
there is a generation of high pulse at the site location consequently decreasing the time period throughout which
the structure receives the fault waves. A great pulse in the ground motions speaks to a sudden energy input that
can initiate serious harm with structures [19]. Medium-to-long span decks subjected to pulse like vertical parts
of tremor strong motion could be more endangered than those subjected to the horizontal ground motion only or
combined horizontal and ordinary vertical ground motion. Consequently, fusing a vertical pulse model for the
improvement of close fault ground motions and including pulse like vertical ground motions in the examination
is suggested for solid seismic evaluation and plan of medium-to-long stretch structures in the region of active
faults [20].
This impact applies a essential effect on cable stayed bridges which have various long period modes [21]. In
ongoing decades, the examples of some near field ground motion earthquakes are Chi-Chi (1999) , San
Fernando (1971), Kobe (1995), Parkfield (1966), and South Napa (2014) [22]. Cable-stayed bridges are
sensitized to three dimensional excitations because of the solid coupling between methods of vibrations. The
presence of a vertical solid speed pulse has a tendency to lessen the high frequency substance of the record and
thus increment the seismic interest on medium-to-significant lot structures. For the structures considered and the
seismic tremor record set utilized, the bending moment and rotation requests along the bridge deck are
considerably expanded as for most extreme dead and live load requests because of pulse like vertical ground
motion records [20].
3.2 Examination between Far-field and Near-field Effect
Near field earthquakes have particular qualities not quite the same as far-field ones. There is involvement of
higher acceleration and more constrained frequencies in near-field earthquakes as compared to far-field
48
earthquakes. Numerical researches have concluded that near fault ground motions can introduce more
destructive reaction with isolated bridges than an equivalent or bigger higher vertical acceleration. For a similar
ductility factor, near fault ground motions imply a larger strength demand in their acceleration sensitive area
contrasted with far-fault motions, with both requests communicated as a small amount of their separate elastic
demands.
Discussion
Fig: 7 Span Length Vs Flutter Speed [23]
Fig: 8 Bar Chart Chi-Chi Earthquake [18]
49
Fig: 9 Bar Chart Kobe Earthquake [18]
It is observed in the figure that for the spans reaching up to 750 m there is potential to show larger
displacements and effecting the seismic response, as the span length is gradually reduced the displacement is
gradually decreased. Thus the magnitude of seismic effect is also minimized.
This result shows that for reliable designs of cable stayed bridges the main span length is a crucial factor in
governing the displacement of the deck during seismic activity.
In the case of aerodynamic nature of cable stayed bridge it is seen that the critical flutter speed gradually
increases with the increase in main span of the bridge up to around 1000 m but as the span increases greater
value the flutter speed decreases and almost becomes constant.
Conclusion
Behavior of cable stayed bridge under aerodynamic and seismic load has been reviewed in this present study.
Under aerodynamic load vortex induced effect, buffeting, fluttering and galloping effects are reviewed. Near
field effect and Far field effect are studied under seismic behavior of Cable Stayed Bridge. Following
conclusions are drawn from the study.
It is observed that on increasing span length of the cable stayed bridge up to certain length the vertical
displacement is increased and then displacement becomes almost constant on increasing more span length.
Critical flutter speed gradually increases with the increase in main span of the bridge.
Near fault ground motions can introduce more destructive reaction with isolated bridges than an equivalent or
bigger and higher vertical acceleration.
References
1. Ma C., Duan Q., Liao H., (2017) "Experimental investigation on aerodynamic behavior of a long span
cable-stayed bridge under construction", KSCE Journal of Civil Engineering, 22: 2492–2501. doi:
10.1007/s12205-017-0402-7
50
2. Shen, Jerry, Tsai M., Chang K., and Lee G.C., (2004) "Performance of a Seismically Isolated Bridge
under Near-Fault Earthquake Ground Motions", Journal of Structural Engineering, 130:861-868. doi:
10.1061/(ASCE)0733-9445(2004)130:6(861)
3. www.irjet.net
4. Matsumoto, M., (2001) “Vortex-induced cable vibration of cable-stayed bridges at high reduced wind
velocity", Journal of Wind Engineering & Industrial Aerodynamics 89: 633–647. doi:
10.1016/S0167-6105(01)00063-0
5. Darabi S., Aghayari R., (2018) "Investigation of the behavior of reinforced concrete-coupled shear wall
with opening using IDA analysis", Asian Journal of Civil Engineering, 19:319-332.
doi:10.1007/s42107-018-0030-7(
6. www.ijste.org
7. Larsen A., Larose G.L., (2015) "Dynamic wind effects on suspension and cable-stayed bridges",
Journal of Sound and Vibration, 334:2-28. doi: 10.1016/j.jsv.2014.06.009
8. Dyke, S.J., Caicedo J. M., Turan, G., Bergman, L. A., and Hague, S. (2003). "Phase I benchmark
control problem for seismic response of cable-stayed bridges." Journal of Structural Engineering,
ASCE, 129, 857-872.
9. Saha P., (2015) “Seismic Control of Benchmark Cable-Stayed Bridges Using Variable Friction
Pendulum Isolator”, Advances in Structural Engineering, pp 1271-1282
10. Wu, Q., Takahashi K., Nakamura S., (2003)"Non-linear vibrations of cables considering loosening",
Journal of Sound and Vibration, 261:385-402. doi:10.1016/S0022-460X(02)01090-8
11. www.ijser.org
12. Zhu Q., Xu Y.L., Shum K.M., (2016) "Stress-level buffeting analysis of a long-span cable-stayed
bridge with a twin-box deck under distributed wind loads", Engineering Structures, 127:416-433. doi:
10.1016/j.engstruct.2016.08.050
13. Javanmardi A., Ibrahim Z., Ghaedi K., Jameel M., Khatibi H., Suhatril M., (2017) "Seismic response
characteristics of a base isolated cable-stayed bridge under moderate and strong ground motions",
Archives of Civil and Mechanical Engineering, 17:419-432. doi: 10.1016/j.acme.2016.12.002
14. Soti P.R., Saha P., (2011) “Review of various passive control devices for seismic control of benchmark
cable stayed bridge” International Journal of Earth Sciences and Engineering, 04(06):726-731
15. Patil S S., Patankar J.P., (2017) “Seismic response of cable stayed bridge with different types of pylons
of various heights” International Journal of Advanced Research and Development, 2:204-211
16. www.scilit.net
17. www.iitk.ac.in
18. Soyluk K., Karaca H., (2017) "Near-fault and far-fault ground motion effects on cable supported
bridges", Procedia Engineering, 199:3077-3082. doi: 10.1016/j.proeng.2017.09.421
19. Chopra A.K., Chintanapakdee C.,(2001) "Comparing response of SDF systems to near-fault and
far-fault earthquake motions in the context of spectral regions", Earthquake Engineering & Structural
Dynamics, 30:1769-1789. doi: 10.1002/eqe.92
20. Lin Y.Y., Cheng C.M., Wu J.C., Lan T.L., Wu K.T., (2005) “Effects of Deck Shape and Oncoming
Turbulence on Bridge Aerodynamics”, Tamkang Journal of Science and Engineering 8(1) 43-56
51
21. Zuo, D. (2008) "Field observation of vortex- and rainwind- induced stay-cable vibrations in a three
dimensional environment", Journal of Wind Engineering & Industrial Aerodynamics, 96: 1124-1133.
doi: 10.1016/j.jweia.2007.06.046
22. Ge Y.J., (2016) “Aerodynamic challenge and limitation in long-span cable-supported bridges”,
Advances in Civil, Environmental and Materials Research (ACEM16)
23. Lakshmi A.A.V., Saha P., (2011) “Review of the Performance of Semi-Active Control Systems for
Benchmark Cable-Stayed Bridge”, International Journal of Earth Sciences and Engineering,
04(06):720-725
52
Paper ID- 20
Basalt Fiber and Its Composites: An Overview
Krishan Pareek1, Dr. Purnachandra Saha2* (ORCID-0000-0002-2722-1669)
1
Student, School of Civil Engineering, KIIT Deemed to be University, Bhubaneswar, India
(Email id: karankool7568@gmail.com)
2
Senior Associate Professor, School of Civil Engineering, KIIT Deemed to be University, Bhubaneswar, India
(Email id: dr.purnasaha@gmail.com) *Corresponding Author
ABSTRACT
Basalt fiber (BF) is obtained from basalt rock and can be so finely divided that it can form fiber, which if mixed with polymer
give rise to Basalt fiber reinforced polymer (BFRP). In this paper chemical constituent, mechanical properties and durability of
BF is reviewed. Also, properties of BFRP like temperature variation and adhesion is also discussed. It is observed that BFRP is
better than Carbon fiber reinforced polymer (CFRP) when it comes to cost effectiveness and in comparison, to Glass Fiber
reinforced polymer (GFRP) it got better strength therefore becoming a good alternative to other fiber materials. BFRP has high
flexural strength and better adhesion as compared to CFRP and GFRP, when temperature increases mechanical and thermal
properties of BFRP reduces remarkably.
Keywords—Adhesion, "Basalt Fiber Reinforced Polymer (BFRP)", basalt fiber, bond characterization, flexural strength, tensile
strength.
1. Introduction
The compressive strength of the concrete is higher as compared to its tensile strength, therefore we should provide steel
reinforcement to enhance the properties of reinforced concrete that increases the strength and durability of the structure. When
conventional steel reinforcement is provided in concrete structures, they tend to corrode the structure and thus decrease its
strength. Therefore, the need to endorse the use of natural fibers in polymer reinforcement has seen a widespread recognition
[1]. Henceforth, stress has been given for eco-friendly material that gives strength and does not harm the environment. One of
these materials are natural fiber in polymer-based composites. These materials are fabricated by combining high strength,
thread-like structures with a polymer. The result is a firm material that has high strength, yet lightweight i.e. A kilogram of
basalt reinforcement equals 9.6 kilogram of steel. The strength and stiffness characteristics are landed by the fiber part whereas,
polymer hold the fiber in alignment, known as Fiber Reinforced Polymer. Polymer composites are these days normally
reinforced by Glass strands (GF). GF has high modulus, great strength, and shows adequate attachment towards different
polymer networks, and, in the meantime, it is low estimated [2]. In all the more requesting applications carbon fiber (CF),
outflanking GF in regard with firmness and quality is utilized. Be that as it may, CF is particularly costlier than GF, and the best
possible attachment among CF and polymer framework is likewise difficult to acknowledge by and large [3]. As of late another
kind of fortifying fiber with magnificent mechanical properties and substance, properties seemed known as Basalt Fiber. At the
point when Basalt is utilized as fiber in the composite material, at that point it is known as Basalt Fiber Reinforced Polymer
(BFRP). Basalt is a characteristic substance that is shaped in volcanic rocks affected by solidified magma, with a liquefying
point enveloped somewhere in the range of 1500 and 1700°C [4,5]. The basalt rocks can be converted into a type of strands [6].
Basalt fiber is non-destructive, non-harmful, naturally sheltered, and have great attractive protection properties [7]. They are
described by their high protection from alkalinity in encompassing cement, expelling the general detriment of glass filaments
[8].
Fiber reinforcement is gaining popularity due to their performance in terms of ecological cost i.e. US$ 5 per 1 kilogram, Basalt
fiber (BF) is highly potential alternatives compared to US$ 15 per 1 kilogram of glass fiber and US$ 30 per 1 kilogram of
carbon fiber and better chemical-physical properties [9]. Use of fibers obtained from basalt rocks as reinforcement in polymer
composites is a generally new chapter [10]. Mechanical properties of BF are viable with glass ones, while tensile strength and
displacement at break of BFare very strong. Like some other fibers, the elastic modulus of BFs also depends upon the chemical
constituent and observed that it is typically equal or somewhat superior to that of glass fibers [10]. These properties of BFs help
in swapping glass fiber for impact resistance composite material. Environmental sustainability is achieved through basalt, as it
is chemically inert and has very low thermal conductivity compare to other reinforcement, making it good choice.
53
2. COMPOSITION OF BASALT FIBER

Being rich with oxides of Mg, Ca, Na, K, and Fe, along with traces of alumina and earth crust consist of 33% of basalt
becoming a copious mineral. Table 1 and Fig. 1 depicts chemical content of basalt but can differ according to topographical
distribution [11]. Chemically these fibers are composed of plagioclase, olivine, pyroxene and clinopyroxene minerals [12].
Based on the
Fig. 1 - chemical composition of basalt fiber [4, 9, 11, 25]
above chemistry, there are different type of basalt like Boninite (rich in magnesium), Alkali basalt (rich in sodium) and Tholeitc
basalt (rich in sodium). Figure 1 describes the mineral distribution within basalt and effect of geological changes [11].
Basalt fibers basic mechanical properties are shown in (fig. 2), which shows it (BF) has a good tensile strength and also an
excellent ‘E’ (elastic modulus).
Fig. 1a. Based on differing silicate concentrations, basalt tetrahedron is shown which describes the different phases of basalt mineral [11].
54
Fig. 2. Mechanical properties of Basalt Fiber [11]
3. PROPERTIES OF BASALT FIBER

3.1 Alkali Resistance
Most important role in service life of composite is to balance stress and external load that is done by reinforced fiber [1, 8, and
13]. Durability, maintenance under critical environment and long-term performance are some of hurdles that a composite has to
face and the most important one is chemical corrosion [14, 15]. Toughness, degradation properties of resin and crack-
propagation are the factors on which resistance against corrosion depends. The complete mechanism behind degradation of fiber
is not been understand completely, but structural and morphological changes arises in basalt when it has been exposed to
unfavorable environment for long period of time. For shorter duration of time basalt is chemical stability as compared to other
composite materials [16], therefore extensively used to counter chemical corrosion.
Basalts are more stable in salt solution especially water than glass, but in strong acids these show less stability [5, 17, 18]. In
alkaline solutions, the complete loss of tensile strength deprivation is due to pitting of the fiber in large area. In acid solutions,
the most important cause for the loss of the tensile strength is the damage or change in the chemical constituent of the fibers. For
basalt weight loss in alkali, boiling water, and acid is also considerably lower [19].
From the figure-3 it can be depicted that BFs and their composites shows greater resistance to salt and water solutions as
compared to glass fiber where as in acidic environment glass beat the basalt, from figure 4 it is observed that the properties of
BFs can be modified using different matrix and coupling agents [18]. For example, when Epoxy resins reinforced BFs and glass
fibers were tested with a seawater solution, similar degradation was observed for epoxy based basalt fiber glass fiber. BFRP
exhibit greater strength by reducing the content of Fe2+ in BFs [1, 20].
Water Salt Acid Alkaline
Figure 3-Comparison of tensile strength of different fiber [18]
55
Water Salt Acid Alkaline
Fig 4-Comparison of tensile strength of different FRP [18]
3.2 Thermal Stability

In comparison to E-glass (-60 to 450/460°C) basalt have wider temperature range -200 to about 650/800°C [5, 11, 19,21].
Due to mineral composition and presence of huge number of micro pores, BFs are passive fire protection and thermal insulation
applications [22- 25].
BFs are better than E-glass in terms of residual relative strengths (after heat treatment) [19]. In 300-500°C range, basalt
outperforms E-glass in the stressed condition and can maintain integrity up to 1250°C in unstressed (used as fire/heat barrier)
condition [26].
Crystallization behavior is the key factor in evaluating the heat stability of BFs. It depends on heat treatment condition and
chemical composition of fiber. In basalt crystallization occur in two phases first phase in which spinel structure is formed due to
oxidation of ferrous cation (high of iron oxides in basalt), in second phase the diffusion of divalent cations (Fe2+, Mg2+, Ca2+)
take place where they come to surface to form Nano crystalline layers of CaO, (Mg,Fe)3O4 , MgO [25]. At very high
temperature nucleation site is formed due to crystallization of pyroxene on spinel crystals [25, 27, 28]. Crystallization of basalt
can be controlled by doping elements like zirconium oxide [25,30].
At high temperature, high thermal stability of basalt fibers material is observed due to characteristics of natural basalt rocks,
which nucleate. The basalt rocks, however, do not need but during the melting process, produce a usual nucleating agent such as
Fe3O4 [30].
When partial pyrolysis of basalt polymer matrix composite in nitrogen is done at high temperature, ceramic matrix composite is
obtained which has better resistance to oxidation [25, 31-34].
4. PROPERTIES OF BASALT FIBER REINFORCED POLYMER

Basalt is made of basalt fiber that carry load provide strength and stiffness characteristic whereas polymer distributes load &
protects the fibers from environmental conditions. Polymer can be of different types such as thermosetting (vinylester, epoxy),
thermoplastic (polypropylene, polyethylene etc depending on the properties of polymer polyethylene) etc. Depending on the
properties of polymer properties of BFRP changes.
4.1 Flexural Behavior

Composites are subjected to various loading conditions during their service life, among them, flexure is predominant [35]. When
basalt is mixed with polyester and a three-point flexure test is done and found that the flexural strength and flexural modulus
were highest for the composite material with pure stacking of Basalt has more flexural strength among the other pure glass
sample specimen [36].
56
When series of test (tensile and compression static, defamation and fatigue) were carried on basalt fiber with two matrices of
vinylester and epoxy it was found that basalt reinforced epoxy composites showed higher mechanical properties in comparison
to vinylester [37].
When the effect of surface modification of BF and glass fiber (GF) reinforced with unsaturated polymer matrix in acidic and
alkali environment it was observed that BFRP laminates have superior mechanical properties as compare to GFRP laminates
(higher tensile and impact strength of BFRP composite in acidic environment) [38].
A comparison is done between BFRP having epoxy matrix and BFRP having tourmaline micro/Nano particles it was seen that
as compared to Basalt fiber epoxy composite, Basalt loaded with different tourmaline percentage showed higher tensile and
flexural strength [39].
It was noticed that BF reinforced with poly butylene succinate (PBS) (thermo plastic) composites enhanced the tensile and
flexural properties appreciably. It was also seen that BF/PBS composites may be a possible alternatives of PP or PP composites
to manufacture some daily commodities to solve the “white pollution” in environmental management [40].
When tensile, flexure and compressive test were performed on BF reinforced plastic and GF reinforced plastic, results were
compared it was seen that basalt have better compressive and flexural strength while glass have better tensile strength as seen in
figure [41].
Ul 800
ti Tensile
m
Flexural
at 600 Compression
e
str
en
400
gt
h
(M
200
P
a)
0
Basalt Glass
Fig 5- Comparison of Mechanical properties of basalt and E-glass fiber composites [41].
Basalt fiber reinforced concrete increases the mechanical strength, E, and crack resistance of concrete with increasing both
content and length [42].
4.2 Temperature Variation

Incident that happens in petrochemical and chemical industries are because of sudden ignition of unstable substance prompting
fire breakout eventually leading to death of numerous human life and degradation of environment, therefore need of the hour is
a material that has resistance to fire at high temperature as well as good mechanical properties.
When thermal properties of glass fiber is compared to basalt fiber it was found that tensile modulus of BF was around 60.4 ±
18.9 GPa while tensile strength was found to be 568 ± 267 MPa [43]. There are several properties of basalt that make it a
remarkable fiber such as chemical resistance, thermal resistance, low water absorption and sound insulation ability. When basalt
is mixed with thermoplastic and thermoset matrix it can be a very good reinforcing material [43-44, 46-47].
At various temperature, basalt fiber demonstrates shifted elastic properties because of warm impact, When basalt fiber is
tempered at various temperatures of 300, 200, 100 and 50°C it was seen that there was some basic change inside the fiber, it
indicates
57
At the point when hot wear properties of basalt cross breed composites were examined differing the basalt content from 0 to 40
Volume % keeping the earthenware content consistent (10 Volume %) by utilizing a warmed stick on-circle mechanical
assembly. It was seen that disc temperature increased when coefficient of friction is increased to 300°C. Coefficient of friction is
directly proportional to volume content of basalt fiber hybrid composite [48]. When heat is applied to basalt fiber it is noted that
as compared to heat applied part the non-heated part of is much cooler it shows heat transfer within particle is very less. When
matrix properties are changed it is seen that deformity/damage to composite after the farthest point of warm immersion at
hoisted temperature is accomplished [11, 33, 34, 49, 50, 51].
4.3 Adhesion
The grip among strands and framework influences a definitive mechanical property of a composite, not only in the off-axis
direction but as well as parallel to the fibers. Through many experiments, it is realized that there is an area in structure and
composites near fiber and matrix interface known as inter-phase; i.e., region between bulk fiber and mass matrix which is a
three-dimensional [52]. This inter-phase incorporates the 2D region of contact among fiber and network (the interface) yet in
addition consolidates the area of some limited thickness stretching out on the two sides of the interface in both the fiber and
matrix as appeared in figure 3. Past investigations presumed that if the interface between basalt fiber and matrix, for example,
vinyl ester (VE) is powerless then it isn't protected to utilize basalt. To tackle these issues numerous examinations have been led
and one of them reports that on the off chance that we utilize basalt fiber covered with silica nano-strengthened epoxy resin. The
outcome demonstrated that there was 5-25 % enhancement in mechanical properties. The presence of nano-silica at the inter-
face between the BF and VE matrix prompts a critical improvement of inter-laminar and ultimate tensile strength [53].
Figure 6 - Detailed diagram of composite inter-phase [54]
By the utilization of surface change of the fibers, grip properties among matrix and BF can in like manner be upgraded by
thermal/chemical treatment [55, 56], or by mass alteration of the polymer matrix, which includes the summation of various sorts
of compatibilizers. At the point when the polyamide-6 is reinforced with BF it demonstrates that the inter-facial bond between
BFs and polyamide can be to a greatly extent enhanced by the utilization of silane coupling agents in the whole temperature run
[35,43]. At the point when Polypropylene is reinforced with BF, with the end goal to accomplish an adequate interfacial bond,
the fibers were treated with the response blend of maleic corrosive anhydride and sunflower oil. The test outcome showed that
the mechanical characteristics improved when diverged from the unadulterated polypropylene matrix. Specifically, it was
uncovered that BF hybridization brought about an expansion in the mechanical properties [57].
5. INDUSTRIAL APPLICATION
Mechanical properties, chemical properties, and thermal stability properties, Basalt has wide range applications. Basalt is a pure
substance with no health hazards after being spun with a diameter higher than 6 μm [41]. In USA and Europe is made label free
as it is non-reactive towards water, it also has three times heat insulating capacity as compared to asbestos as well as it does not
cause pollution, due to its resistance against electromagnetic waves it is used as electro technical purpose in automobile
industry, aircraft and ship [58-62].
58
Basalt also have good recyclability as well have good interfacial adhesion therefore used widely in construction industry in
places like rebar for reinforcement in RCC structures, main benefit of basalt are explained below-
 Higher strength to weight ratio than steel;

 Basalt and concrete have the same thermal expansion coefficient ;
 Require no up gradation like coating in glass fiber rod;
 Resistance to salty, acidic and alkali environment
 No heat transfer;
 Does not conduct electricity;
 Can be cut easily by regular tools
Nowadays Fiber reinforced concrete is also getting attention due to excellent impact resistance, resistance to splitting, high
flexural and tensile strength, superb permeability. Fresh basalt fiber has shown good properties such as good workability, anti-
seepage resistance, crack resistance, good stability, impact resistance and thermal stability therefore becoming a feasible
alternative to polyacrylonitrile and polypropylene fibers.
6. DISCUSSION
In this paper, Chemical composition of basalt fiber is studied as well as chemical resistance, thermal stability of BF is discussed,
and properties of BFRP as a substitute of conventional steel are reviewed through various properties that differentiate it from
GFRP and CFRP.
As per the above study, we observe that CFRP composite is the favored material in the construction industry, due to their high
stiffness and strength. As, the durability of carbon fiber is rather less and the resulting damage resistance is poor whereas if
BFRP is used then these problems are solved as it has high damage resistance and good energy absorption capacity. When
tensile, flexure and compressive test were performed on BF reinforced plastic and GF reinforced plastic, results were compared
it was seen that basalt have better compressive and flexural strength as compared to glass fiber. It can also be said that increase
length and content of basalt in R.C.C increases the crack resistance, elastic modulus, and flexural strength of concrete. When the
fiber was tempered at different temperature, it displayed the increase in tensile strength of the fiber when subjected to 300 C for
different time periods, 1, 15 and 60 minutes.
From the adhesion tests, it was demonstrated that the ultimate mechanical properties of a composite get affected by the grip
between fibers and matrix, not only in the off-axis direction but also parallel to the fibers. When basalt fiber covered with silica
Nano-reinforced epoxy resin, we see a remarkable 5 to 25 % improvement in the mechanical properties of basalt.
7. CONCLUSION
In this review, Properties and performance of Basalt Fiber and Basalt Fiber Reinforced Polymer (BFRP) as a substitute of
conventional steel and other composite material is studied. Basalt, when in acidic medium, displayed the most effective results.
The reviews of different experimental data suggest that BFRP resists better chemical attack than CFRP and GFRP in acidic and
salt water solution. Alkali resistance, thermal stability of Basalt fiber as well as flexural strength, temperature variation and
adhesive nature of Basalt Fiber Reinforced polymer is studied. Basalt composites exhibited higher tensile strength when
subjected to higher temperature for different time periods then GFRP. The adhesion between the Basalt fiber and its composites
is better as compared to carbon fiber and its composites. Initial cost of basalt fiber depend chemical composition, quality of raw
material leading to formation of different type having different mechanical, thermal and chemical properties. Due to wide
application in infrastructure and automobile industry and low cost basalt fiber can material for future.
References
[1] Wei, B., Cao, H. and Song, S. 2010, “Tensile behavior contrast of basalt and glass fibers after chemical treatment”. Mater
Des 2010; 31(9):4244–50.
[2] Bunsell, A. R. 1988. “Fibre Reinforcements for Composite Materials” (Elsevier, New York, 1988)
59
[3] Gay, D., Hoa, S. V. and Tsai, S. W. 2003. “Composite Materials, Design and Applications” (CRC Press, New York, 2003
[4] Militky, J., Kovacic, V. and Rubnerova, J. 2002, “Influence of thermal treatment on tensile failure of basalt”. Eng Fract
Mech; 69: 1025-33.
[5] Militky, J. and Kovacic V. 1996, “Ultimate mechanical properties of basalt filaments”. Text Res; 66: 225-9.
[6] Pareek, K., Ansari, M.S., Verma, S. and Saha, P. 2016, “Seismic retrofitting of the structure: An Overview”, International
Conference on Recent in Materials and Mechanics; Burla, India.
[7] Palmieri, A., Matthys, S. and Tierens, M. 2009, Basalt Fibers: Mechanical Properties and Applications for Concrete
Structures, Taylor & Francis Group, London, UK, 165-169.
[8] Sim, J., Park, C., Moon D.Y. 2005, Characteristic of Basalt Fiber as a Strengthening Material for Concrete Structures,
Composites: Part B, 36(6), 504-512.
[9] Deak, T. and Czigany, T. 2009, “Chemical composition and mechanical properties of basalt and glass fibers: a
comparison”, Text Res J; 79: 645–51.
[10] Ross, A. 2006. “Basalt fibers: alternative to glass?” Compos Technol; 12:44–8.
[11] Dhand, V., Mittal, G., Rhee, K.Y., Park S.J. and Hui, D. 2015, “A short review on basalt fiber reinforced polymer
composites”, Composites: Part B 73:166–180.
[12] Sarasini, F., Tirillò, J., Ferrante, L., Valente, M., Valente, T. and Lampani, L. 2014, “Dropweight impact behaviour of
woven hybrid basalt–carbon/epoxy composites”. Compos Part B – Eng;59:204–20
[13] Park, J. M., Shin, W. G. and Yoon, D. J. 1999, “A study of interfacial aspects of epoxy-based composites reinforced with
dual basalt and SiC fibres by means of the fragmentation and acoustic emission techniques”. Compos Sci
Technol,59(3):355–70.
[14] Shan, Y. and Liao, K. 2001, “Environmental fatigue of unidirectional glass-carbon fiber reinforced hybrid composite”.
Composites Part B,32(4):355–63.
[15] Tsotsis, T. K., Keller, S., Lee, K., Bardis, J. and Bish, J. 2001, “Aging of polymeric composite specimens for 5000 h at
elevated pressure and temperature”. Compos Sci Technol, 61(1):75–86.
[16] Berozashvili, M. 2001, “Continuous reinforcing fibers are being offered for construction”, civil engineering and other
composites applications. Adv Mater Com News Compos Worldwide, 21(6):5–6
[17] Stephen Cater. 2002. ‘Editorial’, International Composites News, (40)
[18] Wu, G., Wang, X., Wu, Z., Dong, Z., and Zhang, G. 2014, “Durability of basalt fibers and composites in corrosive
environments”, Journal of Composite Materials 0(0), 1-15
[19] Tengiz, C., ‘Industrial assimilation of the effective type of fibre with multicomponent charge’, http://www.tctv.ne.jp
[20] Wei, B., Cao, H. and Song, S. 2011. “Degradation of basalt fibre and glass fibre/epoxy resin composites in seawater”,
Corrosion Sci ; 53: 426–431.
[21] Sergeev 1994, “Basalt Fibers – A Reinforcing Filler for Composites”, Powder Metallurgy and Metal Ceramics, 33(9-10),
555-557.
[22] Novitskii, A. G. 2004. “High temperature heat insulating materials based on fibres from basalt type rock materials”. Refract
Ind Ceram, 45:144-6
[23] Hao, L. and Yu, W. 2011, “Comparison of thermal protective performance of aluminized fabrics of basalt fibre and glass
fibre”. Fire Mater, 35:553-60.
[24] Hao, L. C. and Yu, W. D. 2010, “Evaluation of thermal protective performance of basalt fibre non-woven fabrics”. J Therm
Anal Calorim, 100:551-5.
[25] Fiore, V., Scalici, T., Di Bella, G. and Valenza A., 2015. “A review on basalt fibre and its composites”, Composites Part B,
74-94.
[26] ‘Basaltex, The thread of stone’, http://www.basaltex.com
[27] Moiseev, E. A., Gutnikov, S. I., Malakho, A. P. and Lazoryak, B. I. 2008, “Effect of iron oxides on the fabrication and
properties of continuous glass fibres”. Inorg Mater, 44:1026-30.
[28] Gutnikov, S. I., Manylov, M. S., Lipatov, Y. V., Lazoryak, B. I. and Pokholok, K. V. 2013. “Effect of the reduction
treatment on the basalt continuous fibre crystallization properties”. J Non-Cryst Solids, 368:45-50.
[29] Lipatov, Y. V., Arkhangelsky, IV., Dunaev, A. V., Gutnikov, S. I., Manylov, M. S. and Lazoryak, B. I. 2014,
“Crystallization of zirconia doped basalt fibres”. Thermochim Acta, 575:238-43.
8
60
[30] Shuni, Y. and Xiaodong, Z. 2013. “Chemical and Thermal Resistance of Basalt Fiber in Inclement Environments”, Journal
of Wuhan University of Technology-Mater. Sci. Ed. Vol 28 No. 3.
[31] Cerný, M., Glogar, P., Golias, V., Hruska, J., Jakes, P. and Sucharda, Z. 2007, “Comparison of mechanical properties and
structural changes of continuous basalt and glass fibres at elevated temperatures”. Ceram Silik, 51(2):82-8.
[32] Cerný, M., Glogar, P. and Sucharda, Z. 2009., “Mechanical properties of basalt fibre reinforced composites prepared by
partial pyrolysis of a polymer precursor”. J Compos Mater, 43:1109-20.
[33] Cerný, M., Glogar, P., Sucharda, Z., Chlup, Z. and Kotek, J. 2009, “Partially pyrolyzed composites with basalt fibres -
mechanical properties at laboratory and elevated temperatures”. Compos Part A Appl S ;40:1650-9.
[34] Cerný, M., Halasova, M., Schwaigstillova, J., Chlup, Z., Sucharda, Z. and Glogara, P. 2014, “Mechanical properties of
partially pyrolysed composites with plain weave basalt fibre reinforcement”. Ceram Int, 40:7507-21
[35] Pareek, K., Tiwari, P., Verma, S. and Saha, P., 2017, “A Review Basalt Fiber Reinforced Polymer and Composites”,
International Civil Engineering Symposium (ICES).
[36] Amuthakkannan, P., Manikandan, V. and Uthayakumar, M., 2014, “Mechanical Properties of Basalt and Glass Fiber
Reinforced Polymer Hybrid Composites”, Journal of Advanced Microscopy Research Vol. 9, 1–6.
[37] Colombo, C., Vergani, L. and Burman, M., 2012. “Static and fatigue characterisation of new basalt fibre reinforced
composites”, Composite Structures 94. 1165–1174.
[38] Manikandan, V., Jappes, J. T. W., Kumar, S. M. S. and Amuthakkannan P., (2012). “Investigation of the effect of surface
modifications on the mechanical properties of basalt fibre reinforced polymer composites”, Composites: Part B, 43 812–818
[39] Ary Subagia, I. D. G., Tijing, L. D., Kim, Y., Kim, C. S., Vista, FG-IV and Shon, H. K. 2014, “Mechanical performance of
multiscale basalt fiber–epoxy laminates containing tourmaline micro/nano particles”. Compos Part B – Eng, 58:611–7.
[40] Zhang, Y., Yu, C., Chu, P. K., Lv, F., Zhang, C. and Ji, J. 2012. “Mechanical and thermal properties of basalt fibre
reinforced poly(butylene succinate) composites”. Mater Chem Phys, 133:845-9.
[41] Lopresto, V., Leone, C. and Iorio, I. D. 2011, “Mechanical characterisation of basalt fibre reinforced plastic”. Compos Part
B – Eng, 42(4):717–23.
[42] Elshafie, S. and Whittleston, G. 2015. “A Review Of The Effect Of Basalt Fibre Lengths and Proportions On The
Mechanical Properties Of Concrete”, IJRET: International Journal of Research in Engineering and Technology, Volume: 04
Issue: 01
[43] Czigány, T. 2006, “Special manufacturing and characteristics of basalt fiber reinforced hybrid polypropylene composites:
mechanical properties and acoustic emission study”. Compos Sci Technol 66:3210–20.
[44] Reddy, B. R., Ravi, K. M. and Waugh, B. K. 2009, “Compositions comprising melt-processed inorganic fibers and methods
of using such compositions”. US patent 7493968B2;
[45] Reddy, B. R. and Heathman, J. F. 2012, “Cement compositions comprising high aspect ratio materials and methods of use
in subterranean formations”. US patent 8123852B2;
[46] Hung, T. D., Louda, P., Kroisová, D., Bortnovsky, O. and Xiem, N. T. 2011, New generation of geopolymer composite for
fire-resistance. In: Tesinova Pavla, editor. Advances in composite materials – analysis of natural and man-made materials,
vol. 4. InTech; p. 73–98.
[47] Wu, Q., Chi, K., Wu, Y. and Lee, S. 2014, “Mechanical, thermal expansion, and flammability properties of co-extruded
wood polymer composites with basalt fiber reinforced shells”. Mater Des 60:334–42
[48] Ozturk, B., Arslan, F. and Ozturk, S. 2007, “Hot wear properties of ceramic and basalt fiber reinforced hybrid friction
materials”. Tribol Int 40:37–48.
[49] Borhan, T. M. 2012. “Properties of glass concrete reinforced with short basalt fiber”. Mater Des 42:265–71
[50] Sinica, M., Sezeman, G. A., Mikulskis, D., Kligys, M. and Cesnauskas, V. 2014, “Impact of complex additive consisting of
continuous basalt fibres and SiO2 microdust on strength and heat resistance properties of autoclaved aerated concrete”.
Constr Build Mater 50:718–26.
[51] Haddad, R. H., Saleh, R. J. A. and Akhras, N. M. A. 2008, “Effect of elevated temperature on bond between steel
reinforcement and fiber reinforced concrete”. Fire Saf J 43(5):334–43
[52] Gauvin, F., Cousin, P. and Robert, M. 2015, “Improvement of the Interphase Between Basalt Fibers and Vinylester by
Nano-Reinforced Post-Sizing”
[53] Cech, V., Palesch, E. and Lukes, J. 2013, “The glass fiber–polymer matrix interface/interphase characterized by nanoscale
imaging techniques”,
9
61
[54] Greco, A., Maffezzoli, A., Buccoliero, G., Caretto, F. and Cornacchia, G. 2013, “Thermal and chemical treatments of
recycled carbon fibres for improved adhesion to polymeric matrix”. J Compos Mater 47(3):369–77
[55] Deak, T., Czigany, T., Tamas, P. and Nemeth, C. 2010. “Enhancement of interfacial properties of basalt fiber reinforced
nylon 6 matrix composites with silane coupling agents”. Express Polymer Lett. 4(10):590–8
[56] Meszaros, L., Gali, I.M., Czigany, T. and Czvikovszky, T. 2011. “Effect of nanotube content on mechanical properties of
basalt fibre reinforced polyamide 6”. Plast Rubber Compos Process Appl 40:289-93
[57] Bashtannik, P. I., Kabak, A. I. and Yakovchuk, Yu.Yu. 2003, “The effect of adhesion interaction on the mechanical
properties of thermoplastic basalt plastics”. Mech Compos Mater 39:85-8
[58] Paivaa, J. M. F. D., Santos, A. D. N. D and Rezende, M. C. 2009. “Mechanical and morphological characterizations of
carbon fiber fabric reinforced epoxy composites used in aeronautical field”. Mater Res. 12(3):367–74.
[59] Taketa, I., Ustarroz, J., Gorbatikh, L., Lomov, S. V. and Verpoest, I. 2010. “Interply hybrid composites with carbon fiber
reinforced polypropylene and self-reinforced polypropylene”. Compos Part A – Appl Sci 41(8):927–32.
[60] Hollaway, L. 1993. Polymer composites for civil and structural engineering. New York: Springer.
[61] García, P. D. L. R., Escamilla, A. C., and García, M. N. G., 2013 “Bending reinforcement of timber beams with composite
carbon fiber and basalt fiber materials”, Compos Part B – Eng 55:528–36
[62] Ary Subagia, I. D. G., Kim, Y., Tijing, L. D., Kim, C. S. and Shon, H. K. 2014, “Effect of stacking sequence on the flexural
properties of hybrid composites reinforced with carbon and basalt fibers”, Compos Part B – Eng 58:251–8.
10
62
Paper ID- 22
Mechanical and Durability Properties of Concrete Using Recycled Aggregate

Srishti Saha 1, Rini Dey2 and Dr. Purnachandra Saha3*(ORCID-0000-0002-2722-1669)
1
M. Tech Student, School of Civil Engineering. KIIT Bhubaneswar, Odisha, India
(Email: srishti.civil94@gmail.com)
2
Assistant Professor, BBIT, Budge Budge, Kolkata, WB, India
(Email: rdey789915@gmail.com)
3
Senior Associate Professor, School of Civil Engineering. KIIT, Bhubaneswar, Odisha, India
(Email: dr.purnasaha@gmail.com) *Corresponding Author
Abstract
Recycled aggregate is a term used to describe crushed concrete or brick masonry from construction debris that is
reused in other construction projects. Recycled aggregate (RA) is produced by crushing concrete, and masonry to
reclaim the aggregate. In recent years the concrete industry has started using Construction and Demolition (C&D)
waste in structural concrete application. This can allow the concrete industry to reduce its natural reduces and thus
help it to continue to grow without harming the environment. The objective is to review the mechanical and
durability properties of concrete using recycled aggregate (RAC) from various sources like brick, mortar, waste
ceramic and precast wastes. The recycled aggregate concrete specimens lost their compressive strength when they
were cured in open-air conditions. With the use of admixtures, super plasticizers and modified mixing approaches,
the desired properties of RAC can be obtained. When recycled aggregate produced with precast concrete waste,
mortar, masonry, waste ceramic etc for different replacement ratio up to 100%, the strength of conventional
concrete can be achieved with some modification. The durability properties such as shrinkage & creep affect
according to the replacement of RAC. Performance of all the construction wastes compiled in this study is suitable
replacement of natural aggregate.
Keywords: Recycled Aggregate, Physical & Chemical Properties, Strength, Durability properties.
1 Introduction
The sustainable construction concept was introduced due to the growing concern about the future of our planet
because construction industry is a huge consumer of natural resources and, simultaneously, waste producer.
Conservation, rehabilitation and strengthening of construction industry may be termed as sustainable development
[1]. Concrete is one of the most widely used construction materials in the world, mainly due to its favourable
features such as durability, versatility, satisfactory compressive strength, cost effectiveness and availability.
However, the use of conventional concrete has been claimed to be not environmentally friendly, manifested by
frequently voiced negative concerns such as the depletion of the reserve of natural resources, high energy
consumption and disposal issues [2]. Since aggregates in concrete comprise about 60% to 75% of the total volume
of concrete any reduction in natural aggregates consumption will have significant impacts in the environment [3].
In the last two decades, a variety of recycling methods for construction and demolition wastes (CDW) have been
explored and well developed. For instance, concrete rubble has been used to replace natural aggregate (NA) after
being treated [1-2]. Crushed clay bricks (CCB) come from either load bearing masonry walls or just cladding or
party walls. Ceramic wastes obtained from broken roof tiles, blocks, bricks, electrical insulators etc. used as
aggregates in concrete [4]. Nowadays, prestressed precast concrete are very popular. Since high grade of concrete
is used, aggregate from these demolition wastes would have high strength [5]. Several researchers investigated the
performance of these wastes as recycled aggregate separately; hence there is a need to compile and analyse the
performance of all these wastes in a single platform. The objective is to review the mechanical and durability
properties of concrete using recycled aggregate (RAC) from various sources like CDW, brick, mortar, waste
ceramic and precast wastes.
63
2 Properties of Recycled aggregates of different source

RAs are extracted through the processing of the debris generated from the demolition of concrete structures and
other construction debris such as waste concrete, rejected precast concrete members, broken masonry, concrete
road beds and asphalt pavement, leftover concrete from ready mix concrete plant and the waste generated from
different laboratories. Reused aggregates are made out of original aggregates and old adhered mortar [6]. Mainly
the different source of recycled aggregate considered for the study are : 1)Mortar, 2) Recycled Brick, 3) Waste
ceramic, 4) Wall and Floor tiles, and 5) Precast RCA
The physical properties of reused aggregates rely upon both adhered mortar quality and the amount of followed
mortar. RA can be comprised of blocks, plates, and different random materials, for example, glass, woodland,
cardboard, resilient and different trash alongside pulverized concrete [7]. Recycled aggregate contains crushed and
uncrushed parent aggregate covered with mortar, and little bits of masonry, waste ceramic and solidified mortar.
Research on reused aggregate demonstrates that the water absorption limit of RA is higher than that of waste clay
bricks, mortar, waste ceramic floor and wall tiles and this is a direct result of porosity of adhered mortar [8-11]. A
few scientists contemplated that the compact density, apparent density and bulk density of NA is 12%, 3.3% &
13.3% higher than that of brick aggregate [12]. Aggregate density was seen to be decrease with increment in
porosity of aggregate. This conduct is ascribed to the higher assimilation rate of the crushing index and water
absorption ratio exhibit in bigger pore size of RA particles. Comparative patterns have been contemplated for the
fineness modulous and thickness of RAs. The fineness modulous of recycling waste was cause due to the lack of
fine material [13-14]. The thickness and ingestion limit of reused aggregates are influenced by adhered mortar and
they should be known before the use of reused aggregates in concrete creation with a specific end goal to control
properties of new and solidified cement [15]. In this way, the mechanical and durability properties of RA are
essential parts of the investigation of aggregate. The physical composition of various source of recycled aggregate
is shown in Table 3.
(a) Mortar (b) Recycled Brick (c) Waste Ceramic (d) Precast RCA
Fig. 1. (a) (b) (c) and (d) shows different types of Recycling aggregate
64
Physical Specific Water Bulk Porosity Fineness

property gravity Absorption(%) Density (%) modulous
(kg/m^3)
Types of RAC
Mortar (Dang et al 2018 & 2.01 3.9 1570-1650 ---------- -------

Martinez 2013) [13-14]
Recycled Brick (Cachim 1.702 15.81 1805 48.6 2.55
2009 & Dang et.al 2018)
[14,16]
Waste ceramic (Anderson et 2.300 5.5 2263 -----
al & Rashid et.al 2016) 7.951
[17-18]
Precast RAC 2.42 33.71 2640 21 --------

( Brandes 2016) [19]
3 PROPERTY OF FRESH CONCRETE
3.1 WORKABILITY
Workability is measured by but in general, is measured by slump test. It can be described as a consistency measure.
Workability is the property of fresh concrete [20-21]. The absorption capacity of RA affects the workability of
RAC deeply. Depending upon the type of crusher used to shape and size of the aggregate also affect workability.As
compare to round aggregate annular shape aggregate have higher contact area& this behaviour also shows the
reclaim waste as precast concrete,recycled brick and recycling plant of rubble [22]. As the contact area is high the
absorption of water is also high therefore the water-cement ratio will become higher in order to obtain similar
workability [23]. So, it can be said that for a constant w/c ratio the workability must be the same for every mix.
Normal mixing approach, Stone enveloped with Portland cement,stone enveloped with pozzolonic powder are
some factor on which workability depends [24].
Workability of ceramic waste, waste clay brick, waste mortar depends on the water absorption, porosity and pore
size of aggregate is the main reason behind the reduction of slump value of waste aggregate comparision to normal
concrete [8,10,25]. When amount of moisture absorption increases upto 30% amount of ceramic waste also
increases 30% [26].
It was observed that when coarse and fine recycled aggregate mixed with ceramic brick workability gets improved
by a significant value. For 100% replacement slump value was increased upto 4.6% & 7% for normal concrete and
ceramic brick respectively [27].
The slump values of the recycled aggregate concrete slightly increased with increase in silane content. The
improvement of workability can be attributable to the thin coating film made from pozzolanic powder, which
hampers the water absorption behavior of RA from fresh mixes and enhances workability. It was suggested that for
replacement of NA by RA requires similar characteristics which can be attained by using a different process or
65
alike crushing equipment. For achievement of the same compressive strength and workability as conventional
concrete, it requires to add more cement to concrete [28-29].
But it was reported that superplasticizers [30-31] increase the workability significantly, using superplasticizers
when replacement ratio was 100% and 25% slump of concrete mixes increases by 25.80% and 7.31%. whereas the
addition of super plasticizer decreases the wet density of the concrete mix. When water repelling agent was used it
was reported that there was 75% and 48% increase in slump value for RAC and NAC respectively with the
addition of silane slump value also increases showing the hydrophobic nature of silane leading to high free water
content. Using method as presoaking [32] of aggregate workability ability of RAC can be solved from an economic
point of view.
4 MECHANICAL PROPERTIES
4.1 Compressive Strength
Strength durability and structural performance of concrete is affected by the compressive strength of concrete,
therefore, becoming one of the most effective properties of hardening of concrete.properties of RAC depends upon
several factors such as w/c ratio, use of super-plasticizer , mineral dust content, water absorption,curing condition
[33-36]. Shape and sizes of aggregate affects the compressive performance of RAC. As compare to smaller
aggregate bigger size aggregate have contains higher amount of adhered mortar. As the amount of adhered mortar
is high the compactness is also high therefore to achieve better quality of cement matrix [22]. When there is a
replacement of RCA by RBA(recycled clay brick aggregate) lower compressive strength was observed and RCA
shows better performance as compared to RBA [37] .
It was observed that when natural aggregate was replaced in brick aggregate by 15% similar strength is observed
[38]. But when 50% replacement of crushed concrete brick was done the compressive strength for 7 days and 28
days are reduces by 14% and 20% respectively.This indifferent behavoiur may be due to the higher absorption
capacity and good control of RA grading to a larger content. Due to this property recycled clay brick aggregate can
absorb large quantity of water which may obstruct the hydration of cement at early age[13].
When there is the replacement of coarse aggregate by recycling waste as floor and wall tiles by 20% it was seen
that compressive strength remains unaffected by this cause.It was observed that when 100% replacement of floor
and wall tiles is done in concrete there is 4.3% and 5.6%decrease as compared to the normal concrete. The
compressive strength of NAC is similar to RAC made with waste materials as floor and wall tiles. As compared to
the virgin tiles waste tiles shows relative results. Presoaking method and water-cement ratio are some factors on
which compressive strength depends [18,30].
The ceramic aggregate shows the minimal effect on the compressive strength because of ceramic particles are
pozzolonic in nature when used in a cement mixture it shows a cementitious property. It was observed that
compressive strength almost remains unaffected as furthermore the ratio was increased up to 100% [39-40].
Rather than the type of aggregate used in RAC the development of compressive strength depends on the type of
binding mortar as compared to mortar is prepared using by 1% vinegar (by weight of cement amount) Strength
increases by 8.47%, 8.18%, 1.07% after 28days, 90days, 180days respectively. It may be because of high strength
bonding between the acid attack and hydrated cement paste which may leads to calcium compounds to calcium salt
of the attacking acids [41-42]. Quality of concrete and solubility of calcium salt depends on the rate of attack [43-
45]. It has been observed that the calcium containing products in the cement paste reacts with the acetic acid and
produced C-S-H gel. Compressive strength has a positive effect when calcium compound is used [42]. Figure 2
shows Compressive strength of various types of waste aggregate. It is observed from the figure that recycled
aggregate from ceramic tiles, crushed brick and precast wastes performed better. Ceramic tiles wastes can be used
as 100% replacement of natural aggregate.
66
60
Compressive Strength(MPa) 50
40
30
20
10
0
RCA Brick Mortar Floor tile Wall tile Precast
0 percent 20 percent 60 percent 80 percent 100 percent
Types of aggregate
Fig. 2. The Compressive Strength of RAC using different types of recycling waste [18,31,46-47]
4.2 Split Tensile Strength
For determining the performance of concrete under tensile stress the split tensile test is one of the most popular
indirect methods for evaluation of tensile strength [48]. water cement ratio, RA replacement ratio, mixing method ,
Shape and sizes of aggregate are some factors on which tensile strength of RAC depends [49-50].
With the increase in RA replacement ratio of RCA or crushed clay brick (CCB) the tensile strength has been
observed to be decreasing. When coarse aggregate was replaced in crushed clay brick at 25%, 50%, and 75% the
split tensile tends to decrease 12%, 10% and 6% respectively [51]. Cement content ,porous structure,curing age are
some factors on which tensile strength depends. Cement content and open enviornment curing at 20%, and 50%
replacement of natural aggregate by crushed clay brick exhibits nearly equal tensile strength after 7days. With the
increase in curing period split tensile strength of recycled concrete aggregate also increases. When recycled
concrete was replaced in recycled clay brick at 20% and 50% the split tensile strength tends to increase 31% and
34% respectively after 28days [52]. Rather than the type of aggregate,the normal concrete made with recycled fine
and coarse aggregate has 20% more splitting strength then RAC made with RA [53].
It was suggested that for replacement of Natural aggregate by ceramic coarse and fine aggregate (CWA & CFA)
requires similar characteristics which can be attained by using a different process or curing period. For
achievement of the same tensile strength as conventional concrete, it requires to add more cement to concrete. But
it was reported that using of ceramic sanitary waste aggregate creates a refinement of the pore system,volume of
capillary pores significantly increases and decreasing the volume of macro pores [54].
The study has shown that when NA replaced by ceramic floor and wall tiles strength can remain unaffected for
concrete made using CWA and conventional concrete [55-58]. When three types of tiles such as ceramic tiles,
waste tiles and wall tiles was prepared at 0.55 w/c ratio with replacement up to 100% by volume of standard
concrete it was founded that tensile strength reduces by 6.5%. Tensile strength of recycling waste depends on the
presoaking method and the aggregate replacement is the main reason behind the reduction of tensile Strength of
wall & tiles specimen in comparison to normal concrete [18,59].
67
When RAC was replaced in concrete by Precast waste recycled concrete aggregate at 20%, 50% and 100% the
strength gain 1.65%,1.86% and 1.94% respectively [19] .
As compared to waste tiles, the strength of clean wall tiles increases with an increase in replacement percentage of
NA by RCA. When there is the replacement of industrial brick waste by 30% it was seen that tensile strength
remains unaffected by this cause [60]. It was observed that when 50% and 100% replaced in concrete the split
tensile strength tends to decrease 9.2% and 8.9% respectively.
Adhered mortar and low adhesiveness are some property on which tensile strength depends. The strength gain of
concrete containing recycled concrete are more as compared to clean and industrial brick waste because of bonding
between the adhered mortar and low adhesiveness to ceramic [61]. It is observed from Figure 3 that splitting tensile
strength of various types of waste aggregate are acceptable range. Precast and ceramic tile produced better strength
when replacement is 100%.
5
4.5
4
Tensile Strength(MPa)
3.5
3
2.5
2
1.5
1
0.5
0
RCA Brick Mortar Floor tile Wall tile Precast
Types of aggregate
0 Percent 20 Percent 40 Percent 60 Percent 80 Percent 100 Percent
Fig. 3. The Splitting tensile Strength of RAC using different types of recycling waste [18, 30, 46-47,49]
4.3 Flexural Strength
Another factor that affects the structural performance of concrete is a flexural strength. It depends on water binder
ratio, RAC replacement ratio, curing of concrete and moisture condition of aggregate. With the increase in RA
replacement ratio of RAC, the flexural strength has been observed to decrease [29,62-65]. The recycled concrete
has 3% and 9% less flexural strength then RCA made without crushed clay brick (CCB) after 7days and 28days
respectively. When recycled clay brick (RCB) was replaced 50% to make RAC 96% and 94% less flexural strength
was recorded as compared to that of Natural Concrete aggregate. This is mainly because of the low relative
Young’s modulus of CCB and interfacial bonding between the concentration of tensile stress and matrix aggregate.
For replacement ratio 50% crushed concrete brick shows significant affect [13].
when the replacement ratio is 30% the strength was slightly lower or remains unaffected for concrete made using
pure cement mortar after 7days [66] . The type of aggregate used in RAC the development of flexural strength
depends on the type of hydrated concrete and cementitious mortar as compared to RAC prepared using mortar the
tensile strength increases by 2.019Mpa, 2.945Mpa, and 3.513 Mpa after 7days, 14days and 28days respectively
[67-68].
68
It was observed that As compared to natural coarse aggregate when 100% replacement was done by waste floor
and wall specimens the flexural strength was observed to maximum reduce by 17.9%. But when 100% replacement
of waste ceramic was done the flexural strength reduced by 25%. Presoaking and aggregate substitution also effects
the flexural strength of concrete. For 50% replacement of aggregate by floor and tile no significant effect on
tensile strength.The main reason behind the weaker bonding between the tiles aggregate and adhered cement
mortar [18].
When effect of Hydration of cement was studied on flexural strength on using ceramic filler content it was
observed that flexural strength using mortar with ceramic fillers with 10%,15% and 20% of cement mass
(M10,M15,M20) strength increases 8.5%,8.8%, 9.2% respectively at 28 days and maximum increases upto 12%
after 56 days. It was observed that using ceramic waste aggregate strength gain upto 50% at 2days. This was
because of with the increase in amount of ceramic waste aggregate has leads to the mechanical performance of
recycled aggregate concrete [69]. It can be deduced from Figure 4, that brick wasted exhibit better flexural
strength.
10
9
Flexural Strength(Mpa)
8
7
6
5
4
3
2
1
0
RAC Brick Mortar Floor tile Wall tile Precast
% of RA replacement
0 Percent 20 Percent 40 Percent 60 Percent 80 Percent 100 Percent
Fig. 4. The Flexural Strength of RAC using different types of recycling waste [18,46,70-72]
5 DURABILITY PROPERTIES
5.1 Shrinkage
Shrinkage is one of the essential factors which impacts the durability properties of reused aggregates. When there
is the replacement of natural aggregate by recycled crushed brick aggregate it was seen that recycled brick
aggregate has a higher shrinkage. Size of aggregate particles, RA replacement, curing age are some other factors on
which shrinkage was corelated [73-75].
Shrinkage of recycled aggregate is deeply affected by curing. It was suggested that wet chamber curing gives good
results for lower replacement ratio whereas use of water immersion curing followed by wet chamber curing gives
better results in higher replacement ratios [76].
69
As compared to natural aggregate, the fine crushed brick aggregate has almost six times higher shrinkage at an
early age and increases with in the same rate after 90days. But according to some researches when both coarse and
fine crushed bricks are replaced by natural concrete aggregate there is no significant change in shrinkage at an early
age [77].
When the replacement ratio upto 10%, mortar containing crushed brick shows an higher shrinkage but increment
in replacement ratio upto 20% it shows lower shrinkage as compared to normal concrete. With the increase in
mortar content, shrinkage value was observed to be decreasing. This similar behaviour also obtained when mortar
containing aggregate was replaced by crushed granite. This condition can occur due to the presence of pozzolanic
gel on which expand the specimens and shows lower shrinkage as the conventional concrete [78-79].
With the increase in ceramic waste powder (CWP) replacement ratio, drying shrinkage strain values is observed to
be decreasing. Because of ceramic particles are very fine in nature and also due to the pozzolanic action waste
ceramic particles are changed pore structure of the specimen as a result water movement through the concrete was
hampered. It was also observed that with the addition of 10% to 20% ceramic powder to the normal concrete no
significant changes can occur. As compared to conventional concrete to the ceramic waste the drying shrinkage
strain was reduced by 29% and 60% for replacement more than 20% respectively [80].
The use of recycled aggregates presents two main benefits. Primarily, there would be a natural sparing on the
grounds that it would not have to store squander in landfills. Then again, there would be a further monetary sparing
in light of the fact that it would diminish the requirement for common aggregate [49].
Drying shrinkage and Autogenous shrinkage are one of most critical property of recycled aggregate concrete which
is associated with decrease in quantity or withdrawal of solidified concrete in light of loss of narrow dampness that
results in the change of fine pressure made inside the meo-pore structure of security system and it exhibit that the
durability properties of bond made with incredible nature of recycled aggregate, can be similar to those made with
NA [30]. Figure 5 shows Shrinkage behaviour of various types of waste aggregate. Brick wastes as RCA undergoes
maximum shrinkage.
Times in days
0
10 30 50 70 90
-100
Shrinkage(mm/m)
-200
-300
-400
-500
-600
RCA Brick Mortar Waste Ceramic
Fig. 5. Shrinkage behaviour of different types of recycling waste [71,73,78,80]
70
5.2 Creep
Creep is another factor influencing the auxiliary execution of cement. Creep are gotten by deducting the damaging
because of shrinkage and momentary detriment created by the pressure stack from the aggregate disfigurement. It
depends upon the mortar content, w/c ratio, recycled aggregate content etc [49,81-83].
Moreover, the substitution level of reused aggregate additionally influences the creep deformation. The particular
creep of reused concrete with a 20% substitution rate was observed to be 25% higher than that of the control
concrete, for 50% substitution level, the creep disfigurement was 29% high, for 100% substitution level the
expansion in the creep deformation was 32%. Substitution level of 100% shrinkage and a creep which were
extensively higher than those of traditional cement, being, separately, 70% and 51% higher for a time of 180 days
[83].
It has been observed that creep conduct is more impacted by the nearness of reused totals than shrinkage, in spite of
the fact that its varieties are fairly restricted contrasted with what happens in conventional concrete. The best creep
conduct has been resolved for CR100 as per its most astounding compressive strength and least pore measure
dissemination [81].
The utilization of fly ash as a limited substitution or expansion of cement could decrease the creep of concrete
because of the more noteworthy long-term strength improvement due to the pozzolanic response of fly fiery debris.
The pick-up in strength was significantly more noteworthy for concrete that contained fly slag. Since the quality
picks up for concrete arranged with fly ash remains was more noteworthy, the real stress/strength proportion was
bringing down contrasted with that for concrete arranged without fly powder amid which the creep test was
performed [84].
The lower estimated creep strain for concrete arranged with fly ash remains was subsequently credited to the lower
stress/strain proportion amid the time of the creep test [82].
6 CONCLUSION
Performance of recycled aggregate from various source like CDW, mortar, brick, ceramic and precast wastes are
compiled and analysed in this study. It is observed from the study that:-
1. Performance of all the construction wastes compiled in this study is suitable replacement of natural aggregate.
However, CDW needs suitable treatment before its use.
2. Ceramic tiles and precast wastes are better replacement as per strength is concerned even up to 100%
replacement.
3. The durability properties such as shrinkage & creep affect the performance according to the replacement of
RAC.
71
7 REFERENCES
[1] Chen HJ, Yen T, Chen KH (2003) Use of building rubbles as recycled aggregates. Cement and Concrete
Research 33: pp 125-132.
[2] Yang J, Du Q, Bao Y (2011) Concrete with recycled concrete aggregate and crushed clay bricks. Construction
and Building Materials 25: pp 1935–1945.
[3] Poon CS, Shui ZH, Lam L (2004) Effect of microstructure of ITZ on compressive strength of concrete prepared
with recycled aggregates. Construction and Building Materials 18: pp 461–468.
[4] Halicka A, Ogrodnik P, Zegardlo (2013) Using ceramic sanitary ware waste as concrete Aggregate.
Construction and Building Materials 48: pp 295–305.
[5] Gonçalves P, De BJ (2010) Recycled aggregate concrete (RAC) – comparative analysis of existing
specifications. Mag Concr Res 62: pp 339–46.
[6] Behera M, Bhattacharya SK, Minocha AK, Deoliya R, Maiti S (2014) Recycled aggregate from C&D waste &
its use in concrete – A breakthrough towards sustainability in construction sector.Construction and Building
Materials 68: pp 501-516.
[7] Oikonomou ND (2005) Recycled concrete aggregates. Cement Concrete Composite 27: pp 315–318.
[8] Brito DJ, Pereira AS, Correia JR (2005) Mechanical behaviour of non-structural concrete made with recycled
ceramic aggregates. Cement Concrete Composite 27: pp 429-433.
[9] Gomes M, Brito DJ (2009) Structural concrete with incorporation of coarse recycled concrete and ceramic
aggregates: durability performance. Mater. Struct. 42 (5) pp 663–675.
[10] Medina C, Sánchez De Rojas MI, Frías M (2012) Reuse of sanitary ceramic wastes as coarse aggregate in eco-
efficient concretes. Cement Concrete Composite 34 : pp 48–54.
[11] Medina C, Sánchez De Rojas M I, Frías M (2013) Properties of recycled ceramic aggregate concretes water
resistance. Cement Concrete Composite 40 : pp 21–29.
[12] Debieb F, Kenai S (2008) The use of coarse and fine crushed bricks as aggregate in concrete. Construction
[13] Dang J, Zhao J, Hu W, Du Z, Gao D (2018) Properties of mortar with waste clay bricks as fine
aggregate.Construction and Building Materials 166: pp 898–907.
[14] Martínez I , Etxeberria M, Pavón E , Díaz N (2013) A comparative analysis of the properties of recycled and
natural aggregate in masonry mortars.Construction and Building Materials 49: pp 384–392.
[15] Suryawanshi SR, Singh B, Bhargava P (2015) Characterization of RAC. Advances in Structural Engineering,
pp 1813–1822.
[16] Cachim PB (2009) Mechanical properties of brick aggregate concrete. Construction and Building Materials 23
: pp 1292–1297.
[17] Rashid K, Razzaq A, Ahmad M, Rashid T, Tariq S (2017) Experimental and analytical selection of
sustainable recycled concrete with ceramic waste aggregate. Construction and Building Materials 154: pp 829–840.
[18] Anderson DJ, Smith ST, Au TKF (2016) Mechanical properties of concrete utilising waste ceramic as coarse
aggregate. Construction and Building Materials 117: pp 20–28.
[19] C Thomas, J Setién, Polanco JA (2016) Structural recycled aggregate concrete made with precast wastes.
Construction and Building Materials 114 : pp 536–546.
[20] Matias D, Brito J, Rosa A, Pedro D (2013) Mechanical properties of concrete produced with recycled coarse
aggregates – Influence of the use of superplasticizers. Construction and Building Materials 44 : pp 101–109.
[21] Martínez-Lage I, Martínez-Abella F, Vázquez-Herrero C, Pérez-Ordóñez JL (2012) Properties of plain
concrete made with mixed recycled coarse aggregate. Construction and Building Materials 37 : pp 171–176.
[22] Corinaldesi V, Moriconi G, (2009) Behaviour of cementitious mortars containing different kinds of recycled
aggregate.Construction and Building Materials 23 : pp 289–294.
[23] Shahidan S, Azmi M, Kupusamy K, Mohd Zuki S, Ali N (2017) Utilizing Construction and Demolition
(C&D) Waste as Recycled Aggregates (RA) in Concrete. Procedia Engineering 174 : pp 1028 – 1035.
72
[24] Li J, Xiao H, Zhou Y (2009) Influence of coating recycled aggregate surface with pozzolanic powder on
properties of recycled aggregate concrete. Construction and Building Materials 23: pp 1287–1291.
[25] Medina C, Frías M, De Rojas MS (2012) Microstructure and properties of recycled concretes using ceramic
sanitary ware industry waste as coarse aggregate. Construction and Building Materials 31: pp 112–118.
[26] Suzuki M, Seddik Meddah M, Sato R (2009) Use of porous ceramic waste aggregates for internal curing of
high-performance concrete. Cement and Concrete Research 39: pp 373–381.
[27] Rahhal V, Irassar E, Castellano C, Pavslík Z, Černý R (2014) Utilization of ceramic wastes as replacement of
Portland cements. International Conference on Construction Materials and Structures (ICCMATS) 2014, pp. 208–
213.
[28] Zhang Z, Zhang Y, Yan C, Liu Y (2017) Influence of crushing index on properties of recycled aggregates
pervious concrete. Construction and Building Materials 135: pp 112–118.
[29] Jian-H X, Yong-chang G, Li-sha L, Zhi-hong X (2015) Compressive and flexural behaviours of a new steel-
fibre-reinforced recycled aggregate concrete with crumb rubber. Construction and Building Materials 79: pp 263–
272.
[30] Brito DJ, Ferreira J, Pacheco J, Soares D, Guerreiro M (2016) Structural, material, mechanical and durability
properties and behaviour of recycled aggregates concrete. Journal of Building Engineering 6 : pp 1–16.
[31] Soares D, Brito D J, Ferreira J, Pacheco J (2014) Use of coarse recycled aggregates from precast concrete
rejects: Mechanical and durability performance. Construction and Building Materials 71: pp 263–272.
[32] Yang S, Lee H (2017) Mechanical properties of recycled aggregate concrete proportioned with modified
equivalent mortar volume method for paving applications. Construction and Building Materials 136: pp 9–17.
[33] Waseem S, Singh B (2016) Shear transfer strength of normal and high-strength recycled aggregate concrete –
An experimental investigation. Construction and Building Materials 125: pp 29–40.
[34] Duan H, Poon C (2014) Properties of recycled aggregate concrete made with recycled aggregates with
different amounts of old adhered mortars. Materials and Design 58 : pp 19–29.
[35] Sunayana S, V. Barai S (2017) Recycled aggregate concrete incorporating fly ash: Comparative study on
particle packing and conventional method. Construction and Building Materials 156 : pp 376–386.
[36] Zhang J, Wang J, Li X, Zhou T, Guo Y (2018) Rapid-hardening controlled low strength materials made of
recycled fine aggregate from construction and demolition waste. Construction and Building Materials 173: pp 81–
89.
[37] Zheng C, Lou C, Du G, Li X, Liu Z,Li L (2018) Mechanical properties of recycled concrete with demolished
waste concrete aggregate and clay brick aggregate.Results in Physics 9 : pp 1317–1322.
[38] Nováková I, Mikulica K (2016) Properties of concrete with partial replacement of natural aggregate by
recycled concrete aggregates from precast production. Procedia Engineering 151: pp 360 – 367.
[39] Ay N, Unal M (2000) The use of ceramic tile in cement production. Cement and Concrete Research 30: pp
497–499.
[40] Puertas F, Garcia-DI, Barba A, Gazulla M, Palacios M, Gomez M, Martinez RS (2008) Ceramic wastes as
alternative raw materials for Portland cement clinker production. Cement Concrete Composites 30: pp 798–805.
[41] Zivica V R, Bajza A (2001) Acidic attack of cement based materials—a review: Principle of acidic attack.
[42] Ivica VZ (2006) Deterioration of cement-based materials due to the action of organic compounds.
[43] Oueslati O, Duchesne J (2012) The effect of SCMs and curing time on resistance of mortars subjected to
organic acids. Cement Concrete Composites 42 : pp 205–214.
[44] Belie ND, Lenehan J, Braam C, Svennerstedt B, Richardson M, Sonck B (2000) Durability of building
materials and components in the agricultural environment. J. Agric. Eng. Res. 76 : pp 3–16.
[45] Bakharev T, Sanjayan J, Cheng YB (2003) Resistance of alkali-activated slag concrete to acid attack. Cement
Concrete Composites 33: pp 1607–1611.
73
[46] Nepomuceno MCS, Isidoro RAS, Catarino J P G (2018) Mechanical performance evaluation of concrete
made with recycled ceramic coarse aggregates from industrial brick waste. Construction and Building Materials
165 :pp 284–294.
[47] Eduardo Bezerra Cabral A, Schalch V, Carpena Coitinho Dal Molin D, Luis Duarte Ribeiro J (2010)
Mechanical properties modeling of recycled aggregate concrete.Construction and Building Materials 24 : pp 421–
430.
[48] Awoyera PO (2016) Nonlinear finite element analysis of steel fibre reinforced concrete beam under static
loading. Journal of Engineering Science and Technology 11: pp 1669-1677.
[49] Pedro D, Brito J, Evangelista L (2017) Structural concrete with simultaneous incorporation of fine and coarse
recycled concrete aggregates: Mechanical, durability and long-term properties.Construction and Building Materials
154: pp 294–309.
[50] Rashid K, Razzaq A, Ahmad M, Rashid T, Tariq S (2017) Experimental and analytical selection of sustainable
recycled concrete with ceramic waste aggregate Construction and Building Materials 154 : pp 829–840.
[51] Aliabdo AA, Abd-Elmoaty AEM, Hassan HH (2014) Utilization of crushed clay brick in concrete industry.
Alexandria Engineering Journal 53: pp 151–168.
[52] Evangelista L, Brito DJ (2007) Mechanical behaviour of concrete made with fine recycled aggregate. Cement
Concrete Compos 29: pp 397–401.
[53] Sri R, Tam CT (1985) Properties of concrete made with crushed concrete as coarse aggregate. Mag. Concr.
Res. 37: pp 29-38.
[54] Medina C, Sa´ nchez de Rojas MI, As MF (2012) Reuse of sanitary ceramic wastes as coarse aggregate in eco-
efficient concretes. Cem. Concr. Comp. 34: pp 48–54.
[55] Senthamarai RM, Manoharan PD (2005) Concrete with ceramic waste aggregate. Cem. Concr. Comp. 27: pp
910–913.
[56] Osama MG (2011) Properties of hardened concrete using crushed clay brick as aggregates. J. Eng. 17 : pp
1022–1038.
[57] Fonteboa BG, Martı nez-Abella F (2005) Recycled aggregates concrete: aggregate and mix properties.
Mater.Constr. 55 : pp 53–66.
[58] Tabsh SW, Abdelfatah AS (2009) Influence of recycled concrete aggregates on strength properties of
concrete. Construction and Building Materials 23: pp 1163–1167.
[59] British Standards Institute, BS EN 12390-6:2009. Testing Hardened Concrete. Tensile Splitting Strength of
Test Specimens, BSI, London, 2009.
[60] Qasrawi H (2014) The use of steel slag aggregate to enhance the mechanical properties of recycled aggregate
concrete and retain the environment. Construction and Building Materials 54: pp 298–304.
[61] Silva R V, Brito DJ, Dhir RK (2015) Tensile strength behaviour of recycled aggregate concrete. Construction
[62] Kumar R (2017) Influence of recycled coarse aggregate derived from construction and demolition waste
(CDW) on abrasion resistance of pavement concrete. Construction and Building Materials 142: pp 248–255.
[63] Surya M, Rao K, Lakshmy P (2013) Recycled aggregate concrete for Transportation Infrastructure. Social and
Behavioral Sciences 104: pp 1158 – 1167.
[64] Pradhan S, Kumar S, Barai S (2017) Recycled aggregate concrete: Particle Packing Method (PPM) of mix
design approach. Construction and Building Materials 152 : pp 269–284.
[65] Ahmadi M, Farzin S, Hassani A, Motamedi M (2017) Mechanical properties of the concrete containing
recycled fibers and aggregates. Construction and Building Materials 144 : pp 392–398.
[66] Zheng L, Ge Z, Yao Z, Gao Z (2011) Mechanical Properties of Mortar with Recycled Clay-Brick-Powder.
ICCTP. pp 3379-3388.
[67] Kabir S, Al-Shayeb A, M. Khan I (2016) Recycled Construction Debris as Concrete Aggregate for
Sustainable Construction Materials.Procedia Engineering 145 : pp 1518 – 1525.
74
[68] Pacheco-Torgal F, Jalali S (2010) Reusing ceramic wastes in concrete.Construction and Building Materials
24: pp 832–838.
[69] Jackiewicz-Reka W, Załęgowskia K, Garbacza A, Bissonnetteb B (2015) Properties of cement mortars
modified with ceramic waste fillers.Procedia Engineering 108: pp 681 – 687.
[70] Kisku N,Joshi H,Ansari M,Panda SK, Nayak S, Dutta SCA (2017) A critical review and assessment for usage
of recycled aggregate as sustainable construction material.Construction and Building Materials 131: pp 721–740.
[71] Martínez I, Etxeberria M, Pavón E,Díaz N (2013) A comparative analysis of the properties of recycled and
natural aggregate in masonry mortars.Construction and Building Materials 49: pp 384–392.
[72] Soares D, Brito DJ, Ferreira J, Pacheco J (2014) Use of coarse recycled aggregates from precast concrete
rejects: Mechanical and durability performance.Construction and Building Materials 71: pp 263–272.
[73] Duan H, Poon C (2014) Properties of recycled aggregate concrete made with recycled aggregates with
different amounts of old adhered mortars. Materials and Design 58 : pp 19–29.
[74] Arivumangai, A, Felixkala T (2015) Fire Resistance Test on Granite Powder Concrete. International Journal
of Earth Sciences and Engineering 8: pp 311–316.
[75] Hoe Kwan W, Ramli M, Kam K, Sulieman M (2012) Influence of the amount of recycled coarse aggregate in
concrete design and durability properties. Construction and Building Materials 26: pp 565–573.
[76] Fonseca N, Brito DJ, Evangelista L (2011) The influence of curing conditions on the mechanical performance
of concrete made with recycled concrete waste. Cement Concrete Composite 33 : pp 637–643.
[77] Thomas C, Cimentada A, Polanco JA, Setién J, Méndez D, Rico J (2013) Influence of recycled aggregates
containing sulphur on properties of recycled aggregate mortar and concrete, Compos. Part B: Eng. 45: pp 474–485.
[78] Mansur MA, Wee TH, Cheran LS (1999) Crushed brick as coarse aggregate for concrete. ACI Mater 96: pp
478–84.
[79] Khatib JM (2005) Properties of concrete incorporating fine recycled aggregate. Cem Concr Res 35: pp 763–
769.
[80] El-Dieb AS, Kanaan D (2018) Ceramic waste powder an alternative cement replacement – Characterization
and evaluation.Sustainable Materials and Technologies 17: e00063.
[81] Kou SC, Poon CS (2012) Enhancing the durability properties of concrete prepared with coarse recycled
aggregate. Construction and Building Materials 35: pp 69–76.
[82] Domingo-Cabo A, Lázaro C, López-Gayarre F, Serrano-López MA, Serna P, Castaño-Tabares JO (2009)
Creep and shrinkage of recycled aggregate concrete. Construction and Building Materials 23: pp 2545–2553.
[83] Lakshmi S, Nivedhitha R (2015) Effect of partial replacement of aggregates by recycled concrete debris
on strength of concrete. Malaysian Journal of Civil Engineering 27: pp 250-259.
[84] Andal J, Shehata M, Zacarias P (2016) Properties of concrete containing recycled concrete aggregate of
preserved quality. Construction and Building Materials 125 : pp 842–855.
75
Paper ID- 31
Seismic Response Control of a Building Using Passive Hybrid Damper under

Near Field Earthquakes
Swabarna Roy1*, Swagato Das2, Dr. Purnachandra Saha3* (ORCID-0000-0002-2722-1669)
1
PhD Research Scholar, School of Civil Engineering, KIIT University, Bhubaneswar, Odisha, India,
E-mail: swabarnaroy0210@gmail.com
2
Assistant Professor, Dept. of Civil Engineering, C.V.Raman College of Engineering Bhubaneswar,
Odisha, India, E-mail: swagatodas83@gmail.com
3
Senior Associate Professor, School of Civil Engineering, KIIT University, Bhubaneswar, Odisha, India.
E-mail: dr.purnasaha@gmail.com * Corresponding Author
Abstract. In structural engineering aspect, the original damping of the building is not sufficient to
control the structural response due to occurrence of strong earthquakes. The common control strategy
used by researchers makes use of energy dissipating devices that absorb the energy imparted to the
structure due to earthquake and dissipate the energy by their hysteresis nature. In this study, seismic
response of a G+4 storey building installed with hybrid damper has been investigated. The hybrid damper
system comprises of Viscous Fluid Damper (VFD) and Shape Memory Alloy (SMA). The performance of
the building under near-field earthquakes, Tabas, Kobe and Gebze earthquakes, has been evaluated. The
dampers have been installed at the base of the super structure. SMA exhibits a good self- centering
capability and is known for its superior super elasticity properties. It is capable of dissipating energy
through its hysteresis nature while maintaining low level of residual displacement. VFD dissipates energy
by the fluid inside it which minimizes both stress and displacement the structure under goes during
seismic movement. The time history analysis results show the effectiveness of hybrid damper for
controlling the seismic forces acting on the structure. In this present study there is a significant reduction
in base shear and displacement by about 16% and 8% by using Hybrid damper when compared to VFD
and SMA dampers. A comparative study is made among the effect of the device in reducing the seismic
responses for the selected input near-field earthquake motion.
Keywords: Shape memory alloy, Viscous fluid damper, Hybrid damper, Near-field earthquakes, Passive
energy dissipation devices
1. INTRODUCTION
In recent years the use of base isolators for seismic protection of structures has attracted interest of many
researchers. During an earthquake, a finite quantity of energy in the form of vibration is stuck directly at
the base of the structure so the technique of isolating the base has been adopted to preserve the stability of
the structure. The flexibility of the connection or dampers between the substructure to superstructure
produces an enough counter response to the ground excitation remaining the structure sustainable. The
performance of the base isolation system depends upon the capacity by which the system fundamental
frequency can be shifted to a value lower than that of the un-modified structure and its capacity to
dissipate energy [1].
With the advancement in technologies there is an increasingly important role of passive energy
dissipation system for seismic protection of structures. The passive dampers when installed on a structure
76
functions by absorbing a part of the input seismic energy and hence, minimizing the amount of energy
needed to be dissipated by the primary structural members, thus reducing possible structural damages.
The passive dampers are mainly categorized into rate-independent devices and rate dependent devices [2].
The mechanical response of rate dependent devices depends on the relative velocity between the ends of
the device. For example, viscoelastic fluid dampers (VFD) are rate dependent devices which have
dynamic behavior characterized by their ability to lower stiffness values within a range of given
frequencies and have negligible influence on the fundamental natural frequency. Shape Memory Alloys
(SMA) dampers are rate-independent devices whose mechanical responses depend on displacement
occurring between both ends of the device and describes by non-linear hysteretic models. SMAs exhibit
several unique properties such as recovery to original shape after large deformation due to effect of
heating (shape memory effect) or due to the object being loaded (super-elasticity) [3]. Re-centering, high
damping capacity, minimal maintenance, high fatigue resistance and durability are the characteristics
which makes SMAs an effective damping device or a base isolator [4, 5]. In case of SMA a hysteretic
cycle is obtained which reduces the transmission of energy to the structure by the help of hysteretic cycle,
which dissipates energy within its own area [6].
In this study, a G+4 storey building has been modeled and the performance of the building under near-
field earthquakes taking into consideration the Tabas earthquake, Kobe earthquake and Gebze earthquake.
The present study aims to compare the performance of isolation system when the individual damper VFD
and SMA and Hybrid damper combination of both are installed at the base of the structure. Generally
near-field earthquake causes higher structural damage but its acceleration can be effectively reduced by
installation of Hybrid damper which shows good dampening performance as compared to that of
individual damper. The time-history analysis results for both the uncontrolled and controlled structure
under the selected ground excitations shows that hybrid damper is better in mitigating floor displacement
and base shear compared to other two dampers.
2. NEAR-FIELD EARTHQUAKES
The earthquakes occurring on earth surface close to the fault are referred to as near-field earthquakes. The
distance ranges from 10-60 km around the fault [6]. These earthquakes have long-pulse periods and high
accelerations. During fault rupture a part of the wave is transferred to the site location and if the site is in
the same direction, they get closer to each other thus generating a large pulse which dampens with time
length during which waves gets to the structure. But in case of the site being located in the opposite
direction to that of the occurred fault, the distance between the site and the waves increases incrementally
and hence longer time is required for the waves to reach the structure [7]. Due to this long period of
oscillation, which may sometimes be to the close to the natural oscillation, the structure tends towards
resonance. This condition causes more structural damages and more fragile behaviour of the structure.
Table 1 below show details of some near-fault earthquakes considered in the present research study.
Table 1. Details of the Near-Fault Earthquakes considered for study
Earthquake Distance (km) PGA (g) Magnitude Year

Tabas (Boshrooyeh Station) 17 .8 7.4 1978
Kobe (Kakogawa Station) 22.5 .8 6.9 1995
77
Gebze (Yarimca Station) 22.7 .3 -.4 7.4 1999
3. SHAPE MEMORY ALLOY (SMA)
Seismic Isolation system using Shape Memory Alloy or SMA has attracted good attention as a smart
material and has been used effectively in passive protection of structures [8]. Shape memory effect and
super-elasticity are two unique properties of SMA, in which the former refers to the property by which the
SMA retains the predefined shape and returns back to it when agitated and the later refers to the property
by which SMA undergoes large inelastic deformations and recovers its own shape upon unloading [9].
The schematic diagram of a SMA damper is shown in Figure 1. When under excitation, the SMA
dampers generate a control force which makes the main structure safe against earthquake. The schematic
diagram of working of SMA damper is shown in Figure. 2. The SMA dampers follow the passive control
concept by which it absorbs major part of input energy due to seismic activity, or undergoes major
damage, keeping the structural members in the elastic limit or makes sure that the structure undergoes
minimal damage.
(a) Cross Sectional View (b) Re-Centering Group
(c) Energy Dissipating Group
Figure.1. Schematic detailed figure of SMA Damper showing (1) Internal Shaft, 2) Shim Plate, 3) Pre- Compressed Spring, 4)
External Tube, 5) Middle Anchor, 6) Side Anchor, 8) SMA wires) [11]
Figure.2. Schematic diagram of performance evaluation passive structure [11]
78
No external power source is required for a passive control system and structural control using SMAs
make use of its damping property to reduce the structural response and subsequent structural damage
subject to severe vibration or loading [9]. The hysteresis loop for the SMA Damper is shown in Figure 3.
The hysteresis loop for energy dissipation and re-centering has been shown in Figs. 3a and 3b
respectively. The combination of the behavior of both these functional groups represents the hysteresis
loop for SMA damper which is represented in Fig 3c. Thus, the graph shows both self-centering
capability and maximum energy dissipating capacity. As discussed, SMA damper, being highly nonlinear
material, is difficult to model based on some fixed rules. However, mathematical model for hysteresis
loop of SMA damper is effective for a damper control structure [11].
(a)Energy Dissipating Part (b) Re-Centering Part (c) SMA Damper

Figure.3. Hysteresis loop for SMA Damper as presented using Bouc-Wen model [11]
The energy dissipating device restoring force is shown as [12]:
(1)
where x is displacement or the elastic part, α indicates ratio between post-yielding to pre-yielding
stiffness, K0 is linear stiffness, and Zs is the hysteresis part of the isolator displacement. The maximum
displacement in austenite phase is denoted by a and b is the displacement responsible for martensitic
transformation. In order to linearize the complex nonlinear equation of hysteresis curve, a linearized
equation is introduced which minimizes the residual error of the non-linear and linear terms of governing
equation of motion. The simplified linearized version of Zs is given as Equation. 2 [10, 13]:
(2)
where sgn(xs) is signum function:
4. VISCOUS FLUID DAMPER (VFD)
The implementation of Viscous fluid damper (VFD) has been recently adapted for seismic protection of
buildings. A Viscous fluid damper comprises of piston within a damper housing which is filled with
silicon oil acting as a fluid [2]. The piston consists of a numerous small orifices through which the
movement of fluid occurs from one part of the piston to the other part. The schematic cross-sectional
figure of VFD is shown in figure 4. When the piston rod is stroked, the fluid is pressurized to flow
through the orifices. This creates a pressure differential across the piston head which produces very large
79
forces thus resisting the relative motion of the damper [14,15]. The fluid flows at a higher velocity thus
resulting in generation of friction force between fluid particles and piston head. This force is responsible
for energy dissipation in the form of heat. This damping force reduces both stress and displacement in the
structure. In a viscous fluid damper, the resistive control force is :
=C | | s n( ) (3)
In the above equation, f indicates damping force, Cd indicates damping coefficient with units of force per
velocity, indicates velocity across both the ends of damper, indicates velocity exponent whose value
ranges between 0.4 to 1.4. The force displacement relation of VFD is shown in Figure.5. Performance of
passive VFD depends upon its velocity exponent ( ) and damping coefficient (Cd ).
Fig. 4. Schematic cross-sectional figure of VFD [14]
Figure. 5 Force Displacement relation incase of VFD[14]
5. NUMERICAL STUDY
A G+4 model structure was analyzed using MATLAB [16] for the three earthquakes to study the
performance of hybrid control system. The building is a 3 * 3 bay-framed structure with ground level
height 5.2m, storey height 3.1m and bay distance 5m as shown in Figure 6. The time history analysis is
performed using the State Space Method on the equations of motion in order to obtain structural response
of the building [17].
80
Figure.6. The Basic Structural Frame of G+4 building
The equation of motion for SDOF system may be written as (Equation 4) [18,19]
MẌ(t) + CẊ(t) +KX(t) = -M g+ FE(t) (4)
where, FE(t) the external force, M indicates the mass matrix, C indicates the damping ratio matrix, K
indicates the stiffness matrix, is ground acceleration, X(t), Ẋ(t) and Ẍ(t) are displacement, velocity and
acceleration respectively.
State Space Equation has been employed for smart base isolation structural problem and is given by
Equation 5.
(5)
where u(t) =fd(t) is the time varying control device force, x consists of the states, and A, B and E are the
system matrices defined as:
, ,
where and are the location of device and earthquake influence vector respectively.
Time history analysis is performed using near-field earthquakes taking into consideration the three
earthquake ground motions namely Tabas (1978), Kobe (1995) and Gebze (1999) earthquake to obtain
the uncontrolled and controlled structural responses of the building. The time-history of the earthquakes
for a time period of 60 seconds are shown in Figure 7. Passive damper SMA and VFD and Hybrid are
installed at the base of the structure. The effectiveness of the Hybrid Damper is investigated against
individual dampers. Time history analysis is performed for the uncontrolled and controlled responses
under the earthquakes. The lumped masses (mi), storey stiffness (ki), and damping ratio (ξi) for every
storeys of the superstructure have been considered identical for the sake of simplicity. The mass ratio (i.e.
ratio between isolator mass to the total superstructure-isolation system mass) is taken as 2%. The
superstructure damping (ξ) assumed is of viscous type and taken as 5% [20]. The time period for the
81
damper was taken as 2s. The value of Cd considered for VFD damper is 1500. The parameters considered
for SMA damper force-deformation hysteresis loop has been shown in Table 2 [21].
.
Table.2. Structural parameters and SMA parameters considered for design
Structural Parameters SMA Parameters
T = 0.55s Tb = 2s
ξ = 2% ξb = 5%
αs = 0.10
a =0.005
b = 0.05
Tabas (1978)
100
Acceleration (cm/s2)
50
0
0 10 20 30 40 50 60
-50
-100
Time(s)
Kobe (1995)
10
0
0 10 20 30 40 50 60
-5
-10
Time(s)
82
Gebze (1999)
200
100
0
-100 0 10 20 30 40 50 60
-200
-300
-400
Time (s)
Figure. 7. Time History of Tabas (1978), Kobe (1995), Gebze (1999) respectively.
6. RESULTS AND DISCUSSIONS
The time history analysis has been performed in order to calculate the structural responses of the structure
for the three specified near-field earthquake ground motions. The three specified near-field earthquakes
are: (1) 1978 Tabas (Boshrooyeh station) with PGA (peak ground acceleration) equals to 0.8g having a
hypocentral distance of 17km; (2) 1995 Kobe (Kakogawa station) with PGA value equals to 0.8g and
hypocentral distance 22.5km; (3) 1999 Gebze (Yarimca station) with PGA value equals to 0.3-0.4 and the
hypocentral distance of 22.7km.
From the earthquake data a comparison has been carried out between base shear and displacement for
uncontrolled structure and structure fitted with dampers. The time variation of the displacement response
of the earthquakes for a damping ratio of 1.5% for VFD and 2% for SMA and Hybrid damper
combination of SMA and VFD are shown in Figure 8. In this figure, the displacement response of the
structure has been compared for both uncontrolled structure and structure individually incorporated with
SMA, VFD and next by Hybrid of SMA and VFD. From the figures it can be observed that when the
model was applied to Tabas earthquake, the reduction in displacement by Hybrid damper is 4% more than
that of VFD and 8% more than that of SMA. Then in case of Kobe earthquake acceleration, the reduced
value of displacement by Hybrid damper are about 2% and 7% more than that of VFD and SMA. While
considering the Gebze earthquake acceleration datas, the Hybrid damper reduces displacement by about
4% more than of VFD and 7 % more than of SMA. Thus, the control system has helped to reduce the
structural displacement thus preventing damage to the structure.
Uncontrolled VFD SMA Hybrid

0.3
0.2
Displacement(m)
0.1
0
0 10 20 30 40 50 60
-0.1
-0.2
-0.3
Time(s) (a)
83
0.2 Uncontrolled VFD SMA Hybrid
0.1
Displacement(m) 0
0 10 20 30 40 50 60
-0.1
-0.2
Time(s) (b)

0.6
0.4
Displacement (m)
0.2
0
-0.2 0 10 20 30 40 50 60
-0.4
-0.6
Time (s)
Fig. 8 Uncontrolled and displacement controlled by the dampers for (a) Tabas (1978), (b) Kobe (1995), (c) Gebze
Earthquake (1999).
The base shear of the structure has been compared for the uncontrolled and SMA, VFD and Hybrid
damper incorporated structure and shown in Figure. 9. From the figure, it can be observed that the model
when incorporated with Hybrid damper under Tabas earthquake motion datas reduces the base shear
response by 11% more than that of VFD and 16% more than that of SMA. In case of Kobe earthquake
acceleration, the Hybrid damper reduces the base shear responses by about 10% and 13% more than that
of VFD and SMA. Further, under Gebze earthquake motion the Hybrid damper is effective in reducing
the base shear response by 4% more than VFD and 7% more than SMA. Thus, the hybrid damper is
successful in absorbing the input seismic energy and dissipating the same more effectively than VFD and
SMA, thus protecting the main structure from damage. As a whole, considering all the three earthquake
motions, we can say that maximum reduction in displacement and base shear using Hybrid damper is
obtained as 8 % and 16% respectively when compared to VFD and SMA. With the application of Hybrid
damper both the displacement and base shear has been controlled and the damage to the structure due to
long duration near-field earthquakes can be prevented. Thus, the structural responses of the structure have
been reduced, hence proving the effectiveness of the Hybrid damper.

150
Base Shear(kN)
100
50
0
-50 0 10 20 30 40 50 60
-100
-150
Time(s) (a)
84

150
Base Shear (kN)

100
50
0
-50 0 10 20 30 40 50 60
-100
-150
Time(s) (b)
40
20
Base Shear (kN)
0
0 10 20 30 40 50 60
-20
-40
Time (s)
Fig. 9 Uncontrolled and controlled base shear responses by the dampers for (a) Tabas (1978), (b) Kobe (1995), (c) Gebze
Earthquake (1999).
7. CONCLUSION
In this paper, a G+4 building has been modeled and an attempt is made to compare the effectiveness of
Hybrid damper. The performance of the building is under different near-field earthquake are investigated
and the results are graphically represented. Based on the investigation performance of Hybrid damper is
found to be better than individual VFD and SMA performance in regard to seismic response control of the
building. The hybrid damper is efficient as it is a combination of energy dissipation property from VFD
and re-centering property from SMA. The following conclusions are drawn from the research paper: -
1. With the application of Hybrid damper, the G+4 structure was seismically controlled and hence
damage may be reduced.
2. Despite the structure being a flexible one, the reduction in response of seismic vibration of building
can be obtained by installation of VFD, SMA individually and Hybrid damper in the base of the
building.
3. Reduction in seismic response displacement of the building using Hybrid Damper is obtained as
about 4% more than that of VFD alone and 8% more than that of SMA alone.
4. The reduction in the shear response of base using hybrid damper has been observed to be about 11%
more than that of VFD and 16% more than that of SMA for all the three earthquakes.
REFERENCES
[1] Ibrahim, R.A.: Recent advances in nonlinear passive vibration isolators. Journal of Sound and
Vibration, vol. 314, pp. 371-452 (2008). doi: 10.1016/j.jsv.2008.01.014
85
[2] Castaldo, P.: Passive energy dissipation devices. Integrated seismic design of structure and control
systems, pp. 21-62. Springer, Switzerland (2014). doi: 10.1007/978-3-319-02615-2_2
[3] Sepulveda, J., Boroschek, R., Herrara, R.: Steel beam-column connection using copper-based
shape memory alloy dampers. Journal of Constructional Steel Research, vol. 64, pp. 429-435
(2008). doi: 10.1016/j.jcsr.2007.09.002
[4] Motahari, S. A., Ghassemieh, M., and Abolmaali, S.A.: Implementation of shape memory alloys
dampers for passive control of structures subjected to seismic excietations. Journal of
Constructional Steel Research,vol.63, pp.1570-1579(2007).doi: 10.1016/j.jcsr.2007.02.001
[5] Des Roches, R, Mc Cormick, J., Delemont MA.: Cyclical properties of superelastic shape memory
alloys. ASCE Journal of Structural Engineering, vol. 130(1), pp. 38-46 (2004).
[6] Heydari, M and Mousavi, M.: The Comparison of seismic effects of near-field and far-field
earthquaks on relative displacement of seven-storey concrete building with shear wall. An
International Research Journal of Environmental Science, Iran.
http://dx.doi.org/10.12944/CWE.10.Special-Issue1.07
[7] Taheri, J.S., and Anderson, J.G.: The 1978 Tabas, Iran earthquake: An interpretation of the strong
motion records Bulletin of the Seismological Society of America, Vol. 78(1), pp. 142-171, (1988).
[8] Ozbulut, E., and Hurlebaus.S.: Evaluation of the performance of a sliding-type base isolation
system with a NiTi shape memory alloy device considering temperature effects. Engineering
Structures, vol. 32, pp. 238-249 (2010).doi: 10.1016/j.engstruct.2009.09.010
[9] Song, G., and Li,H.N.: Applications of shape memory alloys in civil structures. Engineering
Structures, vol.28, pp. 1266-1274 (2006).
[10] Shinozuka, M., Chaudhuri S.R., Mishra S.M.: Song, G., and Li,H.N. (2006). "Applications of
shape memory alloys in civil structures." Engineering Structures 28: 1266-1274.Shape-Memory-
Alloy supplemented Lead Rubber Bearing (SMA-LRB) for seismic isolation. Journal of Sound and
Vibration, Vol. 41, pp. 34-45 (2015).
[11] Ma H., Yam M.C.H.: Modelling of a self-centring damper and its application in structural control.
Elsevier, Journal of Constructional Steel Research, vol.67, pp. 656-666 (2011).
[12] Ikhouane F., Rodellar J.: Systems with hysteresis: analysis, identification and control using the
Bouc–Wen model. John Wiley & Sons: Chichester (England, Hoboken, NJ), (2007).
[13] Yan X., Nie J.: Response of SMA super elastic systems under random excitation. Journal of
Sound and Vibration, vol. 238(5), pp. 893-901 (2000).
[14] Kumar, P. S., Naidu, M.V., Mohan, S.M., and Dr.Reddy, S.S.: Application of Fluid Viscous
Dampers In Multi-Story Buildings. International Journal of Innovative Research in Science,
Engineering and Technology, vol.5(9), pp. 17064-17069 (2016).
[15] Marko, J., Thambiratnam, D., and Perera, N.: Study Of Viscoelastic And Friction Damper
Configurations In The Seismic Mitigation Of Medium-Rise Structures. Journal of Mechanics of
Materials and Structures, vol.1(6), pp.1001-1039 (2006).
86
[16] MATLAB. The Math Works Inc. 2010. Natick Massachusetts.
[17] Saha, P., and Jangid, R.S.: Seismic control of benchmark cable stayed bridge using passive hybrid
systems. The IES Journal Part A Civil and Structural Engineering, vol. 2 (1), pp 1-16.
[18] Chopra, A.K. “Dynamics of structures –Theory and Applications to earthquake engineering”,
prentice hall.
[19] Elnashai, A.S., and Sarno, L.D. “Fundamentals of earthquake engineering” A john wiley & sons
ltd,2008.
[20] Soong TT, Spencer B F.: Active, Semi-Active and Hybrid Control of Structures.12 WCEE
(2000), pp.1-16.
[21] Shinozuka, M., Chaudhuri, S.R., Mishra, S.M., Shape-Memory-Alloy supplemented Lead Rubber
Bearing (SMA-LRB) for seismic isolation. Probabilistic Engineering Mechanics, Vol.41, pp. 34-45
(2015).
87
Paper ID- 32
Structural Modification of a 65 Degree Delta

Wing using a Step-Wise Discontinuity
Nishit Sanil ,Raza Khan and Carine D’Souza
PES University, Mechanical Department,

Bengaluru-560085,India
https://pes.edu
Abstract. Increasing lift effectively at higher angles of attack has al-

ways been a daunting challenge in aviation especially on a delta wing.
They have some undesirable characteristics, notably flow separation at
high angles of attack and high drag at low speeds. In order to solve this
problem a design modification in the form of a step-wise discontinuity
is modeled at the upper surface of the delta wing to attain an increase
in the lift of a 650 delta wing at all angles of attack ranging from 00 to
400 . A conventional delta wing is validated for comparison which would
thereby give us a measure of flow separation and the coefficient of lift
affected by the modification. The results obtained deliver an overall in-
crease of coefficient of lift by 6.32% thereby delaying stall. Hence the
benefits of the modification would aid the potential designs of aircrafts
in the time to come.
Keywords: Angle of Attack, coefficient of lift ,flow separation, step-wise

discontinuity
1 Introduction
A delta wing is a wing whose plan form is shaped in the form of a delta (Greek
symbol ). It sweeps sharply back from the fuselage with the angle between the
leading edge (the front) of the wing often as high as 600 and the angle between
the fuselage and the trailing edge of the wing at around 900 .Earlier combat jet
designs involved straight wings which were not able to provide sufficient lift at
higher speeds and elevations which paved the way for a large delta wing. Al-
though the delta wings are crucial to achieving high lift at supersonic velocities,
they also have number of disadvantages when it comes to the amount of drag
produced to keep the plane level rendering it unstable at high angles of attack
and require high landing and takeoff speeds. Currently, small horizontal fins are
mounted on the fuselage ahead of an aircrafts main wings to offer better control
which when incorporated into a delta wing aircraft improves its stability and
maneuverability.The main advantage observed on a pure delta wing due to its
long root chord and short span are that it results in a lighter wing compared to
a swept wing of the same lifting capacity while also being stronger and stiffer.
It has also been observed that at low speeds an aircraft with a delta wing must
88
2 National Conference on Advances in Structural Technologies
maintain high angles of attack to generate lift. The main purpose of a delta wing
is to provide lift at high speed flow regimes but the aircraft must still take off and
land at lower speeds thus it is ideal to have a wing with good aerodynamic char-
acteristics at all flow regimes. The lift generated by a delta wing at low speeds
can be understood by looking into two types of lift generating phenomenon.
1.Potential Flow lift
2.Vortex flow lift
1.1 Step-Wise Discontinuity

Demeter G Fertis in his book Nonlinear mechanics explored the possibility of an
airfoil design modification which improves the lift and increases the stall angle
by delaying flow separation to allow higher angle of attack. In existing designs
movable flap or slats on the leading and trailing edges of the wing are usually
introduced in order to improve the overall aerodynamic characteristics of the
airplane wing. The disadvantage due to the slats and flaps is an increase in drag
which makes it applicable only for relatively low air speeds and therefore are un-
able to improve lift and stall characteristics of the wing at high cruising speeds.
The purpose of the experiment conducted by Demeter G Fertis was to introduce
an airfoil design concept with improved aerodynamic characteristics by the in-
troduction of a step-wise aerodynamic discontinuity on the upper surface of the
airfoil. The benefits from this step-wise discontinuity is to improve lift to drag
ratio while simultaneously retaining some of the important functional character-
istics of the existing shape of the airfoil. As an added advantage the turbulence
introduced by the discontinuity will either eliminate stall for angles of attack up
to at least 400 and will provide a much smoother stall and thus smoother flow
separation for greater angles of attack.The disadvantage due to the slats and
flaps is an increase in drag which makes it applicable only for relatively low air
speeds and therefore are unable to improve lift and stall characteristics of the
wing at high cruising speeds.
1.2 Kline-Fogleman Design
Richard Kline and Floyd Fogleman designed an airfoil in 1960 which had some
single or multiple steps along the length of the wing. The purpose of the step
was to make a paper airplane that could handle strong winds, climb high, level
off by itself and then enter a long downwards glide. The modification allows
some of the displaced air to fall into a pocket behind the step and become part
of the airfoil shape as a trapped vortex or vortex attachment. This purportedly
prevents separation and maintains airflow over the surface of the airfoil.
89
National Conference on Advances in Structural Technologies 3
2 Methodology
This project will investigate the improvement in aerodynamic characteristics on
a 650 delta wing with a step-wise discontinuity inspired by the book Non-Linear
Mechanics by Demeter G Fertis. The modification will involve a cut section made
on the upper surface of the wing leading to the formation of a turbulent boundary
layer that will delay flow separation as it does on a regular airfoil. A wing
without the modification will be tested and compared with the modified wing to
show the improvement. The numerical analysis will be done on a computational
fluid dynamics software called Ansys Fluent 18.2 Experimental and Numerical
Investigation of 650 Delta and 650 /400 Double-Delta Wings (Journal of Aircraft,
Vol.45, No.1,2008) by Ahmad Z. Al-Garni et al The research conducted in King
Fahd University in Riyadh studied the lift generated on a 650 delta wing and
a 650 /400 double delta wing at different angles of attack. They conducted a
detailed study of literature by Earnshaw and Lawford (1961), Polhamus (1971)
and by Wentz and Kohlman (1971) which calculated the coefficient of lift at
various angles of attack among other aerodynamic characteristics. The aim of
their research was to conduct an experimental verification of literature and then
compare it with results obtained from CFD analysis of the same wings. They
started by setting up of the wind tunnel using the inhouse wind tunnel facility
of King Fahd University. A low speed blowdown wind tunnel of open return
type was available. It had a maximum free stream velocity of 35 m/s and 1%
turbulence level operated by a centrifugal blower. The temperature was kept at
300K and operated at atmospheric pressure. It covered angles of attack ranging
from 00 to 400 To begin the procedure of modifying the delta wing, first an
existing delta wing having a basic functional design must be validated on the
same software and results compared to that of an existing simulation to serve
as a basis for applying the modification and comparison of the improvements.
The objective is to create a method for analysing different types of wings by
following the set of steps in order to achieve a positive result in terms of the
lift coefficient and thereby proving that the concept works. Several research
papers from various journals were investigated and the paper by Ahmed Z.Al
Garni et al titled Experimental and Numerical Investigation of 650 Delta and
650 /400 Double-Delta Wings published in the AIAA journal of aircraft in 2008
was selected for validation and comparison.
90
Model Design
The wing used in the paper was a 650 leading edge sweep delta wing with a root
chord length of 30cm, a wing span of 28cm, a bevel angle of 8.50 and a thickness
of 1cm. The wing area was 0.04197 with an aspect ratio of 1.865.These dimen-
sions were chosen to allow the model of the wing to be housed in the wind tunnel
facility available in King Fahd University where this research was conducted. For
this project a wing of the same dimensions was modeled using SOLIDWORKS
and SolidEdge.
Geometric Configurations
The configurations have been modelled based on the findings of Demeter which
have been to create a step-wise discontinuity. The step-wise discontinuity was
created on the upper surface of the Delta wing because the discontinuity creates
a turbulent boundary layer which has a small thickness compared to laminar
boundary layer which results in high wall shear stress which prevents flow rever-
sal due to adverse pressure gradient resulting in smoother flow separation ergo
delaying stall.
The discontinuity depends on these three factors:
Chord length:
The length of the root chord from the tip of the delta wing is varied in percent-
ages ranging from 30% to 60% of the total length of the delta wing.
Depth of the Step:
After the chord length of the step is fixed, the depth of the step is similarly
varied in steps ranging from 30% to 50%. The depth of the step isnt varied more
than fifty percent to maintain the structural integrity of the model.
Step Length:
The length of the step is reliant on the chord length and correspondingly varied
in a similar manner as mentioned above.
91
2.1 Modification Types
Demeter Type.:
This type of modification features the step-wise discontinuity in the form of a
tapered cut across the cross-sectional model of the delta wing.
Rectangular Type.:
This type of modification features the step-wise discontinuity in the form of a
rectangular cut across the cross-sectional model of the delta wing.
Kline-Fogleman Type.:
This type of modification features a step-wise discontinuity based on the KF
theory across the cross-sectional model of the delta wing.
2.2 Models
MDW 444:
This modification of the delta wing has the following specifications:
Chord length = 40% of the total length of the wing (12 cm)
Depth = 40% below the upper surface of the wing (0.4 cm)
Step length = 40% of the length considered after the chord length (12 cm)
MDW 453:
MDW 345:
MDW 362:
MDW 353:
MDW 535:
MDW 463:
92
Depth = 40% below the upper surface of the wing (.4 cm)
MDW 443:
MDWr 54:
MDWr 64:
KF 2102:
Depth1 = 20% below the upper surface of the wing (0.2 cm)
Step length1 = 3.33% of the length considered after the chord length (1 cm)
Step length2 =33.3% of the length considered after the chord length (10 cm)
KF 222:
Step length1 = 6.67% of the length considered after the chord length (2 cm)
Step length2 =33.3% of the length considered after the chord length (10 cm)
KF 2752:
Step length1 = 25% of the length considered after the chord length (7.5 cm)
93
MDW 444 MDW 453
MDW 345 MDW 362
MDW 353 MDW 535
MDW 463 MDW 443
MDWr 54 MDWr 64
KF 2102 KF 222
KF 2752
94
2.3 Mesh
The mesh for the wing model in the research paper by Ahmed Z.Al Garni was
created using a software called GAMBIT. They modeled and meshed half of the
delta wing to reduce computational time.
Initially they constructed a mesh consisting of tetrahedral cells using the size
function available in GAMBIT. Two size functions were used, one to capture
the effects near the wall and the other to minimize the mesh density in the rest
of the enclosure. The one equation Spallart-Allamaras turbulence model with
vorticity-based production was used. Due to the lack of computational power to
use the above mesh along with the required turbulence model a mesh of hexa-
hedral cells was used which consisted of 1,230,968 cells.
Numerous turbulence models were tested for this type of mesh namely k-, k-,
SST k- , SST k-, and of which SST k- yielded the best correlation between ex-
perimental and CFD analysis.
Since over 20 modifications of the wing had to be modeled, meshed and analyzed
the above method was not feasible as it would require more time and computa-
tional power. To be able to test numerous modifications in the required time a
simpler mesh was used and fine-tuned to achieve similar results.
The mesh created used hex dominant cells using the hex dominant method
provided in Ansys meshing. A sizing function was applied near the walls to ac-
curately capture the effects on the wing.
The sizing mesh was given an element size of 0.00175m and the rest of the en-
closure was assigned an element size of 0.015m. This allowed the mesh to have
a gradient which would allow a higher density of cells near the wing and lower
density for the enclosure away from the wing. This mesh resulted in approxi-
mately 1.2 million cells which was close to the number of cells achieved by the
research paper.
Despite using this type of mesh, the procedure is very tedious so a simpler mesh
with the same parameters was created but with an enclosure element size of
0.05m which resulted in a mesh of approximately 250,000 cells and achieved
similar results.
3 Results and Discussion

The results have been classified on the type of modification incorporated with
the 650 Delta wing. All the results obtained are plots of coefficient of lift vs
angle of attack. Each configuration has been tested for angles of attack ranging
from 00 to 400 .The corresponding values of coefficient of lift are obtained by the
numerical analysis. The results of the validation are tabulated and compared
for its deviation from the paper. Additionally, the results for the modified delta
wing are tabulated and classified based on the type of modification.
3.1 Validation Case Study

The validation was conducted based on the paper written by Ahmad Z. Al-Garni
et al. The results were compared based on the on the number of mesh elements
95
and the plot of the lift coefficient vs the angle of attack.

The mesh obtained during the conducted analysis produced 1.17 million ele-
ments compared to that of the expected 1.23 million elements in the paper. The
results of the validation were successful with an overall deviation of 5% from the
expected results.
Fig. 2: Validation Data
96
Fig. 3: Graph 1: CL vs AOA
97
3.2 Modification Results
The results shown below are of 8 configurations of the Demeter type. Each graph
is a plot of coefficient of lift vs angle of attack ranging from 00 to 400 .Each
configuration is compared to a regular delta wing and the results are tabulated
with colour coding which is followed in a similar manner in the rectangular and
Kline-Fogleman configurations.
Blue denotes the angle of attack
Golden denotes the coefficient of lift for a regular delta wing
Red indicates a decrease in the coefficient of lift whereas Green indicates the
increase in coefficient of lift.
Fig. 4: CL of the Modified Models
3.3 Pressure Contours
The pressure contours signify the turbulent intensity across the region of the
wall. The delta wing has a region of high pressure at the leading edge and low
pressure at the upper surface since at the leading edge a stagnation point is
created which reduces the velocity of the flow to nearly zero. Over the upper
surface of the delta wing the flow accelerates which depicts the change in colour.
The step-wise discontinuity creates a pocket because some of the displaced air
falls into the step which creates a trapped vortex which is shown in a light green
hue of the contour. The yellow hue which appears to be sticking to the upper
surface of the delta wing are an indication of low pressure whereas the amber
hue at the lower surface of the delta wing indicates a region of high pressure.
98
contour kf.jpeg contour kf.jpeg
Fig. 5: Pressure Contour for KF type
4 Conclusion and Future Scope
The configurations that were tested showcased higher lift than a regular delta
wing at high angles of attack initially. This is due to the modification that in-
corporates Demeters concept which creates a trapped vortex at the step which
in turn causes an increase in lift.
The rectangular type of modification showed an increase in lift at low angles of
attack while the lift at high angles decreased and the KF step configurations
showed similar results.
Based on this trend the best option was incorporate two types of modifications
into the wing. Thereby using the rectangular and KF modification the modifi-
cation KF 2752 was able to achieve the best results as it provides greater lift
at all angles of attack, delayed flow separation resulting in smoother stall at the
operating conditions that were tested.
Although the overall lift to drag ratio increases, the lift is higher at low angles
of attack while there exists a lower lift to drag ratio at higher angles of attack
because of the increase in the amount of drag produced.
References
1. Fertis, D. G., 1999. New Airfoil Design Concept with Improved Aerodynamic Char-
acteristics. In: Non-Linear Mechanics Press, p. 534.
99
2. Wilcox, D. C., Turbulence Modeling for CFD, DCW Industries, La Canada, Cali-
fornia, 1998.
3. Jr., J. A., 2017. Introduction to Flight. 6th ed. McGraw-Hill Education.
4. Al-Garni, A. Z., Saeed, F. Al-Garni, A. M., 2008. Experimental and Numerical
Investigation of 65 Degree Delta and 65/40 Degree Double Delta. Journal of Aircraft,
45(1), p. 6.
5. Unoguchi, T. et al., 2012. Aerodynamic Characteristics of a Delta Wing with Arc
Camber for. JAXA, pp. 343-351
6. Polhamus, E., A Concept of the Vertex Lift of Sharp-Edge Delta Wings Based on
a Leading-Edge Suction Analogy, NASA TN D-3767, Dec. 1996.
7. Polhamus, E. C., Predictions of Vortex-Lift Characteristics by a Leading-Edge Suc-
tion Analogy, Journal of Aircraft, Vol. 8, No. 4, April 1971, pp. 193199.
8. Hsu, C.-H., and Liu, C. H., NavierStokes Computation of Flow Around a Round-
Edged Double-Delta Wing, AIAA Journal, Vol. 28, No. 6, June 1990, pp. 961968.
9. https://www.cfd-online.com
100
Paper ID- 34
SOCIO-ECONOMIC EFFECT OF EARTHQUAKE: A CASE

STUDY OF BHUJ EARTHQUAKE 26th JAN 2001
Anusha Rani, Assistant Professor, Civil &Environmental Engineering, BIT MESRA, Ranchi
ABSTRACT
India has a history of devastating earthquakes, since medieval time many earthquakes had devastated many
cities of India. Before 1900, instrumentation had not done for seismically active region so we do not have
record of earlier earthquakes. After 1900 India has suffered 16 major earthquakes and it has found that about
60% of country is vulnerable to moderate to great earthquakes. This case study specially focused on 26 th Jan
2011 earthquakes in which 15.9million people out of 37.8 had been affected. Due to this earthquake, there was
loss Rs 106 million. This earthquake was so devastating in its scale and suffering that the likes of it had not
been experienced in past 50 years .The main aim of case study is to find socio economic impact of Gujarat
earthquake in all the phases from emergency to rehabilitation.
Keywords
Earthquake, Seismic, Death, Causalties,Loss.
1. INTRODUCTION
India has a long history of devastating earthquakes, primarily because the country continues to drive into
Asia at a rate of approximately 47mm/year. India is currently divided into four seismic zones, but statistics
show that almost 60% of the country is vulnerable to moderate to strong earthquake shaking.
The 1993 Killari earthquake that killed close to 10,000 occurred in an area considered to be a non-seismic
zone. Since 1988, there have been five moderate earthquakes (ranging from 6-6.4M) and the destructive
Gujarat earthquake of 2001, which killed more than 20,000 people. Indian history of loss of life during an
earthquake since 1900, has been tabulated below:
Table1: DAMAGES TO LIFE DURING AN INDIAN EARTHQUAKE SINCE 1900[4]
Indian Earthquakes Since 1900 With 1,000 or More Deaths
Date: 4 April 1905 City: Kangra Deaths: 19,000 Magnitude: 7.5
Damage in the Kangra area and at Dehra Dun.
Date: 15 Jan 1934 City: Bihar Deaths: 10,700 Magnitude: 8.1
Extreme damage (X) in the Sitamarhi-Madhubani, India area, where most buildings tilted or sank up to 1 m
(3 ft) into the thick alluvium. Sand covered the sunken floors up to 1 m deep. This liquefaction damage
extended eastward through Supaul to Purnia, India. In the Muzaffarpur-Darbhanga area south of the zone of
liquefaction most buildings were shaken apart by "typical" severe earthquake damage. Two other areas of
extreme damage (X) from shaking occurred in the Munger (Monghyr) area along the Ganges River, India
and in the Kathmandu Valley, Nepal. Large fissures occurred in the alluvial areas.
Date: 20 August 1988 Nepal–Indian border Deaths: 1,000 Magnitude: 6.8
101
Seven hundred twenty-one people killed, 6,553 injured and 64,470 buildings damaged in eastern Nepal,
including the Kathmandu Valley. Maximum intensity VIII. Liquefaction observed in a 5,500 sq. km area of
southern Nepal. At least 277 people killed, thousands injured and extensive damage in northern Bihar,
India, particularly in the Darbhanga-Madhubani-Saharsa area. Damage in the Gangtok area, Sikkim and in
the Darjiling area, India. Felt in large parts of northern India from Delhi to the Burma border and in much of
Bangladesh.
Date: 19 October 1991 Northern India Deaths: 2,000 Magnitude: 7.0
Two events about 1.6 seconds apart. At least 2,000 people killed, more than 1,800 injured and 18,000
buildings destroyed in the Chamoli-Uttarkashi area. Some damage occurred at Chandigarh and New Delhi.
Felt in northern India, western Nepal and northeastern Pakistan. Landslides occurred in the epicentral area.
A 30-meter deep crack was noted in the Uttarkashi area.
Date: 29 September 1993 City: Latur–Killari Deaths: 9,748 Magnitude: 6.2
This earthquake was centered about 70 kilometers northeast of Shoapur and 230 kilometers west-northwest
of Hyderabad, in a region where earthquakes are infrequent. At least 9,748 people were killed; about 30,000
were injured and extreme devastation in the Latur-Osmanabad area. Nearly all buildings were destroyed in
the village of Killari. Felt in large parts of central and southern India, including Bangalore, Bombay,
Hyderabad and Madras. This earthquake was the largest known earthquake to occur in the area. Many
aftershocks, some large enough to cause additional damage and deaths, followed the main shock.
26 January 2001 City: Gujarat Deaths: 20,085 Magnitude: 7.6
At least 20,085 people killed, 166,836 injured, approximately 339,000 buildings destroyed and 783,000
damaged in the Bhuj-Ahmadabad-Rajkot area and other parts of Gujarat. Many bridges and roads damaged
in Gujarat. At least 18 people killed and some injured in southern Pakistan. Felt throughout northern India
and much of Pakistan. Also felt in Bangladesh and western Nepal. The earthquake occurred along an
approximately east-west trending thrust fault at shallow depth. The stress that caused this earthquake is due
to the Indian plate pushing northward into the Eurasian plate. Complex earthquake. A small event is
followed by a larger one about 2 seconds later.
2.Seismic Zoning Map

2.1Seismic Zone of India
Bureau of Indian Standards[5], based on various scientific inputs from a number of agencies including
earthquake data supplied by IMD, has grouped the country into four seismic zones viz., Zone-II, -III,
-IV and -V. Of these, zone V is rated as the most seismically prone region, while zone II is the least.
The Modified Mercalli (MM) intensity, which measures the impact of the earthquakes on the surface
of the earth, broadly associated with various zones, is as follows:
Table 2: Seismic Zones of India
102
Seismic Zone Intensity on MMI scale % of total area
II (Low intensity zone) VI (or less) 43%
III (Moderate intensity zone) VII 27%
IV (Severe intensity zone) VIII 18%
V (Very severe intensity zone) IX (and above) 12%
2.2 Seismic Zoning Map Of India

The Geological Survey of India (G. S. I.) first published the seismic zoning map of the country in the
year 1935. With numerous modifications made afterwards, this map was initially based on the
amount of damage suffered by the different regions of India because of earthquakes. Color coded in
different shades of the color red, this map shows the four distinct seismic zones of India. Following
are the varied seismic zones of the nation, which are prominently shown in the map:
 Zone - II: This is said to be the least active seismic zone.

 Zone - III: It is included in the moderate seismic zone.
 Zone - IV: This is considered to be the high seismic zone.
 Zone - V: It is the highest seismic zone.
Fig1: Seismic Zoning Map of India
3.Vulnerability Assessment of an Earthquake

The damage to built environment during recent earthquake in India has demonstrated the need for
seismic risk assessment that is capable of predicting the consequences of earthquakes. The collapse
of man-made engineered and non-engineered buildings during an earthquake is the chief contributor
to the loss of lives and injuries to the people. Vulnerability Atlas of India states that there are about
11 million seismically vulnerable houses in seismic zone V, while the corresponding figure for
seismic zone IV is 50 million. In all, there are about 80 million building units in India, which are
vulnerable, and pose unprecedented risk, if earthquake strikes. The greatest challenge, therefore, is
not only to rehabilitate these vulnerable houses so as to reduce considerable loss to human life and
103
property but also to evolve an accepted methodology in Indian context to estimate/quantify the
seismic vulnerability of the existing built environment, which will be provide a useful information
for policy making. “Earthquakes alone don't kill people; collapsed buildings do.”
Earthquake shaking is a complex process and the chain of causation from earthquake source
magnitude through infrastructural damage to human harm involves factors like the type of
earthquake at fault, its orientation, the hardness of bedrock or presence of wet soil, and so on. A lot
also depends on the time of day the earthquake strikes in terms of how many people are inside
buildings that could collapse. Population density, distance from the epicenter, and the depth of the
rupture are the most important factors of all.
For socio economic aspects it is most important thing is that know number of people vulnerable for
particular disaster. For Indian scenario till now no processing method has developed to know the risk
in terms of life loss and economic loss.
4. CASE STUDY: GUJARAT EARTHQUAKE
A Powerful Earthquake of magnitude 6.9 on Richter-Scale rocked the Western Indian State of
Gujarat on the 26th of January; 2001.It caused extensive damage to life & property. This earthquake
was so devastating in its scale and suffering that the likes of it had not been experienced in past 50
years. Leaving thousands seriously injured, bruised and handicapped; physically, psychologically
and economically.
The epicenter of the quake was located at 23.6 north’s Latitude and 69.8 east’s Longitude, about 20
km Northeast of Bhuj Town of the Kutch district in Western Gujarat. At a depth of only 23 kms
below surface this quake generated intense shaking which was felt in 70% region of India and far
beyond in neighboring Pakistan and Nepal too. This was followed by intense aftershocks that
became a continued source of anxiety for the populace. The Seismicity of the affected Area of Kutch
is a known fact with a high incidence of earthquakes in recent times and in historical past. It falls in
Seismic Zone V, only such zone outside the Himalayan Seismic Belt. In last 200 years important
damaging earthquakes occurred in 1819, 1844, 1845, 1856, 1869,1956 in the same vicinity as 2001
earthquake.
Twenty-one of the total 25 districts of the state were affected in this quake. Around 18 towns, 182
Talukas and 7904 villages in the affected districts have seen large-scale devastation. The affected
areas even spread up to 300 km from the epicenter. In the Kutch District, four major urban areas –
Bhuj, Anjar, Bachau and Rapar suffered near total destruction. The rural areas in the region are also
very badly affected with over 450 villages almost totally destroyed. In addition, wide spread
damages also occurred in Rajkot, Jamnagar, Surendranagar, Patan and Ahmedabad districts. Other
Urban areas such as Ganhidham, Morvi, Rajkot and Jamnagar have also suffered damage to major
structures, infrastructure and industrial facilities. Ahmedabad the capital was also severely affected.
Gujarat Earthquake is very significant from the point of view of earthquake disaster mitigation in
India. The problems observed in this disaster are no different from other major recent earthquakes in
the world. The issues in the recovery and reconstruction phase are: the proper understanding risk
among different stakeholders, training and confidence building among the professionals and masons
with appropriate development planning strategies.
104
4.1. Socio Economic Impact

Based on body count, until midday February 15, 2001, the total casualty was 18,602. As of February
15, 2001, the number of deaths reported in Kutch itself was 17,030. Extensive losses have been
reported from the affected areas. On the basis of preliminary estimates, the severity of the quake and
its devastation can be judged by a comparison of some natural disasters in recent times (Table 1).
Table 1, however, underscores a fundamental problem of inter-country comparison of natural
disasters. Normalization of disaster induced deaths, homelessness, and economic loss by the
population or GDP of a country can reduce the apparent scale of a disaster in a large country. This
can be illustrated with the example of the district of Kutch – a district with a population close to that
of Nicaragua – worst affected by the recent quake. While the inherent magnitude of the loss and the
tragedy is a given quantum, it looks very different depending on whether the loss to life and income
are expressed as a proportion of the population of the district itself, or of Gujarat (the Indian State to
which it belongs) or India.
The economic impact of an earthquake or any natural disaster can be classified as: (i) losses to
immovable assets, (ii) losses to movable assets (iii) economic losses due to business interruption, (iv)
public sector economic costs, and (v) household income losses due to death, injury, and job
disruption. The first impact consists of the direct economic losses due to destroyed or severely
damaged buildings and other structures (such as power substations). Losses to movable assets
consist of economic losses due to damaged or destroyed contents of buildings and other private
property. Public sector economic costs accrue because of loss of revenues and increases in expenses
for the public sector.
I. Deaths and Injuries

Geographic distribution of deaths and injuries
More than 18 thousand people have perished in the earthquake, with nearly 1.66 lakh people
injured (Table 2). Kutch has been devastated by the quake while five other districts –
Ahmedabad, Jamnagar, Patan, Surat and Surendranagar – have been severely affected. These
six districts account for as much as 99 1/3 per cent of the deaths. Towns and villages in the
areas ravaged by the quake in these severely affected districts have turned into heaps of stone
and concrete, and the livelihood of people has been destroyed. The affected people have
been living in tents provided by the State Government and other donors, including NGOs and
other governments, both from inside and outside India. There are a few reports of mental
disorders from the trauma of destroyed homes and death of near and dear ones. The district
of Kutch alone accounts for 92 per cent of the deaths and 82 per cent of injuries.
Arid Kutch is the most sparsely populated district in Gujarat. With a density of 28 per sq. km.
relative to the state average of 211, Kutch accounts for only 3.06 per cent of Gujarat’s population.
Among the other affected districts, Ahmedabad and Surat have population shares of 11.11 per cent
and 6.23 per cent, respectively. Fifteen other districts of Gujarat have also been affected by the
quake. Although, unlike the seriously affected districts, with 123 casualties they have been spared
the trauma of a very large number of deaths, with 6,730 injured they account for over a good 4 per
105
cent of the quake-induced injuries. Only four districts, namely Dahod, Dangs, Narmada, and Panch
Mahals, appear to have escaped the wrath of nature.
Table 3. Gujarat January 26, 2001 Earthquake: Deaths and Injuries[3]
Districts No. Population1 Death Injuries Death Percen Percentage share of
of (September Rate (per tage total injuries
Affe 2000) thousand) share
cted of total
Talu death
kas
Seriously affected districts
Ahmedabad 11 5,367,464 750 4037 0.14 4.03 2.42
Jamnagar 11 1,829,397 119 4930 0.07 2.96
0.64
Kutch 10 1,477,160 17030 136000 11.53 81.57
91.55
Patan 8 1,516,584 34 1695 0.02 1.02
0.18
Rajkot 14 2,941,577 422 11946 0.14 7.16
2.27
Surendranag 10 1,414,406 112 2907 0.08 1.74
ar 0.60
Other Districts
Amrelli 11 1,544,746 - 5 - 0.00
-
Anand 8 1,927,914 1 20 - 0.01
0.01
Banaskantha 8 2,318,413 32 2770 0.01 1.66
0.17
Bharuch 8 1,343,480 9 44 0.01 0.03
0.05
Bhavnagar 11 2,408,570 4 45 - 0.03
0.02
Gandhinagar 4 1,034,505 8 241 0.01 0.14
0.04
Junagadh 14 2,446,752 8 89 - 0.05
0.04
Kheda 10 2,098,010 - 28 - 0.02
-
Mehsana 9 1,827,351 - 1339 - 0.80
-
Navasari 5 1,270,283 17 52 0.01 0.03
0.09
Porbandar 3 549,269 9 90 0.02 0.05
0.05
Sabarkantha 8 2,060,509 - 56 - - 0.03
Surat 8 3,975,616 46 184 0.01 0.11
0.25
106
Vadodara 6 3,554,674 1 256 - 0.15

0.01
Valsad 5 1,272,960 - - - -
-
Total 182 44,179,641 18,602 166,734 0.42 100 100
Loss from an earthquake depends on the interaction of the geographical pattern of ground
motion with the spatial array of the population and properties at risk and their loss
vulnerabilities. Although Gujarat has a high urbanization ratio of 34 per cent compared to the
all-India average of 26 per cent, the location of the epicenter of the quake in the relatively
sparsely populated northwestern part of the State averted a disaster that could have been of an
even higher magnitude.
The enormity of the disaster in Kutch can be gauged by the quake-induced high death rate of
11.53 per thousand people relative to the death rate of the State of 7.9 per thousand according
to last 1991 census.
II. Social security
The injured and the next of kin of the deceased from the earthquake were entitled to two social
security and insurance benefits. First, the Government of Gujarat had announced the direct
payment of ex-gratia amounts varying between Rs.2, 000 and Rs.1, 50,000 for the injured and
the dead, with a ceiling of Rs.5, 00,000 per family (Table 4). These compensations are in
addition to the immediate relief of cash doles of Rs.15 per person per day and a onetime lump
sum Rs 1250 per family for household kits provided to the quake affected families. The doles
are for a maximum of five persons per family for fifteen and thirty days in affected and
severely affected areas, respectively.
Table 4: Ex-gratia Payments to Victims from Government of Gujarat during Bhuj Earthquake
2001[3]
Death relief: Rs.1 lakh for a major and Rs.60, 000 for a child
Additional death relief: In addition to the death relief, Rs.50, 000 to government employees and
school children.
Relief in the case of injury
Injury above 40 per cent: (permanent disability): Rs.50, 000
Less than 40 per cent: Rs.25, 000
Major Surgery: Rs.10, 000
Minor Surgery: Rs.5, 000
Minor Injury: Rs.2, 000
4.2. Economic Impact and GDP
107
The economic impact of the earthquake includes: (a) asset losses (b) output losses damage and
(c) fiscal costs .Destruction of private assets in the affected districts had been massive. Of the
estimated Rs. 9,900 crore total asset losses, Rs. 7,400 crore were private assets. While the impact
of the earthquake on India’s gross domestic product was insignificant, in the areas where it
struck, the earthquake devastated lives, social infrastructure, and economic foundations.(a)
Preliminary estimates total Rs. 9,900 crore in asset losses and Rs. 10,600 crore in improved-
standard reconstruction costs. Losses in public sector assets were estimated to total Rs. 2,500
crore, and include asset losses in health (Rs. 220 crore), education (excluding private schools,
some 70 percent of the total, a loss of Rs. 470 crore), municipal infrastructure (Rs. 140 crore),
public buildings and monuments (Rs. 339 crore), rural water supply (Rs 233 crore), irrigation
(Rs. 186 crore), power (Rs. 186 crore), roads, bridges, railways, and airports (Rs. 321 crore ),
ports (Rs. 98 crore), and telecommunications (Rs. 51 crore). In addition to the public and private
losses set out above, there were some $55 million of identified damages to the environment.
These consist of the cost of rubble removal in urban and rural areas, and further environmental
impacts may later be identified. The costs of tackling these problems were likely to accrue to the
public sector.
5. Conclusion
The dominant impact of the earthquake will be on the financial and asset market behavior of the
affected people. The wealth loss sustained as a result of the quake would lead to a major
reshuffling of their portfolios. The loss being primarily of physical assets, the people in the
affected regions would draw down their financial assets to restore some of the physical assets
lost. While the asset loss will impose substantial hardships on the people affected, in financial
markets in aggregate terms the effect will be of second order of magnitude. The total deposits in
the affected districts of about Rs. 173 billion constituted only 2.2 per cent of the aggregate
deposits in the country of Rs.799 trillion. The large deposit base of the affected regions indicates
that the banks had ongoing relations with many of the affected people. This would help the
banks to do the credit appraisal for restoration and rehabilitation, and together with the low
credit-deposit ratio, would enable credit to flow to the affected regions. The chain of cooperatives
would have also contributed to the recovery and restoration.
6. Lessons Learnt
Natural disasters like earthquake cannot be predicted. The foremost requirement is construction
of seismically safe structures in earthquake prone areas. A large number population of the
country live in the highly seismic active regions, to better deal with the hazard the art of living
with earthquake has to be learnt. The Gujarat earthquake has given a sort of “Nation Wakeup
call”. There should be a national learning to take appropriate and specific steps for disaster
reduction, mitigation, prevention and preparedness. This earthquake can be regarded as an
opportunity to learn from this disaster and to work to never let it happen. Therefore, there is need
to learn lesson this event and a plan must be designed for the areas located in high seismic zones
to mitigate the impacts of future earthquakes. Development and disaster go hand in hand.
Development models must have built in components of disaster reduction, mitigation and
preparedness.
108
REFERENCES
[1] A,Gupta 2001. “The Great Gujarat Earthquake Lessons Learnt”. A Paper presented at 22
nd
Asian conference on Remote Sensing, 5-9 November 2001.

[2] Lecy, J.D. “Aid Effectiveness after the Gujarat Earthquake: A Case Study of Disaster Relief”.
Journal of Development and Social Transformation Volume 4, November 2000.
[3] Lahiri A.K., Sen T.K., Rao R.K., and P.R.Jena. “Economic Consequences of the Gujarat
Earthquake”. A Paper written during a visit to Gujarat as part of the Gujarat Earthquake
Assessment Mission of the Asian Development Bank and the World Bank, during February 11-
22, 2001.
[4] Jain S.K. and Lettis W.R. “Preliminary Observations on the Origin and Effects of the January
26, 2001 Bhuj (Gujrat, India) Earthquake”. A report published on EERI Special Earthquake
Report April2001.
[5] IS 1893 (Part 1) (2002), “Criteria for Earthquake Resistant Design of Structures General
Provisions and Buildings”, Bureau of Indian Standards, New Delhi.
[6] Danbar P.K., Bilgham R.G.,and Laturi M.J.. “Earthquake Loss Estimation for India Based on
Macroeconomic Indicators” Risk Science and Sustainability Volume 112, Kluwer Academic
Publishers, Bouston.
109
Paper ID- 35
Effect of Triaxiality and Lode Angle on Failure Behaviour of Riveted Lap

Joints
Abhishek Sarmah1 (0000-0002-5080-8812), Jyoti Mukhopadhyay2
1
Assam Don Bosco University, Guwahati, India
abhisheksarmah90@gmail.com
2
Indian Institute of Technology, Gandhinagar, India
jm@iitgn.ac.in
Abstract. Failure prediction is of utmost importance to achieve optimal design of complex mechanical systems.
The second stress invariant and triaxiality factor [1] have substantial effect on the elasto-plastic behaviour of
ductile materials specifically metals. Study of the aforementioned parameter is carried out in conjunction with
Lode [2] parameter to determine whether they can be used as an effective indicator of failure. Numerical
simulation is carried out for lap joints with different riveting patterns under tensile loads. An elastic-plastic
material model with strain hardening is adopted for accurate description of material behaviour. The simulated
behaviour is validated through corresponding experiments. It is found that Lode angle can describe the state of
stress prevalent in the specimen while triaxiality can give an indication of the dominating failure regime.
However, an accurate numerical prediction on initiation of fracture was possible only when the effect of triaxiality
as well as Lode angle was combined and studied as a part of the Mohr-Coulomb failure model [3-4]. It was evident
from the triaxiality values that the failure regime was predominantly of shear type in the riveted lap joints. The
fracture onset could be accurately predicted by the Mohr-Coulomb fracture model and the failure initiated from
the rivet hole and grew along the pitch or towards the margin depending on the riveting pattern.
Keywords: Mohr-Coulomb failure. Triaxiality. Lode angle. Damage
1 Introduction
Rivets are permanent mechanical joints having wide mechanical and civil engineering applications such as
automobile chassis, aerospace industry, framed buildings, structures etc. As such multiple research initiatives have
been taken up previously to determine the integrity of riveted structures both experimentally and numerically.
Sakiyama et al [5] outlined the use of both blind rivets and self-piercing rivets (SPR) in automobile industry.
While the strength of SPR may sometimes be inadequate, the strength of blind rivets is comparable to spot welds.
Rivet joints have an advantage over welded joints especially when two dissimilar metals are being joined. The
deformation in thin walled aircraft parts was carried out by Zhang et al [6] to determine whether the physical
phenomena can be accurately modelled through numerical simulations as well as to establish the stresses
generated during the process. Fung et al [7] studied the fatigue failure of countersunk riveted lap joints
experimentally in relation with the stress concentration around the rivet hole. However, they had left out the
plasticity effects which may have far reaching consequences in dictating the failure behaviour. Hoang et al [8]
studied the structural behaviour of self-piercing rivets to validate a resultant based point connector model for self-
piercing rivets. Experimental and numerical investigation was carried out by Porcaro et al [9] on riveted
aluminium joints of AA6060, to study the influence of thickness of plates, specimen geometry and loading
conditions. Hoang et al [10] tried to provide an experimental database on the riveting process using an aluminium
SPR and studied the effect of quasi-static tensile loading on the SPR connected joints. Shahani and Kashani [11]
modelled three dimensional fatigue crack growth profiles in riveted lap joints using fracture mechanics concepts.
They showed use of fracture mechanics concepts yielded numerical results that are much closer to experimental
values. It is seen that although substantial work has been done in simulating the riveting process, the amount of
work done in prediction of fracture in riveted joints is quite scant. It is primarily due the complex stress regimes
prevalent in a riveted joint. Due to this, generalized stress and strain behaviour is inadequate in predicting the
behaviour of a riveted joint. To address this inadequacy, a closer look must be taken on the effect of individual
parameters on the behaviour of a ductile material under a complex stress-strain regime.
The failure of material cannot be attributed to only one or two factors only; instead a host of different mechanisms
are responsible for it. However, it is seen that under specific conditions of stress and strain, one of the mechanisms
will dominate over the others. Predominantly, ductile metal failure occurs due to void coalescence or formation
of shear bands that evolve within the metal. Discontinuities introduced by the presence of foreign particles such
as inclusions act as sites of initiation. Initial damage models were based only on the effect of equivalent plastic
strain while mostly ignoring the effect of the prevalent stress regime. Subsequent works [12-13] proposed
involving the effect of stress triaxiality along with equivalent plastic strain. Most of these models had considerable
experimental research [14-15] lending credence. Although exhaustive studies validated that the models were good
enough for simple scenarios, but it also laid bare their inability to capture the effects release of elastic strain energy
110
during crack initiation. Continuous damage mechanics based models [16-21] incorporating elastic strain energy
release are now available. However, these had certain drawbacks relating to softening of metals in the necking
regions. On the other hand, damage theories based on Gurson model [22] such as proposed by Tvergaard and
Needleman [23] and Hsu [24] considered void growth coupled with yielding. They were better suited for
prediction of shear failure, but was constrained by the number of material parameters needed for accurate
description. Models proposed by Le Roy [25] and other researchers [26-28] were based on phenomenology along
with energy release parameters. In mid 2000s, it was proposed [29-32] that incorporating the third invariant of the
deviatoric stress tensor yielded far better results with minimal material constants needed for accuracy. The Mohr-
Coulomb is one of the more promising failure models incorporating the effect of the third deviatoric invariant in
the form of Lode angle along with stress triaxiality to determine the equivalent failure stress.
The stress vector t  nˆ  can be written as
t  nˆ   n.
ˆ (1)
Where σ is the Cauchy stress tensor and n̂ is the unit normal vector. The Cauchy stress can be defined as current
forces per unit deformed area. In indicial form, the Cauchy stress tensor can be represented as
  ijeˆ i eˆ j (2)
The eigenvalues and eigenvectors of the Cauchy stress tensor are the principal stresses and principal vectors
respectively. A better way to state of stress on a point is through the use of the invariants of the stress tensor since
it gives a measure of stress which is independent of any change in point of reference or co-ordinate axes. The
invariants of the Cauchy stress tensor is given by equations 3.
I1  tr
1
2 
2
 
I2   tr   tr 2 

I3  det (3)
The tr and det represents the trace and determinant respectively. The mean stress or hydrostatic stress, which
of great physical significance, can be defined in terms of stress invariants as
1 1
m  tr  I1 (4)
3 3
Along with the mean stress, the deviatoric stress, which is responsible for plastic deformation in body, is also
very important. The deviatoric stress is defined, in terms of invariants, as follows:
1  1 
    I1I  ij  ij  ijkk  (5)
3  3 
The ij and I represents the Kronecker delta and a second order identity matrix respectively. The third invariant of
the deviatoric stress tensor is given by
1
J3  ij  jk ki (6)
3
The deviatoric stress is particularly handy in finding out the principal stresses of the stress tensor. The first
invariant and the second invariant are directly related to the mean stress and von Mises stress respectively, which
makes them parameters of great importance. Comparing the von Mises stress or the deviatoric stress with the yield
stress is one of the frequently adopted techniques to determine whether yielding has taken place in a material. The
von Mises stress is an indicator of the plastic deformation a system has undergone due to development of stress
upon loading. However, in complex loading scenario von Mises stress may not be sufficient for complete
description of the system. In terms of the principal stresses, the von Mises stress is given by equation 7.
1
VM  1   2 2   2   3 2   3  1 2  (7)
2 
It was observed by Rice and Tracey [13] as well as by McClintock [12] that hydrostatic stress also plays a very
important role in determining the material response. It was found that a high hydrostatic stress contributed to void
growth which later resulted in void coalescence. In other words, a high mean stress means a lowering of ductility
due failure by void coalescence during plastic deformation in elasto-plastic materials.
The ratio of the hydrostatic stress to the von Mises or equivalent stress is termed as the triaxiality factor,
η. It is widely accepted that moderate to high triaxiality leads to accelerated degradation of the material under
plastic deformation. It was shown that in a remote strain rate field, the enlargement rate of spherical voids is
dependent on triaxiality [33]. This rapid void growth leads to accelerated degradation of the material leading to
failure at moderate to high triaxialities. Bao and Weirzbicki [28] established that for large triaxiality, void growth
is the dominant mode while at negative stress triaxialaities, shear failure is the dominant fracture mode. At low
triaxiality between the aforementioned regimes, a combination shear and void coalescence causes fracture.
Moreover, Bao and Weirzbicki [33] further proved, both analytically as well as experimentally, the existence of
a cut off value of triaxiality below which failure never takes place. This cut-off value was determined to be -1/3.
Kweon [34] developed a theoretical framework based on crystal plasticity and void growth for failure at negative
111
triaxiality. Therefore, it is seen determining the triaxiality regime in a stressed body can provide an insight into
the state of stress dominant as well as indicate the possible mode of fracture should failure initiate at any point.
Moreover, existence of points where the triaxiality value is below the cut-off value essentially eliminates them as
possible point of fracture initiation.
However, more insight can gleaned about the state of stress in any body by finding out the Lode angle. In simple
terms, the Lode angle can be described as the angle that the line of pure shear makes on the projection of the stress
tensor on the deviatoric plane. In terms of principal stresses, the Lode parameter, initially proposed by Lode [2],
can be defined as
2    
  2 1 3 (8)
1   3
Where the value of the Lode parameter varies between -1 and +1. However Wierzbicki [29] proposed a different
expression in terms of the third deviatoric invariant and von Mises equivalent stress.
27 J 3
X  (9)
2  3
VM
The parameter X also varies between -1 to +1. The parameter X has been named as the normalized third invariant
[35]. From the expression of the normalized third invariant, the normalized Lode angle [36] has been derived as
2
  1  arccos X (10)

The value of the normalized Lode angle also varies between -1 and +1. The significance of the normalized Lode
angle is that when  =1, the state of stress prevalent is one of axisymmetric tension while  =0 corresponds to
a state of generalized shear. If  = -1, then the state of stress is axisymmetric compression or equibiaxial tension.
Therefore, we can glean a lot about the state of stress prevalent in the system from the value of the Lode angle. It
was proved conclusively by multiple researchers [28, 29, 37-38] that material ductility and fracture strain is
heavily dependent upon the triaxiality factor and the Lode angle.
The Mohr-Coulomb failure model [3-4] is very frequently used in soil mechanics. It is a model which
combines the effects of material plasticity along with response to hydrostatic stress to predict failure. However,
its applicability extends beyond soil and rock to failure of a wide range of materials. It has been used as model for
fracture prediction for ceramics and brittle materials [39]. Bai and Weirzbicki [40] showed the applicability of the
M-C model to ductile materials with excellent results. One of the primary advantage of the M-C model is that
since it is an extension of the maximum shear stress failure theory, it is well equipped to predict shear fracture
along with ductile fracture. In fact, the maximum shear stress model is a special case of the M-C model.
The M-C model in terms of equivalent failure stress can be written as a function of triaxiality and normalized
Lode angle as
c2
f  (11)
2
1  c1    1   
cos     c1   sin    
3  6   3  6 
Where the co-efficients c1 and c2 are material constants relating the friction and cohesion of the material. The
failure takes place wherever the von Mises stress exceed the failure stress predicted. For low carbon steels, the
values of the constants are c1 =0.32 and c2 =405MPa [41]
2 Experimentation
Blind riveting was done on three different specimen with different riveting patterns. The joint made was lap joint
in all the three specimen. The different riveting pattern adopted was chain riveting, diamond riveting and zig-zag
riveting. The number of rivets in each specimen was kept same for the sake of uniformity. The diameter of the
rivet was of 9 mm while the rivet head was of 16nn. The pitch was kept 27mm in all the specimens while the
margin was of 13.5 mm as per design consideration taking into account the plate thickness and rivet diameter.
The schematics of only one plate the specimen are shown in the following figures, as the other plate of the
specimen are symmetric. Figure 1 shows the schematic of chain riveting while figure 2 and figure 3 shows the
schematic of diamond riveting and zig-zag riveting respectively. The dotted lines in the figure denote the fracture
locus in all the experimental specimens. The plate thickness in all the specimens were of 5mm. The specimens
were subjected monotonic quasi-static tensile loads until fracture in a 400kN Universal Tensile testing machine
and the corresponding load-displacement data were recorded for validating the simulated models.
112
Fig. 1. Chain riveted specimen dimensions
Fig. 2. Diamond riveted specimen dimensions
Fig. 3. Zig-zag riveted specimen dimensions
Fig. 4. Experimental true stress vs true plastic strain for mild steel which is used as an input for numerical simulation
Additionally tensile dog-bone standards of the same material were obtained to determine the material properties
of the material. Upon testing, the Young’s Modulus was determined to be 208GPa with a yield stress of 254MPa.
The experimental stress-strain curve was converted to a true stress- plastic strain curve for purposes of using it as
material input for simulated FEM studies. Figure 4 shows the true stress-plastic strain curve for the material.
Figure 5 shows the failure pattern in a chain riveted joint, where the failure has taken place along the locus
connecting the rivet hole and the margin of the joint. The diamond riveted joints demonstrate a similar failure
pattern. However, the failure pattern is different in zig-zag patterned riveted joint in the sense that failure occurred
along the pitch length.
113
Fig. 5. Fracture from the rivet hole towards the margin in chain riveted specimen
3 Numerical Simulation
The quasi-static and monotonic tensile loading of the three different riveted lap joints were numerically simulated.
The numerical scheme adopted was Finite Element (FE) method. The simulation was carried out in the software
ABAQUS 6-14. The element type adopted is C3D8R which is an eight noded brick element with reduced
integration and hourglass control. To avoid instability in the model due to excessive local elemental distortions,
hourglass control is activated. Hourglassing is essentially a faulty distortion of FEM mesh wherein zero degree of
freedom elements are spuriously excited. It can be anticipated in complex simulations such as the current work,
where some elements may deform excessively while its neighbouring elements remain relatively undistorted. In
the current simulation, all degrees of freedom in one end of the model is prohibited while the other end is subjected
to monotonic and quasi-static loading.
Fig. 6. Depiction of directions of interest in the FE model
Fig. 7. Methodology of extraction and comparison from simulated model
The material behaviour is assumed to be homogeneous and isotropic. The material is assumed to behave
elasto-plastically under loading beyond yield stress. From the tensile test carried out, the true stress vs plastic
strain curve is used as an input for accurate modelling of the material behaviour under tensile loading. Basically,
by using the true stress vs plastic strain curve, the elastic recovery associated with material deformation in plastic
region is eliminated from the simulation results resulting in higher accuracy. Another advantage of using true
stress and corresponding plastic strain is that the effect of any residual stress that may have been present in the
experimental samples will be reflected in the flow stress results of the simulation. Figure 4 shows the variation
114
of true stress with respect to plastic strain for mild steel. From the simulated models, triaxiality is plotted against
distance from the rivet hole in three different directions viz., Direction A: along the axis of application of load,
Direction B: perpendicular to the axis of application of load towards the margin and Direction C: along the pitch.
The directions are chosen since the experimental specimen failed primarily along rivet hole to margin or along
the pitch. Apart from the triaxiality, the stress at which the material fails as predicted from the M-C model is
plotted along with the Mises stress with respect to distance from the rivet hole. All the directions of interest are
shown for a diamond patterned lap joint in fig. 6. The M-C failure law has been computed using UVARM
subroutine in ABAQUS so as to get the history output as well as field output at all time increments. An implicit
integration scheme is used for current simulation so as to avoid any stability issues. Figure 7 shows the
methodology of extracting and comparing the failure stress from the simulated models. An element set considering
only elements near the rivet holes, excluding the rivets and clamped parts, are pre-defined. The subroutine is
employed only on this elemental set so as to increase computational efficiency as well as to avoid inclusion of
singularities in the calculation as Mises stress tends to zero, the triaxiality tends to infinity.
Fig. 8. Comparative load vs displacement for chain riveting
Fig. 9. Comparative load vs displacement for diamond riveting
Fig. 10. Comparative load vs displacement for zig-zag riveting
115
Fig. 11. Triaxiality vs distance from rivet hole for direction B

4 Results and discussions
Figure 8, fig. 9 and fig. 10 shows the comparison between the experimental and numerically predicted load vs
displacement behaviour. The behaviour of the lap joints under tensile loads in the current study is found to be
consistent with the similar experimentation done by Porcaro et al [9] on ductile riveted lap joints. It quite evident
from the figure that the diamond riveting shows maximum strain to fracture as well as is able to withstand the
maximum stress. On the other hand zig-zag riveting shows the minimum resistance to fracture among all the three
specimens. The behaviour of the monotonic increase of the load vs displacement curve for the riveted lap joint is
also found to be consistent with previously established behaviour. It is evident from the fig.8, fig.9 and fig.10
that numerical prediction is conservative meaning it shows a lower load value for corresponding displacement.
The difference in behaviour can be attributed to the fact that in numerical simulations, only the tensile behaviour
is incorporated. On the other hand, the real behaviour of the material during experimentation is a combination of
material response to tensile, shear and to an extent bending forces. Moreover, the although the loading is modelled
to be predominantly of Mode I in simulated study, but the actual loading condition during the experimentation is
a combination of Mode I, Mode II and Mode III loading although Mode I loading predominates. However, the
difference is the maximum deviation between the simulated values and experimental values are within acceptable
range of less than 20%. Due to the varying loading modes, traditional means of studying stress regime is
inadequate. It may addresses by taking a closer look on the orientation of the individual stress components using
the M-C criterion.
Plastic deformation is restricted to an extent in diamond riveting as evident form the lower Mises values,
which makes it the ideal pattern for lap joints under tensile loads. Relatively lower absolute value of triaxiality
coupled with lower Mises stress component means a relatively lower mean stress component for diamond riveting,
which makes the material resist deformation resulting in higher strength. In case of zig-zag, riveting, plastic
deformation progressively increases and mean stress also correspondingly increases before failure making the
joint to fail relatively easily as compared to diamond riveting.
The triaxiality as a function of distance from the rivet hole is plotted for all the three pre-defined
directions for all the three specimens. Figure 11 and Figure 12 depicts the change in triaxiality with distance for
direction B and direction C respectively for all the three specimen. It is well documented that fracture is restricted
from occurring at traixialities below -1/3 [33]. Therefore, it can be seen from the figures depicting the triaxiality
variations that, points which falls below the cut-off value are sites from which fracture could not have initiated.
Fig. 12. Triaxiality vs distance from rivet hole for direction C
116
Fig. 13. Mises stress and predicted failure stress for Direction A in chain riveting
Fig. 14. Mises stress and predicted failure stress for Direction B in chain riveting
Fig. 15. Mises stress and predicted failure stress for Direction C in chain riveting
Fig. 16. Mises stress and predicted failure stress for Direction A in diamond riveting
117
Fig. 17. Mises stress and predicted failure stress for Direction B in diamond riveting
Especially in the case of direction B in chain riveting as well as diamond riveting, where the experimental
specimens had failed, only the region near the rivet hole and the margin allows for fracture initiation. On the other
hand, zig-zag rivets in the direction of C only allows the rivet holes as failure initiation points. However, had the
fracture been unknown, it would have not been possible to determine the fracture direction just from the values
of triaxiality. It is because values had exceeded the cut-off triaxiality values at a number of points in all the three
directions. Although, the stress regime can be assumed to be of shear dominated since most of the triaxiality values
are negative, especially in the directions of known fracture locus. The shear regime domination can be validated
by taking looking at the Lode angle values. The close-to-zero values of Lode angle reinforces our earlier findings
that the loading conditions prevailing in all the specimens is of generalized shear.
Fig. 18. Mises stress and predicted failure stress for Direction C in diamond riveting
Fig. 19. Mises stress and predicted failure stress for Direction A in zig-zag riveting
It is quite clear that while triaxiality and Lode angle can determine the state of stress prevalent in the
system, it falls of short of accurately predicting the fracture initiation points conclusively, when considered
individually. However, when considered in conjunction as a part of the M-C model, it is seen that the fracture
initiation points are clearly predicted eliminating the ambiguity associated with the study of individual parameters.
Figures 13-21 plots the Mises stress and predicted failure stress against distance from rivet hole. It is clear from
the plots that failure occurs initially at the rivet hole in the direction towards the margin for chain riveting, while
in all other direction the generated Mises stress fails to exceed the cut-off values set by the M-C model. The
118
diamond pattern riveting also shows a similar trend in regards to points of failure initiation. At the rivet hole, the
Mises stress generated is 791 MPa where the failure stress predicted is only of 783.33 MPa. However, the zig-zag
patterned riveting differs from the other two specimens in a sense that while failure initiates at the rivet hole, but
the direction is towards the pitch length with both the rivet holes acting as fracture initiation points. While the
triaxiality values give an indication of possible site for fracture initiation, it fails to point out the exact regions.
The M-C stress values addresses this concern by giving an indication of the sites where fracture has most likely
initiated. However, it must be kept in mind that while studying triaxiality along with M-C model gives an
indication of fracture initiation point, but exact determination of the point of initiation is impossible. It is due to
the fact that initiation fracture depends on a host of metallurgical properties, such as presence of inclusions,
precipitates etc., apart from the stress regime. Inclusions and precipitates acts as stress risers which exponentially
increases the stress value locally under the action of concentrated stress. This sudden increase in stress value
around inclusions or precipitates helps in creating micro-cracks through void coalescence or severe micro-
shearing. As the distribution of precipitates and inclusions varies from specimen to specimen, it effectively
eliminates the possibility of identifying the exact points of initiation. However, the M-C model has proved quite
efficient in identifying possible zones of failure initiation as well as predicating the mechanism which will
predominate in creation of micro-cracks.
Fig. 20. Mises stress and predicted failure stress for Direction B in zig-zag riveting
Fig. 21. Mises stress and predicted failure stress for Direction C in zig-zag riveting
5 Conclusions
It is apparent that the study of the parameters like triaxiality and Lode angle can give an indication of the stress
regime predominating in a stressed body, especially in a complex system where geometry as well as loading
condition make logical approximation impossible. Study of triaxiality and von Mises stress can be used as an
approximate indicator of failure behaviour. For a more accurate prediction, it is advisable to go for a model like
Mohr-Coulomb model, depending upon the type of material used, which coupled the effects of triaxiality, Lode
angle and material properties with one another to give a more accurate prediction of failure. In terms of strength
of the riveted lap joints, diamond pattern riveted lap joints resists failure the most while zig-zag pattern resists the
least. It is advisable to employ diamond riveting for lap joints when the loading is tensile. The failure mode is
predominantly of shear type since the triaxiality regime is negative along the fracture locus. Rivet hole and edge
119
of the plates act as failure initiation sites when subjected to tensile loading. Lower hydrostatic stress, along the
fracture locus contributes to the higher strength of the joint.
6 References
[1] Thomason P.F.: Ductile fracture of metals. Pergamon Press, Oxford (1990)
[2] Lode, W.: Versuche über den Einflubder mittleren Hauptspannung auf das Flieben der Metalle Eisen. Kupfer
und Nickel, Zeitschrift für Physik. 36, 913–939 (1926). doi:10.1007/BF01400222
[3] Mohr O (1900) Welche Umstaende bedingen die Elastizitaetsgrenze und den Bruch eines Materials, Zeitschrift
des Vereins Deutscher Ingenieure 44, pp 1524–1530.
[4] Bigoni, D., Piccolroaz, A.: Yield criteria for quasibrittle and frictional materials. Int. J. Solids Struct. 41, 2855–
2878 (2004). doi:10.1016/j.ijsolstr.2003.12.024
[5] Sakayama, T., Murayma, G., Naito, Y., Saita, K., Miyazakki, Y., Oikawa, H., Nose, T.: 自動車ボディにお
ける鋼板とアルミニウム合金板との異種金属接合技術 Dissimilar metal joining technologies for steel sheet
and aluminum alloy sheet in auto body. 新日鉄技報. 393, 91–98 (2012). doi:C621.791/.792:621.88:669.14-
415:669.715:629.11.011.5.
[6] Zhang, K., Cheng, H., Li, Y.: Riveting process modeling and simulating for deformation analysis of aircraft’s
thin-walled sheet-metal parts. Chinese J. Aeronaut. 24, 369–377 (2011). doi:10.1016/S1000-9361(11)60044-7
[7] Fung, C., Smart, J.: An experimental and numerical analysis of riveted single lap joints. Arch. Proc. Inst.
Mech. Eng. Part G J. Aerosp. Eng. 208, 79–90 (1994)
[8] Hoang, N.H., Hanssen, A.G., Langseth, M., Porcaro, R.: Structural behaviour of aluminium self-piercing
riveted joints: An experimental and numerical investigation. Int. J. Solids Struct. 49, 3211–3223 (2012).
doi:10.1016/j.ijsolstr.2012.05.039
[9] Porcaro, R., Hanssen, A.G., Langseth, M., Aalberg, A.: The behaviour of a self-piercing riveted connection
under quasi-static loading conditions. Int. J. Solids Struct. 43, 5110–5131 (2006).
doi:10.1016/j.ijsolstr.2005.10.006
[10] Hoang, N.H., Porcaro, R., Langseth, M., Hanssen, A.G.: Self-piercing riveting connections using aluminium
rivets. Int. J. Solids Struct. 47, 427–439 (2010). doi:10.1016/j.ijsolstr.2009.10.009
[11] Shahani A.R., Kashani H.M.: Fracture mechanics-based life prediction of a riveted lap joint. Journal of
Computational and Applied Research 4, 1-17(2014)
[12] McClintock F.A.: A Criterion of Ductile Fracture by the Growth of Holes. J. of Appl. Mech. 35, 363-371
(1968)
[13] Rice, J.R., Tracey, D.M.: On the ductile enlargement of voids in triaxial stress fields*. J. Mech. Phys. Solids.
17, 201–217 (1969). doi:10.1016/0022-5096(69)90033-7
[14] Hancock, J.W., Mackenzie, A.C.: On the mechanisms of ductile failure in high-strength steels subjected to
multi-axial stress-states. J. Mech. Phys. Solids. 24, 147–160 (1976). doi:10.1016/0022-5096(76)90024-7
[15] Mackenzie, A.C., Hancock, J.W., Brown, D.K.: On the influence of state of stress on ductile failure
initiation in high strength steels. Eng. Fract. Mech. 9, (1977). doi:10.1016/0013-7944(77)90062-5
[16] Lemaitre, J.: A Continuous Damage Mechanics Model for Ductile Fracture. J. Eng. Mater. Technol. 107,
83 (1985). doi:10.1115/1.3225775
[17] Chaboche, J.L.: A review of some plasticity and viscoplasticity constitutive theories. Int. J. Plast. 24, 1642–
1693 (2008). doi:10.1016/j.ijplas.2008.03.009
[18] Bonora, N.: Industrial Engineering Department, University of Cassino, Via G. De Biasio 43, 03043 Cassino
(FR), Italy. Eng. Fract. Mech. 58, 11–28 (1997)
[19] Tie-jun, W., Tib-jun, W.: Unified Cdm Model and Local Criterion. 42, 177–183 (1992)
[20] La Rosa, G., Mirone, G., Risitano, A.: Effect of stress triaxiality corrected plastic flow on ductile damage
evolution in the framework of continuum damage mechanics. Eng. Fract. Mech. 68, 417–434 (2001).
doi:10.1016/S0013-7944(00)00109-0
[21] Haddag, B., Abed-Meraim, F., Balan, T.: Strain localization analysis using a large deformation anisotropic
elastic-plastic model coupled with damage. Int. J. Plast. 25, 1970–1996 (2009). doi:10.1016/j.ijplas.2008.12.013
[22] Gurson, A.L.: Continuum Theory of Ductile Rupture by Void Nucleation and Growth: Part I—Yield
Criteria and Flow Rules for Porous Ductile Media. J. Eng. Mater. Technol. 99, 2 (1977). doi:10.1115/1.3443401
[23] Tvergaard, V., Needleman, A.: Analysis of the cup-cone fracture in a round tensile bar. Acta Metall. 32,
157–169 (1984). doi:10.1016/0001-6160(84)90213-X
[24] Hsu, C.Y., Lee, B.J., Mear, M.E.: Constitutive models for power-law viscous solids containing spherical
voids. Int. J. Plast. 25, 134–160 (2009). doi:10.1016/j.ijplas.2007.11.003
[25] Le Roy, G., Embury, J.D., Edwards, G., Ashby, M.F.: A model of ductile fracture based on the nucleation
and growth of voids. Acta Metall. 29, 1509–1522 (1981). doi:10.1016/0001-6160(81)90185-1
[26] Johnson, G.R., Cook, W.H.: Fracture characteristics of three metals subjected to various strains, strain
120
rates, temperatures and pressures. Eng. Fract. Mech. 21, 31–48 (1985). doi:10.1016/0013-7944(85)90052-9
[27] Schiffmann, R., Heyer, J., Dahl, W.: On the application of the damage work density as a new initiation
criterion for ductile fracture. Eng. Fract. Mech. 70, 1543–1551 (2003). doi:10.1016/S0013-7944(02)00151-0
[28] Bao, Y., Wierzbicki, T.: On fracture locus in the equivalent strain and stress triaxiality space. Int. J. Mech.
Sci. 46, 81–98 (2004). doi:10.1016/j.ijmecsci.2004.02.006
[29] Wierzbicki, T., Bao, Y., Lee, Y.W., Bai, Y.: Calibration and evaluation of seven fracture models. Int. J.
Mech. Sci. 47, 719–743 (2005)(b). doi:10.1016/j.ijmecsci.2005.03.003
[30] Barsoum, I., Faleskog, J.: Micromechanical analysis on the influence of the Lode parameter on void growth
and coalescence. Int. J. Solids Struct. 48, 925–938 (2011). doi:10.1016/j.ijsolstr.2010.11.028
[31] Bigoni, D., Piccolroaz, A.: Yield criteria for quasibrittle and frictional materials. Int. J. Solids Struct. 41,
2855–2878 (2004). doi:10.1016/j.ijsolstr.2003.12.024
[32] Bai, Y., Wierzbicki, T.: A new model of metal plasticity and fracture with pressure and Lode dependence.
Int. J. Plast. 24, 1071–1096 (2008). doi:10.1016/j.ijplas.2007.09.004
[33] Bao, Y., Wierzbicki, T.: On the cut-off value of negative triaxiality for fracture. Eng. Fract. Mech. 72, 1049–
1069 (2005). doi:10.1016/j.engfracmech.2004.07.011.
[34] Kweon, S.: Damage at negative triaxiality. Eur. J. Mech. A/Solids. 31, 203–212 (2012).
doi:10.1016/j.euromechsol.2011.02.005
[35] Dunand, M., Mohr, D.: On the predictive capabilities of the shear modified Gurson and the modified Mohr-
Coulomb fracture models over a wide range of stress triaxialities and Lode angles. J. Mech. Phys. Solids. 59,
1374–1394 (2011). doi:10.1016/j.jmps.2011.04.006
[36] Nahshon, K., Hutchinson, J.W.: Modification of the Gurson Model for shear failure. Eur. J. Mech. A/Solids.
27, 1–17 (2008). doi:10.1016/j.euromechsol.2007.08.002
[37] Bao, Y.: Prediction of Ductile Crack Formation in Uncracked Bodies. PhD Thesis. 252 (2003)
[38] Wierzbicki, T., Bao, Y., Bai, Y.: A new experimental technique for constructing a fracture envelope of metals
under multi-axial loading. Proc. 2005 SEM Annu. Conf. Expo. Exp. Appl. Mech. 1295–1303 (2005)(a)
[39] Fossum, A.F., Brannon, R.M.: On a viscoplastic model for rocks with mechanism-dependent characteristic
times. Acta Geotech. 1, 89–106 (2006). doi:10.1007/s11440-006-0010-z
[40] Bai, Y., Wierzbicki, T.: Application of extended Mohr-Coulomb criterion to ductile fracture. Int. J. Fract.
161, 1–20 (2010). doi:10.1007/s10704-009-9422-8
[41] Ebnoether, F., Mohr, D.: Predicting ductile fracture of low carbon steel sheets: Stress-based versus mixed
stress/strain-based Mohr-Coulomb model. Int. J. Solids Struct. 50, 1055–1066 (2013).
doi:10.1016/j.ijsolstr.2012.11.026
121
Paper ID- 36
Effect of superplasticizer and mineral admixtures on mechanical and

durability properties of geopolymer concrete: A review
Priyanki Das1 and Prasanta Roy2 and Purnachandra Saha2*(ORCID-0000-0002-2722-1669)
1
M. Tech Student, School of Civil Engineering. KIIT Deemed to be university, Bhubaneswar, Odisha, India
(Email: das.priyanki1995@gmail.com )
2
Faculty, School of Civil Engineering. KIIT Deemed to be university, Bhubaneswar, Odisha, India
(Email: prasanta.royfce@kiit.ac.in, dr.purnasaha@gmail.com) *Corresponding Author
ABSTRACT
Geopolymer concrete is an innovative, typically ceramic and eco-friendly construction material which is mainly
produced with the help of chemical action of inorganic molecules and a viable alternative to traditional Portland
Cement Concrete used in civil construction industry. This type of geopolymer concrete is starting to revolutionize
concrete. It has significant prospective towards numerous structural applications. It also has rapid strength gain and
cures very quickly, has high tensile strength, very low creep and shrinkage, reduces the demand of OPC which leads
CO2 emission. The main objective of the paper is to review the effects of superplasticizers and mineral admixtures
on mechanical and durability properties of geopolymer concrete. Various types of superplasticizers used in
geopolymer concrete are sulphonate melamine formaldehyde condensate (SMF), Sulphonate naphthalene
formaldehyde (SNF), Lignosulfonate (LS) and Polycarboxylic Ether (PCE) respectively. Different types of mineral
admixtures such as Rice Husk Ash (RHA), GGBS, Silica Fume and recycled aggregates like concrete, mortar,
recycled glass, pumice are used in geopolymer concrete. For improving the workability of fresh concrete and to
reduce the highest range of water, a commercially available naphthalene based superplasticizer, naphthalene
sulphonate based superplasticizer can be used respectively. Melamine and modified Polycarboxylate
Superplasticizers are also used, whereas modified Polycarboxylate based superplasticizers is the most effective type
at the time of using multi-compound activators. The performance of fly-ash based geopolymer concrete
supplemented by means of mineral admixtures are reviewed and it is observed that GGBS and RHA impart
substantial strength to geopolymer concrete at ambient temperature and Silica Fume improves the durability of
geopolymer concrete.
Keywords: Geopolymer Concrete (GC), Superplasticizers (SPs), Mineral Admixtures, GGBS, Rice Husk Ash
(RHA), Fly Ash (FA).
1. Introduction
Geopolymer is stated by the means of the newest trend of adhesive, afterward gypsum cement and OPC. It can be
referred as one of the major construction material and an amorphous alkali alum inosilicate or alkali cements. With
the use of polymerization process, the aluminosilicates such as fly ash, metakaolin, slag, rice husk ash geopolymer
concrete can be produced through activation using alkaline solution [1]. The economical and the environmental
benefits of using geopolymer materials are because of the utilization of wastes from energy manufacture, out of
which 80% are not utilized and have to be filled in land[2]. Superplasticizers are of different types , among all the
various types of SPs , Naphthalene based and Polycarboxilate based SPs are most essential and usable. To deal with
different drawbacks regarding mechanical properties of concrete SPs used. Polycarboxilate based SPs mixed with
Class F and Class C fly ash both, are more effective than naphthalene based SPs [3]. the disposal process of waste
materials is one of the most effective and efficient matter in our country. The applications of solid waste materials
present in the construction industry are used nowadays as it presents as a usable construction materials. Concrete
technology usually establishes suggestions for recycling industrial wastes like FA and GGBFS.To upgrade the
122
utilization of such waste materials like rice husk ash, ground granulated blast furnace slag ,fly ash and various kinds
of SPs and its acknowledgment as a manageable development material, the examination of mechanical and
durability properties are important which will help pick up certainty with respect to its application and lead us all
together more like a perfect sheltered and economical answer for our requirement for green framework and will
reduce the high emission of Carbon-di-oxide in the environment. This review is targeted to contribute an all-
encompassing understanding and assessment on geopolymer concrete [1, 4]. The objective of this study is to review
the mechanical and durability properties of the geopolymer concrete by using various types of superplasticizers,
recycled aggregates and mineral admixtures.
2. Physical and Chemical Properties
The physical and chemical properties of various materials reported by researches are summarized in table 1 and 2.
Table 1Physical properties of various materials
Material Specific Specific Absorption Bulk Fineness

gravity(g/cm surface area (%) density(t/ modulus
2) (m2/g) m3)
Fly ash[5-7] 2.22-2.68 0.3217
GGBS[8-9] 2.89-2.90 1.3128
RHA[5,10] 2.06-2.07 4.091-22.5
River sand[11-12] 2.6-2.63 1.07-3 1.64 2.6
Limestone[11] 2.65 0.61 6.0
Alccofine [7] 2.7 1.2 0.7495
OPC[6,10,13] 3.12-3.15 0.31
Recycled concrete[11] 2.30 5.97 6.0
Fine& Coarse aggregate 2.69-2.73 0.62-0.79 1.4176- 3.54-8.72
[8] 1.7318
Gravels (10mm-20mm) 2.4-2.5 0.17-0.87 2.7-2.8
[14]
Table 2 Chemical compositions of various materials used as mineral admixtues
Material SiO2 Al2O3 Fe2O3 CaO MgO SO3 Na2O K2O LOI
OPC[6,13] 20.80- 4.69- 3.40- 64.46- 0.10- 1.50- 0.10- 0.40-
21.87 4.70 4.27 65.30 0.98 2.69 0.28 0.76
High calcium and 36.02- 20.58- 5.07- 1.29- 0.97- 0.09 1.0 0.07-
low calcium Fly 57.9 31.11 15.91 18.75 2.24 0.8
ash[15-16]
GGBS[9,17] 31.52- 12.22- 0.84- 37.34- 4.62 3.24 0.21 0.33 0.25-
36.25 18.63 1.14 44.53 0.79
RHA[5,10] 91.15- 0.40- 0.21- 0.41- 0.45- 0.045- 0.05- 0.850- 0.45-
92.30 0.41 0.450 0.700 0.850 0.62 0.700 6.25 3.150
Metakaolin(MK)[11] 47.8 37.1 1.48 1.77 0.76 1.68 5.40
Alccofine[7] 35.30 21.40 1.20 32.20 6.20 0.13
Kaolite[17] 45.14 33.32 11.99 4.13 1.37 0.48 0.07 0.13 0.41
Silica fume(SF)[15] 88.30 1.17 4.76 0.48 1.05
123
3. Fresh Properties:-
3.1 Workability
Workability, known as the fresh property that concludes the simplicity and consistency through which the material is
mixed, compacted, transported and placed. It can be described as consistency measure. According to ASTM C1611
by using slump flow method the fresh properties of geopolymer concrete were measured [11]. The content of
sodium silicate enhances the workability by increasing the alkaline activator ratio which contains water
concentration of sodium hydroxide solution of 56%, not only which also fluctuates the workability. It was detected
that slump value decreased by the increasing value of molarity [9].
3.1.1 Effects of superplasticizers on workability of geopolymer concrete:-
Superplasticizer known as high range water reducer, a concrete additive which used to improve workability without
affecting water cement ratio. As compared to Portland Cement Paste, Superplasticizer is less effective in
Geopolymer Concrete [3]. Slump value enhances with the increasing percentage of Superplasticizer [18].
Polycarboxilate based Superplasticizers gives more slump value than Naphthalene-based Superplasticizers [19]. It
was observed by different experimental results with the help of visual observations and quantitative analysis mix
samples of Superplasticizers containing 3%, 4%, 5% respectively fails in exhibiting the necessary workability due to
inadequate quantity of superplasticizer which makes the mixtures less workable however, mix samples of
Superplasticizers with 6%, 7% respectively consist of the desired fresh properties along with the workability
properties that are within the EFNARC limits of SCC.
3.1.2 Effects of mineral admixtures on workability of geopolymer concrete:-
Quantity of fly ash influences the slump value. It is quiet inconvenient to adsorb and bond the anionic
superplasticizer molecules as the fly ash particles are neighboring to Silica Hydroxide and Aluminum Hydroxide
instead of Calcium cations. Naphthalene based superplasticizer, which pH value is 6.0–9.0 where polycarboxylate
based superplasticizer generates a lower pH value of 4.8–6. Among the various chemical compositions of the two fly
ashes it was observed that polycarboxylate-based superplasticizer performs better for Class C fly ash as compared to
fly ash of Class F [3]. It has been observed that 100% garnet mix in geopolymer concrete increased the slump value
of 29% [12]. All mixes of workability is within the acceptable limits of EFNARC. When 100%Fly ash and
100%GGBFS used the slump value increases by 720mm and 710mm respectively. Increase in the percentage of
RHA from 5% to 25% it decreases the slump value of concrete as compared to 15%RHA.Metakaolin is a valuable
admixture for concrete structures, but the effect of incorporating metakaolin in workability of concrete, the slump
value of concrete decreases with increasing the percentage of metakaolin content[11]. The increase in GGBS of fly
ash based geopolymer concrete minimizes the workability [20].
124
Effect on slump flow value with respect to percentage of binders

800
700
slump flow value(mm)
slump flow value(mm)
600
500
400
300
200
100
0
0%MK 10%MK 10%Slag 20%Slag 20%MK 30%Slag 30%MK
with RA
%of binders
Fig (a):-Effect on slump flow value with respect to various percentage of binders [3, 11, 20].
4. Mechanical Properties:-
4.1 Compressive strength:-
Compressive strength is the most effective property for assessing the execution and quality control of all various
kinds of concrete, which impacts the quality, mechanical, durability and structural performance of concrete. The
compressive strength of geopolymer concrete has been greatly influenced by different types of Superplasticizers,
recycled aggregates and mineral admixtures.
4.1.1 Effects of superplasticizers on compressive strength of geopolymer concrete:-
Superplasticizers are required to upgrade the hardened compressive quality of geopolymer concrete. As hardening is
an endothermic procedure so here heat should be given to the concrete specimens during hardening process to
facilitate the process.Napthalene based superplasticizer has negative impact on compressive strength with reference
to those pastes without using any Naphthalene. So ultimately it has been observed compressive strength of pastes
with using various types of Superplasticizer was decreased as compared to the pastes without using any
Superplasticizer. Compressive strength was decreased from 29%, 15%, 16%, 42%, 51% and 43% for
Polycarboxylate, Naphthalene and Melamine based Superplasticizer respectively with respect to those pastes
without using any Superplasticizer[21].When concrete specimens are prepared with various types of superplasticizer
dosage namely 3%, 4%, 5%, 6% and 7% and being cured at temperature of 600C, 700C, 800C and 900C, it was
concluded that 7% gives the maximum compressive strength around 53.80MPa after 28days of testing as compared
to other Superplasticizer dosage, whereas 6% at curing temperature of 700C gives compressive strength of about
51.52MPa.Compressive strength decreased by use of 2% Superplasticizer in Ordinary Portland Cement based
concrete to improve the consistency [19].The content of superplasticizers when increased, a reduction in strength
was observed, but without addition of any extra water with alkaline activator, the increase of strength was
significant. Although, the action of all this additives on reaction mechanism of geopolymer concrete is still not so
much distinct, especially in some cases while Superplasticizer increased strength reduced, so to define the actual
strength capacity further study and research are required [22].
125
4.1.2 Effects of mineral admixtures on compressive strength of geopolymer concrete:-
Compressive strength, the supreme effective hardening characteristics of geopolymer concrete. Properties of
geopolymer concrete depends on several factors such as Rice husk ash, Ground granulated blast furnace slag, Silica
fume, Fly ash/kaolin and many other kind of mineral admixtures. At ambient curing to achieve maximum
compressive strength 5%RHA content kept as the optimum level and by the replacement of Rice husk ash beyond
5% strength decreased. Similarly, fly ash replacement by Ordinary portland cement upto 20% and beyond that,
strength decreased [13]. Increased fly ash content results improving the compressive strength parameter as well as
increased quantity of binder material [7]. Due to incomplete geopolymerization process without heat it was observed
that when geopolymer concrete was blended with 100%flyash it fails to achieve the desired strength at 3, 7 and 28
days of curing [1]. In a long term study, it was reported that mix of fly ash and glass powder where weight of glass
powder used up to 0% to 15% strength increases after 15% it starts decreasing, another mix of glass powder and
Ground granulated blast furnace slag where weight of Ground granulated blast furnace slag used 5-15% compressive
strength increases after 15% it starts decreasing, at 15% for both the two cases above, compressive strength is
maximum [23]. According to some researches it was concluded that on geopolymerization process by replacing fly
ash with 50% Kaolin displayed a constructive result and the compressive strength was detected much higher with
respect to the concrete obtained from Ordinary Portland cement. Alternatively, the compressive strength was
established the lower with reference to that of concrete attained from Ordinary Portland cement when Fly ash was
replaced by 10% Kaolin [24]. Therefore, it was also concluded that addition of silica fume up to 40% and also by
increasing amount of GGBS, compressive strength increased [23,25].
Effect of various SPs and mineral admixtures on Compressive Strength

Ccompressive Strength(MPa)
80
70
60
50
40
30
20
10
0
% of binders
Fig (b):-Effect of superplasticizers and mineral admixtures on compressive strength of geopolymer concrete
[5,19,23,25].
126
4.2 Flexural strength:-
4.2.1 Effects of mineral admixtures on flexural strength of geopolymer concrete:-
Flexural strength, the most common and effective factor on the structural performance and hardened properties of
concrete. It depends on water binder ratio. Increase in the amount of percentage replacement present in fly ash with
silica fume results in the increase in flexural strength [25]. When excessive amount of Rice husk ash replaced
instead of different cementitious material, nevertheless, this reduces the strength as deficiency in the strength caused
for excessively leaching out of silica which effect on the mechanical properties of geopolymers. With reference to
other ratios of Si/Al (1.7, 2.05, 2.41 and 3.6), when the Si/Al ratio is equals to 2.8 gives the higher strength value
[10]. However, it was reported that at 5% replacement of Rice husk ash as related just before at all ages of control
achieves higher value of flexural strength, beyond 5% it starts reducing. It was also illustrated that for dilution of
SiO2 particles of binders in the alkaline solutions at ambient temperature was insufficient which weakens the
geopolymerization process and the geopolymer matrix [5]. Therefore, it was concluded that flexural strength
increased with the increasing percentage of Ground granulated slag, Ordinary Portland cement and calcium
hydroxide used with fly ash, it has also been observed that mixtures having 6% Ordinary Portland cement, 2%
Calcium hydroxide and extra water attained to some extent of flexural strength as compared to geopolymer (A 35
S00) without extra water [22].
9
8
7
Flexural Strength
6
5
4
3
2
1
0
Flexural Strength(MPa) Si/Al ratio
Fig (c):-Effect of various ratios of silica and aluminum on flexural strength of concrete [22].
127
4.3 Split tensile strength:-
4.3.1 Effects of superplasticizers on split tensile strength of geopolymer concrete:-
Split tensile test is one of the most popular methods for evaluation of tensile strength. This one remains an
essential mechanical property, utilized as various project features of concrete constructions, for example, individuals
identified with inception in addition to proliferation of splits, shear, then anchorage of strengthening steel in
concrete [26].To determine split tensile strength of geopolymer concrete least bearing area 150*150*150mm size
cubes and 12*150mm area were provided to give the load over the whole length [7]. Further, a significant increment
on tensile strength was found when concrete mixes with superplasticizers as compared to the mixes of concrete
without using any superplasticizers [23].
4.3.2 Effects of mineral admixtures on split tensile strength of geopolymer concrete:-
On splitting tensile strength of geopolymer concrete different types of mineral admixtures has incredible impact.
Several factors used to improve the indirect tensile strength of concrete are generally named as Ground granulated
blast furnace slag, Rice husk ash, Fly ash, Silica fume and many other admixtures. It was investigated that 100%
Ground granulated blast furnace slag achieved the maximum split tensile strength on geopolymer concrete at all
ages of curing. Similarly, mixes with 5% Rice husk ash also develop the splitting tensile strength of concrete
structures; beyond 5% increasing contents of Rice husk ash reduces the strength of geopolymer concrete. In few
researches it was resulted that at 5% Rice husk ash mix increasing percentage on split tensile strength values were
such as 208%, 121.67% and 105.04% at 3, 7 and 28 days of ambient curing. When 100% fly ash used in the mixes
very less amount of indirect tensile strength was achieved. The bond between Silica and Aluminum monomers
achieved from Fly Ash particles was unbreakable which begins the reaction process and also gains the split tensile
strength blended with Fly ash on ambient temperature [5]. Rather than all the control mix replacement percentage of
fly ash with silica fume increases the split tensile strength [25]. When Geopolymer concrete with 8M, 12M, and
16M were compared substantial augmentation remained. Consequently, split tensile strength is improved by
intensification of Fly Ash particles and molarity of the above solution. Further, addition of alccofines increases the
split tensile strength increases [7].
4.3.3 Effects of recycled aggregates on split tensile strength of geopolymer concrete:-
Generally, an indirect method is used to determine the split tensile strength of geopolymer concrete. In comparison
with OPC concrete containing 50% recycled aggregates the reduction in indirect tensile strength is high with
reference to geopolymer concrete containing 50% recycled aggregates. Therefore, it was observed that geopolymer
concrete which contains 50% of recycled aggregate can be a better bond of the geopolymer binders. Split tensile
strength decreases by excessive intensification of RCA contents [27]. Contracting the above fact, it was also
concluded that during production of light weight geopolymer concrete pumice and crushed clay bricks exhibits
lower splitting tensile strength with reference to individuals of geopolymer concrete having normal aggregates [28].
5 Durability Properties:-
5.1 Sorptivity:-
Sorptivity communicates the impulsion of a material for retaining and transmitting of water and different fluids by
capillarity. When 50% recycled aggregates used the sorptivity of geopolymer concrete increases about 26% which
was observed as lower percentage of increase as compared to Ordinary Portland cement containing the same amount
and type of recycled aggregates. During the manufacturing process of recycled aggregates crack and fissures were
formed which contributes to increased sorptivitty of geopolymer concrete [27]. Furthermore, an increase of about
128
26% to 102% in sorptivity of geopolymer concrete containing natural and recycled aggregates occurred due to the
porous nature of the recycled aggregates. In few researches it was observed that recycled aggregate concrete using
metakaolin has higher sorptivity than concretes containing natural aggregates. Therefore, it was resulted that
increase in metakaolin sorptivity decreases [29]. OPC being included as a fly ash replacement, sorptivity was found
to be reduced. It was observed that at all ages, the values of sorptivity decreases up to the OPC content of 20% and
beyond that to an increase on 30% OPC [13]. In an aggressive and adverse environment if the value of sorptivity is
lower it shows that the concrete is performing well and it’s more durable. It was investigated that with consideration
of 5%RHA mix the lowest value of sorptivity was found and the value was 0.069. When replacement of RHA was
done with the considered percentage of 15% and 25%, the values of sorptivity observed were higher one; as a result
of moderate presence of Silicon-di-oxide this impacted the improvement of heavier aluminosilicate in addition of C-
S-H gel. Therefore, the maximum sorptivity was measured, by the control blends with the 100% Fly Ash content.
Sorptivity of geopolymer concrete depends upon the absorption of water, they are co-related, and it was found that
sorptivity and absorption of water are directly proportional to each other. If the absorption of water will be the less,
then sorptivity will also be the lesser, which helps to break the continuity of pores present in the concrete [5].
5.2 Acid attack:-
5.2.1 Effect of addition of 3% sulfuric acid solution on geopolymer concrete:-
Geopolymer Concrete cubes of dimension (100*100*100) mm were tested for sulfuric acid attack as per guidelines
of ASTM C267. The cubes were cured for 7 days and exposed to 3% sulfuric acid solution instead of going for the
conventional 28 days curing period for hydraulic cement speeding up the procedure. Change in weight of the test
specimens were noted at an interval of 7, 14, 28, 56, 84 & 120 days. Freshly prepared sulfuric acid solution was
replaced against the previous solution for each successive test that was performed [11]. Geopolymer Concrete
shows insignificant weight change results after 28 days while after immersing for 56 days weight of the particles
start to decrease remarkably. 10-OPC & 15-OPC cubes portrayed almost the same weight losses than 5-OPC which
proved to be having the lowest resistance to sulfuric acid attack. Loss in weight was lesser in CON-R concrete than
those concretes blended with RAGCs. Partially replaced OPC geopolymers with fly ash provides poor resistance to
acid corrosion due to C-A-S-H gel formation. Susceptibility in acid attack was more in Ca-rich gel than
aluminosilicate gel or N-A-S-H gel. 84 days after immersing in acid solution, 5-OPC, 10-OPC & 15-OPC mixture
cubes showed 25%, 22.4% & 22.6% loss in weight. CON-R concrete loosed 19.3% of weight after 84 days of
immersion. 120 days of immersing of RAGC specimens containing 1%, 2% & 3% nano-SiO2 showed 31%, 32.5%
& 30.3% loss in weight, while CON-R concrete portrayed 28.7% loss in weight. In case of silica fume added
concrete low resistance to acid attack in concrete is due to enormous openings which were fully occupied through
unreacted nS particles and decrease all available space intended for extensive response by increase of inner
compression, which causing disintegration of concrete [30-31]. 20 MK-L concrete specimens showed 0.5%, 0.9%,
1.0%, 5.2%, 11.7%, and 16.3% loss in weight for 7, 14, 28, 56, 84, and 120 days respectively. Sulfuric acid
resistance of NAGC was more than RAGC while both had the same MK content. Loss in weight in concrete is a
reason for presence of CaO in HCF which effects the development of Calcium Hydroxide, CASAH & Al-substituted
CAAASAH gel, which has been percolated with development in acid attack. Loss in weight for 30 MK-L samples
when exposed for 120 days was found to be more than that of 0 MK-L cubes. Reason behind this can be MK
particles acting as a filler of the pores reducing chances of entry of expansive products. 10 MK-L & 20 MK-L
samples showed high resistance towards attack apart from 0 MK-L samples.
5.2.2 Effect of addition of 5% sulfuric acid solution on geopolymer concrete:-
MK geopolymer contents portrayed a loss in mass of 7% when the contents were exposed to 5% sulfuric acid for 30
days. Fly ash-based geopolymer’s resistance to acid and alkaline attack depends on its chemical composition. After
exposing the geopolymer contents blended with Fly Ash into 5% of acetic acid and 5%of H 2SO4 solutions, it turned
out more superior as compared to OPC content. Furthermore, due to exposing the particles for duration of 12months,
reduction in strength demonstrated around 33% which is less as compared to OPC concrete. Compressive strength of
129
FA-GPC reduced to 20% when samples were exposed to 0.5% H2SO4 solution after getting exposed to 12 months.
Degradation in geopolymer matrix and the aggregates which lead to opposing and destruction happening on surface
of concrete decreases the strengthening power of concrete cubes. This correspondingly proves that GPC has more
resistance to acid than OPC. After exposing of GPC blended with palm oil fuel ash & pulverized fuel ash to 2%
sulfuric acid solution for 18 months, it shows a weight loss of 8% whereas OPC concrete shows a loss of 20% in
weight in the same conditions. Strength reduction of OPC content was found to be 68% and for GPC 35% [32].Bio-
additives mixed with geopolymer specimens appeared to slightly corrode the surface and less damage around cube
edges, less loss in weight and better acid resistance.
6 CONCLUSION:-
Behavior of geopolymer concrete under different types of super-plasticizers and mineral admixtures are discussed in
this study. Following observations are made from this study:
1. 6-7% of superplastisizer gives the maximum compressive strength as compared to other Super-plasticizer dosage,
2. As the Si/Al ratio increases, the flexural strength of geopolymer concrete increase.
3. Performance of geopolymer concrete in acid sorptivity test is better than conventional concrete
7 REFERENCES:-
[1] Awanga Z A, Maa C.K, Omar W (2018) “Structural and material performance of geopolymer concrete”. Construction and
Building Materials, 186: pp 90–102.
[2] Bakharev.T (2005) “Durability of geopolymer materials in sodium and Magnesium sulfate solutions”. Cement and Concrete
Research, 35: pp 1233– 1246.
[3] Xie J, Kayal O (2016) “Effect of superplasticizer on workability enhancement of Class F and Class C fly ash-based
geopolymers”. Construction and Building Materials, 122: pp 36-42.
[4] Ozakca M, Their Musdif J (2018) “Developing geopolymer concrete by using cold-bonded fly ash Aggregate, Nano-silica,
and steel fiber”. Construction and Building Materials, 180: pp 12-22.
[5] Shah Patel J Y (2018) “Enhancement of the properties of Ground Granulated Blast Furnace Slag based Self Compacting
Geopolymer Concrete by incorporating Rice Husk Ash”. Construction and Building Materials, 171: pp 654-662.
[6] Posi P, Thongjapo P, Thamultree N, Boontee P (2016) “Pressed lightweight fly ash-OPC geopolymer concrete containing
recycled lightweight concrete aggregate”. Construction and Building Materials, 127: pp 450-456.
[7] Parveen,Singhal D,Junaid M.T, Jindal B.B,Mehta A (2018) “Mechanical and microstructural properties of fly ash based
geopolymer concrete incorporating alccofine at ambient curing”. Construction and Building Materials, 180: pp 298-307.
[8] Pandurangana K,Thennavanb M,Muthadhic A (2018) “Studies on Effect of Source of Fly ash on the Bond Strength of
Geopolymer Concrete”. Proceedings, 5: pp 12725-12733.
[9] V.K.Nagaraj, Venkatesh Babu D.L (2018) “Assessing the performance of molarity and alkaline activator ratio on engineering
properties of self-compacting alkaline activated concrete at ambient temperature. Journal of building engineering.
[10] Zabihi M, Tavakoli H, Mohseni E (2018) “Engineering and Microstructural Properties of Fiber-Reinforced Rice Husk–Ash
Based Geopolymer Concrete”. Journal of Materials in Civil Engineering.
[11] Nuaklong P, Sata V, Chindaprasirt P (2018) “Properties of metakaolin-high calcium fly ash geopolymer concrete containing
recycled aggregate from crushed concrete specimens”. Construction and Building Materials, 161: pp 365-373.
[12] Muttashara L.H, Ariffinb Mohd A Mohd, Husseinc Noori M (2018) “Self-compacting geopolymer concrete with spend
garnet as sand Replacement”. Journal of Building Engineering, 15: pp 85-94.
[13] Mehta A, Siddique R (2017) “Properties of low-calcium fly ash based geopolymer concrete incorporating OPC as partial
replacement of fly ash”, Construction and Building Materials, 150: pp 792–807.
[14] Singh N.B, Durgprasad .J, Okoye N.F (2015) “Mechanical properties of alkali activated fly ash/Kaolin based geopolymer
concrete”. Construction and Building Materials 98: pp 685-691.
[15] Sukontasukkul P, Pongsopha P, Chindaprasirt P, Songpiriyakij S (2018) “Flexural performance and toughness of hybrid
steel and polypropylene fiber reinforced geopolymer”. Construction and Building Materials 161: pp 37-44.
[16] Shehab K. H (Ph. D.), Eisa. S. A (Ph.D.) (2016) “Mechanical properties of fly ash based geopolymer concrete with full and
partial cement replacement”. Construction and Building Materials, 126: pp 560-565.
[17] Akbarnezhad. A, Huan. M, Mesgari. S, Castel.A (2015) “Recycling of geopolymer concrete”. Construction and Building
Materials, 101: pp 152-158.
130
[18] Nuruddin Fadhil M, Demie S, Ahmed Fareed M, Shafiq N (2011) “Effect of Superplasticizer and NaOH Molarity on
Workability, Compressive Strength and Microstructure Properties of Self-Compacting
Geopolymer Concrete”. International Scholarly and Scientific Research & Innovation, 5(3): pp 187-194.
[19] Demie S, Nuruddin Fadhil M, Shafiq N (2013) “Effects of micro-structure characteristics of interfacial transition zone on the
compressive strength of self-compacting geopolymer concrete”. Construction and Building Materials, 41: pp 91-98.
[20] NathP,Sarker Kumar P (2014) “Effect of GGBFS on setting, workability and early strength properties of fly ash geopolymer
concrete cured in ambient condition”. Construction and Building Materials, 66: pp 163-171.
[21] Nematollahi B, Sanjayan J (2014) “Effect of different superplasticizers and activator combinations on workability and
strength of fly ash based geopolymer”. Materials and Design, 57: pp 667-672.
[22] Nath P, Sarker Kumar P (2017) “Flexural strength and elastic modulus of ambient-cured blended
low-calcium fly ash geopolymer concrete”. Construction and Building Materials, 130: pp 22-31.
[23] Mithanthaya I. R, Marathe Shriram, Rao N,Bhat V(2017) “Influence of superplasticizer on the properties of geopolymer
concrete using industrial wastes”. Materials Today: Proceedings4: pp 9803-9806.
[24] Okoye .N.F, Durgaorasad.J, Singh .B.N (2015) “Mechanical properties of alkali activated fly ash/Kaolin based geopolymer
concrete”. Construction and Building Materials, 98: pp 685-691.
[25] Durgaprasad. J, Okoyea. N. F, Singh. B.N (2016) “Effect of silica fume on the mechanical properties of fly ash based-
geopolymer concrete”. Ceramics International 42: pp 3000-3006.
[26] Naskara S, Kumar A (2016) “Effect of Nano materials in geopolymer concrete”. Perspectives in Science 8: pp 273—275.
[27] Shaikh Ahmed Uddin F (2016) “Mechanical and durability properties of fly ash geopolymer concrete containing recycled
coarse aggregates”. International Journal of Sustainable Built Environment, 5: pp 277-287.
[28] Wongsa A, Sata V, Nuaklong P, Chindaprasirt P,2018, “ Use of crushed clay brick and pumice aggregates in lightweight
geopolymer concrete”, Construction and Building Materials,188: pp 1025-1034.
[29] Nuaklong P, Sata V, Chindaprasirt P (2018) “Properties of metakaolin-high calcium fly ash geopolymer concrete containing
recycled aggregate from crushed concrete specimens”. Construction and Building Materials, 161: pp 365-373.
[30] Nuaklong P, Sata V, Wongsa A, Srinavin K, Chindaprasirt P (2018) “Recycled aggregate high calcium fly ash geopolymer
concrete with inclusion of OPC and nano-SiO2. Construction and Building Materials, 174: pp 244-252.
[31] Nuaklong P, Sata V, Wongsa A, Srinavin K, Chindaprasirt P (2018) “Recycled aggregate high calcium fly ash geopolymer
concrete with inclusion of OPC and nano-SiO2”. Construction and Building Materials, 174: pp 244-252.
[32] Singh. B, Ishwarya. G, Gupta M, Bhattacharyya.S (2015) “Geopolymer Concrete: A review of some developments”.
Construction and Building Materials, 85: pp 78-90.
131
Paper ID- 37
Innovative Optimization Techniques of Time and Cost Trade Off
Paromik Ray1, Rajarshi Patty1, Dillip Kumar Bera2, A.K. Rath2
1
M.Tech Students, School of Civil Engineering, KIIT Deemed to be University, Bhubaneswar, India
(Mail id:-paromikray@gmail.com, patyrajarshi@gmail.com)
2
Faculty, School of Civil Engineering, KIIT Deemed to be University, Bhubaneswar, India
(Mail id:-dkbera.fce@kiit.ac.in), (akrath.fce@kiit.ac.in)
Abstract. Cost and time optimization method is the most effective and time efficient method with highest
achievable performance under specific condition in a construction project. This method is mainly required for cost
and time optimization in a construction project. In this paper the various innovative techniques are being reviewed
which can be used by the construction companies. Genetic Algorithm(GA) as well as Particle Swarm
Optimization(PSO) methods are considered the advanced innovative techniques which are being used
continuously by the construction companies for cost and time optimization. This innovative techniques includes
the advance work of Genetic Algorithm(GA) method in the form of GA with Dev-C++ 4.9.9.2, GA with Line of
Balance(LOB), GA with Modified Adaptive Weight Approach(MAWA), GA with Critical Path Method (CPM)
along with new methods Linear Programming(LP), Non-Linear Integer Programming Model(NLIP), Discounted
Cash Flow Method(DCF), Maximum Flow-Minimal Cut Theory and Artificial Neural Networks Method(ANN).
From this, the GA method is classified into two where the global parallel GA method provides more effectiveness
and efficiency than coarse-grained parallel GA method. The Non Linear Integer Programming (NLIP) method and
Line of Balance(LOB) with GA method both have an efficient and optimum solution for time cost trade off
problem, along with PSO method which is best for Pareto-compromise solution and Direct Cash Flow(DCF)
method which optimizes cost and time within the project boundaries. Finally it is observed GA along with its
advanced parameters, ANN method and NLP technique are better for solving time cost trade off problems.
Keywords: Cost and Time Trade off, Cost and Time Optimization, Innovative Techniques, Project
Management.
1 Introduction
According to today’s market driven condition, construction industries are growing rapidly. Construction
companies including private and government sectors faces numerous challenges at the time of construction. The
utmost difficulties faced by the companies is to finish or complete the project within scheduled time and estimated
budget. At the time of construction, cost and time of different activities has faced consequential changes owing to
unpredictable factors such as price escalation, labour performance, labour strike, economical and social issues,
design errors, execution errors of contractors, climatic changes etc. These factors are mainly responsible for the
delay of duration and addition of extra cost liabilities in the budget of the project. This type of problem is usually
known as Time Cost Trade off(TCO). To overcome these situations, many project planners and researchers are
132
still encountering to find the best possible solution of time cost trade off problem. As a result, several innovative
techniques and models are being developed for finding the optimal solution.
Genetic Algorithm (GA) method, Advanced Neural Network (ANN) method, Critical Path Method, Line of
Balance (LOB) method, Modified Adaptive Weight Approach(MAWA) Method, Discounted Cash Flow(DCF)
Method, Maximum Flow- Minimal Cut Theory Method are the advanced and widely used techniques which
generates the most optimal outcome for construction TCO, especially for large and complex projects. The current
work represents the review of the various innovative techniques that has been used in different construction project
for optimizing TCO problem. The main objective of the paper is to overview the various work done by the
researchers along with the applications of different techniques used in the field of construction industry for solving
the TCO problems. Initially, the various innovative techniques processes has been discussed followed by the
various applications used of individual techniques in the field of construction industry for optimizing cost and
duration. Moreover, an overview of the procedure of the techniques has also been included to highlight the
working process of optimization. Further, a critical review of the work done has also been done for the readers to
concentrate on the development of the various processes for future work.
1.1 Brief classification of innovative techniques
The below table1 represents the different classification of optimization techniques used for cost and time
optimization process. These methods are derived from various papers as used for various types of application of
constructions.
Different Optimization Methods

Manual Methods Mathematical Technique Metaheuristic Approach Others
 Maximum Flow  Non-Linear  Evolutionary  Artificial Neural

Minimal Cut Theory Programming (NLP) Algorithm(EA) Network (ANN)
(MFMC)  Linear Programming I. Particle Swarm  Discounted Cash Flow
 Least-Cost (LP) Optimization (PSO) Method (DCF)
Scheduling Model  Integer Programming II. Ant Colony
 Chance-Constraint (IP) Optimization(ACO)
Programming Model  Dynamic Programming III. Modified Shuffled frog
etc. (DP) Leaping(SFL)etc.
 Line of Balance  Inventory Models  Genetic Algorithm(GA)
Method (LOB)  Stochastic Models  Genetic
 Simulation Model etc. Programming(GP) etc.
Table 1. Classification of Different Optimization Techniques
2 Various Innovative Techniques In Optimization Process
Optimization is defined as the process of finding the maximum / minimum values of an objective function to
satisfy a particular constraint specified within a certain range of set. The main goal of optimization process is to
yield the best value of function present within a range of variables by formulating some mathematical
programming problems. The Time Cost Trade off depends on the Critical Path Method(CPM) and it has been
used for a long period of time. In a construction project, increase in cost, increase in scheduled time, presence of
risks, unavailability of resources and quality are the main reasons for the optimization process. Various innovative
techniques are being used for finding the optimize solution in an optimization problem along with various
parameters are optimized including cost, time, risk, resources, quality etc. in a construction project. The different
types of techniques includes Manual techniques, Mathematical techniques and metaheuristic method for
133
optimization of cost and time of a project. Lastly, the presence of uncertainty in some optimization techniques or
control decision making of a project, the fuzzy logic was developed.
2.1 Manual methods
In this method, the main objective is to find the optimum time-cost point between normal activity and crashing
activity. This method is required for performing each activity of alternative plans where a time cost curve is
plotted based upon the relationship between activity duration and direct cost [1]. Maximum flow Minimal Cut
Theory is one type of manual method which has been used for scheduling and optimizing the construction project.
This theory states that “in the network from start to end the maximal flow is equal to the minimum cut set
capacity”. This technique can be used for shortening the total duration of the project and time-cost optimization[2].
Least- cost scheduling is another manual optimization technique where different cost and time related activities
has been considered for range estimating and probabilistic scheduling and the data is analyzed in a statistical
method for obtaining the optimal project cost and duration[3]. Similarly, chance constraint method is another
manual optimization technique where a programming model has been developed which comprise of varieties of
funding further transforming into a similar deterministic approach at a presumed confidence level. This
deterministic approach is integrated in the optimization process of time-cost trade off problem[4].
2.2 Mathematical Optimization Technique
Mathematical Optimization Technique is one kind of programming method which converts the Time-Cost Trade-
Off (TCO) problems to mathematical programmed models. This mathematical approach solves the decision
making problems by utilizing linear programming, integer programming, non-linear programming, Dynamic
Programming, Stochastic Programming, Simulation Programming[5]. This approach assume that the relationship
between time cost is either linear/ non-linear, discrete/continuous, fixed/ not-fixed, or hybrid[6]. The main three
techniques of mathematical optimization process used in construction sectors for optimization process are:-
2.2.1 Linear Programming
This technique was developed by Leonid Kantorovich in 1939 for finding the maximum and minimum output of a
function. Apart from TCTO optimizing solution, it also can be used for scheduling, production, transportation and
distribution, inventory management, telecommunication [7]. This method consist of linear relations between
variables which should be linear and continuous subjected to linear constraints. Linear programming problems can
be solved by graphical approach, simplex approach, transportation approach, assignment approach where the
simplex approach is mostly preferred. For Optimization process, an exact and an approximate method has been
developed, the exact method consist of the linear programming method which generate an project time-cost curve
and minimum cost schedule in relation with all the time-cost activities. The result of the method specifies the
approximate method which gives an accuracy surpassing 99% of cost-time trade off[8].
2.2.2 Non-Linear Programming
An objective function subjected to non-linear constraint demonstrates a mathematical problem used for solving
equality and inequality systems of maximize/minimize problems refer to the non-linear programming method. This
method is used to analyze the solutions which include local minima and local maxima present in the feasible region
along with the boundary region where the optimal solution is considered. The widely used varieties of application
such as resource allocation, production, planning, data analysis etc. [6].
2.2.3 Integer Programming
134
Integer Programming is a mathematical optimization technique where one or more decision variables are taken
into integer variables in the ultimate solution for obtaining optimal solution. This method has various applications
such as time cost optimization, capital budgeting, scheduling of the project etc.
Many researchers have used different types of programming models of integer type for solving TCTO problems. A
NLIP programming model which promotes a relationship constraints between project activities, activity duration,
project duration and crashing method of both time and cost of a project. The results obtained suggest minimum
computational time at a lower cost for optimization process and can be accessed by project managers for large
network projects[9].A hybrid method is developed using linear and integer programming to find an efficient
optimal solution for TCTO problem of construction project[10]. Another model was developed using linear
programming technique for solving the TCTO problems consisting with overlapping precedence networks.
Moreover, this model is applicable for vast network of multi-objective projects[11].
2.2.4 Non-Linear Integer Programming Method
Non-Linear Integer Programming Method is a basic version of general non-linear integer problems where all
objective functions subjected to non-linear or linear constraints. In this method, an NLP optimization model is
formed which is used for generating the optimized decision of time and cost in project management. This
optimized model enables a direct relationship with the project duration and total project cost of a project. Since the
Linear Programming optimization method can only handle linear relations between the variables, hence a non-
linear optimization model have been developed to act as a discrete relationship between the variables of linear
functions. A case study has been made in which NLIP technique is used for generating best optimal time cost
value. Result indicates that using NLIP technique shows better results when compared with for TCTO problems.
Moreover it is also suggested by using meta-heuristic techniques can compute ideal optimal solutions [12]. A NLP
model based on precedence relationship of project time span activities has been developed for obtaining the
optimal solution of time and cost[13].
2.3 Metaheuristic Optimization Techniques
Metaheuristic optimization technique is one type of optimization method based upon metaheuristic algorithms to
find or generate or select a heuristic that provides a good solution. It is mainly classified into three types which
includes EA method, GA method and GP method. The Evolutionary Algorithm(EA) method further is
distinguished to Particle Swarm Optimization(PSO) method, Ant Colony Optimization(ACO) method, Modified
Shuffled Frog Leaping(SFL) method which are used to find best possible time-cost solution and decision making
problem solution[6].
2.3.1 Particle Swarm Optimization Technique (PSO)
Particle Swarm is one type of stochastic optimization technique based on population which was developed by Dr.
Eberhart and Dr. Kennedy in 1995. The concept of PSO technique comprises of, at each time steps, the each
particle velocity is changed towards its local location. A multi-attribute utility function model has been developed
based on inter-related linear programming function. An algorithm based on improved PSO has been used to obtain
the optimal solution of TCTO problem[14]. For decision making process, contractual time, cost and quality used
as a benchmark for assessing of the optimal solution. Finally, a new algorithm is developed by the combined
process of an immunity general algorithm with a construction factor particle swarm optimization(IGPSO). The
optimized results signify that IGPSO creates a better efficiency of converging and easily can overcome premature
status of the project[15].
135
2.3.2 Ant Colony Optimization Technique (ACO)
Ant Colony Optimization(ACO) is a population based probabilistic technique searching for optimal solution to
difficult optimization problems. Ant Colony optimization approach was used to solve multi-objective TCTO
problems to generate best solution without much requirement of computational resources to help project managers
and planners in decision making process[16]. A new approach for the discrete TCTO problem has been developed
in which activities have multi-objective discrete cost function whereas activity durations are normally distributed.
Finally, ACO has been applied to the TCTO model for measuring overall efficiency. The result obtaining
demonstrates the improvement of the project completion coming from the initial uncertain value to the maximum
possible value[17].
2.3.3 Modern Shuffled Frog Leaping Model Optimization Method (SFL)
This optimization technique is one type of metaheuristic optimization technique which is a repetitive technique
originated from a group of frogs searching for food in a pond of same memetic evolution. This method has been
developed for solving combinatorial optimization problems. A multiobjective shuffled frog leaping model has
been formed to solve complex TCTO problems where three important objective functions has been considered
simultaneously in project planning. It is found that the proposed shuffled frog leaping model generating a better
project optimal solution with minimum total project cost, less total project duration[18].
2.4. Genetic Algorithm
The Genetic Algorithm belongs to the heuristic techniques which is an advanced method uses non-traditional
search algorithms for solving single as well as multi-objective optimization problems and it is mainly based on
natural evolution. At each going step, this method select values randomly and iterates further to give an optimal
value. The accurate solution of the problem using GA method is dependent on the fitness level and the number of
populations developed[19].This method consists of four main processes- Initialization, Selection, Crossover,
Mutation. These four main processes are mainly involved for optimization as follows:-
Initialization: - It is the process where the values from the fitness function is initialized.
Selection: - It is the process in which the optimized values are selected randomly. It is often referred as Random
Selection Process.
Crossover: - It is the process where the random values interchange their positions to generate the closet optimized
value.
Mutation: - It is the process where one or more than one random values are swapped or flipped to generate higher
optimized value. This process mainly occurs after Crossover process.
Various other processes along with Genetic Algorithm method works together to generate a higher optimized
value. These processes are as follows:-
2.4.1 Dev-C++ 4.9.9.2
Dev- C++ 4.9.9.2 version is a fully integrated developed programming software used in C and C++ programming
languages. This software along with the parameters of Genetic Algorithm is used as a technique for optimization
of cost and time in a construction project. This process involves the use of Genetic Algorithm (GA) method as a
searching tool, along with the present of different combination of GA parameters for finding the perfect values of
those parameters. In this method, the software Dev-C++ 4.9.9.2 has been applied as a code solver[20].
2.4.2 Line Of Balance Method
136
Line of Balance is a technique used for representing the repetitive works of a project which are represented in a
single line on a graph[21]. It also incorporates a sequence of inclined lines to represent the working rate between
the repetitive activities in any project[22]. Line of Balance Method with the presence of Genetic Algorithm
generate mathematical operations which are essential for optimizing the cost and time in a construction project.
The process involves the use of genetic algorithm in Line of Balance Method for optimizing in a repetitive
construction project. This process mainly consist of multi-objective Genetic Algorithm which is essential for
construction of multi-storey building. A separate software Evolver 4.0 is used which act as an addition program to
MS-Excel software. An optimization- model based on LOB technique has been developed as a result of searching
and optimizing tool (GA) in the software which helps in assisting the decision makers to figure out the most ideal
solutions in the problem with large search space[23]. This method creates robustness and efficient optimized
output in the construction field.
2.4.3 MAWA Approach
Modified Adaptive Weight Approach(MAWA) is a mathematical programming operating method based on

Adaptive Weight Method(AWA) consisting of a derivative function used for optimizing process. MAWA is an
advanced version of Adaptive Weight Approach Method(AWA) to overcome the existing deficiency of the AWA
method. This process mainly involves the use of improved GA based multi-objective model to generate the best
time-cost optimal solution in a construction project[24]. The derivative function generated continuously iterates to
give a highest optimal value of a parameter. Initially, the adaptive weight is generated by choosing the initial
population randomly and fitted into equations for computing the fitness of each present chromosomes followed by
using crossover, mutation and initialization operation of Genetic Algorithm Method to the current population to
update the new population[25]. This method creates manual simulation based on application which helps in
incorporating GA-multi-objective time cost optimization solution. This also acts as an aid to decision-makers in
project to assess the optimal total time and cost simultaneously. It is further seen that the MAWA method consists
of certain weakness which requires some further refinements to improve the steady performance accuracy. More
experimentation and development works are required to assure effective result when applied to large-scale
projects[26].
2.4.4 Critical Path Method
Critical Path Method is the longest duration of each task required for the completion of the project. Critical Path
Method represents a steady approach of project management which signifies the activities present in the critical
path. It is anapproach that calculates the total duration of the activities based on estimated durations of each
activities to complete the project.Critical Path Method along with the presence of Genetic Algorithm Method
generate a mathematical operation which is significant for cost and time optimization in a construction
project[27]. An enhanced version of CPM method which is Resource Activity Critical Path Method (RACPM)
optimized by GA method coded into a computer program for finding the optimized duration of project subjected
to availability of resource constraints and precedence relationship .It is found that GA-RACPM constantly achieve
an average of 10 to 15 percent reduction in the total duration of the projectGA-RACPM method also provides
effective time and cost efficient solution[28].
2.5 Other Optimization Methods
The other optimization methods are as follows :-
2.5.1 Discounted Cash Flow Method
137
Discounted Cash Flow Method is an estimation method used for reckoning the funding property using
mathematical formulation process. This method is used for analyzing as well as optimizing solutions of a Time
Cost Trade off (TCTO) problem in a construction project using mathematical algorithms. This process involves
the formation of a model of ant colony optimization technique for multi-objective purposes based on TCTO
problem and finally tested for minimizing the discounted cash flow of direct cost. The application of this method
in the model shows that the insertion of DCF would change the mode of construction, activities duration and costs
for optimal solution of project span[29]. It is found that DCF method can be implemented in the analysis of TCTO
problem of a project whose time period exceeds for more than a year to get realistic results and consequent sound
decisions[30].
2.5.2 Artificial Neural Network Method
ANN is a mathematical programming method is used for increasing as well as decreasing the data-set on duration
basis. In Construction Management, lot of uncertainties related to time, cost, quality and safety are present. Due to
this the entire construction process becomes unpredictable. Hence the Artificial Neural Network Method (ANN)
gives highly beneficial in interpretation and adequate input information [31]. ANN method includes development
of a neural network architecture using multi-layered feed forward network for optimization process. A classic feed
follower ANN consist of an input layer, followed by a single or more hidden layers and an output layer. Every
layer consists of a numeral of neurons which is interconnected with all the neurons in the next layer[32]. There
should be no connection among the neurons present in the same layer[33]. Several techniques are present to train a
neural network among them is the back propagation technique is powerful and widely used in ANN
applications[34]. It has been found that using ANN method many industrial professionals, academicians have
optimized the different cost and time of a construction project. One of these investigated that the overhead cost of
a construction project situated in Saudi Arabia. It showed that in an unstable and competitive market in Saudi
Arabia, how a construction companies decide on the level of overhead costs that enables to control effectively a
large construction projects[35]. Using ANN method, a neural network a construction cost has been formed and
developed a parametric cost model of a highway project by introducing two alternative techniques to train network
weights:- simplex optimization and genetic algorithm method[36]. Moreover, ANN method is used as a tool for
optimizing cost and time in a residential building, apartment projects[37], cost of a school building[38], highway
tunnel project[39].ANN Method was used for estimating the optimal contingency of an owners funding of a
transportation project and it achieved the solution which are closer to the optimum than existing tools[40].
2.6 Summary
The below table refers to the innovative techniques that has been done by the researchers and investigators over
the recent years along with the scope of future work for optimization purpose in the construction industry.
Authors Techniques Problem Statements Results
Aagrama.A. A multi-objective optimization model A multi-storied building was

F based on LOB was developed with the analyzed to find the optimal
GA with LOB help of GA and Evolver 4.0 solution of the building which
proves the robustness and
consistency.
138
Zheng.X.M A Multi-objective TCO model powered The MAWA method could help
et al by GA technique is proposed in the decision makers in finding
GA with paper Which includes the modified optimal total cost and optimal
adaptive weights approach. total time simultaneously.
MAWA
Yi.P et al The GA optimized resource activity GA provided a achievable

critical path method (GA-RACPM) has optimization solution for
GA with CPM been developed and coded into a shortening total cost of the project
computer program. and reckoning time for
optimization process.
Aamar. A. A non-linear optimization problem has The results shows that if DCF is
Mohammad been developed by taking into account considered, then simultaneously
DCF discounted cash flows (DCF) the project activities duration, cost
along with resultant optimal
project duration varies from
traditional method. It can be
concluded that DCF should be
applicable for the project span
exceeding one year.
Liu. J et al A technique is developed based upon The technique can be used for
“maximal flow minimal cut” (MFMC) shortening the total duration of
MFMC for a large project. the project and time-cost
optimization.
Arafa. M et ANN can be used for the highway cost, It have used for predicting the
al residential and apartment projects, estimated cost and duration of the
ANN school buildings, tunneling projects project. It is highly recommended
etc. with different values and data. that more studies should be
carried out in this field for
optimization process for future
projects.
A Programming model of non-linear NLIP technique gives remarkable
Haj.Al.A.R integer type has been developed result and it can be used by the
et al NLIP project managers for large
networks project to access new
and efficient type of cost and time
trade off problems.
Table 2. Summary of the various innovative optimization techniques.
3. Discussion
It is examined that for optimization of cost and time, various innovative techniques and developing models by the
researchers, investigators and planners have been used. It is seen that Line of Balance(LOB) method with Genetic
Algorithm Method(GA) is used for multi-objective and repetitive projects similarly the MAWA method with GA
method is used for same along with giving decision-making optimized result. It is also found that Critical Path
Method(CPM) with GA is used for repetitive work along with multi-objective project by giving cost-effective
optimized solution. The Direct Cash Flow (DCF) method along with Minimal theory method is used for
optimizing solution on the basis of duration. The DCF method optimizes the solution for project exceeding the
one-year duration whereas minimal theory is used for large projects for shortening the total duration of the project.
The Artificial Neural Network(ANN) method has various applications that in optimization in school buildings,
residential buildings, tunneling projects, highway projects and large-sized projects whereas the Non-Linear Integer
139
Programming technique is also used for large-network project by giving minimal cost and time optimization of the
project.
4. Conclusion
This paper reviews the various innovative techniques that are demonstrated for cost and time optimization of a
construction project. Many researchers and investigators has tried to formulate the GA model based upon different
techniques like LOB, CPM, Dev-C++ 4.9.9.2 software, MAWA etc. for finding the best feasible optimal result of
time and cost trade off and resource constrained scheduling. However, GA method guide the decision makers,
planners, project managers for finding the most appropriate possible solution of different projects and also avoids
the complex formulation of mathematical models. Moreover, the NLIP method is preferred by the researchers and
investigators as compared to deterministic approach like heuristic model for obtaining the optimal solution. The
DCF method is advantageous for contractors and owners for estimating the priced bill of the project and considering
the time value of money. The ANN method is the powerful and mostly preferred by the researchers and
investigators among the above mentioned methods for optimization problem. This method is used infrequently in
construction industry as a optimizing tool, thus this method has large scope to be used in future research work
including the work with training algorithms, variety of network architectures and heuristic hybrid model.
References
[1]. Reda R., Carr R., : Time-cost tradeoff among related activities. Journal of Construction Engineering and
Management, Page 475-486(1989).
[2].Liu. J, Rahbar.F,: Project Time-Cost Trade-Off Optimization by Maximal Flow Theory ” Elsevier Publications.
Page 607-609(2004).
[3]. Isidore L., Back W.E., : Probabilistic optimal-cost scheduling. Journal of Construction Engineering and
Management, Page 431-437(2001).
[4]. Yang T.I., : Chance-constrained time- cost trade –off analysis considering funding variability. Journal of
Construction Engineering and Management, Page 1002-1012(2005).
[5]. Trivedi K.M., Namdev S., : Use of optimization techniques in time-cost trade off (TCT) in civil construction:
An Overview. International Journal of Civil Engineering and Mechanics, Page1-13(2015).
[6]. Tatar V.B., Patil. S. R., : Optimization tools for time cost trade off applicable in construction project
management. International Journal of Science Technology & Management, Page230-237(2015).
[7]. Meggiddo N., :Microcomputers, Class Lecture, Topic “Linear Programming”. School of Mathematical Science,
Page 1-8(1991).
[8]. Chassiakos P.A., Sakellaropoulos P.S., : Time-Cost optimization of Construction Projects with generalized
activity constraints. Journal of Construction Engineering and Management, Page1115-1124(2005).
[9]. Haj Al. A. R., SayeghEl.M.S,: Time-cost optimization model considering float consumption impact. Journal of
Construction Engineering and Management, Page1-10(2015)
[10]. Lu L, Burns A.S., Feng W.C., : Construction Time-Cost Trade-Off analysis using LP/IP Hybrid Method.
Journal of Construction Engineering and Management, Page446-454(1995).
[11]. Perera S., : Linear Programming Solution to network compression. Journal of Construction Division,
Page315-326(1980).
[12]. Tatar V.B., Patil. S. R., : Time-Cost optimization techniques in Construction Project Management.
International Journal of Latest trends in Engineering and Technology, Page 230-237(2015).
[13].Klansek U. Psunder M., : Cost optimization of time schedules for project management. Economic Research
Journal(Taylor and Francis group) Page22-36(2015).
[14]. Yang Q., : Application of Time-Cost-Quality tradeoff optimization model based on improved PSO Algorithm
to construction project. Asia-Pacific Conference of Information Processing, Page(2009).
[15].Zhang L., Du J., Zhang S., : Solution to the Time-Cost-Quality Trade-off problem in construction projects
based on Immune Genetic Particle Swarm Optimization. Journal of Management in Engineering, Page 163-
172(2014).
140
[16]. Ng T.S., Zhang Y., : Optimizing construction time and cost using Ant Colony Optimization Approach. Journal
of Construction Engineering and Management, Page 721-728(2008).
[17].Mokhtrai H., Bardaran R., Salmasnia A., : Time Cost Trade Off analysis in project management : An Ant
system Approach. IEEE Transaction on Engineering and Management, Page 36-43(2011).
[18]. Ashuri B., Tavakolan M., : Shuffled Frog –Leaping Model for Solving Time –Cost- Resource optimization
problems in construction project planning. Journal of Computing Civil Engineering Page 1-11(2015).
[19].Bhosale. V, Shastri.S.S, Khandare.M.A ,: A Review of Genetic Algorithm used for optimizing scheduling of
Resource Constraint construction projects. International Research Journal of Engineering and Technology (IRJET)
Publications, Page 2869-2872(2017).
[20]. Khan Md., Haque A., HasinA.A.Md., : Genetic Algorithm for project time-cost optimization in fuzzy
environment. Journal of Industrial Engineering and Management, Page 365-381(2012).
[21].Su. Y, Lucko. G ,: Linear scheduling with multiple crews based on line-of-balance and productivity scheduling
method with singularity functions.Journal of Automation in Construction, Page 546-556(2016).
[22].Su. Y, Lucko. G,: Comparison and renaissance of classic line-of-balance and linear schedule concepts for
construction industry.Creative Construction Conference 2015 (CCC2015), Page 38-50(2015).
[23]. Aagrama.A.F,: Multi-objective genetic optimization for scheduling a multi-storey building. Elsevier
Publications. Page 119-128(2014).
[24].Zheng.X.M.,Ng.T, Kumarswamy.M,: Applying a Genetic Algorithm-Based Multi-objective Approach for
Time-Cost Optimization . Journal of Construction Engineering and Management, Page 168-176(2004).
[25].Adrian J. J.,:Construction accounting: Financial, managerial, auditing and tax. Prentice-Hall, Englewood Cliffs
Publications, 1979
[26]. Alkass, S., Mazerolle, M., Harris, F.,:Construction delay analysis techniques. Construction Management Econ.
Publication, Page 375–394(1996).
[27].Yi.P, Lu.M,: Genetic Algorithm Optimized Resource Activity Critical Path Method. First International
Conference Publications, Page 1978-1982(2002).
[28]. Lu. M, Li. H , : Resource-activity Critical Path Method for Construction Planning. Journal of Construction
Engineering and Management, ASCE, (2001).
[29] Aldani.A, Afsar. A, Kalhor. E ,: Discounted Cash Flow Time Cost Trade-Off Problem Optimization. ACO
Approach. Asian Journal of Civil Engineering , Page 511-522(2011).
[30]. Aamar. A. Mohammad ,: Optimization of Project Time-Cost Trade-Off Problem with Discounted Cash Flows.
Elsevier Publications Page 65-71(2011).
[31].Kulkarni. P.S., Londhe.S.N.,Deo. M.C. :Artificial Neural Networks for Construction Management: A Review.
Journal of Soft Computing in Civil Engineering , Page 70-88(2017).
[32]. Arafa. M, Alqedra. M.: Early Stage Cost Estimation of Buildings Construction Projects using Artificial Neural
Networks. Journal of Artificial Intelligence, Page 63-75(2011).
[33]. Hagan M.T., Demuth H.B., Beale M.H.,: Neural Network Design. First Edition PWS Publishing Co. Boston,
MA, USA, Page 2-7 (1996).
[34]. Setyawati B.R., Sahirman S., CreeseR.C.,:Neural Networks for cost estimation. AACE International
Transactions. Page-595-602 (2002).
[35]. Assaf. S, Bubashait. A, Atiyah. S, Al-Shahri. M,: The management of construction company overhead costs.
International Journal of Project Management, Vol.19, No.5, Page- 295-303(2001).
[36]. Hegazy T., Ayed A.,: Neural network model for parametric cost estimation of highway projects. Journal of
Construction Engineering and Management, ASCE 124(3), Page 210-218(1998).
[37]. Kim G., An S., Kang K. , :Comparison of construction cost estimating models based on regression analysis,
neural networks, and case-based reasoning. Building and Environment Publications, Page 1235-1242(2004).
[38]. Kim G., Shin J., Kim, S., Shin, Y, : Comparison of School Building Construction Costs Estimation Methods
Using Regression Analysis, Neural Network, and Support Vector Machine. Journal of Building Construction and
Planning Research, Page 1-7(2013).
[39]. Luu V.T, Kim, S.Y. , : Neural Network Model for Construction Cost Prediction of Apartment Projects in
Vietnam. Korean Journal of Construction Engineering and Management, Page 139-147(2009).
[40]. Lhee S.C., Flood I. and Issa R., : Development of a two-step neural network-based model to predict
construction cost contingency. Journal of Information Technology in Construction, Page 399-411(2014).
141
sourav201ghosh@gmail.com
Paper ID- 39
OIL CONTAMINATED SAND: TOWARDS CLEANER FUTURE

Sourav Ghosh 1, Dillip Kumar Bera2
1
M.Tech Student ; School of Civil Engineering, KIIT (deemed to be University), Bhubaneswar, India
Mail id: sourav201ghosh@gmail.com
2
Faculty; School of Civil Engineering, KIIT (deemed to be University), Bhubaneswar, India
Mail id: dberafce@kiit.ac.in
Abstract
The oil spills of Torrey Canyon and Santa Barbara in the 1960s sparked the study of such disasters in an extensive
manner. India was never an oil producer but by 2010 with economic liberalization oil imports have increased. On
4th August 2011, a vessel MV RAK carrier sunk nearly 20 nautical miles from the coastline of Mumbai. This
tragedy spilled 340 tones of FFO (Furnace Fuel Oil) and Low Sulfur High Flash High Speed Diesel (LSHFHSD)
dragging India into this age old ball game. Coastal and terrestrial oil spills contaminates the sand and soil in the area
and thus affecting the ecosystem as a whole. With advancement of engineering technologies new application is to
use the contaminated sand as an engineering sustainable material for the construction works. . Strength properties of
concrete have been found to be enhanced when oil contaminated sand and aggregates had been added in certain
quantities. However, the increasing trend holds up to a certain contamination level after which the properties show
steady decline. Still higher contaminated sand can be used for low-load bearing structures and works. Such works
include landfills, production of bricks, precast masonry units. With the presence of such broad spectrum of uses and
possibilities, Oil Contaminated Sand can be used as a sustainable material in construction thus paving the path for a
cleaner future.
Key words: Oil Contaminated Sand, Oil Spill, Compressive Strength, Sustainable Materials, Remediation.
1. Introduction
The end of the past century saw the advancement of technology due to which the world energy demand has
risen through the roof. The shortage in energy rose initially within the time frame of 1918 – 1973. With the increase
in energy demand oil production increased which finally led to increased oil vessel traffic and increased oil
production. It has been reported by OPEC that oil production has reached 1.4 million barrels per day. The rise in
energy demand and supply started from the 1900s after the industrial revolution. However, due to recurring cost
increments and enactment of various laws the decade has seen a lower drilling and production of fossil fuels.
The drilling and production of oil leads to supply chain of oil. It is the most vulnerable part of an oil
demand and supply chain. The supply chain contains a various process after crude oil is extracted Refining,
transportation, storing are a few stages where the general probability of oil spillage is high. Oil spillage can be
caused due to natural disasters like that of MV RAK near the coast of Mumbai, or it can be human influenced like
the intentional sabotaging of pipelines in Nigeria. This has led to contamination of soil and surroundings in the
environment. Moreover when the spilled oil is ignited it causes air pollution, water pollution as well as land
pollution.
Oil spillage categorically are dived into two broad spectrums being offshore and onshore spills. Spills are
generally hazardous and needs to be taken care of as it disrupts ecological balance and harm the local flora and
fauna. The presence of oil seriously affects the ecosystem and nearly makes the area inhabitable. It has been found
that oil sediments into the lower layers reducing soil moisture, affecting plant toxicity and even change in species
composition when subjected for a longer period of time.
There are various decontamination processes present in the market for oil contaminated soil and water that
have been discussed later in the paper. With various decontamination processes come various challenges. Some are
too expensive processes, while some processes partially decontaminate the soil of the affected area. Some processes
are too slow and time consuming and on the other hand some process produces other type of pollutants in order to
remedy the soil of the affected area. So it becomes imperative to find a way that can decontaminate or replace the
soil of an oil spilled area, in-expensive a method, time efficient, and does not produce other pollutants. Usage of oil-
contaminated sand in the field of construction is a relocation and replacement process that satisfies all of the above
criteria making it a viable solution to tackle the problem of pollution by oil spills.
142
2. Study of previous researches
Global oil requirement scenario
The period of 1918-1973 saw the highest production of crude oil in the world. The production of crude oil
steadily increased at the rate of 7.1% per year i.e. 56.4 million barrels per year from a measly 1.4 million barrels per
year in 1918. A major part of this uncontrolled growth was contributed by OPEC (Organization of Petroleum
Exporting Countries), for whom the daily output was clocked to be 12.6% per year. Non-OPECs showed a much
less production of only about 5.7%. The first crude oil price shock in 1973 initiated recurrent changes in oil
production and in the breakdown between OPEC and Non-OPEC producers. This was due to the radical change of
the oil exporting countries from an organizational form to a cartel form of approach on the market. OPECs started
curtailing crude oil production within the period of 1973-1986, to influence the price of oil to go rise or at least
contain the ever decreasing oil prices due to the huge demand and production.[1]. Now due to this staggering
increase in oil production pre 1973, oil demand and supply grew extremely high which finally led to some massive
tragic disasters. However no such disasters were profoundly studied as the market value was much lucrative and
pressingly important. The disasters of Torrey Canyon and Santa Barbara in 1967 and 1969 respectively sparked the
scientific community to study the disasters and comprehend the environmental impact. The present state of oil
demand is at an alarming rate 1.4 million barrels per day. Figure 1 depicts the consumption and demand of oil over
the world.
Figure 1: Oil demand and consumption of the present decade
Effect on the ecology
The effect on the ecology due to oil spills is a broad spectrum. So it can be better studied and understood by
dividing them under- Terrestrial oil spills and Off-shore oil spills. Both the forms come with their own set of
problems. Terrestrial spills come with ponding effect and sand tailings, but those from off shore spills moves and
spreads faster. Effects are varied regarding the types but the common problem being both types pollute the
sedimentary soil layer at some point of time. One affects it faster while the other takes time to reach the sedimentary
layers.
Terrestrial oil spill: In 4th August, 2011 a Panama flagged carrier vessel MV RAK sunk nearly 20 nautical
miles of the coast of Mumbai. According to Indian Coast Guard 340 tones of Furnace Fuel Oil (FFO) and Low
Sulfur Flash High Speed Diesel (LSHFHSD) were on that vessel. The spill affected a significant portion of the
coastline of Mumbai. The impact was felt on both infauna and epifauna. Some places continue to go on a cycle of
deterioration leading to complete loss of habitat. The oil in coastal beds have taken months to penetrate months and
thus caused delayed plant toxicity. It was observed that affected shoreline and tidal movement resulted into mixing
of oil at deeper levels at impacted areas increasing the environmental and biological hazards [2].
143
Figure 2: Effect of terrestrial oil spill in Israel
In 1970-1974 a research by U.S Army Cold Region Research and Engineering Laboratory conducted a
terrestrial oil pollution research, the focus being to understand the effects of terrestrial oil pollution in cold regions
particularly plant communities. It was found that petroleum contaminants were someway detrimental to plant life as
well as the animal community especially burrowing animals. Environmental factors dictate the extent of oil
contamination and provide different symptoms in different situations. The degree of the oil spill and it’s longevity is
also governed by environmental and climatic factors. High soil moisture tends to less damage by oil spills. Plants
with long rooting system with high soil moisture have a relatively better chance at survival in an oil contaminated
soil. Oil spills causes temporary changes within a species composition thus making it imperative to find a better way
of clean up procedures [3].
Ocean oil spill: Moore and Dwyer [4] concluded that it takes 1-2 days of weathering to reduce the toxicity
of oil spills but Teal and Howarth [5] reviewed some much recent and promising events like Florida, Arrow, Agro
Merchant etc. They finally concluded that oil may persist in water column for about half a year. It may then be
diluted and finally sediments down to the background levels. Florida oil spill showed presence of oil in 7 years after
the spill and even 12 years after oil contaminants were still found within sediments. It was finally concluded that
contaminants from spills always transports to the sediments and is found in the bed or in the benthic animals.
Figure 3: Eared grebe slicked during the 2007 Cosco Busan spill in San Francisco Bay
For understanding the effects of oil contamination Cheston T. Peterson[6] used catch data from two fishery
independent boats used for shark surveys in Florida Big Bend. It was found that that there was no substantial change
to the tropic pathways used by various species of the area. However it was further concluded that any absence of
pattern derived from their methodology does not imply that there was no Deep Water Horizon effect. Owing to the
strong currents in the concurrent years the spills could have sedimented or moved away from the test site.
Remedial Procedures
Establishing the fact that oil spills- be it terrestrial or offshore is detrimental for flora and fauna of the
surrounding areas it is required to study about the clean up procedures. The basic clean up procedures are primarily
of two types:
144
 PHYSICO- CHEMICAL CLEAN-UP METHODS

o Soil excavation
o Soil washing
o Soil Vapor Extraction(SVE)
 THERMAL CHEAN-UP METHODS
o Thermal Desorption
o Incineration
 BIOLOGICAL CLEAN-UP METHODS
o Microbial degradation of crude oil
o Bio remediation
The above methods provide promising or at least satisfactory results in cleanup of the oil from soil. Among the
above mentioned procedures, the most effective is the process by Bioremediation. This procedure enacts the addition
of fertilizers in the soil thus attempting an accelerated biodegradation process. In case of a failed attempt a micro
organism community is introduced in the affected areas, which in time degrades the hydrocarbons present in the
crude oil contaminants. However the procedure is time consuming and takes a lot of effort, monitoring and very
precise levels of temperature, pH etc. Bioremediations leave behind an aggravated colony of the introduced micro-
organism which has to be controlled even after the remediation process, thus changing soil chemistry [7]. Ganey and
Boyd[8] concluded that it may be possible that chemical products unrelated to the crude oil contaminants may be
produced which may cause unexpected biological activity.
Xiaoci Ji [9] studied the effects of bioremediation process and even 14 years after the introduction of the
genetically modified micro organism, traces were form in soil chemistry and microbial community structures. The
study proves that even after the removal of the oil contaminants the long term soil bacterial characteristic changes
when bioremediation is used. There can be lasting impact on the organic matter with in soil and bacterial community
restructure.
Behavioral Change Of Material After Contamination
Chemical composition of contaminants primarily governs the extent of the extent of the contamination of a
particular area. A study of the Persian Gulf Spill showed that contamination of soil or sand by crude oil changes its
physical and chemical composition.
 Atterberg limits decrease with increasing oil contamination in clayey soil.
 Maximum dry density reduces as the levels of contaminations increase.
 Oil contaminations results in decreased permeability and strength in soil samples.
Water Content: Mashalah Khamehchiyan [10] suggested that since oil had a complex structure and it
evaporates even under room temperature, the regular formula for water content is not applicable for oil contaminated
sand. The addition of oil contamination levels reduces the water content within soil.
The regular formula:
W% = 𝑊𝑤 ∗ 100
𝑊𝑠
Should be replaced by:

𝑊𝑡
W% = (1 + 𝑀𝑁) − (1 + 𝑁)
𝑊𝑑
Where,
W% = water content; Ww = weight of water
Ws = weight of solids in soil Wt = wet weight of contaminated soil
Wd = dry weight of contaminated soil/ sand M = oil residual after drying
N = oil content before drying
145
The generic trend of reduction of water content with the increase in oil contamination is given by the following
table.
Table 1: Relation between oil contamination levels with that of water content in the material.
Oil contamination % Water content %
CL SM SP
0 16 13.5 15
4 13.5 8.9 10
8 9.5 8.4 7.5
12 7.3 5.9 5.5
16 3.2 2.2 2.9
Angle of Friction and Cohesion: It have been observed by Hasan A. Al-Sanad[11] that the presence of oil
decrease with time. Within a sample of 6% contamination a decrease of 1% contamination have been recorded. The
rate of decrease also decreases with time. Uncontaminated sample of Kuwaiti sand have an angle of friction of 35°
where as a contaminated sample shows angle of friction to be around 28° but with time as the contamination reduces
and the hydro-carbons evaporates the angle of friction increases up to 32°. Compression index for uncontaminated
Δ𝑝
sand have been found to increase to 0.065 for contaminated sand. The average constrained modulus, 𝑀 =
Δ€
decreases from 20.3 MPa to 9.2 MPa in case of Kuwaiti sand.
Mashalah Khamehchiyan [10] proved that the cohesion of poorly graded sand rise with increasing
contamination level up to a level between 12% and 16% after which it starts to fall. In case of sand with silt in it the
cohesion level remains the same and the cohesion reduced with increasing contamination for clayey soil.
Table 2: Relation between oil contamination levels with that of cohesion within the material.
Oil contamination % Cohesion (kg/cm2)
CL SM SP
0 0.75 0.27 0
4 0.295 0.2 0.04
8 0.21 0.21 0.08
12 0.195 0.195 0.09
16 0.19 0.35 0.06
Porosity: When soil is contaminated by any form of oil or hydrocarbons, it slowly loses soil moisture
levels. Capillary water content increases and the bulk density increases with the increase in soil contamination.
There is also an increase in total porosity. In a research by Anna Klamerus-Iwan et. al[12] it was observed that 100
g/m2 contamination decreased the air-filled porosity by 4% and 200 g/m2 decreased the porosity by 10%.
Since most of the oil fields and terrestrial oil drilling stations are located in areas of scanty vegetation and
are mostly desert areas, attention is required on the properties of oil contaminated sand. Desert sand contaminated
with various levels and proportions of light and heavy crude oil show very interesting properties. Such an example is
of Jahra sand, from Al Jahara, a town near Kuwait city. Direct shear tests in normal sand provides a straight line
relation of relative density and frictional angle, but for contaminated Jahra sand the results and relationship pointed
to a concave downwards curve [13].
Shear Stress: Glacially deposited sand of northern England removed from crude oil contaminated sites that
had been contaminated with various levels of diesel fuel and petrol were first screened and subjected to a thermal
disorption treatment process. It was found that partial thermal treatment of the sand provided acceptable shear
stresses with results around 26.2 – 67.1 kPa. The strength and the compaction behaviors are quite similar to that of
uncontaminated sand and are believed to be a potential aggregate of highway construction [14].
Hydraulic conductivity: E. C. Shin and B. M. Das [15] studied the properties of unsaturated oil
contaminated Jumoonjin sand to understand the physical properties. Jumoonjin sand is poorly graded sand rich in
silica from South Korea. It was found that the maximum Dry Unit Wt. and the Degree of Saturation of the sand
increases with the simultaneous increase of Kinematic viscosity. This property was due to the fact that higher
viscous crude oil effectively reduces the inter-granular friction within sand particles. It was further concluded that:
146
 Increasing oil content decreased the hydraulic conductivity.

 Decreasing the oil’s kinematic viscosity decreases the hydraulic conductivity, and
 Relative density of compaction increases with reducing hydraulic conductivity.
So if we conclude the previous researches, we can corroborate the data in the following table.
Table 3: Changing trend of properties with increase in oil contamination level

Author Properties Relation
Mashalah Atterberg limits Increases
Khamehchiyan Maximum dry density Decreases
Optimum water content Decreases
Water permeability Decreases
Hasan A. Al-Sanad Angle of friction Decreases
Compression index Increases
Increasing Constraint Modulus Decreases
Anna Klamerus-Iwan levels of Oil Capillary water content Increases
contamination Total porosity Increases
Bulk density Increases
Air-filled porosity Decreases
Robert D. Andrea Shear stress
Degree of saturation Decreases
Max. dry unit Wt. Decreases
E. C. Shin and B. M. Hydraulic conductivity Decreases
Das Kinematic viscosity Decreases
Relative density of Decreases
compaction
Behavioral change of concrete after contamination.
As the energy demand of the world increases the technologies related to the oil extraction also increases.
However there remains greater challenges regarding the waste management and disposal of oil extraction based
wastes. Having established the fact that oil contaminated soil or sand is one of the greatest and largest headache of
oil extraction and production. Various methods and techniques have been implemented to tackle the control and
dispersion of the petroleum production wastes into environment. Such a process is called “Stabilization and
Solidification”. S/S is a process that incorporates the use of additives like fly ash, lime and Portland cement to
stabilize the the waste materials and renders them into and harmless non-toxic form. This step is called as
stabilization and the reaction is generally Pozzolanic in nature. The solidification of the stabilized product generates
a monolithic solidified mass of structural integrity. Ahmet Tuncan[16] deduced that addition of stabilizing agents
like fly ash, lime and cement increases pH value. The oil content of 1% is the most effective and produced the
highest structural integrity.
Flowability: A. Mneina[17] studied the oil sand tailing left behind after that effectively has to be land
filled. It has been found that Treated Oil Sand Wastes generally increases the flowability of the mixtures thus
reducing the demand of higher water to cement ratio which in turn increases the compressive strength of the
mixture. Lower dry density of mixtures with Treated Oil Sand Wastes makes it extremely suitable for field
applications.
Table 4: Various flowability values from previous researches.

Author Methodology / Standards Flowability Category
A. Mneina ASTM standards D6103-04 185 to 250 mm High flowability
R.M. Abousnina ASTM C1437-07 150 ±10mm High flowability
Bilal S. Hamad ASTM standards 160 to 220 High flowability
147
Compressive strength: A general study by R.M. Abousnina et al.[18,19] deduced that with increasing
percentage of oil in sand increases the compressive strength and then decreases with further inclusion. Such a trend
has been shown in Table-2.
Table 5: Change in compressive strength with changing levels of oil contamination

Oil Contamination (%) Compressive Strength (MPa)
0 25 – 27
0.5 25 – 30
1 30 – 33
2 27 – 30
4 25 – 27
Table 2:
The use of oil contaminated aggregates can effectively reduce the pollution levels of environment by its use
in the field of construction. It can also be used for the production of construction ceramics. Brick formation is a high
energy consuming and polluting process. It consumes a large amount of energy thus increasing the demand of
energy production. L.L. Maslennikova[20] proposed the use of oil contaminated stone screenings for production of
construction ceramics. It was concluded that screening lower than 5mm in size having a contamination level up to
8% provides greater strength than conventionally made bricks.Table 3 compares the enhancement of the strength
compared with conventional bricks
Table 6: Comparison of compressive strength between oil contaminated bricks and conventional bricks
Name Compressive strength Quality Water
(Mpa) absorption %
Conventional bricks 10.8  Un-uniform color 12.0
 Cracks during drying
Oil contaminated (5%) 14.8  Uniform color 7.0
stone screening bricks  Intense color
 No cracks visible with naked
eyes
Previous researches have proved that certain wastes in petroleum wastes possess admixture properties and
can be used as additives, which in turn either can alter or enhance fresh properties of concrete. The levels of
contamination are significantly dependent on the required properties of the final product. In concrete technologies,
compressive strength is considered to be the most important property to be kept in mind. It depends on the properties
of the ingredients involved in the concrete mixture, mix proportions, compaction methology involved, contaminants
involved and the levels of contaminants. W.O. Ajagbe[21] deduced that contamination level of 2.5% by weight of
sand lowers or reduces the strength properties of Oil Contaminated Sand Concrete (OCSC) by 18%. Contamination
levels up to 5% is favorable in areas where strength is an important parameter but further effort should be made to
enhance the strength properties, one being by the addition of stabilizers.
The increasing strength and the non formation of cracks is a result of the presence of oil and other
petroleum contaminants that burns during the burning process and the gases formed acts as space holders. Moreover
the inner oil content leads to less fuel consumption and the burning of the clay sinters the clay in the bricks thus
providing increased flexural strength and reduced water absorption.
Air voids: Further research by H.F. Hassan [22] on the compressive strength of petroleum contaminated
sand proves that the compressive strength of concrete increases with increasing cement content however there
remains an adverse effect on cement hydration. It is recommended that crushed stone with 10% oil contaminated can
be used for road sub base. In higher concentration of oil contamination of about 30% to 40% air voids have been
found to be in an aggravated level.
Workability: Researches on mechanical properties of oil contaminated sand have found various properties
like cohesion, compaction increases up to a certain level. A study by Hamad and Rteil [23,24] proved that oil
148
improved fluidity and nearly doubled the slump but at the same time maintained the compressive strength. Oil
contamination in sand or other forms of fine aggregates acts as air entraining agents in cement concrete mixtures.
Other authors also stated that contamination has led to increased workability. Addition of treated oil contaminated
aggregated increases the surface area of the aggregates which leads to increased water demand but the contaminated
aggregates acts as a better packer of the materials within the mix due to its lubricating nature. The water trapped
between the cement particles are released and helps in further lubrication of the concrete mixture.
Application areas
According to guidelines of various organizations, government and non-government organizations like United
States Environment Protection Agency (USEPA) permissible compressive strength after 28 days should be 0.35
MPa for landfill disposal site. Similarly Wastewater Technology Centre (Canada) demands 3.5 MPa for sanitary
disposal; British Standards incorporate 2.8 and 7 Mpa for bricks and building blocks [18]. So, various areas like
sanitary landfill, construction ceramics, brick production, highway sub-base material etc can incorporate the use of
oil contaminate sand and aggregates as a viable sustainable material.
The various Application areas include:
 Highway Construction Material
 Engineering Construction Material
 Production of construction Ceramics
 Production of bricks and blocks.
 Landfill Material
 Precast concrete masonry units
 Sub-base and Base materials
3. Summary
From the previous researches we can summarize the following points that can lead to future research.
 Demand of energy increases the drilling and production of crude oil.
 With increased production increases the tendency of tragic accidents to happen by natural and man-made
interventions.
 The tragic accidents can lead to massive extent of pollution in the form of spills and contamination of soil.
 The contaminated soil can either be subjected for remediation process or it can be used up in the construction
industry thus replacing the soil of the affected area.
 Oil contaminated sand is a hazardous material but after the stabilizing process by the addition of additives
renders it harmless and even enhances certain properties.
 Change of the properties depends on various factors like the moisture content, level of contaminants, type of
contaminants etc.
 Certain properties like the Atterberg limits, hydraulic conductivity and Degree of saturation decreases but when
used in a cement concrete paste properties like Flexural strength and compressive strength increases up to a
certain level.
 Due to the increases fluidity properties like workability increases requiring less admixtures.
 Contamination levels up to 5% by weight of aggregates should be considered for load bearing construction.
 The highest strength is found when aggregates are contaminated by 1 to 2 % oil.
 Higher contamination levels can be considered and used for non-load bearing constructions like landfill, brick
production, tiles production, highway sub-grade etc.
4. Conclusion
The result of the study deduced that there is a high possibility and potential of oil contaminated sand and
other aggregates to be used in the field of construction. If the engineering structure is load bearing in nature the oil
contamination levels have to be kept low to about 1 – 2%. The lower strength generating high contamination level
concrete can also be used in various other ways.
149
This would reduce the problem of remediation of soil after pollution and would in fact be used as a material
for construction. Since contaminants from petroleum production possess an admixture type effect, when used in
construction it increases the flowability of the concrete paste reducing the amount and cost of admixtures required.
Furthermore it would also deduct the cost of soil remediation. It is a cost effective and time efficient process to
change and replace the soil altogether rather than remedy it, whereas the contaminated soil can thereafter be
inspected and used in any form of construction.
The only unknown variable in this process is the identification of hazardous nature of the contaminants
which have to be inspected before it can and should be used in construction. The leachate effect should be taken into
account and its behavior should be taken into account during designing of the concrete mixture.
5. References
1. Noureddine Krichene (2002), World Crude Oil And Natural Gas: A Demand And Supply Model, Energy
Economics 24 557 – 576.
2. Maharashtra Pollution Control Board (MPCB) (April, 2013), Environmental Assessment Of Mumbai Oil Spill
From MV RAK Carrier.
3. Corps. Of Engineers (1975), U.S Army, Biological Aspect Of Terrestrial Oil Spills: Usa CRREL Oil Research
In Alaska 1970-1974, U.S Army Research Office; Research Report- 346.
4. S.F Moore, R.L Dwyer (1974), Effects Of Oil On Marine Organisms: A Critical Assessment Of Published Data,
Water Research Volume 8, Issue 10, Pages 819-827
5. John M. Teal (1984), Oil Spill Studies: A Review Of Ecological Effects, Environmental Management, Vol. 8,
No. 1, Pp. 27-44,
6. Cheston T. Peterson, R. Dean Grubbs And Alejandra Mickle, An Investigation Of Effects Of The Deepwater
Horizon Oil Spill On Coastal Fishes In The Florida Big Bend Using Fishery-Independent Surveys And Stable
Isotope Analysis, DOI: Http://Dx.Doi.Org/10.1656/058.016.0101
7. Uchechukwu E. Ezeji, Sylvia O. Anyadoh, Vincent I. Ibekwe (2007), Clean Up Of Crude Oil-Contaminated
Soil, Terrestrial And Aquatic Environmental Toxicology 1(2), 54-59.
8. Patricia E. Ganey1 And Steven A. Boyd(2005), An Approach To Evaluation Of The Effect Of Bioremediation
On Biological Activity Of Environmental Contaminants: Dechlorination Of Polychlorinated Biphenyls,
Environmental Health Perspectives, Volume 113, Number 2.
9. Xiaoci Ji, Steven A Ripp, Alice C Layton, Gary S Sayler And Jennifer M Debruyn(2013), Assessing Long
Term Effects Of Bioremediation: Soil Bacterial Communities 14 Years After Polycyclic Aromatic Hydrocarbon
Contamination And Introduction Of A Genetically Engineered Microorganism, Journal Of Bio Remediation &
Biodegradation, DOI: 10.4172/2155-6199.1000209
10. Mashalah Khamehchiyan, Amir Hossein Charkhabi, Majid Tajik (2007), Effects Of Crude Oil Contamination
On Geotechnical Properties Of Clayey And Sandy Soils, Engineering Geology 89 220–229.
11. Hasan A. AI-Sanad And Nabil F. Ismael/ Members, ASCE (1997), Aging Effects On Oil-Contaminated Kuwaiti
Sand, Journal Of Geotechnical. Geoenvironmental. Engineering.123:290-293.
12. Anna Klamerus-Iwan & Ewa Błońska & Jarosław Lasota & Agnieszka Kalandyk & Piotr Waligórski (2015),
Influence Of Oil Contamination On Physical And Biological Properties Of Forest Soil After Chainsaw Use,
Water Air Soil Pollut 226: 389;DOI 10.1007/S11270-015-2649-2.
13. C. Mirza(1996), Geotechnical Properties Of Oil-Contaminated Kuwaiti Sand, Journal Of Geotechnical
Engineering.
14. Robert D. Andrea Et Al. (1996)., Geotechnical Properties Of Oil-Contaminated Kuwaiti Sand, Journal Of
Geotechnical Engineering.
15. E.C. Shin T, And B. M. Das (2000), Some Physical Properties Of Unsaturated Oil- Contaminated Sand,
Advances In Unsaturated Geotechniques (American Society Of Civil Engineers).
16. Ahmet Tuncan, Mustafa Tuncan, Hakan Koyuncu (2000), Use of petroleum-contaminated drilling wastes as
sub-base material for road construction, Waste Management & Research 18: 489-505.
17. Mneina, A.M. Soliman, A. Ahmed, M.H. El Naggar(2018), Engineering Properties Of Controlled Low-Strength
Materials Containing Treated Oil Sand Waste, Construction And Building Materials 159: 277–285.
18. Rajab M. Abousnina, Allan Manalo, Weena Lokuge, Jim Shiau, (2015) Oil Contaminated Sand: An Emerging
And Sustainable Construction Material, Procedia Engineering 118: 1119 – 1126.
19. Rajab M. Abousnina, Allan Manalo, Weena Lokuge(2016), Physical And Mechanical Properties Of Cement
Mortar Containing Fine Sand Contaminated With Light Crude Oil, Procedia Engineering 145 250 – 258.
150
20. L. L. Maslennikovaа, M. S. Abu-Khasan, N.А. Babak(2017), The Use Of Oil-Contaminated Crushed Stone
Screenings In Construction Ceramics, Procedia Engineering 189: 59 – 64.
21. Wasiu O. Ajagbe , Olusola S. Omokehinde, Gabriel A. Alade, Oluwole A. Agbede(2012), Effect Of Crude Oil
Impacted Sand On Compressive Strength Of Concrete, Construction And Building Materials 26: 9–12.
22. Hossam F. Hassan, Ramzi Taha , Amer Al Rawas , Badr Al Shandoudi ,Khalfan Al Gheithi , Ahmed M. Al
Barami (2005), Potential Uses Of Petroleum-Contaminated Soil In Highway Construction, Construction And
Building Materials 19 646–652.
23. Bilal S. Hamad, Ahmad A. Rteil, Mutassem El-Fadel(2003), Effect Of Used Engine Oil On Properties Of Fresh
And Hardened Concrete, Construction And Building Materials 17: 311-318.
24. Bilal S. Hamad, Ahmad A. Rteil(2003), Effect Of Used Engine Oil On Structural Behavior Of Reinforced
Concrete Elements, Construction And Building Materials 17: 203- 211.
25. https://www.timesofisrael.com/israel-to-spend-nis-17-million-to-clean-southern-oil-spill/
26. https://www.kcet.org/redefine/10-california-oil-spills-bigger-than-the-refugio-oil-spill-we-think
151
Paper ID- 43
Wall Effects on Terminal Velocity of Test Fuel Bundle in the Fuel Test
Loop of High Flux Research Reactor
G. Verma1*, S. Sengupta1a, S. Mammen1b, P. Mukherjee1c, P. V. Varde2
1
Research Reactor Design and Projects Division, Bhabha Atomic Research Centre, Mumbai-400085, India
*
gaurav91verma@gmail.com, asamsen@barc.gov.in, bmshaji@barc.gov.in, cpradipm@barc.gov.in,
2
Research Reactor Services Division, Bhabha Atomic Research Centre, Mumbai-400085, India
varde@barc.gov.in
Abstract. Fuel Test Loop (FTL) is a self-contained independent experimental loop which is designed for testing of
nuclear fuel materials under simulated power reactor conditions in High Flux Research Reactor being developed at
BARC. It is a high pressure loop with maximum operating pressure up to 17.5 MPa and maximum operating temperature
up to 330°C. The In-Pool Test Section of the FTL consists of a series of concentric tubes with different thickness and
functionality. The test fuel bundle resides within the innermost Fuel Tube which is further enclosed within an internally
insulated Pressure Tube which acts as pressure boundary enclosure to the test fuel bundles.
The present work deals with the investigation of wall effects on terminal velocity of test fuel bundles falling under
gravity within the In-Pool Test Section. While loading/unloading of the test fuel bundle, accidentally the fuel bundle may
fall under the influence of gravity within fluid filled Fuel Tube. The minimum gap between the fuel bundle and the fuel
tube is 2 mm. Thus, the wall effects on the velocity of the falling fuel cannot be neglected. The purpose of the present
work is to estimate this wall effects on the terminal velocity of the falling fuel in terms of Drag Coefficient. This
involves, initially estimating drag characteristics and terminal velocities with wall effects for different geometries and
finally predicting the overall characteristics with the test fuel bundle as the geometry of interest. To fulfill this objective,
an analytical methodology is established which has been validated through a commercial code.
Keywords: Terminal velocity, Test fuel bundle, Wall effect, Drag coefficient
1. Introduction
The FTL in the proposed High Flux Research Reactor (HFRR) is a high temperature and high pressure
experimental loop to perform testing of nuclear fuel of power reactors under similar conditions. The
maximum operating pressure and temperature of the FTL is 17.5 MPa and 330°C respectively. The FTL
consists of two sections, the In-Pool Test Section and Out-of-Pool Section. The In-Pool test section provides
a pressure boundary enclosure to the test fuel bundles and also a separation between test section high
temperature/pressure water with the pool water. It is designed such that it maintains its mechanical strength
and structural integrity over the lifetime of the reactor. The Out-of-Pool section consists of various process
systems (such as Main Loop System, Purification and Sampling System, Pressurizer System, Jacket Coolant
System etc.), piping and equipment necessary to maintain the required temperature, pressure and flow
conditions inside the In-Pool test section.
The In-Pool Test Section of the FTL consists of a series encapsulation of four concentric tubes (Fuel Tube,
Pressure Tube, Inner Jacket Tube and Outer Jacket Tube) with different thickness and functionality. The
present work investigates wall effects on terminal velocity of test fuel bundles accidentally falling under
gravity within the 4.5 m long fuel tube of the In-Pool Test Section while loading/unloading of the test fuel
bundle. Since, minimum gap between the falling fuel bundle and the fuel tube is very small, the wall effects
on the falling fuel velocity need to be established. The present work estimates this wall effects on the
terminal velocity of the falling fuel in terms of Drag Coefficient. For this, drag characteristics and terminal
velocities with wall effects for different geometries are estimated which is further utilized to estimate these
characteristics for the geometry of interest using an established analytical methodology which is
subsequently validated with fluid-solid interaction coupled techniques through a commercial code.
152
2. Analytical Methodology
According to Gabitto and Tsouris [1], an object falling in an infinite fluid medium under the influence of
gravity will accelerate until the gravitational force is exactly balanced by the resistance force that includes
buoyancy and drag. This constant velocity hence achieved is termed as “terminal velocity”. Co-relations are
developed by various researchers (Clift et al. [2], Kahn and Richardson [3], and Haider [4]) that relates drag
coefficient (CD) and the Reynolds Number (Re) for objects of spherical shape falling at their terminal
velocities in an infinite medium. Further, there have been literatures suggesting the retarding effects of the
nearby wall on an object settling in a liquid. Most of these investigations have involved a single sphere
settling in cylindrical/ triangular/ rectangular ducts in either Newtonian or Non-Newtonian fluids [5-7].
However, not much literature is available for complex geometries falling in a confined flow passage under
the influence of gravity.
To introduce the wall effects on the motion of a sphere falling axially in a cylindrical tube filled with a
stagnant fluid, as per Arsenijevic et al. [8], velocity ratio (f) or the wall factor is defined as,
f 
Ut
U t

 1  1.121.26 
0.7
(1)
Where U t is the terminal velocity of the falling sphere (diameter d p and density  p ) settling along the axis of
a tube of diameter, D, in a fluid medium of density (  f ) and viscosity (  ), whereas U t denotes the
unbounded velocity of the same sphere in the same liquid in the absence of walls.  denote the ratio of the
sphere-to-tube diameter ( d p D ). Other widely used correlations in the turbulent flow regime are those of
Newton [9] as shown in Eqn. (2) and of De Felice [10] as presented in Eqn. (3).
 
f  1  2 1  0.52 
0.5
0.11    0.83 (2)
0.85
 1  
f   (3)
 1  0.33 
For non-spherical particles (such as cylinders), Chhabra [11], correlated the wall factor as,
f  1  1.33( d s / D ) L / D   10 (4)
f  1  3.58( d s / D ) L / D   10 (5)
where d s is the diameter of a sphere with volume equal to that of a non-spherical particle and D is the fall
tube diameter through which the cylinder is falling. Length of the cylinder is taken as L.
In order to calculate the terminal velocity of cylindrical object, various empirical relations have been
proposed. One such factor is the “volumetric shape factor” defined by Heywood [12] as,
V
k (6)
d A3

Where, V is the volume of the object and d A  4 Ap /  0.5 is the projected area diameter, which is calculated
as the diameter of a sphere with equal projected area as that of the particle, and Ap is the projected area of the
object.
The degree of sphericity is given by Wadell [13] as,
AV
 (7)
A
where AV is the surface area of a sphere having the same volume as that of the particle, and A is the actual
surface area of the particle. The sphericity of a true sphere is equal to 1.
For spherical objects, a number of CD-vs-Re correlations are available depending on the Reynolds number
range. In the present study, a two termed correlation developed by Cheng [14] is used to estimate the drag
coefficient. According to this relation,
153
CD 
24
Re
 
1  0.27 Re 0.43  0.47 1  exp  0.04 Re0.38  (8)
The first term is for Re < 100 and is an extension of Stokes law. The second term estimates drag coefficients
for high values of Reynolds number.
Drag coefficient of a non-spherical object explained by Haider and Levenspiel [15] as a relation between CD-
vs-Re in the form,
73.69 Re exp 5.0748 
CD 
24
Re
  
1  exp 2.3288  6.4581  2.4486 2 Re 0.09640.5565   
Re  5.378 exp6.2122 
(9)
which predicts higher accuracy up to a   0.67 .
Terminal velocity for any object can be given as
Ut 

2m p g  p   f  (10)
 f  p ApCD
Where U t is the terminal velocity, CD is the drag coefficient based on projected area, m p is the object mass,
 p is the object density,  f is the density of the surrounding fluid, Ap is the projected area of the object in
the direction of the motion and g is the gravitational acceleration.
For cylinders, Coulson and Richardson [16] developed k as,
k   4 2.5 dc L 0.5 (11)
Reynolds number is calculated for non-spherical particles using Galileo Number (Ga) number using
Heywood’s [12] method as
  
Ga  4 kdA3  f g  p   f /  2 (12)
where  is the fluid viscosity. Haider and Levenspiel [15] developed a correlation for modified Galileo
number ( Ga ) as
  
Ga  CD Re 2  d c3  f g  p   f / 2  2 (13)
Reynolds numbers within the Stokes and transitional flow regimes can be calculated using the following set
of equations:
Log10 ReL   a0  a1w  a2 w2  a3w3 , (14)
where
w  Log10 Ga 3  ,
1
(15)
 
a0  0.81824  0.55689 / E , (16)
a1  2.41227  1.54674 / E  0.53872 / E 2 , (17)
a2  0.2056  1.34714 / E  0.65696 / E 2 , (18)
a3  0.82343  0.40625a0  0.5625a1  0.75a2 (19)
where E is the aspect ratio L d c  . The procedure presented here allows for the explicit calculation of the
Reynolds numbers in the Stokes and transitional flow regimes for which Ga extends up to 200,000. For the
estimation of Reynolds number in the Newton flow regimes (ReT) as a function of the ratio of object-to-water
densities and of the cylinder aspect ratio.

ReT   p  f 0.06 E0.04Ga0.5 for ( Ga >100,000) (20)
154
Gabitto and Tsouris [1] provides the following equation for all the flow regimes and for aspect ratio E >1.
Re  ReL T1  ReT T2 , 
for 0.01  Ga  108  (21)
Where
T1  1 for Ga  2  103 , (22)
  3
T1  Ga , for 2  103  Ga  2  105 , (23)
And
T1  0 for Ga  2  10
5
(24)
Where
 
Ga  2  105  Ga 2  105  2  103 ,  (25)
And
T2  1  T1 (26)
3. Benchmarking
In an effort to validate the code for the modelling procedure and further analyses with result verification, drag
coefficient of a spherical object free falling in an unbounded medium is estimated through numerical
simulation using a commercial code and is compared with the analytical results. For this purpose, an
Aluminium sphere of 10 mm diameter is allowed to free fall in a cylindrical tank of 200 mm diameter and
400 mm length such that wall effects are negligible. Cheng correlation Eqn. (8) is used to estimate the drag
coefficient and is compared with the numerical simulation results in Table-1. Fig. 1 displays the flow velocity
contour across the sphere at terminal velocity.
Fig.1. Flow velocity contour across the sphere at terminal velocity in infinite
medium
155
Table 1. Comparison of drag coefficients and terminal velocity between analytical and numerical procedures for sphere
in infinite medium
Methodology Drag Coefficient (CD) Terminal Velocity (Ut)
Analytical (Cheng model) 0.3505 1.067 m/s
Numerical Simulation 0.3354 1.08 m/s
Error (%) 4.3 -1.21
4. Results & Discussion

In order to investigate the wall effects on terminal velocity and drag forces for the test fuel bundle (19
cylindrical pin geometry-Aluminium dummy), multiple analytical and numerical case studies are performed
for cylindrical geometry by varying the nearness of the wall.
4.1. Wall effects on terminal velocity and drag coefficients of cylinder
A number of case studies are performed in which an Aluminium solid cylinder of dimensions comparable to
that of test fuel bundle with an aspect ratio (E) of 4.615 is allowed to fall under the influence of gravity in a
fall tube filled with water. In order to study the wall effects, a reduced fall tube diameter (D) is chosen for
each case study. The observations are presented below.
Ratio of cylinder diameter to fall tube diameter (dc/D = 0.05). In this case study, drag coefficient and
terminal velocity of the falling cylinder is estimated such that the wall effects of the fall tube diameter are
negligible. Both analytical and numerical procedures are employed in this case study and its results are
compared in table-2. Equations presented by Gabitto and Tsouris [1] are used for analytical determination of
the results. The analytical terminal velocity is observed to be 4.33 m/s whereas the numerical results imply a
value of 4.18 m/s having error of 3.46%. Similarly, drag coefficients observed through analytical and
numerical methods are 0.789 and 0.846 respectively with an error of 7.17%. This suggests that both the
analytical and numerical results are in well agreement. Fig.2 displays the flow velocity contour across the
cylinder at terminal velocity.
Ratio of cylinder diameter to fall tube diameter (dc/D = 0.1). In this case study, drag coefficient and
terminal velocity of the falling cylinder is estimated for reduced diameter of the fall tube in order to
understand the influence of wall effects on the characteristics of the falling cylinder. Different analytical
models and numerical procedure are employed in this case study and its results are presented in table-3. It is
important to note that not many models are available in literature for predicting the wall effects on the free
fall of cylindrical objects (only Chhabra [11]), Newton [9] and De Felice [10] models which are for spherical
objects are used for comparison. As can be seen from table-3, the velocities predicted by the analytical model
are lower (especially Chhabra) in comparison to our numerical prediction. This can be explained on the basis
of changing orientation during the drop through the fall tube. For all the numerical case studies, it is assumed
that the principal axis of the cylinder is parallel to that of the fall tube. However, there has been no mention
as such of the falling orientation in the literature by Chhabra. Since, fall orientation plays a key role in
changing the drag because of the higher resistance caused by increased surface area, therefore, lower velocity
is predicted by Chabbra.
Table 2. Comparison of the drag coefficients and terminal velocities between analytical and numerical procedures for
cylinder in infinite medium
Analytical (Gabitto & Tsouris) 0.789 4.33 m/s
Error (%) -7.17 3.46
156
Fig.2. Flow velocity contour across the cylinder at terminal velocity in infinite
medium
Ratio of cylinder diameter to fall tube diameter (dc/D = 0.167). In this case study, the fall tube diameter is
reduced even further in order to understand the influence of wall effects on the characteristics of the falling
cylinder. Results from different analytical models and numerical method are tabulated and its results are
presented in table-3. As explained in 4.1.2, Chhabra model continues to estimate lower values of the wall
factor and the terminal velocity in comparison to that of the numerical values.
Ratio of cylinder diameter to fall tube diameter (dc/D = 0.25). The ratio of cylinder diameter to fall tube
diameter is further increased to 0.25 with increased wall effects on the fall object characteristics. Results
obtained from different analytical models and numerical methods are compared in table-3. The terminal
velocity and wall factor values obtained from Chabbra are much lower than the numerical values.
diameter is now increased to 0.5. Results obtained from different analytical models and numerical method is
compared in table-3. For these diameter ratios, Chhabra and Newton model are not applicable. Further, De
Felice model predicts a much lower value of the wall factor and the terminal velocity compared to the
numerical results.
Ratio of cylinder diameter to fall tube diameter (dc/D = 0.556). Beyond this ratio of cylinder diameter to
fall tube diameter, results from analytical models ceases to exist as no literature is available for prediction of
wall factor and terminal velocity for these diameter ratios which was the primary objective of the present
investigation on wall effects of falling objects. Results obtained from numerical method are presented in
table-3.
diameter is further increased to 0.625 with increased wall effects on the fall object characteristics. Results
obtained from numerical method are presented in table-3. The terminal velocity and wall factor values
obtained are 2.21 m/s and 0.528 respectively.
Ratio of cylinder diameter to fall tube diameter (dc/D = 0.714). In this case study, terminal velocity and
wall factor of the falling cylinder is estimated for dc/D = 0.714. Numerical results for this case study are
presented in table-3. The terminal velocity and wall factor values obtained are 1.585 m/s and 0.379
157
respectively. It can be seen from these results that with the decreasing fall tube diameter, the wall effects are
significantly affecting the characteristics of the falling cylinder.
Ratio of cylinder diameter to fall tube diameter (dc/D = 0.833). As the fall tube diameter is approaching
the cylinder diameter, the wall effects on the terminal velocity reduction is gaining more significance. In this
case study, terminal velocity and the wall factor of the falling cylinder is estimated for an increased dc/D of
0.833 through numerical analysis. The terminal velocity and wall factor values obtained are 0.83 m/s and
0.198 respectively as shown in table-3.
diameter is further increased to 0.91. Results obtained from numerical method are provided in table-3. The
terminal velocity and wall factor values are reduced to 0.335 m/s and 0.08 respectively.
Ratio of cylinder diameter to fall tube diameter (dc/D = 0.952). In this last case study for cylindrical object
falling freely in a fall tube, terminal velocity and wall factor is estimated for dc/D = 0.952. Numerical results
for this case study are presented in table-3. The terminal velocity and wall factor values obtained are 0.115
m/s and 0.0275 respectively. It is observed that there has been a considerable reduction of the terminal
velocity and the wall factor when compared to case study 4.1.1 where cylinder was falling freely in an
unbounded or infinite medium.
Fig. 3 and Fig. 4 display the variation of terminal velocity and wall factor with respect to the diameter ratio
(dc/D). Both terminal velocity and wall factor follow the same trend as wall factor (f) is directly proportional
to the terminal velocity (Eqn. 1). It can be seen from Fig. 3 & 4 that the available analytical models do not
predict the terminal velocity and wall factor for higher diameter ratios. At lower diameter ratio values, the
error between the results of the analytical values and numerical values is large. This difference is explained in
terms of the orientation of the falling objects. Therefore, in the absence of appropriate analytical models, the
results obtained from numerical simulations are employed for future uses.
4.2. Wall effects on terminal velocity and drag coefficients of fuel bundle
The following case studies estimate the terminal velocity, drag coefficients and the wall factor for gravity-
influenced fall of the fuel bundles.
Ratio of cylinder diameter to fall tube diameter (d/D = 0.05). In this case study, drag coefficient and
terminal velocity of the falling fuel bundle is estimated with large fall tube diameter when compared to the
fuel diameter. Both analytical and numerical procedures are employed in this case study and its results are
compared in table-4. The analytical terminal velocity is observed to be 3.45 m/s whereas the numerical
results imply a value of 3.31 m/s having an error of 4.05%. However, drag coefficients observed through
analytical and numerical methods are 0.789 and 1.73 respectively with a large error estimate of 119.2%. This
suggests that the analytical expressions in the literature available ineffectively estimates drag coefficients for
complex geometries like that of the test fuel bundle. Also, in compared to solid cylinder with similar
dimension and aspect ratio, the terminal velocity obtained is lower. This attribute can be explained in terms
of the higher surface area of the fuel bundle in comparison to that of the solid cylinder. Fig.5 displays the
flow velocity contour across the cylinder at terminal velocity.
4.5 1.0
Numerical Numerical
4.0 Chabbra [4] Chabbra [4]
Wall Factor or Velocity Ratio (f)
Newton [2] 0.8 Newton [2]

3.5
De Felice [3] De Felice [3]
Terminal Velocity (m/s)
3.0
0.6
2.5
2.0
0.4
1.5
1.0 0.2
0.5
0.0 0.0
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
Cylinder Diameter / Fall Tube Diameter (dc/D) Cylinder Diameter / Fall Tube Diameter (dc/D)
Fig.3. variation of terminal velocity with respect to the Fig.4. variation of wall factor with respect to the diameter
diameter ratio ratio
158
Table 3. Comparison of the drag coefficients and wall factor between analytical and numerical procedures for various
cylinder diameter to fall tube diameter ratios (for cylinder diameter, dc = 65 mm)
Sr. D dc/D Methodology Terminal Velocity (Ut) Wall factor (f)

No.
Chhabra model* 3.23 m/s 0.7465
1 650 mm 0.1
+
Newton model 4.13 m/s 0.954
De Felice model+ 3.82 m/s 0.883
Numerical Simulation 4.14 m/s 0.99
Chhabra model* 2.5 m/s 0.578
2 390 mm 0.167
Newton model+ 3.79 m/s 0.876
+
De Felice model 3.43 m/s 0.794
*
Chhabra model 1.58 m/s 0.366
3 260 mm 0.25
Newton model+ 3.15 m/s 0.728
+
+
4 130 mm 0.5
5 117 mm 0.556
6 104 mm 0.625 Numerical Simulation 2.21 m/s 0.528
9 71.5 mm 0.91 Numerical Simulation 0.335 m/s 0.08
10 68.25 mm 0.952 Numerical Simulation 0.115 m/s 0.0275
Note: * Analytical model for cylindrical Particle; + Analytical model for spherical particle
Table 4. Comparison of the drag coefficient and terminal velocity for analytical and numerical procedures for fuel
geometry in infinite medium (d/D = 0.05)
Analytical 0.789 3.45 m/s
Error (%) -119.2 4.05
159
Fig.5. Flow velocity contour across the fuel bundle at terminal velocity in infinite medium
Ratio of cylinder diameter to fall tube diameter (d/D = 0.94). In this case study, terminal velocity and wall
factor of the falling fuel bundle is estimated for reduced diameter of the fall tube in order to understand the
influence of wall effects on the characteristics of the falling cylinder. The dimensions used are comparable to
the dimensions in the actual geometry of the FTL. Fig. 6 shows the flow velocity contour across the fuel
bundle for diameter ratio of 0.94. Because of the higher diameter ratio, no analytical model is available in the
literature for prediction of results for this case study. As can be seen from table-5, the velocities predicted by
the numerical procedure are higher in comparison to numerical prediction of the solid cylinder with similar
diameter ratios. In solid cylinders, presence of fall tube wall in the vicinity considerably reduces the terminal
velocity; however, such is not the case with fuel having comparable dimensions. With the presence of an
open area between the pins (for cooling purposes), the influence of the fall tube wall diminishes.
Table 5. Comparison of the drag coefficients and terminal velocity between analytical and numerical procedures for fuel
geometry (d/D = 0.94)
Methodology Wall factor (f) Terminal Velocity (Ut)

160
Fig.6. Flow velocity contour across the fuel bundle at terminal velocity for
diameter ratio of 0.94
5. Conclusions
In the present work, the investigation of wall effects on terminal velocity of test fuel bundles falling under
gravity within the In-Pool Test Section of the FTL was carried out both analytically and numerically. In order
to achieve this, multiple case studies estimating drag characteristics and terminal velocities with wall effects
were performed on comparable dimension cylinder having similar diameter and aspect ratio. It was observed
that at higher diameter ratios, the reduction in the terminal velocity and the wall factor was very large
attributing to considerable influence of the wall in the vicinity. Further, in the absence of analytical models in
the literature for higher values of the diameter ratio, results obtained through numerical simulations could not
be compared. Subsequently, estimation of terminal velocity, drag coefficient and wall factor was carried out.
It was observed that the velocities predicted are higher in comparison to prediction of same characteristics of
the solid cylinder with similar diameter ratios. As In solid cylinders, presence of fall tube wall in the vicinity
considerably reduces the terminal velocity; however, such is not the case with fuel having comparable
dimensions. With the presence of an open area between the pins (for cooling purposes), the influence of the
fall tube wall diminishes.
161
Nomenclature
CD Drag coefficient dA Projected Area Diameter
Re Reynolds Number Ap Projected Area
f Velocity ratio or wall factor k Volumetric Shape Factor
Ut Terminal velocity in confined medium  Sphericity
U t Terminal velocity in infinite medium Av Surface area of a sphere having the same
volume as of the particle
dp Diameter of sphere mp Mass of the particle
p Density of sphere g Gravitational acceleration
D Fall Tube Diameter dc Cylinder diameter
f Fluid Density Ga Galileo Number
 Fluid Viscosity Ga  Modified Galileo Number
 Sphere-to-tube diameter E Aspect ratio of the non-spherical particle
ds Diameter of sphere with volume equal to that ReT Reynolds Number in the Newtonian Regime
of non-spherical particle
L Length of the Cylinder ReL Reynolds Number in the Stokes and

Transitional flow regime
V Volume of the object T1, T2 Weighting Factor
162
References
[1] Gabitto J, Tsouris C (2008) Drag coefficient and settling velocity for particles of cylindrical shape
jointed plates. Powd. Tech. 183:314-322. doi: 10.1016/j.powtec.2007.07.031
[2] Clift R, Grace JR, Weber ME (1978) Bubbles, Drops and Particles. Acad. Press, New York.
[3] Kahn AR, Richardson JF (1987) The resistance to motion of a solid sphere in a fluid. Chem. Eng.
Comm. 62-135. doi:10.1080/00986448708912056
[4] Haider AM (1987) M.S. Project, Oregon State University.
[5] Miyamura A, Iwasaki S, Ishii T (1981) Experimental wall correction factors of single solid spheres in
triangular and square cylinders, and parallel plates. Int. J. Multiph. Flow 7 (1): 41-46. doi:10.1016/0301-
9322(81)90013-6
[6] Machac I, Lecjaks Z (1995) Wall effects for a sphere falling through a non-newtonian fluid in a
rectangular duct. Chem. Eng. Sci. 50:143-148. doi:10.1016/0009-2509(94)00211-9
[7] Balaramakrishrna PV, Chhabra RP (1992) Sedimentation of a sphere along the axis of a long square duet
filled with non-newtonian liquids. Can. J. Chem. Eng., 70: 803-807. doi:10.1002/cjce.5450700427
[8] Arsenijevic Z Lj, Grbavcic ZB, Garic-Grulovic RV, Boskovic-Vragolovic NM (2010) Wall effects on
the velocities of a single sphere settling in a stagnant and counter-current fluid and rising in a co-current
fluid. Powd. Tech., 203: 237-242. doi: 10.1016/j.powtec.2010.05.013
[9] Newton I (1687) Philosophiae Naturalis Principia Mathematica, Lib. II, Prop. XXXIX, Theor. XXXI,
Camb. Univ. Press, Cambridge.
[10] De-Felice R (1996) Experimental wall A relationship for the wall effect on the settling velocity of a
sphere at any flow regime. Int. J. Multiph. Flow, 22:527-533. doi:10.1016/0301-9322(96)00004-3
[11] Chhabra RP (1995) Wall effects on free-settling velocity of non-spherical particles in viscous media in
cylindrical tubes. Powd. Tech., 85:83-90. doi:10.1016/0032-5910(95)03012-X
[12] Heywood H (1962) Uniform and non-uniform motion of particles in fluids. In: Proc. Symp. interaction
fluids and particles, Inst. Chem. Eng., London. 1-8
[13] Wadell H (1934) The coefficient of resistance as a function of Reynolds number for solids of various
shapes. J. Franklin. Inst., 217: 459-490. doi:10.1016/S0016-0032(34)90508-1
[14] Cheng NS (2009) Comparison of formulas for drag coefficient and settling velocity of spherical
particles. Powd. Tech., 189 (3): 395-398. doi: 10.1016/j.powtec.2008.07.06
[15] Haider A, Levenspiel O (1989) Drag coefficient and terminal velocity of spherical and non-spherical
particles. Powd. Tech., 58, 63-70. doi:10.1016/0032-5910(89)80008-7
[16] Coulson JM, Richardson JF (1977) Chemical Engineering, Vol.2, Perg. Press, Oxford, Chapter 4.
163
Paper ID- 44
Risk Assessment and Management in Construction Projects
Lisyna Priyadarshini1, Paromik Ray1, Prasanta Roy2, Dillip Kumar Bera2

1
M.Tech Students, School Of Civil Engineering, KIIT Deemed to be University, Bhubaneswar, India
(Mail id:-lovelylisyna@gmail.com), (paromikray@gmail.com)
2
Faculty, School Of Civil Engineering, KIIT Deemed to be University, Bhubaneswar, India
(Mail id:-prasanta.royfce@kiit.ac.in),(dkbera.fce@kiit.ac.in)
Abstract. Risk management is the identification, evaluation, and prioritization of risks and necessary steps
required to minimize, monitor, and management control of the probability of risk. This is considered to be an
integral component of project management. Risk assessment is the determination of quantitative or qualitative
estimate of risk related to a well-defined situation and a recognized threat .This generally occurs when the project
exceeds the estimated or planned duration and cost overrun takes place. For any construction project, increase in
cost and time plays a significant role for the outcome of Risks in that project. The objective is to manage the risks in
a project which will result in reduction of the cost as well as scheduled completion of the project. The present work
aims at controlling the outcome of the project comprising of various factors including increase in time and cost trade
off and involvement of various categories of risks. Risk Management also possesses an advantage of analyzing as
well as minimizing the amount of risk that occurs in a construction project. This concludes that management of risk
is essential for efficient and timely completion of project.
Keywords: Project Management, Risk Management, Risk Assessment, Time and Cost Trade off, Construction
Projects.
1 Introduction
At present, the construction industry is one of the most dynamic, risky, and challenging business. Due to complexity
and uniqueness of such construction projects, the risks prevail in the system. The risk management has become the
important part of project management. In each industries risk management practice indicate that the risk
management has the obvious function for improving the efficiency of project implementation, saving the cost and
improving the project profitability [1]. It is a systematic process of identifying, assessing, analyzing, responding to
project risk and setting preference for risk mitigation [2]. Generally, risk is a choice in an environment rather than a
fate. Project risk can be defined as an uncertain event that leads to failing to achieve at least one project objective [3-
5]. The risk management process can improve project performance by controlling the consequences of risky events
on project objectives [6]. It is recognized that it is possible to manage risks but not eradicate them [7]. Systematic
and proactive risk management practices are needed to handle and manage risks so that the success of projects can
be ascertained [8].
164
2 Definition of Risk
Risk is defined as “a situation where there exists no knowledge of its outcomes”
In Macquarie dictionary, it is defined as “Exposure to the change of injury or loss; a hazard or dangerous chance, to
run risks”.
Uncertain events with negative impacts are called risks.
Risk is the chance of something happening that will have an impact on objectives.
2.1 Concept of risk and risk management

Risk management has become an increasingly challenging activity [9]. It has become an essential requirement for
construction projects. Risk management process includes Hazard identification, Risk assessment and Risk control.
Risk is assessed by Qualitative Methods and Quantitative Methods. Risk management is the systematic process of
identifying, analyzing, and responding to project risk, and it includes maximizing the probability and consequences
of positive attributes and minimizing the probability and consequences of attributes adverse to project objectives[10-
11].In current construction management literature, various risk identification and assessment methodologies have
been offered within which the common acknowledged methodology is identifying risks that may emerge during the
execution of projects, assigning ratings to identified risk factors considering their likelihood of occurrence and
impact, and finally calculating the overall risk rating to estimate the overall impact of risk factors on the project
success[8].For many years, risk management in construction projects has been approached using a reductionist
approach that produces poor results and limits the quality of project management. For example, most of the times
risk is handled through the application of contingencies (money) or floats (time) that are not determined based on a
comprehensive analysis of the risks that can affect a particular project, and that in many cases are clearly insufficient
to cover the consequences of risks that do occur during the project realization. Then, in most of the cases projects
end with costs overrun and late [12].
To make an effective and efficient risk management it is necessary to have a proper and systematic methodology
and, more importantly, knowledge and experience of various types. For example, it requires knowledge of the
unforeseen events that may occur during the execution of a project, on the actions that work well or not when one of
these events happens, on ways to assess a risk or estimate the likelihood that it will occur, and so on.
2.2 Sources of Risk in Construction Projects
There are many sources of uncertainty in construction projects, which include the performance of the performance
of the contractor(s), consultant(s), owner(s), availability of resources, environmental conditions, involvement of
other parties (sub contractors, suppliers and vendors), contractual relations, Design changes and errors, Poorly co-
ordinated work, Poor estimates, Poorly defined roles and responsibilities, Unskilled staff ,Natural hazards ,Political
and legal problems[11].
165
2.3 Risk Management process

Manager can plan their strategy based on four step of risk management process.
1. Risk identification
2. Risk assessment
3. Risk response
4. Risk monitoring and control
Fig. 1. Flow chart of Risk Management process
2.3.1 Risk Identification

The first main step in the process of risk management is risk identification. Before management of risk, the probable
risks must be properly identified in a construction project before any difficulties and adverse factors affecting the risk
arises [13]. Without appropriate identification of risks involved in the project, the management and controlling of the
project would be difficult and it cannot be avoided in the project lifecycle [13-14]. During the identification process of
risk, certain statement factors are present which likely to occur at the end of scheduled project duration. This
identification process consists of various techniques through which the risks are identified from the following:- the
experts’ evaluation, the checklists, the periodic document reviews, the internal audit in a company, the periodic
document reviews, the brainstorming and the Delphi technique. The risks which are identified at the time of
construction project are presented in a tabular form in the name the risk register or Ishikawa's diagram [15]. Risk
identification is done by the following methods:-
2.3.1.1Brainstorming:- This technique is most popularly used for risk identification process. This process involves the
gathering of all related persons in a single place. In the presence of facilitator, the persons involved take turns to give
their opinions regarding various aspects of the topic where the facilitator notes down the important factors followed by
reviewing the factors. In the end, the unnecessary factors are eliminated [16].
166
2.3.1.2 Delphi Technique:- This technique possess same similarities with the brainstorming technique but the
difference lies that the participants are unaware of each other and not present in same place. In this factor, the factors
are identified without any permission from the other contributors. The co-ordinator in this method adds up all the
factors and reviews it [16]. This technique is being supported by knowledge of structure along with experience and
creativity being collected from an expert panel supporting the opinion of collective individual rather than individual
opinions[17-18].
2.3.1.3 Interview or Expert Opinion:- This technique consist of the opinions of experts and personnel in a project of
sufficient experience. The participants involved in this process is interviewed for the factors being identified affecting
the risk arise[16].
2.3.1.4 Past Experience:- Past experience is another method of same experienced project where the identification of
risks are formed. The characteristics of the project is used to provide insight about common factors[16].
2.3.1.5 Checklists:- This method consist of useful pre-determined factor list that are used for risk identification in a
construction project that undertakes the risks in the past along with provides a head-start for risk identification
method[16]. This method also consist of the lists of items being marked positive or negative and it could be used in a
group in an interview or by an individual project member[17].
2.3.2 Risk Assessment/Analysis

Following risk identification process, risk assessment is the next step for process of risk management [13]. This process
can also be termed as quantification or analysis stage [15]. Risk assessment is defined as the process in risk
management where the data collected about the likely risks that occurred in the project are analyzed. It can be also
described as the listing of short risks that creates a maximum impact on the project from the identified risks [19-20].
This analysis is done generally by two methods:- qualitative analysis and quantitative analysis. Qualitative analysis
consist of the risks that usually placed in a description scale of a project from level higher to lower. Quantitative
analysis is used for defining the impact created by the risk and the probability identified on numeric estimations [21]. In
construction companies, the qualitative analysis is more suitable than quantitative analysis due to detailed description of
the risk occurred in the site [20,22].
2.3.2.1 Qualitative Analysis

Qualitative Analysis is important to determine the specific risks that are labeled along with the risk responses that are
directed in the project. It also determines the potential effect and possibility of the risks occurred in the objective of the
project. The analysis of this method is dependent on QRA sheet which determines the outcome of the risks occurred on
a five point scale namely:- very low, low, medium, high, very high along with the impact created by the risks namely:-
very low, low, medium, high, very high[23].
Risk Probability Impact

A. Physical Very Low Medium High Very Very Low Medium High Very
low high Low high
Risk 1
Risk 2
B. Financial
Risk 1
Risk 2
Fig. 2. Format of the QRA sheet
167
Further the qualitative analysis is sub-divided into categories namely:- Risk probability and impact
assessment, Probability Impact and Risk Rating Matrix, Risk Categorization and Risk Urgency Assessment.
a) Risk Probability and Impact Assessment:- In the risk probability and impact , the possibility of risk
evaluated is likely to occur along with the impact created in the purpose of the project assessed a positive
effect for opportunity as well as a negative effect coming from threats[24].
b) Probability Impact and Risk Rating Matrix:- This method represents the base of qualitative analysis
where the risks identified are assessed in the previous step where the risk identified are made important
by using various techniques for calculation. A priority score is calculated which identifies the average of
impact and probability occurred in the risk. The features of priority score including the rating ,range and
color are assigned to a particular risk for denoting the impact created [25]. The results obtained shows the
impact and probability created by the risk.
c) Risk Categorization and Risk Urgency Assessment:- The sources obtained for the risk identified is
dependent on two main structures namely:- Work Breakdown Structure(WBS) or Risk Breakdown
structure(RBS) for developing effective risk response. WBS is used for breaking down huge activities
into minor, manageable risks and also generate a hierarchical, linked series of individualistic activities
[26]. RBS classifies the risk on the basis of dependent activities [27].
The second method Risk Urgency Assessment method main concern of risk is how they require the quick
response [24].
2.3.2.2 Quantitative Analysis

This method requires a specific amount of work to be performed for the analysis. It is completely dependent upon the
outcomes and benefits that are weighed from the chosen method which includes the small projects can sometimes need
only identification process and processing actions on the risk identified whereas the bigger projects need more analysis
on depth [20,24]. Moreover, due to presence of more essential resources including complex software and skilled
personnel, this method is perfectly suitable for medium and large size projects [28]. This method can be further
classified into several types:-
a) Scenario technique - Monte Carlo simulation:- The Monte Carlo simulation method is statistically
dependent used for simulation process for assessing the risks. This simulation process is generally used
for risk analysis, estimations and forecasting in the form of different scenarios [29]. This simulation
process is repeated for a number of times until the average obtained from the data collected which
constitute the risks identified. The result obtained from this method is calculated in percentage for the
probabilities of the risk to occur [30].
b) Modeling technique - Sensitivity analysis:- This method represents the analysis process of those
risks that creates a large or great impact in the construction project. The high value of the level of
unresolved particular risk that occurred, the maximum sensitive the risk would be as concerned with
the project objective [28]. Alternatively stated, the specific risks considered to be more critical for the
construction projects which are greatly perceptive and needs sequential actions for mitigation process.
The result generated from this method possess one disadvantage stated that the variables are separately
considered and have no network among them [31-32].
c) Diagramming technique- Decision Trees:- Diagramming Technique or commonly known as
Decision Trees are used when a specific risk identified creates maximum impact on the main two
objectives of the project:- cost and time. This technique is further classified into two types of decision
trees:- Fault tree analysis (FTA) and Event tree analysis (ETA)[28].
The FTA method determines the probability of risk that likely to occur along with the identification of
the risks that can cause damage or failure in single event of the project. The main purpose is to
determine the causes underlined related to the event [19]. Similarly, the ETA method possesses
similarities with the FTA method, but what differentiates between them is the outcome or the result
obtained of the project. The branches of the decision trees has own particular cause of outcome or
concern which ultimately leads to the need of risk assessment in the project [33]. In both the method of
decision trees, the effective causing skills are essential for the possibility of understanding the failure
occurrence and to analyze which failure modes causes what situation in a project[33].
168
2.3.3 Risk Response

The third main step involved in Risk Management process (RMP) signifies the type of action that has to be taken for the
threats processed by the identified risks. The strategy of response being created depends on the kinds of risks that are
concerned [21]. The lower impact risk can be managed systematically than the higher impacted risk with aid from
strategies involved. The strategies involves:- avoidance, reduction, transfer and retention. Apart from these mentioned
responses, the presence of appropriate information should be present to deal the risk occurred. This process is termed
as “Delay the decision” approach which is not permissible for all the situations that arises in a construction project,
especially at the time of handling of critical risks [34].
2.3.3.1 Avoidance/prevention
The risk identified brings negative consequences to the whole construction project, thus being important for reviewing
the purpose of the project. Taking by example, the risk identified creates a significant impact on the construction
project, hence in order to avoid the project changed scope or in worst case for canceling it, the potential risks of a
project exposes can create an impact of success [34]. The word avoidance is termed looking for alternative result in
which the risks may be eliminated. In other words, avoidance is mainly about refusal to accept the risk. This is either
done by refusal of the project that is considered very risky for proceeding clauses with exception present with the tender
[ 35-36].
2.3.3.2 Reduction
This method can be justified when it becomes easy to detect as well as identify the risk causing problems leading to
damage of the project. For reducing the level of causing risk, the areas exposed has to be changed as per the situation
[34]. This process is similar to minimize the potential risks by monitoring the main objectives of the project [37]. The
main way of reduction of risks is by adding expenditure that aim to provide benefits for long term future projects. The
reduction process strategies includes [19] includes:- Quality assurance, Contingency planning, Separation or relocation,
Crisis management, Contract terms and conditions.
The risks can be reduced by sharing among the parties that consist more than suitable resources as well as
information regarding the outcome occurred [37]. This results in taking advantage of ones resources to other. It act as a
mode of sharing responsibilities concerning about the risks of the project [38].
2.3.3.3 Transfer
This method is managed by another person having a greater capability or a capacity for transferring. It is stated that the
risk is usually transferred to those persons who has an expertise of managing it . The person involved usually includes
designer, client, contractor, sub-contractor; etc. depending on the risk characteristic involved resulting in the
reorganization of the risk being eliminated leading to additional work and higher costs involved termed as risk premium
[34]. The recognized risk is only transferred to the other party for managerial purpose [24].
2.3.4.4 Retention
The risk obtained cannot be transferred or obtained; the best possible way is the retention process. For a particular case,
the risk are able to be controlled, for minimizing the impact of the occurrence of the risk [34].
2.3.5 Risk Monitoring and Control
This is last stage of risk management. Risk monitoring is a integral part of any construction risk management process. It
is the important step of risk management process where all possible and potential risk are frequently monitored to
identify any change in status or in case they turn into an issue [21-22]. The aim of this method involves the
determination of cost and time, previously selected arbitrary factors, , quantification of undesirable on time and cost
factors in a construction project along with the progress of different alternative realization factors, the actions causing
minimal damages and the urgent time schedule[15,39-40]. Continuous monitoring of the plan and modifying it
169
appropriately relying on the condition of the project, thus leading to proper evaluation of the design and a back up of
the plan .This is the major standard method in developing a mitigation approach [41].
3 Case Study
The main objective in the case study of an creation of an electric grid in a small town of Batalha signifies the main EDP
distribution system of the particular project along with analysis of the risk identified as well as assessed at the time of
project. This concerned project deals with many uncertainties occurred along the project along with the unpresence of
any historic data available which determines the known risks identified and also the unknown risk that tends to occur
during construction process.
3.1 Development of Risk Management

The development of the risk management process involves the main methods or techniques involved along with the
main Practice Standard of Risk Management process [42-43]. These standards are mainly selected due to task execution
process involved in a particular construction organization [44]. These construction organization uses several benefits
with aid from internationally recognized organizations considering as one of the best methods which is further divided
into:- a) assurance given to the organization that the practice is referred as one of the best practice; b) the main demand
from the external customers[45]; c) assistance as well as involvement from the external recruiters[46]; d) removing of
barriers and extinction contained in the project for the designing as well as development of the project and risk
management[47].
3.2 Risk Management Plan

The development involved for the plan of risk management process is considered the most valuable activity involved in
project risk management, as it confirms the activities that a management group requires for handling the uncertain risks
of the project. Moreover, it also acts as a guide by providing the main tools, documents and procedures required by the
management team for the management process.
3.2.1 Risk Identification

The plan involving management risk defines the techniques, tools and inputs that are required for the identification of
the project risks. Along with the activities involved, the identification as well as potential risk responses are identified
which provides a clear understanding of the risk involved. The review process involving in the risk identification act as
an aid for better understanding of the reality as well as management of the project along with the identification of the
strengths and weakness of the project. The main process includes the brainstorming process where the meetings
organized require scheduled preparation for both the participants and the meetings moderator. A special agenda has to
be present for every meeting. This agenda must provide details about the purpose and subject of the meeting along with
the concepts involved and set of objectives. However, the detailed information should be present as a person to all
participants before, for them to come with relevant ideas and doubts to clarify.
3.2.2 Qualitative Analysis

This process of qualitative analysis is main for the accomplishment in corresponding with the risks identified involved
in the project. In this process, the probability impact matrix is used. Each risks identified in the project is classified
under probabilities of individual occurrence along with the impact created by the risk on the aim of the project through
probability impact matrix results in the classification of the risks occurred as low, medium and high impact risk. Further
for proper identification, a specific color code is present as yellow for moderated, green for low and red for high
impacted risk.
170
3.2.3 Quantitative Analysis

The risk management involved in the process of quantitative analysis is not compulsory for all the identified risks in
the project. This process analysis is correctly done when the impact matrix is categorized above 0.14 along with the
possessor of the project is a distribution of EDP owner. In case, the possessor of the project is a project supplier, the
risks likely to be occurred would have a higher impact on the project resulting in the distribution of EDP not taking any
additional actions in this process. Additionally, even for EDP distribution assessing the detailed risk information, the
expected money should be evaluated for all the project risks occurred.
3.2.4 Plan Risks Response

A Plan Risk Response is mainly identified for all the risks involved even at the acceptance of the risks that does not
take any action. This process possess the information and is attached to the risk register of the project possessing the
records of risk response consisting of individual risk identification and description along with assumptions that involves
the risk causes, possible risk results on schedule, scope, quality and costs, risk responsibility, intervention action,
communication along with risk response interaction.
3.2.5 Risk Monitoring and Control

This risk process has the purpose of evaluating the risk management process for excess controlling the project risk. At
the time of project control progress, the information is recorded on the risk register which is further divided into four
classifications stated:- irrelevant, controlled, requires attention and critical. This analysis involving the risk assessment
is mainly done for controlling the evolution of risk exceeding time along with anticipating the behavior preceding the
actual occurrence.
3.3 Advantages of Risk Management
Most of the organizations tends to recognize the advantages of enterprise risk management. The following are the
benefits of risk management in a construction organization [11].
 Minimum uncertainty
 Successful objectives
 Accuracy
 Decrease in capital cost
 Value creation
 Benefits of identified risks
 Benefits of assessment of risks
 Better treatment of risk
 Reduction of risk
 New opportunities
 Awareness about the risk
 Successful business strategies
3.4 Limitations of Risk Management
 Complex calculations:- complex calculations are involved for management of risk in risk management.
Without the presence of the automatic tool, the calculation becomes difficult.
 Difficulty in implementing:- A long time is taken for gathering and managing the information regarding the
strategic plans in Risk Management. Too much time on assessment and management results in diversion of
resources that can be more profitable.
171
 Performance:- Since the risk management is dealing with only with subjectivity, holding the power of
prospects present in every issue. The difficult implementation of controls can be identified with this process.
 Unlikely events to occur:- If the risk is unlikely for occurrence, retention and dealing with the risk results is a
better option.
4 Conclusion
This paper reviews the various risk management processes which includes the identification, assessment, response ,
monitoring and control that occurs in a construction project. By applying the above processes, it is possible to manage
the potential risk in a simple way. Moreover, it assigns a probability for detecting risk identified that creates a huge
impact on cost, time and quality. These identified risk should be mitigated or eliminated by taking suitable action.
Moreover, it is proved that the results obtained from the impact and probability method may varies from one project to
another considering each projects aim and scope are unique. This risk management process brings profit to the
organization.
References
[1]. Zhou L.F., Yuan G.H., : The risk assessment model of BT construction engineering project financing. 2011
International Conference on Risk and Project Management, Page 169-173(2011).
[2]. Jayasudha K., Vidivelli B., Surjith E.R.G., : Risk Assessment and Management in Construction Projects.
International Journal of Scientific & Engineering Research, Page 387-396(2014).
[3].Mark W., Cohen P.E., Glen R.P., : Project risk identification and management. AACE Int. Trans. INT.01, Page
1–5(2004).
[4].Morote A.N., Vila F.R., : A fuzzy approach to construction project risk assessment. International Journal of
Project Management, Page 220–231 (2011).
[5]. Islam M. S., Nepal M. P., Skitmore M., Attarzadeh M., : Current research trends and application areas of fuzzy
and hybrid methods to the risk assessment of construction projects. Advanced Engineering Informatics, Page 112–
131(2017).
[6]. El-Karim M. S. A., Nawawy O.A. M.El., Alim A.M.A., : Identification and assessment of risk factors affecting
construction projects. Housing and Building National Research Center, Page 202–216(2017).
[7]. Farooq M.U.,Thaheem M.J., Arshad H., : Improving the risk quantification under behavioural tendencies: A tale
of construction projects. International Journal of Project Management,Page 414–428(2018).
[8]. Yildiz A.E., Dikmen I., Birgonul M.T., : Using expert opinion for risk assessment: a case study of a
construction project utilizing a risk mapping tool. Procedia - Social and Behavioral Sciences, Page 519–528( 2014 ).
[9]. Hsueh S.L., : On-Line Multi-Criterion Risk Assessment Model for Construction Joint Ventures in China.
Automation in Construction, Page 607-619(2007).
[10]. Azevedo R.C., Ensslin L., Jungles A E., : A Review of Risk Management in Construction: Opportunities for
Improvement. Modern Economy, Page 367-383(2014).
[11]. Taroun A., : Towards a better modelling and assessment of construction risk: Insights from a literature review.
International Journal of Project Management, Page 101–115(2014).
172
[12]. Salawua R.A.,, Abdullahb F., : Assessing Risk Management Maturity of Construction Organisations on
Infrastructural Project Delivery in Nigeria. Procedia - Social and Behavioral Sciences, Page 643 – 650 ( 2015 ).
[13]. KarimiAzari A.R., Mousavi N., S. Mousavi F., Hosseini S.B., : Risk assessment model selection in
construction industry. Expert Systems with Applications, Page 9105–9111 (2011).
[14]. Bajaj J., : Analysis of contractors approaches to risk identification in New South Wales, Australia.
Construction Management and Economics, Page 363–369(1997).
[15]. Dziadosz A., Rejment M., : Risk analysis in construction project - chosen methods. Procedia Engineering, Page
258 – 265( 2015 ).
[16]. Mhetre K., Konnur B.A., Landage A.B., : Risk Management in Construction Industry. International Journal of
Engineering Research, Page 153-155(2016).
[17]. Ruotolo M.C.A., Garrido M.C., Ribeiro F.M.L., :Application of techniques for the identification of risk in the
E & P ventures Engevista, Page 120-133(2006).
[18]. Garrido M.C., Ruotolo M.C.A., Ribeiro F.M.L.,Naked H.A., : Risk identification techniques knowledge and
application in the Brazilian construction. Journal of Civil Engineering and Construction Technology, Page 242-
252(2011).
[19]. Cooper D.F., Grey S., Raymond G.,Walker P., : Project Risk Management Guidelines : Managing Risk in
Large Projects and Complex Procurements. Chichester . John Wiley & Sons, Ltd, Page 45-59(2005).
[20]. Gajewska E., Ropel M., : Risk Management Practices in a Construction Project – a case study. Civil and
Environmental Engineering Chalmers University of Technology, Page 11-60 (2011)
[21]. Winch G., : Managing construction projects, an information processing approach. Oxford Blackwell
Publishing, Pag 145-149(2002).
[22]. Elkingtin P., Sallman C., : Managing project risks: a case study form the utilities sector. International Journal
of Project Management, Page 49-57(2002).
[23]. Pawar C.S., Jain S.S., Patil J.R., : Risk Management in Infrastructure Projects in India. International Journal of
Innovative Research in Advanced Engineering, Page 172-176(2015).
[24]. PMI (Project Management Institute), : A guide to the project management body of knowledge: PMBOK. 3rd
edition Pennsylvania, Page 99-101(2007).
[25]. Westland J., : Project Management Life Cycle: A Complete Step-by-step Methodology for Initiating Planning
Executing and Closing the Project. Kogan: Page Limited, Page 159-164(2006).
[26]. Maylor H., : Project Management. Edinburgh: Pearson Education, Page 223-230(2005).
[27]. Dallas M.F., : Value and risk management; a guide to best practice. Oxford: Blackwell Pub, Page 41-48(2006).
173
[28]. Morgan G.,Smircich L., : The case for qualitative research. The Academy of Management Review, Page 491-
500(1980).
[29]. Öztaş A., Ökmen Ö., : Judgemental risk analysis process developement in construction projects. Building
Environmental journal, Page1244-1254(2005).
[30]. Darnall R.,Preston J.M., : Project Management from Simple to Complex. Flat World Knowledge, Inc Project
Management Journal, Page 266-270(2010).
[31]. Perry J., : Risk management – an approach for project managers. Butterworth & Co. Vol. 4, Page 211-
216(1986).
[32]. Smith N.J., Merna T., Jobbling P.,. Managing Risk in Construction Projects. 2nd edition Oxford: Blackwell
Publishing, Page 78-85(2006).
[33]. Akintoye A.S., MacLeod M. J., : Risk Analysis and Management in Construction. International Journal of
Project Management, Page 31-38(1997).
[34]. Potts K., : Construction cost management, learning from case studies. Abingdon: Taylor Francis, Page,108-
114(2008).
[35]. Simu K., : Risk management in small construction projects. Department of Civil and Environmental
Engineering, Page 19-25(2006)
[36]. Chapman C.B.,Ward S.C., : Project risk management: Process, techniques and insights. 2 nd Edition.
Chichester: John Wiley and Sons, Page 1-16(2003).
[37]. Rao T.V.M., Burchett J. F., : Applying a Risk Management Process (RMP) to manage cost risk for an EHV
transmission line project. International Journal of Project Management, Page 223-235(1999).
[38]. Szymański P., : Risk management in construction projects. Procedia Engineering, Page 174–182(2017).
[39]. Skorupka D., : Identification and initial risk assessment of construction projects in Poland. Journal of
Management in Engineering, Page 120-127(2008).
[40]. Skorupka D., : Modelling of risk in the building projects. Operations Research and Decisions, Wroclaw
University of Technology, Page 133- 143(2006).
[41]. Suchith Reddy A., : Risk Management in Construction Industry - A Case Study. International Journal of
Innovative Research in Science, Engineering and Technology, Page 10058-10067(2015).
[42]. PMI(Project Management Institute), : A Guide to Project Management Body of Knowledge: PMBOK Guide.
Fifth Edition ed. Pensylvania – USA, Page 98-102( 2013).
[43]. Peixotoa J., Teresoa A., Fernandesa G., Almeidab R., : Project Risk Management Methodology: A Case Study
of an Electric Energy Organization. Procedia Technology , Page 1096 – 1105 ( 2014 ).
174
[44]. PMI(Project Management Institute), : A Guide to Project Management Body of Knowledge: PMBOK Guide.
First Edition ed. Pensylvania – USA, Page 98-102( 2009).
[45]. Morris P., : Managing the Front End: how project managers shape business strategy and manage project
definition. Project Management Institute EMEA Symp, Page 2-8( 2005).
[46]. McHugh O.,Hogan M., : Investigating the rationale for adopting an internationally-recognised project
management methodology in Ireland. International Journal of Project Management, Page 637-646 (2011).
[47]. Haji-Kazemi S., Bakhshehsi A. H. F., : Impact of Standards on Project Management. PM World Today, Page
15-22(2009).
175
Paper ID- 45
Influence of Material Damping on Dynamic Response of Foundation
Suvendu Kumar Sasmal1*, Pradip Kumar Pradhan2
1
Ph.D. Student, Department of Civil Engineering, National Institute of Technology Rourkela, Email:
suvendukumarsasmal@gmail.com
2
Professor, Department of Civil Engineering, Veer Surendra Sai University of Technology, Burla, Sambalpur,
Email:pkpradhan1@yahoo.co.in
*Corresponding author
Abstract. The lifetime of any structure is heavily dependent on the part that holds the structure, i.e. the
foundation. Machine Foundations and foundations in the vicinity of industrial areas are often subjected to
vibratory shocks generated from different machineries. One of the key steps in the current methods of dynamic
analysis of foundation soil system under seismic or machine type loading is to estimate the dynamic impedance
functions i.e., spring coefficient and the damping coefficient. In the present study dimensionless spring
coefficient [k(a0)] and damping coefficient [c(a0)] are computed using cone model for a foundation resting on
homogeneous elastic half space and on a layered soil system. Two modes of vibration i.e., vertical and
horizontal are considered for carrying out the analysis. The material damping is then introduced in the elastic
response using correspondence principle. Using the Kelvin model, the material damping is introduced on the
elastic response and the results thus obtained are compared with two other damping models i.e. Hysteretic
model and Voigot model. The effect of change in damping ratio on the response of foundation based on above
three models is illustrated in the form of dimensionless plots. The study indicates that, by comparison to
Hysteretic and Kelvin model, the Voigot model results less spring coefficients and more damping coefficients. It
is also observed that irrespective of damping models the dynamic impedance functions are strongly influenced
by the modes of vibration.
Keywords: Foundation; frequency domain; cone model; impedance function; material damping
1 Introduction
With the rapid increase in industrialization in a developing country like India, a significant amount of land area
comes nearby different industries. The machine foundations and any other foundation in the close vicinity of
machines are generally subjected to machine- induced dynamic load in addition to the static load. In contrast to
foundations under static load, analysis of foundations subjected to dynamic load, though complicated is an
important task for the sake of lifetime prediction of that foundation. The dynamic response of any foundation is
dominated by the properties of foundation soil which may be homogeneous half space or layered soil consisting
of different layers. One of the key steps to find the dynamic response of foundation is to determine the
impedance functions, i.e. spring/stiffness coefficients and damping coefficients. The impedance functions were
formulated and popularized by Gazetas [4] and Gazetas [3]. Sridharan et al. [9] reported dimensionless graphs
for spring constant using weighted average method. Meek and Wolf [5] formulated the impedance functions of
foundation resting on homogeneous elastic half space, subjected to translational and rotational motion using the
cone model. Also, the model was extended for embedded foundation (Meek and Wolf [6]). Wolf and Preisig
176
[12] determined the spring and damping coefficients of foundation using strength of material approach. Wolf
and Deek [11] observed the response of foundations embedded in incompressible and nearly-incompressible
homogeneous and layered half-spaces with the help of strength of material approach using cones. The
impedance functions for foundation resting on a soil layer underlain by rigid base, using the cone model were
reported by Pradhan et al. [7] and Pradhan et al. [8]. Baidya and Mandal [2] reported that dynamic response of a
foundation resting on a layered soil system depends on both thickness and position of the layer.
Various researchers have derived the expressions for spring and damping coefficients for foundations
resting on homogeneous elastic half space as well as layered soil. However, most of these solutions are elastic
solutions without considering the effect of material damping which is an important phenomenon that takes place
beneath the foundation. In order to fill the void, an attempt has been taken to study the influence of material
damping on dynamically excited foundation using three different models i.e. Hysteretic model, Voigot model
and Kelvin model. Foundations resting on both homogeneous half space and layered soil are analyzed for two
different modes of vibration i.e. vertical and horizontal translation in the present study to minutely observe the
effect of material damping. The major outcomes from the present study may be helpful in designing the machine
foundations.
2 Analytical Methodology
In the present study the elastic response of foundation is formulated in terms of impedance functions using the
cone model. Then, material damping is imposed on the elastic solution using the correspondence principle.
Fig. 1. Cone model for translational motion of foundation resting on homogeneous soil (Meek and Wolf, 1992)
The schematic diagram of cone model for foundation under translational vibration is shown in Figure 1, where
A0 = area of the massless footing, r0 =equivalent radium of circular disk, z0=apex height, z= depth below the
apex, u= displacement, N = reacting force and P0=harmonic load. From the analysis of cone, the governing
equation for impedance function was found to be
𝑆(𝑎0 ) = 𝐾[𝑘(𝑎0 ) + 𝑖𝑎0 𝑐(𝑎0 )] (1)
177
Where Spring coefficient

𝜇 𝑧0 𝑐𝑠 2
𝑘(𝑎0 ) = 1 − 𝑎0 2 (2)
𝜋 𝑟0 𝑐 2
And damping coefficient

𝑧0 𝑐𝑠
𝑐(𝑎0 ) = (3)
𝑟0 𝑐
In the above formulations, µ = Poisson’s ratio, c= appropriate wave velocity, cs= shear wave velocity,
a0=dimensionless frequency, z0/r0 = aspect ratio.
In contrast to homogeneous half space when a foundation rests on layered soil the effect of each layer of soil
generally comes into the act. In this condition a simple approximation is to consider the layered soil as a set of
homogeneous soil layers.
Fig. 2. Cone model for translational motion of foundation resting on layered soil (Pradhan et al. 2004)
The schematic diagram of cone model for foundation resting on layered soil is shown in Figure 2, where d=
depth of soil layer,
The governing equation for dynamic impedance function is represented according to Pradhan et al. by Eq. (4)
𝜇𝑧0 𝑐𝑠 2 2 𝑧 𝑐
[1− 𝑎 + 𝑖𝑎0 0 𝑠 ]
𝑝0 𝜋𝑟0 𝑐2 0 𝑟0 𝑐
̅ (𝑎0 ) =
𝐾 =𝐾 2𝑗𝑑 (4)
𝑢0 ∞ 𝐹 −𝑖𝜔( 𝑐 )
1+2 ∑𝑗=1 𝐸𝑗 𝑒
178
2.1 Material damping
Material damping in general is the loss of energy due to friction, which is independent of frequency of loading.
The damped solutions presented in the study are obtained using the correspondence principle, i.e. the elastic
solutions obtained from the cone model are multiplied by some complex factors as per Eq. (5). These complex
factors are selected as per Avilés, and Pérez-Rocha [1]), which are tabulated as Table 1.
𝑆𝜉 (𝑎0 ) = 𝐾 [(𝑘(𝑎0 ) + 𝑖𝑎0 𝑐(𝑎0 )](𝜂) (5)
𝑘(𝑎0 ) = 𝑟𝑒𝑎𝑙 (𝑆𝜉 (𝑎0 )) (6)
𝑐(𝑎0 ) = (1⁄𝑎0 )𝑖𝑚𝑎𝑔𝑖𝑛𝑎𝑟𝑦 (𝑆𝜉 (𝑎0 )) (7)
Table 1. Complex factors for different damping models
Model Complex factor (η)

Hysteretic 1 + 2𝑖𝜉
Voigot 1 + 2𝑖𝑎0 𝜉
Kelvin (1 + 2𝑖𝑎0 𝜉)/(1 + 𝑖𝑎0 𝜉)
The impedance functions for foundation resting on homogeneous elastic half space, obtained after imposing the
material damping on the elastic solutions, for vertical and horizontal modes of vibration are shown in Figure 3
and Figure 4 respectively along with elastic solutions. As observed from both the figures when Hysteretic model
is imposed, the spring coefficient becomes a linear function of dimensionless frequency whereas the damping
coefficient becomes a hyperbolic function of dimensionless frequency.
From Figure 3 it is observed that for a particular mode of vibration of footing, as the damping ratio decreases the
spring coefficient attains higher value and damping coefficient attains lower value for a particular dimensionless
frequency.
It is observed that when material damping is imposed the Voigot model produces smaller spring coefficients and
larger damping coefficients. When damping coefficients are concerned, the Voigot model although produces
greater damping cofficients than the elastic solution, it remains constant throughout, i.e. the Voigot model is
independent of dimensionless frequency. By comparison of Figure 3(a) with Figure 4(a) it is noted that the
damped solutions for spring coefficient attain higher value with dimensionless frequency when the foundation is
subjected to horizontal translation. The opposite phenomenon is observed for damping coefficients i.e. damping
coefficient attains higher value when the foundation is subjected to vertical translation. Hence it can be inferred
that keeping all other parameters constant, more spring coefficients are obtained when the foundation is
subjected to horizontal translation and more damping coefficients are obtained when the foundation is subjected
to vertical translation.
179
(a)
(b)
Fig. 3. Effect of material damping on impedance function of foundation resting on homogeneous half space, under vertical
motion (a) ξ = 0.25, (b) ξ = 0.10
180
2.0
 Elastic
 Hysteretic
1.5
Voigot
Kelvin
1.0
Spring coefficient kh(a0)
0.5
0.0
-0.5
-1.0
-1.5
-2.0
0 1 2 3 4 5
Dimensionless frequency, a0
(a)
(b)
Fig. 4. Effect of material damping on impedance function of foundation resting on homogeneous half space, under
horizontal motion (a) ξ = 0.25, (b) ξ = 0.10
181
Fig. 5. Effect of material damping on impedance function of foundation resting on layered soil, under vertical motion
Fig. 6. Effect of material damping on impedance function of foundation resting on layered soil, under horizontal motion
The variations of impedance functions with respect to dimensionless frequency for foundation resting on layered
soil are presented in Figure 5 and Figure 6 for vertical and horizontal modes of vibration respectively. It is
observed that irrespective of the modes of vibration the Voigot model results in less spring coefficients and the
Hysteretic model results in more spring coefficients. Comparing Figure 5 and Figure 6, it is also observed that
for same value of dimensionless frequency, more spring coefficient is obtained when the foundation is excited
by horizontal dynamic load. Similar trends are observed in damping coefficient for all the three models with no
significant deviation at lower frequency range. However, at higher frequency Voigot model predict larger values
of damping coefficient.
182
4 Conclusions
From the analysis of influence of material damping on the dynamic response of foundation using three different
models i.e., Hysteretic model, Voigot model and Kelvin model, the following major conclusions are inferred.
1) For foundation resting on homogeneous half space, spring coefficient is a linear function of
dimensionless frequency and damping coefficient is a hyperbolic function of dimensionless frequency,
when hysteretic model is introduced.
2) For foundation resting on layered soil all the three damping models have very good similarity for lower
range of dimensionless frequency i.e. below 1.5 for vertical mode of vibration and below 1 for
horizontal mode of vibration.
3) By comparison to Hysteretic and Kelvin model, Voigot model results less spring coefficient and more
damping coefficient for all the cases investigated.
4) Introduction of material damping produces less spring coefficient and more damping coefficients when
the foundation is subjected to both horizontal and vertical translation.
References
1. Avilés J, Pérez-Rocha LE (1996) A simplified procedure for torsional impedance functions of

embedded foundations in a soil layer. Computers and Geotechnics 19(2): 97-115.
2. Baidya DK, Mandal A (2006) Dynamic response of footing resting on a layered soil system. West
Indian J Eng 28(2): 65-79.
3. Gazetas G (1983) Analysis of machine foundation vibrations: state of the art. International Journal of
Soil Dynamics and Earthquake Engineering 2(1): 2-42.
4. Gazetas G (1991) Foundation vibrations, Foundation engineering handbook: 553-593
5. Meek JW, Wolf JP (1992) Cone models for homogeneous soil I. Journal of geotechnical
engineering 118(5): 667-685.
6. Meek JW, Wolf JP (1994) Cone models for embedded foundation. Journal of geotechnical
engineering 120(1): 60-80
7. Pradhan PK, Baidya DK, Ghosh DP (2004) Dynamic response of foundations resting on layered soil by
cone model. Soil Dynamics and Earthquake Engineering 24(6): 425-434.
8. Pradhan PK, Mandal A, Baidya DK, Ghosh DP (2008) Dynamic response of machine foundation on
layered soil: cone model versus experiments. Geotechnical and geological engineering 26(4): 453-468.
9. Sridharan A, Gandhi NSVVSJ, Suresh S (1990) Stiffness coefficients of layered soil systems. Journal
of Geotechnical Engineering 116(4): 604-624.
10. Wolf JP (1994) Foundation vibration analysis using simple physical models. Prentice-Hall, Englewood
Cliffs, NJ
11. Wolf JP, Deeks AJ (2004) Cones to model foundation vibrations: incompressible soil and axi-
symmetric embedment of arbitrary shape. Soil Dynamics and Earthquake Engineering 24(12): 963-978.
12. Wolf JP, Preisig M (2003) Dynamic stiffness of foundation embedded in layered half space based on
wave propagation in cones. Earthquake engineering & structural dynamics 32(7): 1075-1098.
183
Paper ID- 48
Comparison of Seismic Performance of Composite (RCS) Frame with RC

Frame using Pushover Analysis
OinamManoranjan Singh1 and S.S. Ningthoukhongjam2
1
M.Tech. Student,Department of Civil Engineering, Manipur Institute of Technology, Imphal,
oinammanoranjan@gmail.com
2
Department of Civil Engineering, Manipur Institute of Technology, Imphal, sukunao@gmail.com
Abstract. In this paper, comparison of seismic performance has been investigated between RCS (Reinforced
Concrete column- Steel beam) composite frame and RCC (Reinforced cement concrete) frame using SAP2000
software. Nonlinear static analysis (pushover) is employed for evaluation of seismic performanceof bothRCS and
RC frames.Both the RCS and RC frames are similar in geometry and as well as sectional members used except the
beams in RCS frame are replaced by steel member of equivalent flexural capacity of beams in RC frame. Seismic
performances of both RCS and RC frames in terms of base shear, spectral displacement, etc. are investigated. It has
been observed that RCS frame is more cost effective than RC frame of equivalent seismic performance.
Key words: RCS frame• Pushover analysis• Performance point• Pushover curve.
1 Introduction
Reinforced concrete and steel structures have been used as the general trend of materials for the construction
ofbuilding for centuries. In the past few decades, numerous researches, [1],[2],[3],[4] started working on optimum
combination of steel and concrete in construction of building which then introduced Reinforced Concrete-Steel
(RCS) frames to the engineering society.RCS composite moment resisting frames consist of Reinforced Concrete
(RC) columns and Steel (S) beams. RCS frames have manyadvantages than traditional RC or steel frame from
structural, economical and construction points of views. From construction view point, RCS frames are easy in
construction since erection of steel beam is simple as compared to casting of RCC beam which includes form work,
reinforcement detailing, casting of concrete, etc. Further, erection of steel beam requires much less time than casting
of RCC beam since casting of RCC beams requires curing times in terms of weeks. From structural point of view,
the lateral stiffness of RCS frame is greater than that of equivalent strength of RCC or steel frame since weight of
the building is reduced in RCS frame due to the use of composite floor slabs which are lighter in weight than RCC
floor slabs. Hence, all the member sections, foundation size, inertial forces, etc. in RCS frame will be reduced which
will minimise the cost of the building construction. Another advantage of RCS frame is that there is provision for
providing long-span (column free) steel beams due to lighter weight of composite floors. Further, these composite
systems have the ability to accommodate various innovative construction techniques which can lower the overall
cost and expedite the process of erecting buildings.
Different authors have been working on different areas of RCS frame. Mirghaderi et al. [1] recommended a
new moment connection between steel beams and a reinforced concrete column (RCS). In this proposed connection,
two parallel beams pass from both sides of the column and are welded to the cover plates near the concrete column
in the joint area. Rathod et al. [2] performed seismic analysis using pushover analysis on RCC, steel and composite
high rise building (G+11) and different parameters like displacement, storey drift, Performance point, and base shear
are plotted. It is concluded thatcomposite sections are more preferable than RCC for high rise building. Prakarsh et
al.[3] performed a comparative study of analysis and design of R.C. and steel structures using ETABS. Storey drifts
of steel structures are found comparatively more than RC structures (greater weight) within the permissible limit.
And if bare frame and infill frame are compared then bare frames are having greater values of storey drift due to
184
their lesser weights. Cholekar et al.[4] conducted comparative analysis of multi-storeyed RCC and Composite
building due to mass irregularity. It is observed that the mass irregularity can affect all the parameters of storey drift,
base shear, dead weight, shear force and joint displacement during earthquake. From the study of various research
papers it can be concluded that Composite RCS structures are best solution for medium and high rise buildings and
they resulted in speedy construction. Steel option is better than RCC but the composite option for medium and high
rise building is best. Composite structures resulted into lighter construction than traditional concrete construction as
well as speedy construction. So completion period of composite building is less than RCC building (Rathod et al.
[2]). Most of the research works on RCS structure focuses on the connection joint between the reinforced concrete
column and steel beam. Experimental studies have shown that RCS joints have excellent strength and stiffness
(Sheikh et al. [5]). And the connection joint between the steel beam and reinforced concrete column can be assumed
rigid as proven by EysaSalajegheh and HoomanHabibAgahi[6].Many authors have been investigating RCS frame in
different areas but as per the author knowledge, none of them performed effective cost analysis of RCS frame over
equivalent RC frame.
In this paper, a residential of G+3 multi-stories RCC building located at Nagamapal area, Imphal, Manipur is
studied for Pushover Analysis using SAP2000. The same building is converted into RCS frame in SAP2000
software by replacing the beam members with I-section structural steel so as to model a composite RCS building
and studied by pushover analysis. The connection joint between the steel beam and reinforced concrete column is
assumed rigid as suggested by Eysa Salajegheh and Hooman Habib Agahi [6]. The basic loading on both types of
structure is kept equal. The structural steel for beam is selected so as to have identical flexural strength as that of the
RCC beam such that seismic performance and cost comparison can be made between the two structures.
2 Methodology
Nonlinear static analysis known as pushover analysis has been adopted for the analysis of both RCC and RCS
frames in this paper. In pushover analysis, a structure is subjected to gravity loading and a monotonic lateral load
pattern which is constantly increasing over elastic and inelastic behaviour until the target displacement is attained. It
can be executed as force-controlled (used when the load is known, for instance gravity loading) or displacement-
controlled (used when specified drifts are known). A curve is obtained between base shear and roof displacement in
pushover analysis which is known as the capacity curve.
Figure 1. Pushover analysis [7].
Pushover analysis has been the preferred method for seismic performance evaluation of structures by the major
rehabilitation guidelines and codes because it is conceptually and computationally simple. Pushover analysis allows
185
tracing the sequence of yielding and failure on member and structural level as well as the progress of overall
capacity curve of the structure. The nonlinear static pushover procedure was originally formulated and suggested by
two agencies namely, Federal Emergency Management Agency (FEMA) and Applied Technical Council (ATC),
under their seismic rehabilitation programs and guidelines. This is included in the documents, FEMA356 and
ATC40.
Figure 2. Different stages of plastic hinge [7].
The figure above describes the typical force-deformation relation. Five points labeled A, B, C, D and E are
used to define the force deflection behavior of the hinge and these points labeled A to B – Elastic state, B to IO-
below immediate occupancy, IO to LS – between immediate occupancy and life safety, LS to CP- between life
safety to collapse prevention, CP to C – between collapse prevention and ultimate capacity, C to D- between C and
residual strength, D to E- between D and collapse >E – collapse. In SAP2000, those points could be identified by
color bands to understand how plastic hinges form in each stage Figure 3 where IO, LS and CP mean immediate
occupancy, life safety and collapse prevention respectively.
3 Structural Frame Modelling
An idealized G+3 RCC building has been considered for investigation using Pushover analysis in SAP2000
software.The building is four storeys high. The bay lengths are 3.5 m in both the directions as shown in the figure 3.
The plan dimension is 14m X 10.5m. The same building is converted into RCS building in SAP2000 software by
replacing the beam members with I-section structural steel so as to model a composite RCS building. The structural
steel for beam of RCS building is selected so as to have identical flexural strength as that of the RCC beam such that
seismic performance and cost comparison can be made between the two structures. The structural data of both RCS
and RCC building is given in table 1 and 2. The basic loading on both types of structure is kept equal.
186
Figure 3. Plan layout of proposed building.
3.1 Flexural strength of beams
Figure 4.Flexural strength of beams
Since the flexural strength of the ISMB225 is nearest to the flexural strength of the existing RCC beam, the RCS
frame is modeled by replacing the RCC beam with the steel beam of ISMB225 structural steel. The connection joint
between the steel beam and reinforced concrete column is assumed rigid. The basic loading on both types of
structure is kept equal. The structural steel for beam is selected so as to have identical strength as that of the RCC
building such that a cost comparison can be made between the two structures.
3.2 Material Properties
The material used for RC construction is reinforced concrete with M-20 grade concrete and Fe-500 grade
reinforcing steel. The structural steel used for steel beam is I- section ISMB 225. The Stress-Strain relationship used
is as per IS 456:2000 and IS 800-2007. The basic material properties used are as follows:
187
• Modulus of Elasticity of steel, Es = 210000 MPa

• Modulus of Elasticity of concrete, EC = 24890 MPa
• Characteristic strength of concrete, fck = 20 MPa
• Yield stress for reinforcing steel, fy = 500 MPa
• Yield stress for structural steel, fy = 250 MPa
3.3 Structural data for RCC building
Table 1.Structural data for RCC building

Plan dimension 14m x 10.5m
Total height of building 12.5 m
Height of each storey 3.5m for ground floor, 3 m for above ground
floors
Beam 0.35 X 0.3 m
columns 0.35 X 0.35
Thickness of slab 130 mm
Live load (as per IS: 875 part-2 2.5 kN/m2 (for typical floors), 1.5 kN/m2
1987) (roof)
Seismic zone V
Importance Factor 1
Soil condition Medium soil (assumed)

Response Reduction Factor 5
Depth of foundation 1.5m
Grade of concrete M20
Grade of steel Fe500
Damping ratio 5%
3.4 Structural data for composite RCS building
Table 2.Structural data for RCS building

Description RCS
Plan dimension 14m x 10.5m
Total height of building 12.5 m
Height of each storey 3.5m for ground floor, 3 m for above ground
floors
Beam ISMB 225 (I section)
columns 0.35 X 0.35
Thickness of slab 120 mm
Live load (as per IS: 875 part-2 2.5 kN/m2 (for typical floors), 1.5 kN/m2
1987) (roof)
Seismic zone V
Importance Factor 1
Soil condition Medium soil (assumed)
Response Reduction Factor 5
188
Depth of foundation 1.5m

Grade of concrete M20
Grade of steel Fe500
Damping ratio 5%
4 Building analysis
The general finite element package SAP 2000 (Version.19) has been adopted for the analyses. A three dimensional
model of each structure has been modelled to undertake the non linear analysis. Beams and columns are modelled
asnonlinear frame elements with lumped plasticity at the start and the end of each element. SAP 2000
providesdefault hinge properties and recommends M3 hinges for columns and P-M-M hinges for beams as described
in FEMA 356. Both the building frames that is the existing RCC building and the RCS composite building are
analyzed first by considering linear static analysis for defining gravity load case and then a lateral pushover analysis
was performed in a displacement control manner.
Figure 5. 3D model of the proposed building developed in SAP2000.
5 Result and discussion
The results from the pushover analysis of the two buildings are as follows.
5.1 Base shear and displacement
Table 3. Base shear and displacement

Building Target displacement Elastic base shear Inelastic base shear
RCC building 0.052 m 900.64 KN 2283.016 KN
RCS building 0.066 m 803.56 KN 2111.209 KN
From table 3, the target displacement is more in RCS frame than the RCC frame. The base shear of RCS having
structural beam ISMB 225 is more than the base shear of RCC building. In case of RCC building, the inelastic base
189
shear is 2.53 times of the elastic base shear which suggests that the building is seismically safe and has good
resistance against earthquake. Similarly, in case of RCS building, the inelastic base shear is 2.62 times the elastic
base shear which also shows that the building has good resistance against earthquake. The base shear of RCS
building is less than RCC building owing to reduced dead load in case of RCS frame. The RCC building has more
stiffness as compared to RCS building which results in less roof displacement than RCS building (as shown in table
3).
5.2 Demand- Capacity curve (pushover curve)
The resulting capacity curves for the two buildings are shown in figure 6 and figure 7 for RCC and RCS composite
building respectively. They are initially linear but start to deviate from linearity as the beams and the columns
undergo inelastic actions. When the buildings are pushed well into the inelastic range, the curves become linear
again but with smaller slope. For a target displacement of 0.052m for the RCC building, the base shear of whole
structure is 2283.016 KN. For RCS composite building, for a target displacement of 0.066m the base shear is
2111.209 KN.
Figure 6.Performance point for RCC frames
190
Figure 7.Performance point for RCS frames
From the figure 6 and 7, it is obvious that the performance point (intersection of capacity and demand curve) of
the RCS building is nearer to the yield point of the capacity curve as compared to the performance point of RCC
building. This suggests that the RCS building has more reserve strength and stiffness left after being pushed into the
inelastic range.
5.3 Plastic Hinge Mechanism
Table 4.Hinge states in each steps of pushover analysis (RCC building)
191
Table 5.Hinge states in each steps of pushover analysis (RCS building)
Table 4 and 5 shows the hinge state details at each steps of the analysis of RCC and RCS composite building
respectively. In case of RCC building, it can be seen that for the performance point taken as step 7 (which actually
lies between steps 6 and 7), 71.8% of the hinges lies within B and only 28% of hinges lie between B and IO. Hence
the damage will be limited. As for the RCS building, at the performance point taken as step 3, 89.1% of the hinges
lies within B and 10.8% of the hinges lies between B and IO. Hence the damage level in case of RCS building will
be more limited than RCC building.
Figure 8 and 9 shows the hinge states during various stages in course of the analysis.
Fig.8. (a) Hinge state in the structure model of RCC frame at step 2.Fig. 8. (b) Hinge state in the structure model of RCC frame at step 5.
192
Fig. 8. (c) Hinge state in the structure model of RCC frame at step 8.Figure 9. (a) Hinge state in the structure model of RCS frame at step 2.
Figure 9. (b) Hinge state in the structure model of RCS frame at step 5.Figure 9. (c) Hinge state in the structure model of RCS frame at step 8.
5.4 Cost Analysis
From the cost analysis of both the buildings that is RCS and RCC building, the following result is obtained. The cost
analysis excludes the cost of formwork and the loading cost of structural steel in RCS building and other
miscellaneous cost.
Building cost
RCC 2002441
RCS composite 1545203
Difference 457238
193
Figure 10.Cost comparison of RCC and RCS building
6 Conclusions
Comparison of seismic performances between RCC and RCS frames has been investigated using SAP2000 software.
Nonlinear static analysis, pushover method, is employed for evaluation of seismic performance of both the frames.
Based on the analysis the following conclusions are drawn:-
• The base shear of RCS composite building is observed to be less than that of RCC building since light
weight composite floors are used in RCS frames. Light weight and inherent ductility characteristics of the
RCS building gives better seismic performances than RCC building during earthquake.
• RCS building possesses more reserve strength than RCC building since the performance point of RCS
building occurs near the yield point in capacity curve than that of RCC building.
• RCS composite structures are found to be more economical than RCC structure since all the member
sections, foundation size, inertial forces, etc. in RCS frame will be reduced due to lighter seismic weight
which will minimise the cost of the building construction.
• The construction time for RCS building is less than that of RCC building since erection of steel beams and
composite floor slabs require much less time as compared to that of RCC beam and slabs.
REFERENCE
1. Seyed Rasoul Mirghaderi, Nasrin Bakhshayesh Eghbali, Mohammad Mehdi Ahmadi(2016)

Moment connection between continuous steel beams and reinforced concrete column under cyclic loading.
Journal of Constructional Steel Research 118: 105-119.
2. Amol Vansing Rathod, Mohammed Ishtiyaque(2017) Pushover Analysis of Building Structure.
International Journal of Innovative Research in Sciences, Engineering and Technology. Vol. 6, Issue 3.
194
3. Prakarsh Sangave, Nikhil Madur, Sagar Waghware (2015) Comparative Study of Analysis and Design of
RC and Steel Structures. International Journal of Scientific and engineering Research. Volume 6, Issue 2,
256.
4. Swapnil B. Cholekar, Basavalingappa S. M. (2015) Comparative analysis of multi-storeyed RCC and
Composite building due to mass irregularity. IRJET, Vol. 2, Issue 4.
5. Sheikh T. M. (1987) Moment Connection between Steel beam and concrete columns. Bureau of
Engineering Research, University of Texas, Austin.
6. Eysa Salajegheh, Homan Habib (2008) Performance based design of RCS frames. 14 th World Conference
on Earthquake Engineering, Beijing, China.
7. FEMA 440 (2005) Improvement of Non linear Static Seismic Analysis Procedures. Department of
Homeland Security Federal Emergency Management Agency, Washington.
8. Anish N. Shah and Dr. P. S. Pajgade(2013)Comparison of RCC and Composite Multi-storied
Building.International Journal of Engineering Research andApplication, Vol.3, Issue 2, March-April 2013,
pp.534-539.
9. IS 456-2000, ‘Indian Standard code of practice for Plain and Reinforced concrete’, Bureau of Indian
Standards, New Delhi, India.
10. IS 1893 (2002) Criteria for Earthquake Resistant Design of Structure. BIS, Fifth revision.
195
Paper ID- 50
sutrupta.132@gmail.com
A REVIEW ON GEO-POLYMER PERVIOUS CONCRETE BY USING

RECYCLED COARSE AGGREGATE
Sutrupta Mohapatra1, Dillip Kumar Bera3, Ashoke Kumar Rath2
1
M.tech student; School of Civil Engineering, KIIT (deemed to be University), Bhubaneswar, India
Mail id: sutrupta.132gmail.com
3
Mail id: dberafce@kiit.ac.in
2
Mail id: akrathfce@kiit.ac.in
Abstract
Green concrete is the future of the extensive study on construction materials as it
is environmentally safe and by making it pervious it serves a very good purpose
of allowing precipitated rain water to pass through. In this review paper use of
recycled coarse aggregate for making of geo-polymer pervious concrete (GPC)
was studied. Fly ash (FA) & Bottom ash are the by-product of coal combustion.
By using sodium silicate and sodium hydroxide as binding material we can
replace conventional cement and form GPC. The ratio of FA & recycled coarse
aggregate (RCA) is 1:8 by weight. NS/NH ratio is 0.50. The study also
encompasses the durability properties of GPC. Maximum compressive strength as
per the studies performed is 13.6 Mpa curing under temperature of 60-90 degrees
of centigrade. It shows satisfactory mechanical and durability properties same as
ordinary pervious concrete. The oven curing is the most important part for
strength development. The water permeability and void content are tested in
accordance to ASTM C 1688 and calculated. Highest water permeability
coefficient is calculated as 5.96 in the studies performed. In mechanical properties
highest compressive strength is shown as 13.6Mpa and a good split tensile
strength similar to ordinary pervious concrete. Recycled coarse aggregate water
absorption is 4.9%and obtained from common demolition sites. These RCA are
the common by-product of construction demolition and waste material by using
this RCA partial replacement of Natural coarse aggregate is done in studies and it
shows acceptable properties of pervious concrete. Due to high permeability and
water absorption properties efficient conservation of rain water can be done as it
may be the solution for the ground water recharge globally.
Keywords- Geo-polymer pervious concrete (GPC) ;Fly Ash(FA) ;Recycled coarse aggregate(RCA).
1. Introduction
`Concrete is the most diverse and popular building material in the whole world. And construction is a very
fast growing industry worldwide. 260 Metric Tons of cement is required per year and in 10 years the demand will
increase by 25%. As lime stone is the key component of the cement and it is a scarce material obtained from natural
sources in 25-50 years people will suffer from acute lime stone shortage. Again if world considers the CO 2 emission
then it can be seen that that the world is also suffering from the adverse effects of green house gas emission.
Production of Ordinary Portland Cement (OPC) contributes 7% of green house gas to the world wide total production
of green house gas. This 7% CO2 emission by production of cement includes the calcinations process, fuel burning &
196
energy and transportation. Also conventional cement concrete infrastructures maximum survival chance in adverse
environment (acidic environment) is 20-30 years. So now a day’s researchers are emphasizing on production of more
green concrete which is environmentally friendly and in which partially or fully replacement of conventional cement
can be done. In their way for production of more environmentally friendly construction material inorganic alumino-
silicate polymer, known as Geo-polymer is a big step towards a hope for protection of our planet.
Geo-polymer is a binder material which can be used instead of conventional cement paste. It binds loose
course aggregates, fine aggregate and un-reacted material and produces concrete type dense material. In Geo-polymer
concrete Si-Al in fly-ash activated by alkali solution made with the mixture of Sodium silicate and Sodium
hydroxide. This is an inorganic polymer binder whose micro structure is amorphous. Pervious concrete is a type of
concrete which sanctions dihydrogen monoxide and air to pass through it. Opportune utilization of pervious concrete
is apperceived by U.S Environmental Protection Agency (EPA). As a lot of storm comes in U.S so the runoff water
quantity is more and this water makes the real estate development costly, due the introduction of more drainage water
lines and expensive sewage system. Pervious concrete is used for fast flush, pollution control, storm and natural
precipitation water management. It provides less use of drainage lines and helps in ground water recharge as water
can pass through it and can be infused with soil easily [13]. According to Tawatchai Tho-in et al [2] the pore size in
GPC is 2-8 mm; void content can be seen 18-35%, compressive strength can be seen as 2.8-28 Mpa.
Very less authors reported upon Geo-polymer pervious concrete (GPC) as it is a very good environmentally
friendly construction material and have a good strength than conventional pervious concrete. This topic needs to be
explored more and more research work needs to be done in this field. This is a review on the work done till now and
the application, advantages and limitations of GPC.
2. Geo-polymer Concrete
An alternative binder like “Geo-polymer “a new type of materials was established by Joseph Davidovits in
St. Quentin, France, in the 1970s, since 1990s, Fly ash based geo-polymers are playing a high attention and become a
part in the Geo-polymer family. As an unique binder, the behaviour of fly ash based geo-polymer is encouraging, in
a challenging platform where Portland cement concretes are not environmentally acceptable.
In the present scenario of research field, utilization of 100 % fly ash is possible as a binder for both mortar and
concrete by activating them with an alkali solutions, like; Na/ K OH, silicate salts, and non silicate salts.
Z.T. Yao et al [9] reported that alkali activation mechanism of two types, [3] first mechanism is in the
presence of low silicious and aluminous materials with alkali solutions has produced calcium silicate hydrate gel (C-
S-H), i.e analogous to hydrated product obtained from Portland cements concrete [3,4,5]. The second mechanism
involves the high silicious and aluminous materials activation with a highly alkaline solution produces amorphous
alumino-silicate cementitious hydrates, which is an inorganic binder through a polymerization process.
197
2.1 Geo-polymerization
According to Davidovits [1] the polymerization process necessitates strongly alkaline-solutions to react the silica &
alumina ions present in the FA, amount of dis-solution Si to Al ratio in FA are major factors in Geo-polymerization
clarified that Si and Al rich materials are the base material in the formation of Geo-polymers. The majority of studies
reported on the use of alkaline-silicate solutions for the solution of raw materials for the formation of Geo-polymers.
It has been also mentioned that dissolution process is enhanced due to silicate activation of the basic materials and
gives rise to constructive strength properties. Class F fly slag is ideal as a source material than Class C type fly slag as
of high percentage of SiO2 and Al2O3. The polymerization process may hamper due to the content of CaO in the base
material and also change the microstructure of geo-polymer. The majority of the modern process of geo-polymer
paste involved a sequential process of operations like firstly mixed the source materials in dry state and after that
alkaline solution is added for a specified period of time.
In Geo-polymerisation, the role of alkali solution is more predominant as it reacts with source materials on
development of strength with respect to cement concrete production where the presence of alkali solution is
unwanted.
2.2 Necessity of Geo-polymer Pervious Concrete
In the last decade Geo-polymer binders are emerging technology, which are the possible alternatives of
conventional concrete use. By combining Geo-polymer binders with pervious concrete by replacing conventional
concrete system we can achieve many solutions for general construction and real estate development. GPC can
efficiently provide ground water recharge as well as provides a good less costly efficient sewage system. No
excessive expense on providing costly sewage system for developing a real estate. Geo-polymer binders provides
good bonding strength as well as making pervious concrete with Geo-polymer binder serves the purpose of a good
eco-friendly construction material with acceptable properties. Reduction of conventional cement use can be achieved
by using Geo-polymer binder. GPC (Geo-polymer Pervious Concrete) provides a path way for rain/storm water to
pass through it making the water infiltrate the soil providing good ground water recharge locally. This also allows the
trees to grow properly as it can pass water and air through.
As per the recent news in 2014 in United States 1 million tons of demolition waste is produced, in which 76.6 million
tons of asphalt were found. Demolition wastes are creating problem worldwide as they need proper attention for
managing them. But it can be reused and can be constructed again with benefits to eco-system. By reusing demolition
waste cleaning of waste material is done also it need less managing.
2.3 Factors affecting geo-polymer concrete

• Molarity: Strength of the geo-polymer mortar directly relates with Molarity of NaOH solution. Achievement of high
comp. str. is based on higher concentration of NaOH.. A numbers of tests with different molarities of NaOH solution
starting from 4M to 20M have been conducted by various authors.
198
• Temperature imposed to Alkali Activator : Most of the author have investigated geo-polymer material in the
elevated temperature environment after a rest period of 3 hour to 48 hours . Various authors are adopted temperature
varied from 350C to 1000C in heat curing.
• Fineness of Fly ash: The fineness related to size and specific surface of fly ash-shows a most important parameter on
the strength parameter of the geo-polymer concrete, if fineness is more the strength gain is also more.
• Curing: It takes a key role to develop the strength of geo-polymer material either by heat or ambient curing
(atmospheric curing). Authors achieved early strength of GPC while exposed to elevated temperature curing ranging
from 450C to 1150C .Authors also reported good strength of GPC in ambient curing, daylight curing. The most
important factors in heat curing are temperature and time which determines the behavior of the geo-polymer materials
made from Si and Al rich source materials. Palomo et al [10] reported that maximum comp. str. is achieved in high
curing temperature. Hake et al studied the Geo-polymerization process by storing the mixture recently for inclusion in
the ambient for 60 minutes, followed by curing at 650C for 90 minutes, and then drying at 65 0C .
• Fly ash and Alkaline-activator ratio : The increase in comp. str. depends on the higher fly ashcontent with a higher
alkaline solution content than the lower one. The better results are obtained by adopting a ratio of 3.3 to 4.0.
• Water to geo-polymer solids ratio : Total water content includes water used in the sodium-silicate solution, sodium-
hydroxide solution and the extra-water. A ratio of 0.17 to 0.18 gives a good result, above 0.18 the results decline
continuously.
• Sodium-hydroxide to Sodium-silicate (NH to NS) ratio : Authors reported that NH to NS ratio varied between 0.4
to 3. Higher the ratio lesser is the strength.
2.4 Materials Used:
• Pozzolanic materials: Authors used various materials as source materials which are rich in Si–Al minerals for
suitable making of geo-polymerisation. Metakaolin and fly ash are the most suitable unprocessed materials in the midst
of aluminosilicate materials for geo-polymer production. Early studies of Geo-polymerization was formed by using
metakaolin as a base materials rich in Al-Si content. After that most of the fresh papers reported for the production Geo-
polymer utilizes fly ash as the binder. Some of the authors are taking high calcium fly ash. silica fume, nano silica, rice
husk ash (RHA) , red mud slag , bagasse ash , saw dust ash , black coal ash, palm fuel ash, clay, and natural pozzolana as
the base materials for the geo-polymer concrete.
• Alkali activator: According to Davidovits [1] the polymerization process necessitates strongly alkaline-solutions to
react the silica & alumina ions present in the FA, amount of dis-solution Si to Al ratio in FA are major factors in geo-
polymerisation. This alkaline solutions are used by various authors of sodium and potassium hydroxide and silicates in
different molarities and ratio.
• Aggregates: Various authors reported various types of coarse aggregates for the experimentation regarding
obtaining Geo-polymer pervious concrete. According to Vanchai Sata, et al [3] the experiment regarding the use of
recycled coarse aggregates author used three types of aggregates to obtain different data s, they were Natural aggregates
(NA), Recycled aggregates from crushed structural concrete (RC) and Broken clay bricks (RB). The specimens were
named according and the data obtained from these specimen are recorded in table1 and 2 accordingly.
199
The Geo-polymer pervious concrete uses the Geo-polymer binder material as the replacement of
conventional cement. Geo-polymeric binder can be formed by chemical reaction between alkali activators and
alumino-silicate oxide. This is a rapid chemical reaction. In which the key components are Sodium hydroxide,
sodium silicate as alkali activator, fly-ash alumino-silicate oxide. The coarse aggregate is taken from different
demolition sites.
According to Vanchai Sata, et al [3] fly-ash had a particle size of 50µm, blain fineness of 2250cm3/gm as 45%
retained in 45µm sieve. Percentage of SiO2, Al2O3 and Fe2O3 in fly-ash is 71.7% where as the CaO content was
19.4%.Three concentration of sodium hydroxide solution were utilized at 10,15,20M. Commercial grade sodium
silicate had been utilized with 15.32, 32.87, 51.81 percentages of Na2O, SiO2 and dihydrogen monoxide
respectively. NaOH solution was yare by deliquesce commixing NaOH pallets in distilled dihydrogen monoxide.
Coarse aggregate which were utilized had diameter of 4.5-9.5mm. The aggregates were obtained from prevalent
demolition sites. Natural aggregate (NA), Recycled aggregates from crushed structural concrete (RC) and Broken
clay bricks (RB) were utilized in the experiment for engenderment of GPC.
Again according to the experiment done by Tawatchai Tho-in et al [2], Fly-ash used in experiment had
36.8,15.2,19.7,19.4 percentages of SiO2, Al2O3, Fe2O3 and CaO respectively having particle size distribution of 50µm
with blain fineness of 2250 cm3/gm. In this experiment also the commercial grade sodium silicate had been used
having 15.32, 32.87, 51.81 percentages of Na2O, SiO2 and water respectively. Specific gravity of Fly-ash was 2.23.
Aggregates used had diameter of 12.5-20 mm and fineness modulus was7.1. And specific gravity was 2.74 in
saturated surface dry condition [17,18 ] .
Figure-1 (comparison of coarse aggregates obtained from demolition sites)
200
3. Mix Proportion, mixing, curing &Testing
Following Table-1 shows us the base mix designs used in various experiments done by various authors .
MIX MIX PROPERTION(Kg/m3) COURSE

AGGREGATE(Kg/m3)
Author
FA(Kg) NA2SiO3 NaOH(Kg) NA RC RB
(Kg)
10M 15M 20M
35PGC10 221 25.5 51 _ _ 1768
35PGC15 221 25.5 _ 51 _ 1768
35PGC20 221 25.5 _ _ 51 1768
40PGC10 221 29.5 59 _ _ 1768
Tawatchai Tho-in et 40PGC15 221 29.5 _ 59 _ 1768

al [2]
40PGC20 221 29.5 _ _ 59 1768
45PGC10 221 33.0 66 _ _ 1768
45PGC15 221 33.0 _ 66 _ 1768
45PGC20 221 33.0 _ _ 66 1768
NA10 221 33 66 _ _ 1768 _ _
NA15 221 33 _ 66 _ 1768 _ _
NA20 221 33 _ _ 66 1768 _ _
RC10 221 33 66 _ _ _ 1768 _
Vanchai Sata, et al RC15 221 33 _ 66 _ _ 1768 _

[3]
RC20 221 33 _ _ 66 _ 1768 _
RB10 221 33 66 _ _ _ _ 1768
RB15 221 33 _ 66 _ _ _ 1768
RB20 221 33 _ _ 66 _ _ 1768
10M0C90T 320 36 60 _ _ 1175
10M5C90T 304 36 60 _ _ 1175
10M10C90T 288 36 60 _ _ 1175
10M15C90T 272 36 60 _ _ 1175
15M0C90T 327 37 _ 61 _ 1175
Yuwadee Zaetang et 15M5C90T 311 37 _ 61 _ 1175

al[4]
15M10C90T 294 37 _ 61 _ 1175
15M15C90T 278 37 _ 61 _ 1175
10M10C60T 288 36 60 _ _ 1175
10M10C90T 288 36 60 _ _ 1175
10M10C120T 288 36 60 _ _ 1175
201
Here the mixes shown are the experimental details of various authors. The name of each specimen defines
the quantity and proportions of materials used. 35PGC10 shows that 35 is the ratio of Alkaline liquid/Fly-ash (l/a),
similarly 10 signifies the molarity of NH concentration. Again NA, RB and RC are the course aggregates and 10, 15,
20 M are the concentration of sodium hydroxide. The specimen which are named as 10M10C20T here 10M is the
concentration of NH; 10C signifies the supersession percentage of OPC with Fly-ash. And at last 20T signifies the
curing temperature [2].
According to Vanchai Sata, et al [3] the specimens which were produced have a fly-ash/coarse aggregate
ratio of 1-8 by weight, Na2SiO3/NaOH ratio of 0.50 and alkaline liquid (lA/FA) ratio of 0.45.In a typical mix given in
the table-1 FA is used 221Kg /m3, Na2SiO3 quantity used was 33Kg/m3 and the volume of coarse aggregate used was
1768 Kg/m3. Three types of coarse aggregates were used in the prospective of this report which are NA, RC and RB
[3].
Again with respect to the study done by Tawatchai Tho-in et al [2] similar proportions had been used.
FA/CA ratio was 1.0-8.0 by weight, NH/NS ratio was 0.50 which is constant. Now Alkaline liquid to fly-ash ratio
was (L/A) varied from 0.35, 0.40 & 0.45, the curing temperature taken as constant 25 0c (room temperature).
Salaheddin Abdulsalam Arafa et al [5] experimented with variance in FA/CA ratio, size of aggregates and
alkaline liquid/FA ratio. The base ratios are FA/CA was 1:6, 1:7, 1:7.5, 1:8 and 1:9, AL/FA (AL=Alkaline Liquid)
0.4,0.5 and 0.6. The coarse aggregate sizes were 5-10, 10-14 and 14 -20 mm. The curing temperature was taken as
800 c and kept in that temperature for 24 hours.
Yuwadee Zaetang et al [4] designed the ratios of experiment such as designed void ratio was 35%, paste
volume was 22%. The ratio between alkali liquid to binder was 0.3, NH solution used of two types of strength 10
&15M. Again replacement of conventional OPC was done in 0, 5, 10, 15 % by weight. The curing temperature also
varied in this experiment they were maintained at 60, 90 and 120 0c.
3.1 Mixing and Curing
There are variations in experimenting on Pervious Geo-polymer concrete. Authors reported many types of
experimentation to obtain a acceptable PGC.
According to Vanchai Sata, et al & Tawatchai Tho-in et al [2,3] the PGC commixing was done in an air
conditioned room consisting temperature of 250c. The Fly-ash and NAOH mixed in pan mixture for 5 minutes then
Coarse aggregate was added and mixed for 4 minutes again NS is added and mixed for 10 minutes to obtain designed
mix which was proposed. Casting was done in 100×200 mm cylindrical moulds and compacted by vibration table.
After finishing the casting the specimen are wrapped with vinyl sheets to protect from the moisture loss and kept foe
1 hour in 250c.Then after specimen were placed in oven for oven curing at 60 0c for 48 hours. After the curing process
they were placed 1 day for cooling then de-molded.
202
With respect to the experimentation of Salaheddin Abdulsalam Arafa et al [5] curing was done in 250c maintained
temperature room. FA mixed with Alkaline Liquid for 5 minutes in a pan type mixture, after that mixing of coarse
aggregate occurred for 1 minute. Then casted in a 100×100×100 mm cubes and wrapped in plastic sheets to contain
moisture loss, then cured for 24 hours at 800c by oven curing.
Again the experimentation of Yuwadee Zaetang et al [4] ; mixing of OPC and FA was done such a way that
would maintain homogeneity. Mixing was done in a pan mixture NH was mixed with OPC for 5 minutes then NS
was mixed for 2 minute more. Then the bottom ash was added at its saturated surface dry condition for 1 minute.
Molded using standard tamping rod and vibrating table then casted and wrapped in plastic sheet to contain the
moisture loss. Curing was done at 60, 90, 120 0c for 48 hour by oven curing. 100×200 mm cylindrical specimen were
tested for density, water permeability, compressive strength & splitting tensile strength; whereas 150×150×60 mm
prisms were tested for thermal conductivity & surface absorption resistance test.
3.2 Testing
The basic testing of GPC consists of void content test, water permeability test, ultrasonic pulse velocity test
and thermal conductivity test.
3.2.1 Void content and water permeability test
According to Tawatchai Tho-in et al [2] the 100×200 mm cylindrical specimen were tested foe void content and
water permeability. Void content was tested in accordance to ASTM C1688 [22] and calculated by the equation 1 and
equation-2.
(𝑇−𝐷)×100
𝑉𝑇 = …………………………….(Eq-1)
𝑇
𝑀𝑠
𝑇= ……………………………………… (Eq-2)
𝑉𝑠
In the above equation Vt is the void content (%), T is the theoretical density of PGC on free air basis
(Kg/m ), D is the density of PGC (Kg/m3), Ms is the total mass of every element batched and Vs is the sum of
3
absolute volume of component ingredient in batch [2,3] .
After the void content test was performed all specimen were placed in a PVC pipe tightly and clamps were
provided so that the specimen stay steady. To test water permeability the water head was kept constant, steady state
flow achieved then the water permeability co-efficient was found by using equation 3 in accordance to ASTM C 1754
[2,3,4,22].
203
Figure-3 (Water permeability test set-up)
𝑄𝐿
𝐾 = 𝐻𝐴 ……………………………………………….(Eq-3)
𝑡
In the above equation K signifies the water permeability co-efficient (cm/s), Q is the quantity of water in
3
cm , t is the time in seconds, L is the length of the specimen in cm and a signifies the cross-sectional area of the
specimen in cm.
Vanchai Sata, et al [3] reported another formula for calculation of void ratio which is shown in equation 4.
𝑊2 −𝑊1
𝑉 = (1 − )× 100…………………………… .(Eq-4)
𝜌𝑤 𝑉𝑜𝑙
In the above equation 4 ‘V’ is the void ratio percentage, W2 is the saturated weight of specimen (g),
W1 is the weight of specimen under water (g), 𝜌𝑤 is the density of water (g/cm3) and finally the Vol signifies the
Volume of specimen (cm3).
3.2.2 Ultrasonic pulse velocity test and thermal conductivity test
Yuwadee Zaetang et al [4] ; suggested that an ultrasonic pulse velocity test is needed to obtain the pulse
velocity of PGC and the thermal conductivity of PGC in accordance to ASTM C597 [21]. Ultrasonic wave time
elapsed to pass through a sample quantified by using transducer having 50 mm diameter and frequency of 50 KHz.
So the author reported a formula for calculating pulse velocity of PGC. The formula is given below in equation 5.
𝐿
𝑉 = 𝑇……………………………………………………(Eq-5)
In the above equation ‘V’ is the pulse velocity denoted by (Km/s), again L is the length of sample in mm and time is
denoted as ‘T’ in seconds.
204
Thermal conductivity co-efficient test was done to obtain the thermal conductivity of the GPC specimens. It was
directly measured with measuring instrument called ISOMET 2114 having a probe, whose range is 0.04-6.0 W/K.
before the testing samples were dried in oven for 24 hours to reduce the effects of humidity interference [4].
3.2.3 Compressive strength and splitting tensile strength test
Compressive strength and splitting tensile strength tests should be done after & day period
of oven curing. Compressive strength and splitting tensile strength test can be done in accordance to ASTM C 392
and ASTM C 496 respectively. During the compression test both ends of the specimen were capped with sulfur
capping compound [2, 3, 19, 20] .
4. Properties
4.1 Density, compressive strength and splitting tensile strength
Properties found in the reports on given topic is tabulated and shown in Table-2 [2,3,4] .
AUTHOR SPECIMEN DENSITY(kg/m3) COMPRESSIVE SPLIT-TENSILE VOID CONTENT Water permeability

STRENGTH(Mpa) STRENGTH(Mpa) (%) co-efficient(cm/s)
Tawatchai 35PGC10 1780 8.5 1.0 30.5 2.22

Tho-in et al
[2] 35PGC15 1770 9.3 1.3 30.9 3.61
35PGC20 1680 5.4 0.7 34.4 5.96
40PGC10 1770 10.1 1.1 30.7 2.66
40PGC15 1820 9.9 1.3 28.8 2.91
40PGC20 1740 8.4 1.4 31.8 3.96
45PGC10 1810 10.2 1.1 28.7 3.64
45PGC15 1810 11.4 1.2 28.7 1.92
45PGC20 1780 9.6 1.2 29.9 2.39
Vanchai NA10 1840 13.6 1.6 24.2 1.25
Sata, et al
[3] NA15 1760 13.6 1.8 25.3 1.18
NA20 1810 11.9 1.5 27.4 1.71
RC10 1730 7.0 1.3 26.8 1.56
RC15 1710 10.0 1.4 26.9 1.46
RC20 1720 10.3 1.5 26.4 1.47
RB10 1420 2.9 0.4 23.7 1.12
RB15 1510 4.0 0.7 21.7 0.71
RB20 1520 6.6 0.9 22.4 0.80
Yuwadee 10MOC90T 1496 5.9 0.74 31.6 1.1
Zaetang et
10M5C90T 1468 6.6 0.92 32.5 0.9
al
[4] 10M10C90T 1488 7.2 0.94 33.0 0.9
10M15C90T 1466 7.7 1.01 33.4 1.3
15M0C90T 1487 7.6 0.88 33.7 0.6
15M5C90T 1497 7.9 1.05 33.5 0.8
15M10C90T 1487 7.9 1.08 33.5 0.7
15M15C90T 1502 8.6 1.28 33.5 0.8
10M10C60T 1495 5.7 0.69 32.1 0.8
10M10C90T 1488 7.2 0.94 33.0 0.9
10M10C120T 1492 7.5 1.08 33.5 1.1
10
205
According to Vanchai Sata, et al [3], density of the given samples ranges from 1420 Kg/m 3-1840 Kg/m3, this
proves that the density of PGC is lower than the conventional concrete (nearly 2400 Kg/m 3). Authors also reported
that the concentration of NaOH does not affect the density of PGC, due to the higher void content. In the above table
2 the specimen containing RB has low average density of 1480 Kg/m3; whereas NA &RC containing specimen had
higher density 1800 to 1720 Kg/m3 respectively. Again this can be said that the increase in alkaline concentration
enhances strength but excess hydroxide ion concentration lets the alumino-silicate gel to precipitate at early stage
which causes the low strength in PGC.
4.2. Curing temperature
Yuwadee Zaetang et al [4] ; showed that curing temperature has a significant effect upon strength
properties. The samples provided by Yuwadee Zaetang et al; had the temperature variations of 60,90 and 120 0c.
From the table 2b it can be deduced that increasing in curing temperature drastically increase the strength. The
sample 10M10C90T had the increased strength of 7.2 Mpa from the sample 10M10C60T, which is 126% more
strength than 10M10C60T.
5. Benefits, Applications & Limitations
Pervious concrete was firstly used in 1800 in Europe, because of the shortage of conventional cement. Again
in 1920s after World War 2 there was a need for pervious concrete as the cement was not available at that time. At
that time of need this type of concrete was being used as pavement surfacing and load bearing structure. In India this
was a topic of interest in late 2000. But Geo-polymer is a new concept as it was first invented by Joseph Davidovits
in 1978. Geo-polymeric bonds are basically a product of expeditious reaction between chemicals which gives a very
good strength. That was the reason behind the invention of PGC so the concrete can get acceptable strength and the
replacement of OPC can be done, the opportunities and benefits of PGC firstly considered by U.S Environmental
Protection Agency (EPA) [11, 12, 13] During the storm season the run-off water was producing big problems for the
real-estate developers, because of the mud produced by the water they need to apply costly drainage methods. So the
real-estate became expensive in urban areas, but by providing PGC the water can easily infiltrate the soil and so the
ground water can be recharged locally. Renewed interest on PGC is because of the Federal Clean Water Legislation.
The following points are the benefits of PGC [13, 14].
• Allows the rain/storm water to pass through and to be captured and relocate in to the soil resulting less run-
off and ground water recharge.
• Supporting the sustainable structure.
• Providing solution to the environmental problems.
• Eliminates the need for retention ponds for rain water collection
• Allows the trees to grow properly as this type of concrete can let water and air pass through it.
11
206
The true applications of PGC are that it can be used as a good insulating material (Green house) as the void content is
more, this type of concrete can be used as pavements so that it can cut-off the run-off water and the impact will be
less muddy area near construction development sites. PGC can serve as the drainage media of hydraulic structure, can
be used as walls for a two story building.
5.1. Limitations
While PGC is excellent to use in some circumstances but it cannot be applicable to certain scenarios. PGC has a
very rough texture and honeycomb structure so it cannot be used at heavy traffic roads. PGC cannot be used in road
construction due to its structure during the time of heavy traffic the aggregates on the surface will loosen and so the
integrity of the whole structure will be at question [16].
Special care is needed before construction of PGC structure as it needs special design considerations,
without proper supervision and design consideration the water will not pass through it and will be captured inside the
concrete and this will create structural problems.
6. Conclusion
The result of this above study conclude that Geo-polymer binding pervious concrete has a versatile use and
can be used as a environment friendly building material also it is the probable solution for the pollution cleaning, also
due to the minimal density from conventional concrete it is light weight and in the process of making Geo-polymer
pervious concrete world can save a lot of natural resources and by the use of the recycled aggregates more demolition
by-products can be managed properly.
7. References
1. J. Davidovits; (1991), inorganic polymeric new materials; J Therm Anal 1991;37:1633–56.

2. Tawatchai Tho-in , Vanchai Sata , Prinya Chindaprasirt , Chai Jaturapitakkul; 2012, Pervious high-calcium fly ash
Geo-polymer concrete, Construction and Building Materials ,Volume 30, Pages 366-371.
3. Vanchai Sata , Ampol Wongsa, Prinya Chindaprasirt; (2013), Properties of pervious Geo-polymer concrete using
recycled aggregates; Construction and Building Materials 42 33–39.
4. Yuwadee Zaetang , Ampol Wongsa , Vanchai Sata , Prinya Chindaprasirt ; (2015), Use of coal ash as Geo-polymer
binder and coarse aggregate in pervious concrete; Construction and Building Materials 96 289–295,.
5. Salaheddin Abdulsalam Arafa, Ahmad Zurisman Mohd Ali, Siti Nazahiyah Rahmat, and Yee Loon Lee; (2017),
Optimum Mix for Pervious Geo-polymer Concrete (GEOCRETE) Based on Water Permeability and Compressive
Strength; MATEC Web of Conferences 103, 01024.
6. Nishant Mishra, Shyam Verma ,V. Vasugi; (2015), Experimental Study on Geo-polymer Pervious Concrete; IJSRD -
International Journal for Scientific Research & Developmen | ISSN (online): 2321-0613.
7. Valerie López-Carrasquillo, Sangchul Hwang; (2017), Comparative assessment of pervious concrete mixtures
containing fly ash and nanomaterials for compressive strength, physical durability,permeability, water quality
performance and production cost; ELSEVIER; Construction and Building Materials 139 148–158,
8. Duxson P, Provis JL, Lukey GC, van Deventer JSJ; (2007), The role of inorganic polymer technology in the
development of ‘green concrete’; Cem Concr Res;37:1590–7.
9. Z.T. Yao, X.S. Ji , P.K. Sarker , J.H. Tang , L.Q. Ge , M.S. Xia , Y.Q. Xi ; (2015), A comprehensive review on the
applications of coal fly ash; ELSEVIRE; Earth-Science Reviews 141 ;105–121.
10. VLADIMÍR ŽIVICA, MARTIN T. PALOU, MARTIN KRIŽMA; (2014), GEO-POLYMER CEMENTS AND
THEIR PROPERTIES: A REVIEW; BUILDING RESEARCH JOURNAL; VOLUME 61;NUMBER 2, pp. 85 – 100.
12
207
11. Pervious Pavement Manual, Florida Concrete and Products Association Inc., Orlando, FL. http://www.fcpa.org.
12. Richard C. Meininger, 1988 “No-Fines Pervious Concrete for Paving,” Concrete International, Vol. 10, No. 8,
August, pp. 20-27;.
13. Storm Water Technology Fact Sheet Porous Pavement, United States Environmental Protection Agency, EPA 832- F-
99-023, 1999.
14. Dan Brown, (2003), Pervious Concrete Pavement – A Win-Win System, Concrete Technology Today, CT 032.
15. J.G.S. van Jaarsveld, J.S.J. van Deventer, G.C. Lukey, (2002), The effect of composition and temperature on the
properties of fly ash and kaolinite-based Geo-polymers; Chem. Eng. J. 89 63–73.
16. David Thorpe, and Yan Zhuge; 2010; ADVANTAGES AND DISADVANTAGES IN USING PERMEABLE
CONCRETE PAVEMENT AS A PAVEMENT CONSTRUCTION MATERIAL; Annual ARCOM Conference;
1341-1350.
17. Kim HK, Lee HK; 2010; Acoustic absorption modeling of porous concrete considering the gradation and shape of
aggregate and void ratio; J Sound Vib ;329(7):866–79.
18. Tu T-Y, Chen Y-Y, Hwang C-L. ;2006;Properties of HPC with recycled aggregates; Cem Concr Res;36(5):943–50.
19. American Society for Testing and Materials, Standard Test Method for Splitting Tensile Strength of Cylindrical
Concrete Specimens, ASTM C496/C496M: Annual Book of ASTM Standard, 2011.
20. American Society for Testing and Materials, Standard Test Method for Abrasion Resistance of Concrete or Mortar
Surfaces by the Rotating-Cutter Method, ASTM C944/C944M: Annual Book of ASTM Standard, 2012.
21. American Society for Testing and Materials, Standard Test Method for Pulse Velocity Through Concrete, ASTM
C597: Annual Book of ASTM Standard, 2002.
22. American Society for Testing and Materials, Standard Test Method for Density and Void Content of Hardened
Pervious Concrete, ASTM C1754: Annual Book of ASTM Standard, 2012.
13
208
crajemahadik@gmail.com Paper ID- 51
Investigating load withstand by L-Shape concrete cube, RCC slab and to

safeguard reinforcement of RCC slab in salt-water environment using cathodic
protection.
C. F. Rajemahadik*,1, M. M. Kulkarni2, R. S. Durge1, A. R. Kamble3, S. B. Babar1, P. A. Bansode1
1
Department of Civil Engineering, Sanjay Ghodawat Polytechnic, Kolhapur.
2
School of Technology, Sanjay Ghodawat University, Kolhapur.
3
School of Engineering, MIT-ADT University, Pune.
*Corresponding author: C. F. Rajemahadik (crajemahadik@gmail.com)
Abstract: Structures constructed in reinforced cement concrete (R.C.C.) are currently at risk when exposed to salt-
water condition. Eventually, R.C.C. structure fails after corrosion of reinforcement, reducing strength and its life.
Investigation focuses on load carrying capacity of slab specimen (R.C.C.) and shear strength of L-shape concrete cubes
after exposed to salt-water. The concrete of M20 mix was prepared using 53 grade cement to cast L-shape concrete
cubes and reinforced slab specimen. A reinforced bars of 10 mm diameter were used in slab specimen. The size of
R.C.C. slab was 300x300x100 mm and L-shape cubes were prepared after cut of 90x150x90 mm in regular concrete
cube of M-20 mix. After casting and curing of L-Shape cube and slabs, these specimens are exposed to 0.1M NaCl
solution. The L-Shape cubes after regular curing gave 0.0074 kN/mm2 at 7 days and 0.123 kN/mm2 at 28 days of shear
strength. Furthermore, shear strength increased to 0.0078 kN/mm2 at 7 days and 0.133 kN/mm2 at 28 days after curing
in NaCl solution for 30 hours. A D.C. power supply was used to accelerate corrosion of steel in slab at 0.02 mA/mm 2.
Slab specimens tested for 30 hours were as follows, slab reinforcement as anode (P1), slab coated with sodium silicate
and reinforcement as anode (P2) and slabs reinforcement as cathode (P3) in 0.1M NaCl solution with iron
anode/cathode electrode. The failure loads for slabs were 41.75 kN for P1, 75.35 kN for P2 and 78.5 kN for P3 type.
This study indicates, cathodic protection to reinforcement in slab increases load carrying capacity, ultimately increasing
life of specimen.
Keywords: salt-water, L-shape cubes, loading capacity, cathodic protection.
1 Introduction:
Recently, corrosion of reinforcement in R.C.C. structure and shear strength has boosted many researchers [1]. In state of
corrosion of reinforcement, steel turns to passive state in alkaline environment. Chloride breaks passive state, after its
attack on steel, boosting corrosion [2]. Chlorides accelerate, rate of corrosion with increase in concentration, as shown
in reactions (1) given below.
Fe+3 + Fe+2 + 6Cl- → FeCl3 + FeCl2 (1)
After reaction (1), chlorides separates and stays suspended, free to react again with reinforcement (2) [3].
FeCl3 + FeCl2 → 6Cl- + Fe(OH)2 + Fe(OH)3 (2)
Thus, corrosion caused due to chlorides can affect R.C.C. structures adversely hampering stability and reducing its life
span [3]. This leads to structural deterioration pounding safety and economic concern for owners and its stakeholders
[4]. The present scenario concentrates more on considering salt-water environment for stability of RCC structures.
209
crajemahadik@gmail.com
Another vital parameter for concrete is shear, a major parameter to overcome strength or failure capacity [5]. Among all
parameters shear in concrete is analysis for past 100 years [6]. Protection of reinforcement structural members and
observing shear failure of concrete is a task to look forward too. In addition, more focus is required to concentrate on
shear strength of RCC structure. The observations of such parameters mentioned above are a vital task, as these may
depend on site and environmental conditions [7]. Furthermore, reinforcement of corrosion needs protection or proper
maintenance, if RCC structures constructed in vicinity of salt-water environment. The corrosion of reinforcement in
structure can be controlled using electrochemical method known as cathodic protection (CP). This technique has proved
best for salt-water conditions on site, also known as impressed current method [8]. In this method, potential is provided
to protect reinforcement (metal) in structure to resist corrosion [9]. Moreover, cathodic polarized current is bifurcated as
current cathodic protection (ICCP), with external supply of power defined as sacrificial anode cathodic protection
(SACP), which has separate anode (sacrificial) [10]. In addition, basic method to control corrosion using cathodic
protection is divided as impressed current systems (ICS) and sacrificial anodes [11]. This paper focuses on shear
strength of concrete using L-Shaped cubes and applying cathodic protection to reinforcement in concrete slab specimens
using sacrificial anode to learn the effect on load withstanding capacity in salt-water environment.
2 Materials and Methods:
2.1 Materials
Cement: The ordinary portland cement (OPC) of Ultratech company of 53 Grade was purchased. Cement was tested for
fineness, normal consistency and setting time. Table 1, gives physical properties of cement.
Table 1. Physical properties of cement
Sr.No. Physical test Results obtained IS 12269 : 2013 [12]
Fineness 10 max
1 7.9 %
(retained on 90 μm sieve) (%)
2 Normal consistency (%) 31 % --
Vicat time of setting (min)
3 Initial 29 30 min
Final 517 600 min
Fine aggregate (Sand): The natural sand was tested for fineness modulus, specific gravity and water absorption
satisfying IS-383-1970 [13] as shown in Table 2.
Table 2. Physical test on river Sand
Sr. Result
Type of test As per IS Code IS Code
No. obtained
1 Fineness Modulus 2.60 to 2.90 IS:383-1970 2.79
2 Specific Gravity 2.63 to 2.67 IS:383-1970 2.68
3 Water Absorption 2.5%
Water: Water used was potable from municipal supply satisfying IS-456:2000 [14] for mix of concrete.
210
Coarse aggregate: Natural crushed stone are purchased from locally available quarries of size 20 mm, which are used
as coarse aggregates (Table 3).
Table 3. Basic properties of course aggregate
Properties As per IS Code IS Code Result obtained
Specific Gravity 2.7 - 3.0 IS:2386-1963 [15] 2.94

Fineness Modulus 2.9 - 3.2 IS:383-1970 [13] 3.19
Water absorption Less than 10% IS:2386-1963 [15] 9.43%
Other materials:
Steel Rods – Mahalaxmi TMT 10 mm diameter of length- 25cm
Wooden Block – 90 mm x 150mm x 60mm
Plywood size – 900mm x 900 mm
Slab Cover – 25 mm
Electric Wires sustaining up to 3 amperes
Iron flats of size 150 mm x 150 mm
Chemicals: Chemicals required such as NaCl and sodium silicate for experimentation were procured from Thomas
Baker.
2.2 Concrete mix design
Nominal mix proportion of concrete was 1:1.5:3 with water cement ratio 0.40 based on above listed material properties
is shown in Table 4.
Table 4. Quantities of ingredients of Concrete Mix
Cement Fine Aggregate Coarse Aggregate Water
3.78 Kg 5.67 Kg 11.34 Kg 0.189 Kg
Steps for casting of specimens

 Measuring all ingredients in concrete for mix.
 Mixing of concrete ingredients.
 Proper compaction and finishing of concrete is processed.
 Curing of specimens for 7 and 28 days is carried out.
Casting of L Shape concrete cube
 Cubes were casted using M20 nominal mix in mould 150*150*150 (Fig.1A).
 A wooden block was place in cube mould (60*90*150 mm) and concrete was poured with tamping (Fig.1B )
 After 24 hours demoulding of L-shaped (Fig.1C) concrete bolck was done and immersed in curing tank.
211
Figure 1. Details of L-shaped concrete cube

Preparation of Slab specimen
 Preparing mould for slab specimen of size 300mm x 300mm x 100 mm was prepared in plywood as base and
side panels of fibreglass (Fig.2B).
 Reinforcement of 10 mm dia. for slab in grid was tied with help of steel wire of 1 mm diameter (Fig.2A).
 Electrical weirs were tied to each node of reinforcement in grid.
 Slab specimen was casted using M20 nominal mix in constructed mould (Fig.2B)
 After initial and final setting time slab specimen were demoulded allowed to surface dry and kept for curing in
tank (Fig.2C).
Figure 2. Details of slab specimen used in experiment

Curing: After casting and surface drying of L-shaped concrete cubes and reinforced slab specimens, specimens were
immersed in curing tank for 7 and 28 days, respectively. The Figure 3, shows, slab specimen with wires and L-shaped
cubes kept in tank for curing.
212
Figure 3. Curing of R.C.C. slab specimen and L-Shaped concrete cube

2.3 Experimental Setup
Slabs were casted using cement, fine aggregate, coarse aggregate and water. A mix design of M-20 grade with mix
proportion of 1:1.5:3 was considered. Manually mixed concrete was placed in respective moulds of slab. The
reinforcement was introduced during casting practice of slab with proper cover. Electrical wires connected with nodes
of reinforcements in slab specimen and were covered properly. Then, slab specimens were demoulded after 24 hours.
Further, slab specimens were kept in curing tanks for 7 and 28 days. After curing, slabs are dried and again immersed in
0.1M NaCl water tank having 6 liters volume, maintaining water level at half of slab thickness [16]. A cover of 1 cm
was provided at bottom of slabs and in between tank inner surface. Water filled in tank contained 0.1M of NaCl
mimicking salt water condition [16]/[17]. Wires from nodes were connected to D.C. power supply of range 0 to 3
Ampere and 0 to 30 Volts to anode and cathode of specimen (Fig.4 C), according to cases (P1, P2, & P3). A current
density of 0.02 mA/mm2 was maintained for 30 hours in salt-water condition for all experiments on slab specimens.
Three of cases are studied as 1) P1- reinforcement of slab as anode connected with external cathode, 2) P2-
reinforcement of sodium silicate coated slab as anode, connected with external cathode 3) P3- reinforcement of slab as
cathode, connected with external anode (cathodic protection). After, 30 hours of time slab specimens were allowed to
dry completely to check for load carrying capacity (Fig 4A). A steel frame supported slab specimen for plain and
suspended position. Load was applied on slab under UTM (Fig 4A).
Figure 4. Experimental set up for compression and cathodic protection.
213
Figure 5. Experimental set up of L-shaped concrete cubes for shear testing

The loading arrangement for direct shear for L-shaped cubes and failure load for slab specimens was demonstrated
using universal testing machine as shown in Fig.5, respectively. For shear test of L-shaped cubes, 85 x 10 x 150 mm
size plate is placed on left side portion of 90 mm face. Mild steel bar of 12 mm diameter is kept over centre of plate
with 22 mm diameter was placed at edge of plate. To complete this assembly, plate of size 110 x 50 x 150 mm is placed
over initial arrangements as shown in Fig.5. Load was applied on 110 x 50 mm plate, which forms shear plane below
centre of 22 mm diameter bar. Similarly, a steel support was provided to slab specimens with 25 mm cover at the edges.
Load was applied to check failure of slab specimens experimented in different conditions.
3 Results & Discussion
L-shaped cubes: Experiment analysis to check for effect of 0.1 M NaCl on direct shear strength on L-shape concrete
cubes as shown in Fig.6. Concrete mix of M-20 was prepared for cubes specimens. During test for shear on L-shapes
cube specimens, reveled cubes kept for curing in 0.1M NaCl water for 30 hours, resulted in 5.12% more shear strength
than regular water for 7 days. Similarly, 7.5% of extra shear strength compared to regular water for 28 days curing
period [18].
Figure 6. Shear load failure of cubes after 30 hours of curing in 0.1M NaCl solution
214
Slab specimens: Three type of slab specimens were tested against failure load in universal testing machine. Among
them, P1- slab specimen casted and kept in 0.1M NaCl with reinforcement of slab as anode at 0.02mA/mm2 in water for
30 hours after curing and drying cycle. Similarly, P2- slab specimen with 0.02mA/mm2 of current density to
reinforcement and coated with sodium silicate with 1:3 ratio to water. Solution coating was applied in three layers to
slab specimen. In addition, P3- slab specimen where reinforcement worked as cathode and anode as iron plate provided
at bottom of slab surface [19]. After experiment for 30 hours with current density of 0.02 mA/mm 2 applied to
specimens, P3 resisted higher load of 78.5 kN, followed by P2, sodium silicate coated resisted 75.35 kN and anode
reinforcement in salt water environment failed at 41.75 kN.
Figure 7. Failure load of slab immersed and tested in 0.1M NaCl for cathodic protection
4 Conclusion
Experimental analysis for concrete and RCC structures showed that shear in concrete increases but failure may be
possible for RCC structures like slab. Cathodic protection to reinforcement in RCC structure could gain good strength to
structure. The experiment showed rise in strength to 46.81% with cathodic protection to reinforcement, 44.60% of rise
with sodium silicate coating on slab, compared to slab reinforcement as anode at similar current density. In addition,
structures in salt-water condition could improve shear strength complimenting, combining technique of cathodic
protected reinforcement. Method, would may overpass economic conditions, this could be overcome by renewable
energy sources such as solar or wind power. Cathodic protection of reinforcement in RCC structures could lead to
extension of life of structure.
Reference
1. Woo-Yong J., Young-Soo Y., Young-Moo S.: Predicting the remaining service life of land concrete by steel
corrosion. Cement and Concrete Research. 33, 663–677 (2003).
2. Soÿlev T.A., Richardson M.G.: Corrosion inhibitors for steel in concrete: State-of-the-art report. Construction
and Building Materials. 22, 609–622 (2008).
3. Medeiros M.H.F., Helene P. :Surface treatment of reinforced concrete in marine environment: Influence on
chloride diffusion coefficient and capillary water absorption. Construction and Building Materials. 23, 1476–
1484 (2009).
215
4. Ghods P., Isgor O.B., Pour-Ghaz M.: A practical method for calculating the corrosion rate of uniformly
depassivated reinforcing bars in concrete. Materials and Corrosion. 58, 265-272 (2007).
5. Aman M., Amr E.: Importance of shear assessment of concrete structures detailed to different capacity design
requirements. Engineering Structures. 30, 1590–1604 (2008).
6. Evan C. B., Frank J. V., Michael P. C. : Simplified Modified Compression Field Theory for Calculating Shear
Strength of Reinforced Concrete Elements. ACI Structural Journal. 103, 614-624 (2006).
7. Suhag P., Vikas K., Patel H. K., Patel V. R., Pandya I. I.: Prediction of Ultimate Shear Strength of Moderate
Deep Concrete Beam Including Size Effect . International Research Journal of Engineering and Technology.
04, 813-818 (2017).
8. Xu J., Wu Y. : Current distribution in reinforced concrete cathodic protection system with conductive mortar
overlay anode. Construction and Building Materials. 23, 2220–2226 (2009).
9. Marcassoli P., Bonetti A., Lazzari L., Ormellese M.: Modeling of potential distribution of subsea pipeline
under cathodic protection by finite element method. Mater. Corros. 66, 619–626 (2015).
10. Montoya R., Nagel V., Galván J.C., Bastidas J.M.: Influence of irregularities in the electrolyte on the cathodic
protection of steel: a numerical and experimental study. Mater. Corros. 64, 1055–1065 (2013).
11. MATCOR and CP Masters: www.matcor.com/impressed-current-cathodic-protection; (assessed on
01/12/2018)).
12. IS 12269 (2013); Ordinary Portland Cement, 53 grade — specification, Bureau of Indian Standards, New
Delhi.
13. IS-383 (1970): Indian Standard for specification for coarse and fine aggregates from natural sources for
concrete, Bureau of India Standard, New Delhi.
14. IS 456 (2000): Plain and reinforced concrete - code of practice, Bureau of Indian Standards, New Delhi.
15. IS:2386 (1963): methods of test for aggregates for concrete (Part III) specific gravity, density, voids,
absorption and bulking, Bureau of Indian Standards, New Delhi.
16. Saraswathya V., Ha-Won S.: Improving the durability of concrete by using inhibitors. Building and
Environment. 42, 464–472 (2007).
17. Rodrigo de Almeida Souza L.R.A., Farias de Medeiros M.H., Pereira E., Brandão Capraro A.P.:
Electrochemical chloride extraction: Efficiency and impact on concrete containing 1% of NaCl. Construction
and Building Materials. 145, 435-444 (2017).
18. Nanak P. J., Pamnani A., Kachwala A.: Shear Strength of M30 Grade Self Compacting Concrete with
Different Water Based Curing Techniques. International Journal of advancement in engineering and
technology, management and applied science. 2, 6-14 (2015).
19. Fajardo G., Escadeillas G., Arliguie G.: Electrochemical chloride extraction (ECE) from steel-reinforced
concrete specimens contaminated by ‘‘artificial’’ sea-water. Corrosion Science. 48 110-125 (2006).
216
Paper ID- 52
Performance Evaluation of Two-Way RC Slab Subjected

to Blast Loading Using Finite Element Analysis
Kasturi Bhuyan 1 (0000-0003-1311-0332), Kiran Kumar Jujjavarapu 1, and Hrishikesh

Sharma 1
1 Indian Institute of Technology Guwahati, Guwahati, India
b.kasturi@iitg.ac.in
kiranjujjuvarapu@gmail.com
shrishi@iitg.ac.in
Abstract. Recently, blast and impact have gained more importance in research
work owing to the fact that these high intensity events can completely destroy
any structure, causing tremendous casualties and property loss. Therefore, for
any structure to withstand such extreme events, it is important to consider the
effect of blast and impact loading in their design. Research work on response
and damage analysis of reinforced concrete structural components such as slabs,
columns, beams and walls subjected to blast and impact are limited in the litera-
ture and are essential for assessing its vulnerability. The current work investi-
gates the performance of two way reinforced concrete (RC) slabs subjected to
blast loading. To simulate the field blast tests on the different slabs, initially a
numerical model is established in a finite element preprocessor (HyperMesh).
Later on, with help of a commercial simulation software package (LS-DYNA)
the numerical analysis is carried out. The developed numerical model is vali-
dated with the results of an experimental work done on RC slab subjected to air
blast from a previous research. The results of the developed model are found in
close agreement with actual experimental results. The numerical parameters of
the validated model are used to develop a probabilistic model to investigate the
blast response of the slab designed in current work. The probabilistic model de-
veloped in this research is able to capture the damage phenomena and also ca-
pable enough to reproduce the dynamic response.
Keywords: RC Slab, Blast loading, LS-DYNA, Numerical modelling, Maxi-

mum central deflection.
1 Introduction
Blast resistant design was earlier considered for only limited and significant structures
which were supposed to be subjected to blast loading events such as military con-
struction and nuclear power plant. With the increasing terrorist attacks causing loss of
life and economy, it has now become necessary to include blast resistant designs for
civilian structures as well. Blast triggers the deterioration of individual structural
components succeeded by the progressive collapse of the entire structure. The damag-
217
es to buildings include failure of external and internal structural elements, disintegra-

tion of walls and bursting of windows. The intensity of damage imparted to a building
due to blast load depends on weight of blast charge, stand-off distance, orientation
and configuration of building. Analyzing the structural response with all its complexi-
ties like material nonlinearities and time dependent deformation under blast loading is
of paramount importance.
Considering the ability of reinforced concrete to resist blast, it can be treated as a
key material in alleviation of blast effects. In general, concrete and steel behave dif-
ferently under suddenly applied load. Hence, a reasonable failure assessment of struc-
tural components is of utmost importance in anticipating blast loading. Since RC slab
is a major structural component, evaluation of blast resistance for the same is essen-
tial. The current work focuses on modeling the behaviour of RC slabs under blast
loading which also accounts for RC walls.
2 Literature Review
Many researchers conducted experimental studies on RC slabs subjected to blast load-

ing. With the help of experimental results, the explosive charges weight and standoff
distances required to impose a given damage level on RC slabs using a modified dis-
placement based methodology was predicted [1]. The results of an experimental in-
vestigation conducted on the blast-resistance of concrete panels made of ultra-high-
strength concrete (UHSC) in Woomera, South Australia in May 2004 was presented
and a finite-element computer code was also developed to analyze concrete structures
subjected to blast loading which was validated with the outcome from the experi-
mental test [2]. A series of tests investigating the blast resistances of RC slabs was
also conducted and an ANFO blast tests using ANFO charge was performed on rein-
forced ultra-high strength concrete panels. Rebar and short steel fibers provides suffi-
cient ductility, confer outstanding energy absorption and negated the brittle character-
istics of these materials. Ultra high performance slabs showed more resistance to blast
as compared to conventional concrete slab [3].
Incorporating random changes in the structural and blast loading properties, two
loosely coupled SDOF systems were used to model one-way RC slab subjected to
blast loading for analyzing the its flexural and direct shear response. An experimental
investigation indicated that reinforced concrete structures subjected to distribute load
of short duration may not behave plastically at mid-span and fail there, as large load-
ing with short duration is more likely to cause a slab to fail by its shear failure mode
while a relatively small amplitude load with longer duration will result in flexural
failure [4]. A three-dimensional numerical model of RC plate was simulated for con-
crete material spallation under various blast loading and structural conditions and was
seen that with proper configuration protective structure can sustain a specific level of
blast without global failure [5]. To simulate the dynamic response of a simply sup-
ported RC slab under blast loadings, one finite difference program was developed and
the deformation of slab was due to the duration time and peak value of blast load-
218
ings. SDOF analysis is straightforward and suitable for use in a design office but the
results can be substantially conservative [6].
Research conducted, the damage modes of RC slab under the air blast loadings by
using erosion technique in LS-DYNA was clearly showed [7]. Six one-way RC slabs
of two groups were tested under real blast loads to address the scaling of the dynamic
response [8]. The scaling used for the structure is geometrical, whereas that for the
explosive charge is based on Hopkinson’s law. The research concludes that the local
damage on the concrete slabs with larger-scale factors reduces when compared with
that of slabs with smaller-scale factors; and the normalized damage parameters of the
slabs slightly increase with decreasing scale-down factor.
Dynamic response and the damage mechanism of RC slab was analyzed under
close-in explosion using LS-DYNA considering the strain rate effects, dynamic in-
creasing factor and equation of state for concrete [9]. It was observed that the in-
crease of the amount of the explosive can change the failure mode of the slab. A
parametric study concluded that the damage degree increases when the explosive
weight is increased or the explosive position is moved from center to the bounda-
ry after simulating blast response and damage assessment of a two-way RC slab under
blast loading [10]. In an experimental study and numerical simulations on the dynam-
ic performance of UHPC slab in comparison to normal strength concrete slab under
contact charge explosion, significant reduction in concrete punching and spall damage
in UHPC was observed as compared with the NRC slab. The feasibility and accuracy
of the numerical model were validated by comparing numerical results with the ex-
perimental observations [11].
Limitations associated with the modeling of blast loading using single degree of
freedom analysis causes the response to deviate from the actual behavior of concrete
structures. Moreover, do not account for the inherent uncertainty associated with the
blast loading, geometric and material uncertainty of concrete structure, and the uncer-
tainty in the interaction of blast loading with the concrete structure.
In this study the probabilistic model for an RC slab subject to blast loading is de-
veloped. Since the testing of RC slabs with actual blast event is difficult the probabil-
istic models are developed with the help of numerical simulation results. The data
obtained through numerical analysis is used for the regression analysis and the models
are developed such that it fit best to the data. The developed model incorporates the
multi-model response of the structure, interaction among different components and
inherent uncertainty associated with modeling configuration and material properties.
3 Experimental design for FE Simulation
For developing the probabilistic capacity model for any structural components at ul-
timate limit state sufficient number of data regarding the ultimate deformation capaci-
ty is required from experiments. Due to the limitations of conducting actual blast
experiments, FE simulation results are used to develop the probabilistic capacity
models. The ranges of variables of these developed models are selected in such a way
that it should replicate the practical scenario. Blast loadings on these RC slabs are also
219
designed with varying blast charge weight and stand-off distance. A sufficient number
of data is required for minimizing the statistical uncertainty. Thus the developed
probabilistic model is valid only for the range of data used for developing the model.
The ranges of geometric and material properties in the probabilistic model of the RC
slab are established using the data from literatures and common design practices for
ensuring contribution from both existing and new data of RC slabs. If all the variables
in the experiments are randomly combined the resulting model developed can be un-
realistic. Thus in order to generate the realistic data the design variables are divided
into primary variables and secondary variables. The variables which capture the basic
design requirements are categorized into primary variables (shown in Table 1) and
the rest of the variables derived from primary variables are categorized as derived
variables or secondary variables as given in Table 2.
Table 1. Range of basic variables of RC slab
Variable Symbol Range

Slab Length (m) L 2−15
Slab thickness (m) t 0.2−0.5
Longitudinal reinforcement ratio (%) ρl 0.12−2.47

Transverse reinforcement ratio (%) ρs 0.12−0.94
Compressive strength of concrete (MPa) f ck 20−55
Yield strength of longitudinal and transverse fy 415−550
reinforcement (MPa)
Boundary condition of the slab BT Four side fixed condition
Blast Charge (kg-TNT) Q 150.00−900.00
Stand-off distance (m) R 1.00−12.00
Table 2. Expression and range of secondary variables of RC slabs
Variable Symbol Expression/Range

Diameter of longitudinal bar (mm) dl 6, 8, 10, 12, 16, 20, 22, 25, 30
Diameter of transverse bar (mm) ds 6, 8, 10, 12, 16, 20, 22, 25
Spacing of longitudinal bar Sl

 π d 2 x1000 
 l   ρ BD 100 
4 
l
Spacing of Transverse Bar Ss π 2 

 d s x1000   ρs BD 100 
4 
220
The entire experimental design is divided into the design of RC slabs and the de-
sign of blast pressure. The load variables are designed separately to create almost all
the combinations of blast loading scenarios during a detonation. The ranges of blast
charge (Q) and its stand-off distance (R) used in the current work are confined to a
level of protection required to prevent a building from collapsing or minimize injuries
and deaths. However, this research aims to contribute a low to medium level blast
protection and so, a blast threat level from a bomb or any explosive which can be
carried by luggage, automobile and van (FEMA 428) are accounted. Considering the
slab variables and load cases together in the experimental design, there could have
been cases when some slabs experience the blast events to impractical scenarios. This
is avoided by splitting the experimental design into basic and dependent variables and
then performing a realistic combination of the RC slab parameters as well as blast
loading parameters.
4 Numerical Model
The most acceptable modeling methods of RC structures are the smeared model, em-
bedded model, and discrete model. For performing the dynamic response of individu-
al structural components, the discrete model is the best method [12]. In a discrete
model, the concrete element and steel bar element are treated as different elements. A
one- dimensional slide contact is provided between the concrete elements and rein-
forcement elements to simulate the longitudinal shear behavior, and the differences in
two material mechanical properties can be clearly shown, when the blast loadings
effects the structure. The mechanical behaviors of the slab depend on the aspect ratio
of length to width. When the aspect ratio is greater than 2, the slab can be considered
as a one-direction bending type structure and when less than 2, the mechanical behav-
ior of the slab is two way bending [13]. In the present investigation, an ALE (Arbi-
trary Lagrangian Eulerian) formulation is used to simulate two-way RC slab fixed on
all sides subject to blast loading. A geometric model of the RC slab is made using
HyperMesh [14] and analysis is done in the commercial FE program LS-DYNA [15].
4.1 Material model

A rate dependent material model is used for all the materials due to the sensitivity of
material properties. Concrete is modeled as 3D-solid elements with the formulation of
continuous surface cap model (CSCM) available in the software LS-DYNA. This is a
cap model with a smooth intersection between the shear yield surface and hardening
cap; and this model contributes for erosion in concrete and also takes into account the
strain rate dependency of the concrete strength. The reinforcement bars are modeled
explicitly as one dimensional element with the formulation of an elasto-plastic mate-
rial as it accounts for strain rate dependency and also failure based on plastic strain.
The contact between the concrete and reinforcement is modeled using the Lagrangian
coupling method. This method provides the coupling mechanism for steel and con-
221
crete interaction by allowing the coupling between edges of each part and saves the
effort of matching the nodes of the reinforcement and the concrete which might be
very difficult in some cases. Mesh refinement is done and a convergence is achieved
at 25 mm mesh size and minimization of the hourglass energy is ensured.
The boundary conditions of the RC slab modeled are the four sides fixed boundary
conditions. In order to simulate the exact practical condition non-reflecting boundary
is provided for air. And boundary conditions are applied directly to the eight set of
nodes in the slab rather than assigning only to last set of nodes for realistic bearing
state.
4.2 Blast loading
Whenever the explosion happened inside or outside of the structure, the pressure ex-
erting on the slab is not uniformly distributed. It depends on the relative location be-
tween the charge and the slab, the direction of the shock wave motion, and many as-
pects. At the same time, the slab will be subjected to more than one impact by the
reflected pressure wave. From the software LS-DYNA the pressure of explosive,
which is exerted on the slab was determined and pressure is calculated by considering
ALE formulation, to overcome the current limitations of conventional approach. A
hemispherical surface blast at the mid height of the slab is simulated with a blast
charge (Q) at a stand-off distance (R) as shown in Fig. 1. The blast load is simulated
through an empirical pressure load history due to a conventional chemical explosion,
directly applied to the nodes of a Lagrangian structure.
Fig. 1. Schematic view of blast loading
To overcome the limitations of conventional approach in blast load simulations, in

this approach, an Arbitrary Lagrangian Eulerian (ALE) formulation is used. A single
element layer of ambient formulation makes up the exterior surface of the air domain
which faces the blast. Its function is to receive information from the blast loading as
provided by ConWep formulation (Liver-more Software Technology Corporation),
which is then converted to thermodynamic state data and subsequently applied as a
source to adjoining ALE air elements. The air mesh interacts with Lagrangian struc-
tural elements to apply the load on the elements. Meshed slab with air and ambient
layer is given in Fig. 2.
222
Fig. 2. FE model of RC slab along with air and ambient layer
5 Numerical simulation
The numerical model developed using LS-Prepost and then it is analyzed with the
finite element software LS−DYNA. The software is a general purpose finite element
software capable of simulating the complex realistic problems. The codes origin lies
in highly non-linear, transient dynamic finite element analysis using the explicit time
integration method. To analyze the capacity of the RC slab, the deformation is record-
ed at the ultimate limit state of two way RC slab. For the ultimate performance level,
the values are recorded for the RC slab reaches the prescribed damage. The values of
the RC slab which does not got damaged also recorded. The FE model takes into ac-
count the strain rate effect, the multi modal response and the interaction of the RC
slab with different pressure field during the blast loading. Fig. 3 shows the FE model
of a RC slab at different stages of evolution of blast loading.
Fig. 3. FE model of a RC slab (a) at pressure transfer and (b) at damage state
5.1 FE Validation
The FE model used in this research is validated with the experimental results of a
two-way reinforced concrete plates tested under fully-clamped two-way bending of
223
blast load under shock tube analysis by Jacques Eric in Ottawa-Carleton Institute for
Civil Engineering, University of Ottawa [16].
The clear dimensions of the validated slab are 2440 mm in both length (L) and
width (B) and thickness of slab (t) is 75 mm. The clear span kept in both direction is
2032 mm. So, the aspect ratio of length to width is 1.0, which means that the slab is a
two-way bending slab. The specimen is constructed with 6.3 mm diameter un-
deformed steel wire with a yield strength of 580 MPa at a yield strain of 0.002
mm/mm. The experimental slab was specified to have a 28-day compressive strength
of 40 MPa, using 10 mm crushed limestone aggregate and a slump of 100 mm. How-
ever, the concrete strength for the specimens at the time of testing was approximately
50-60 MPa.
Fig. 4. (a) Two way RC slab subjected to the blast load in the experiment [14] and (b) Plastic
strain developed in slab with FE analysis
The steel bars were layout in two-ways with two layers; the spacing between two
bars in both the directions is 240 mm. The mesh generation of the concrete and steel
bar is 25 mm and one-dimensional slide contact between the concrete and steel bar
nodes was established. During the experiment the RC slab was subjected to a pressure
of 355 MPa and to simulate the exact pressure corresponding blast charge and stand-
off distance is determined and applied in LS-DYNA. ALE formulation is used for
creating realistic environment and practical scenario. CSCM concrete material model
is used for the concrete and the steel bars are modeled with plastic kinematic material
model and beam element formulation. The experimental result is shown in Fig. 4(a)
and Fig. 4(b) exhibits the plastic strain developed in the RC slab with the help of LS-
DYNA.
224
Fig. 5. Comparison of (a) pressure and (b) displacement profile for two-way RC slab
In Fig. 5(a), pressure profile is compared for experimental, simulation as well as ana-
lytical. Hand calculated and simulated pressure is almost close to 0.08 MPa whereas
from experimental data it is found to be 0.065 MPa. The maximum displacement of
the RC slab in experiment was 21.2 mm at 14.7 ms with cracking of the slab at the
center. FE analysis shows a close similarity to the experimental results as from Fig.
5(b). The maximum deflection of numerical analysis of RC is 14.7 mm at 14.9 ms
with cracking at the center and some elements at the support. And from analytical
method the value of deformation is 10.3 mm considering a SDOF analysis. The result
shows that the RC slab model performed well and gave deflection predictions that
were close to the experimental values. Therefore, the modeling technique followed in
this research is able to capture the damage mechanism. The close agreement which is
obtained between the FE simulation and the experiment shows that FE model yields a
reliable prediction of dynamic response.
6 Probabilistic capacity model
Using the results in FE simulations, the probabilistic models are developed for estimat-
ing the probabilistic deformation capacity of a two way RC slab subjected to blast
loading under ultimate performance level. The estimated probabilistic capacity models
take into account the multi-modal response of the structure, the interaction among the
different components and the inherent uncertainty associated with the modeling, con-
figuration and material properties of the concrete structure.
The probabilistic models are constructed by adding correction terms to mechanical
models. Because the data used for the model assessment come from FE simulations,
it is assumed that there is no measurement error in the data. The following equation
shows the model form where a transformation of the quantity of interest is written as
the sum of (in order from left to right) a mechanical model, a correction term to ac-
count for the bias in the mechanical model, and model error [17]. Following the gen-
eral formulation for probabilistic models, the deformation capacity for ultimate limit
state, is formulated as
225
10
ln Pi  x,ΘPi  = ln pi  x  + γPi  x,θPi  + σPi ePi (1)
Where Pi is probabilistic capacity of RC slab for i deformation; pi is the mechani-

cal capacity of RC slab for i deformation; γ Pi  x,θ Pi  is the correction term for the bias
inherent in the mechanical model defined as
n
γ Pi  x,θ Pi  =  θ Pi, j h Pi, j  x  (2)
j=1
Where h Pi, j  x  , j = 1,…,n is explanatory function (or regressors) defined as func-

tions of x, θ Pi, j , j=1,…,n are the parameters associated with explanatory functions,
σ Pi ePi is model error, e Pi is Gaussian error and ΘPi = θPi is set of unknown model
parameters in which θ Pi = (θ Pi,1 , K,θ Pi,k ) .
6.1 Mechanical Model
For a RC slab under ultimate performance level, analytical model for deformation
developed by Newmark is used in this research [18].
Xm t 1 1 
= 1+ d   β (3)
XE β 2 3 
Where, β =  ru P  , is explained as the ratio of yield strength to average applied load,

and XE =  ru K E  is the equivalent maximum elasto-plastic deformation. For calcu-
lating the maximum deformation ( Xm ) from Newmark equation, β is considered as 1;
as capacity is independent of load applied. Therefore,
Xm t
= 1+ d (4)
XE 6
6.2 Model correction

The model correction terms are used to capture the physical phenomena that is
T
not accounted in the case of mechanical model. The first explanatory function h1
which accounts for contribution of span moment and support moment of structure
 M x M y  . The second explanatory function, h 2 accounts for the contribution of
positive blast duration and natural period of structure ( Td Tn ). The following equa-
tion is proposed to estimate the probabilistic deformation ( X p ) of a two way RC slab
subjected to blast loading, under ultimate performance level.
226
11
Xp = x m ×e(0.000215h1 +0.009096h2 +σx ex ) (5)
Mx T
(0.000215 +0.009096 d +σ x ex )
My Tn
Xp = x m ×e (6)
The standard deviation of the model is 8.13. Fig. 6 shows the comparison between
FE obtained results and predicted values of probabilistic deformation capacity model.
The median value is plotted for the probabilistic model. The proposed model is capa-
ble of predicting the deformation as most of the data set points lies within the one
standard error ( ±σ x ) bin as shown by dotted lines.
Fig. 6. FE Deformation- Probabilistic deformation capacity of RC slab
7 Conclusion
The developed probabilistic deformation capacity model is realistic, easy to use and
also comforts in avoiding tedious FE simulations or experiments.
In future, from this laid foundation, work can be focused on improvising the current
proposed model for higher accuracies by increasing the number of data sets and also
reducing the mesh size. In this current work, probabilistic capacity models are de-
veloped only for predicting the deformation, so developing models for the internal
strain energy can be focused in the future which will be helpful for energy based de-
sign. As these models are valid only for prescribed boundary conditions, this work
can further be expanded for other boundary conditions that are in practice. Develop-
ing this model for different performance levels such as crack limit state and post blast
scenario will be also be considered in future as the present model only accounts for
ultimate limit state performance level. Moreover, performance based capacity model
227
12
of RC members can be developed; which leads to further improvement of current

blast code provisions.
References
1. Silva, P. F., & Lu, B. (2007). Improving the blast resistance capacity of RC slabs with in-
novative composite materials. Composites Part B: Engineering, 38(5-6), 523-534.
2. Ngo, T., Mendis, P., Gupta, A., & Ramsay, J. (2007). Blast loading and blast effects on
structures–an overview. Electronic Journal of Structural Engineering, 7(S1), 76-91.
3. Wu, C., Oehlers, D. J., Rebentrost, M., Leach, J., & Whittaker, A. S. (2009). Blast testing
of ultra-high performance fibre and FRP-retrofitted concrete slabs. Engineering struc-
tures, 31(9), 2060-2069.
4. Low, H. Y., & Hao, H. (2002). Reliability analysis of direct shear and flexural failure
modes of RC slabs under explosive loading. Engineering Structures, 24(2), 189-198.
5. Xu, K., & Lu, Y. (2006). Numerical simulation study of spallation in reinforced concrete
plates subjected to blast loading. Computers & Structures, 84(5-6), 431-438.
6. Jones, J., Wu, C., Oehlers, D. J., Whittaker, A. S., Sun, W., Marks, S., & Coppola, R.
(2009). Finite difference analysis of simply supported RC slabs for blast load-
ings. Engineering Structures, 31(12), 2825-2832.
7. Tai, Y. S., Chu, T. L., Hu, H. T., & Wu, J. Y. (2011). Dynamic response of a reinforced
concrete slab subjected to air blast load. Theoretical and applied fracture mechan-
ics, 56(3), 140-147.
8. Wang, W., Zhang, D., Lu, F., Wang, S. C., & Tang, F. (2012). Experimental study on scal-
ing the explosion resistance of a one-way square reinforced concrete slab under a close-in
blast loading. International journal of impact engineering, 49, 158-164.
9. Zhao, C. F., & Chen, J. Y. (2013). Damage mechanism and mode of square reinforced
concrete slab subjected to blast loading. Theoretical and Applied Fracture Mechanics, 63,
54-62.
10. Jia, H., Yu, L., & Wu, G. (2014). Damage assessment of two-way bending RC slabs sub-
jected to blast loadings. The Scientific World Journal, 2014.
11. Li, J., Wu, C., & Hao, H. (2015). Investigation of ultra-high performance concrete slab and
normal strength concrete slab under contact explosion. Engineering Structures, 102, 395-
408.
12. Hentz, S., Daudeville, L., & Donzé, F. V. (2004). Identification and validation of a discrete
element model for concrete. Journal of engineering mechanics, 130(6), 709-719.
13. IS 456. (2000). Plain and reinforced concrete–code of practice. Indian Standard, ICS,
91.100. 30, New Delhi.
14. Altair Computing (2003). HyperMesh Ver. 6.0 Basic Tutorial, Altair Engineering Inc.,
1820 E. Big Beaver, Troy, MI 48083.
15. Livermore Software Technology Corporation (2003). LSDYNA Keyword User’s Manual,
Version 970, Livermore, CA.
16. Jacques, E. (2011). Blast retrofit of reinforced concrete walls and slabs. University of Ot-
tawa (Canada).
17. Gardoni, P., Der Kiureghian, A., & Mosalam, K. M. (2002). Probabilistic capacity models
and fragility estimates for reinforced concrete columns based on experimental observa-
tions. Journal of Engineering Mechanics, 128(10), 1024-1038.
18. Brooks, N. B. and Newmark, N. M. (1953), “The Response of Simple Structures to Dy-
namic Loads”, Technical Report, Office of Naval Research.
228
Paper ID- 55
Organic and Inorganic Elements Used for Co2 Absorption in Concrete

Aishna Singh 1 Amrita Ghoshal1 Aprajita Singh1 and Purnachandra Saha2* (ORCID ID: 0000-0002-2722-1669)
1
Student, School of Civil Engineering. KIIT Deemed to be University, Bhubaneswar, Odisha, India
(Email: aishnasingh04@gmail.com, amrita.diya.95@gmail.com, singhaprajita14@gmail.com)
2
Senior Associate Professor, School of Civil Engineering. KIIT Deemed to be University, Bhubaneswar, Odisha,
India
(Email: dr.purnasaha@gmail.com) *corresponding author
ABSTRACT
CO2 emission during the production of cement is one of the major environmental impact caused by concrete. For a
long time, scientists have been searching for ways to reduce the amount of CO 2 by finding ways of CO2 absorption.
The objective of this study is to review the different organic and inorganic elements used to reduce the CO 2 emission
in concrete. Some of the major elements used are Di-calcium Silicate γ phase: γ-2CaOSiO2, Zeolite, Carbonic
anhydrase microbes, coal ash, concrete slurry waste. The organic bacteria is mixed during the mixing process of
concrete and it creates an outer surface which absorbs CO2 and reduces efflorescence and additionally it also has
self-healing property. Zeolite, on the other hand, is easily available and absorbs a large amount of CO 2. Concrete
slurry waste lowers the drying shrinkage of concrete along with CO 2 absorption.
Keywords: CO2 emission, concrete, zeolite, concrete slurry waste, Di-calcium Silicate
1. INTRODUCTION
CO2 is one of the major contributors among all the greenhouse gases towards the cause of global warming. Every
year the emission of CO2 rises throughout the world. Most of the countries of the world are trying to improve their
economy and thus need more energy. For this reason, a huge exploitation of fossil fuel is taking place which emits
many greenhouse gases. The construction industry which involves the production of concrete from cement is one of
the major contributors of CO2 emission in the current scenario. Recent studies show that the global production of
concrete per year is about 2.0 tonne per capita and the main element of concrete, i.e, cement contributes to about 5-
8% of current CO2 emission. [19] As CO2 emission throughout the world is rising rapidly day by day and as it is a
global crisis, the construction industry being a major contributor needs to reduce the amount of CO 2 emitted. The
use of special additives during the production of concrete mix that may help in the cause can go a long way to help
the situation at hand.
There are different processes for the reduction of CO2 emission from cement and cement-based industry like mineral
carbonation, carbonation curing, absorption, sequestration, etc. Mineral carbonation can be defined as the reaction
between carbon dioxide and minerals to form harmless solid carbonates. It has the potential to store large quantities
of carbon dioxide. Carbon dioxide sequestration also known as CO2 capture and storage (CCS) uses various methods
and approaches that isolate, extract and store carbon dioxide emissions. Carbon dioxide absorption is the process of
capturing and trapping of carbon dioxide from the atmosphere. The above processes involve some inorganic
additives and some organic materials. Various elements like zeolite, γ-C2S, concrete slurry waste (CSW), fly ash
(FA), etc when used as an additive substance in the concrete can reduce the CO2 emission to a certain extent. On the
other hand, organic elements mainly include bacteria like Ureolytic, Bacillus sphaericus, Bacillus pasteurii etc.
Several researchers conducted their research on each of the above processes and published their results. There is a
need to accumulate and analyze those data in a common place to understand the topic effectively. Hence, the
objective of the study is to review the different organic and inorganic elements used to reduce the CO2 emission in
cement and concrete.
229
2. Different Methods of Co2 Emission Reduction in Cement/Concrete

The construction industry which involves the production of concrete from cement is one of the major contributors of
CO2 emission in the current scenario. The various processes through which CO2 emission can be reduced are namely
carbonation, sequestration and absorption to name a few.
2.1 Carbonation
2.1.1 Mineral Carbonation
Mineral carbonation is defined as the reaction between carbon dioxide and minerals to form harmless solid
carbonates. It has the potential to store large quantities of carbon dioxide. Its main advantage is that it is used as a
procedure for CO2 sequestration for the reduction of harmful emissions of carbon dioxide. Research has shown that
materials like cement slurry waste sometimes mixed with fine recycled coarse aggregates (FRCA) and fly ash can be
used to reduce the amount of carbon dioxide. The main steps of the carbonation process include introducing the
CSW into an airtight chamber and starting the inflow of CO2 and flue gas. The reduction of carbon dioxide and flue
gas present in the chamber indicated the occurrence of carbon dioxide absorption. This process can be performed at
atmospheric pressure as well as below vacuum. On the other hand fly ash captures carbon dioxide by two different
mechanisms, carbonation reaction and surface adsorption on the solid sorbent. The solid fly ash after desorption of
carbon dioxide was analyzed using EDTA titration to determine the amount of carbon dioxide captured. [Fig 1] The
results from different experiments show that CSW mixed with FRCA can sequestrate 110-gram carbon dioxide/ kg
dry CSW [2] while by using fly ash as an admixture the amount of carbon dioxide captured was up to 870. 1 µ
mole/gram fly ash [4]. These elements can be applied for the purpose of carbon dioxide sequestration as they are
carbon neutral and environmentally sustainable.
2.1.2 Carbonation Curing
The process of carbonation curing [Fig 2] mostly involves drawing the exhaust gas into the chamber in which
concrete is manufactured and then capturing the carbon dioxide included in the exhaust gas into the concrete. In this
process elements like γ-C2S, coal ash and clinker elements like alite and belite are used to reduce carbon dioxide
emissions. Two phases of calcium silicate, alite, and belite, constitute the main components of Portland cement.
Alite reacts selectively rapidly with water and contributes to the early strength development of concrete, belite reacts
relatively less at early stages but contributes appreciably to strength at later ages. The principal component of the γ
phase of Di-calcium silicate consists of CaO and SiO2. One of the instruments that can we used for evaluation of
carbon dioxide content is and inorganic carbon analyzer called “coulometer”. γ-C2S was cured under 20% carbon
dioxide at 40℃ and 60% relative humidity. [5] Also, alite and belite rich Portland cement underwent carbonation
curing. The carbon dioxide generated can be measured through titration using phenolphthalein indicator. In both
conditions it was seen that by using these elements the carbon dioxide uptake capacity increased as compared to
normal cement. The sample which was cured by carbonation process absorbed around 16.9% carbon dioxide
whereas the carbon dioxide captured by the normal cement sample was 13.5%. [6] Also, the overall carbon dioxide
emission by using γ-C2S as a special admixture was found to be below zero. One of the common advantages seen in
this process is the improvement of the mechanical strength of concrete.
230
MINERAL
CARBONATION
Using
Using FLY ASH
CSW+FRCA
Carbonation
Air tight chamber Surface adsorption
reaction
CO2+ Flue gas

Desorption of CO2
inflow
Measurement of
EDTA titration
CO2 capture
Measurement of
CO2 capture
Fig. 1. The process of mineral carbonation
Carbonation
curing
Using Alite,
Using Cement+
Belite rich
Coal ash
Portland cement
Formation of
Formation of
SUICOM
Cement mix
Concrete
Carbonation
Introduction in
curing through
chamber
“Coulometer”
Carbonation
Phenolphthalein
curing using
indicator
exhaust gas
CO2 emissions Measurement of

below zero CO2
Fig. 2. The process of carbonation curing
231
2.2 CO2 Sequestration
2.2.1 CO2 Sequestration Using Inorganic Admixtures
Carbon dioxide sequestration also known as CO2 capture and storage(CCS) uses various methods and approaches
that can isolate, extract and store carbon dioxide emitted from industrial and energy-related sources. The various
materials used for this purpose are zeolite which is a naturally occurring pozzolanic material used as an alternative
in concrete mixes, fly ash, steel slag, and milled glass. The main steps of the carbon dioxide sequestration include
partial replacement of cement mostly with zeolite or milled glass. And by leaching and precipitation of calcite from
the steel slag. Another process of carbon dioxide sequestration involves using fly ash as a raw material for
mesoporous substances like zeolite and SiO2 which can be the sorbent support for carbon dioxide capture. With
reductions of 10%, 20% and 30% replacement of zeolite in samples resulted in 60.3%, 69.7% and 64.3% global
warming potential reduction respectively. [Fig 3][7] Other results showed that CO2 emission reduction achieved by
FA substitution was 0.207g CO2/g fly ash [8] and the maximum CO2 sequestration of steel slag was found to be
0.25 kg of CO2/kg of Steel Slag [9]. It was found that replacing zeolite helped to maintain the desirable mechanical
properties of the concrete and also by using CCS technology with fly ash harmful components such as Pb, As, Cd,
Se, Cr, S, and Al can be stabilized.
Global Warming Potential Reduction

80
69.7
70 64.3
60.3
% potential reduction
60
50
40
30
20
10
0
0
OPC ZE10 ZE20 ZE30
Fig. 3. Percent of potential reduction through CO2 sequestration
2.2.2 CO2 Sequestration Using Organic Elements
There are some organic elements, mainly bacteria which can be used for the process of CO 2 sequestration. Some of
the popular bacteria used as natural additive material to produce a different type of concrete are Ureolytic, Bacillus
sphaericus [Fig 4], Bacillus pasteurii. These bacteria are used as an additive material in the concrete mix and the
concrete thus formed is known as bio-concrete. Other than these, Bovine CA, Bacillus cereus [Fig 5], etc can be
used to extract carbonic anhydrase (CA) for sequestration purpose of CO 2. [18] Carbonic anhydrase belongs to a
group of enzymes that assist with the rapid interconversion of carbon dioxide and water into carbonic acid, protons
and bicarbonate ions. The rate of hydration is increased when the bio-catalyst CA is used during the hydration
reaction of CO2. Thus in turn CA enzyme makes the reaction much faster, accelerating the process of CO2
232
sequestration. Bacteria like Bacillus mucilaginous L3 can produce carbonic anhydrase (CA) when merged into the
surface of cement walls to absorb CO2 from the surroundings. The microorganism can enhance the absorption of
CO2 and convert it into HCO3- in alkaline pore solution of cementitious materials. Experimental results showed that
the cement based wall of area 10,000m2 absorbed and transferred about 400kg of CO2 to calcite in first 7 days. [17]
Studies have shown that the use of bacteria as an additive material in concrete have the possibility of improving
concrete properties such as self-healing.
Fig. 4. Bacillus sphaericus [18] Fig. 5. Bacillus cereus [19]
2.3 Absorption of CO2
Carbon dioxide absorption [Fig 6] is one of the key technology to control Global Warming. The process of capturing
and trapping of carbon dioxide from the atmosphere is known as carbon dioxide absorption. In those studies the
main component where fly ash and zeolite which was used as an admixture with Portland pozzolana cement
concrete in powder as well as sand form. The process consisted of preparing various blocks of cement with a
varying percentage of zeolite substitution. Calculation of carbon dioxide absorbed by the blocks was calculated by
the following formula:
Carbon dioxide absorbed = (Final Weight - Initial Weight) / Molecular weight of CO2
Whereas in the process involving fly ash to determine the carbon dioxide captured by adsorption over the fly ash the
desorption process was performed at 150℃ to desorb all adsorbed CO2 on fly ash. The maximum amount of carbon
dioxide absorbed using zeolite as an admixture using 30% substitution was found to be 1.61 moles of carbon dioxide
and the carbon dioxide captured by fly ash amounted to 870.1 µmol/g fly ash.[13]. The main advantage of using the
zeolite as substitution material is that it reduces the rate of emission of carbon dioxide as well as absorbs the carbon
dioxide from the surroundings. It can also improve the mechanical strength of the cement.
2.4 CO2 EMISSION REDUCTION USING BLENDED CEMENT
The production of clinker contributes to a very large amount of carbon dioxide emission during the process of
cement manufacturing. Blended cement can be defined as a homogeneous product which is prepared by replacing a
portion of the clinker with industrial by-products such as coal fly ash (residue from coal burning) or blast furnace
slag (residue from iron making) or other pozzolanic materials (example volcanic material). Some of the commonly
used blended cement used for sequestration process are namely calcium sulphoaluminate cement (which contains
yelemite, belite and ferrite), calcium aluminate and calcium aluminosilicate cement consisting of limestone and
bauxite instead of typical calcium silicate found in clay , super sulfated cement (contains 82-85% selected ground
blast furnace slag with 10-15% calcium sulphate and about 5% of Portland cement clinker), magnesium oxide-based
cement and sequestrated carbon cement in which calcium and magnesium are stripped from the sea water and
233
reacted with CO2 to produce high quality cement which is snow white, air permeable and stronger than OPC. The
global potential to reduce CO2 emission by the use of blended cement is estimated to be minimum 5% of total
carbon dioxide emissions from cement manufacturing but it can be as high as 20%. [15]
CO2 Absorption
Using Zeolite
(in powder & sand Using Fly ash
form)
Preparation of Desorption of
cement mix with adsorped CO2 on
zeolite fly ash
Fromation of Measurement of
cement blocks CO2 captured
Measurement of
CO2 absobed
Fig. 6. The process of CO2 absorption
80
Percentage of CO2 emission
70
60
50
reduction
40
30
20
10
0
(a) CSW (b) Belite (c) Zeolite (d) Coal fly ash (e) Blended cement
Fig. 7. Comparison between % of CO2 sequestration through mineral carbonation, carbonation curing, CO2 sequestration,
blended cement
234
3. Discussion
From Fig 7, It was seen that, by the process of CO2 sequestration with the use of inorganic elements zeolite and coal
fly ash, the amount of CO2 sequestrated was 69.7% and 69% respectively. Thus the CO2 sequestrated by the help of
zeolite exceeded the amount of CO2 sequestrated by coal fly ash by a small margin. The CO2 reduction through the
use of various blended cement can vary between 5% to 20%. Through the process of CO 2 absorption, the concrete
mix was prepared with 10%, 20% and 30% zeolite substitution. The highest amount of CO 2 was absorbed by the
concrete block with 30% zeolite substitution and amounted to 1.61mole. Furthermore, through the process of
mineral carbonation and carbonation curing using CSW and belite respectively the amount of CO 2 sequestrated was
found to be 75% and 16.9%.
From these results, it was seen that by using CSW through mineral carbonation the amount of CO 2 emission
reduction was the highest, i.e, 75%. Whereas, the CO2 sequestration through carbonation curing with concrete mix
prepared with belite was found to be the least, i.e, 16.9%.
4. Conclusion
In this paper, several possible methods for CO2 emission reduction in cement and concrete are discussed. Some of
the methods used inorganic elements ( zeolite, Y-C2S, CSW, etc) while some used organic elements (Ureolytic,
Bacillus sphaericus, Bacillus pasteurii, etc). There are different processes for the reduction of CO2 emission from
cement and cement-based industry like mineral carbonation, carbonation curing, absorption, sequestration, etc. From
the data collected using the papers reviewed a bar chart was prepared on the basis of CO 2 emission reduction
potential. The following conclusions can be derived from the above discussion:
(1) The highest percentage of CO2 sequestration was achieved by using concrete slurry waste through the process of
mineral carbonation which amounted to 75%.
(2) The lowest percentage of CO2 sequestration was achieved by using belite as an additive element through the
process of carbonation curing which amounted to 16.9 %.
(3) By using organic elements like bacteria Bacillus mucilaginous L3, it was seen that the cement wall of area
10,000m2 absorbed and transferred about 400 kg of CO2 to calcite in first 7 days.
(4) By using zeolite it was seen that the percentage of CO 2 sequestration was about 69.7 %, for coal fly ash it was
about 69% and for blended cement, it was about 20%.
REFERENCES
1. Uliasz Bochenczyk.A., Pomykala R., (2011) “Mineral sequestration of CO2 with the use of cement waste”,
Procedia Vol. 4, pp. 2855-2860. doi: 10.1016/j.egypro.2011.02.191
2. Xuan D., Zhan B., Poon C. S., Zheng W., (2016) “Carbon dioxide sequestration of concrete slurry waste and its
valorisation in construction products”.Construction and Building Materials Vol. 113, pp. 664-672.
doi:10.1016/j.conbuildmat.2016.03.109
3. Xuan D., Zhan B., Poon C. S., Zheng W., (2016) “Innovative reuse of concrete slurry waste from ready-mixed
concrete plants in construction products”. Journal of Hazardous Materials. doi:10.1016/j.jhazmat.2016.03.036.
4. Siriruang C., Toochinda P., Julnipitawong P., Tangtermsirikul S., (2016) “CO2 capture using fly ash from coal-
fired power plant and applications of CO2-captured fly ash as a mineral admixture for concrete”. Journal of
Environmental Management Vol. 170, pp. 70-78, doi:10.1016/j.jenvman.2016.01.010.
5. Higuchi T., Morioka M., Yoshioka I., Yokozeki K., (2014) “Development of a new ecological concrete with
CO2emissions below zero”. Construction and Building Materials Vol. 67, Part C, pp. 338-343,
235
6. Jang J.G., Lee H.K., (2016) “Microstructural densification and CO2 uptake promoted by the carbonation curing
of belite-rich Portland cement”. Cement and Concrete Research Vol. 82, pp. 50-57,
doi:10.1016/j.cemconres.2016.01.001
7. Valipour M., Yekkalar M., Shekarchi M., Panah S.,i (2014) “Environmental assessment of green concrete
containing natural zeolite on the global warming index in marine environments”. Journal of Cleaner Production
Vol. 65, pp. 418-423, doi:10.1016/j.jclepro.2013.07.055
8. HoWee J., (2013) “A review on carbon dioxide capture and storage technology using coal fly ash”. Applied
Energy Vol. 106, pp. 143-151,doi:10.1016/j.apenergy.2013.01.062
9. Huijgen W. J. J., Comans R. N. J. “Mineral CO2 Sequestration by Steel Slag Carbonation” (2005)
Environmental Science Technology, Vol. 39 (24), pp. 9676–9682, doi: 10.1021/es050795f
10. Mastali M., Abdollahnejad Z., Pacheco-Torgal., F., “Carbon dioxide sequestration on fly ash/waste glass alkali-
based mortars with recycled aggregates: compressive strength, hydration products, carbon footprints, and cost
analysis”, Carbon dioxide sequestration in cementitious construction materials, Vol. 1, pp. 299-348, doi:
10.1016/B978-0-08-102444-7.00013-7, 2018
11. More B., Jadhav P., Jadhav V., Narule G., Mulani S., (2014) “CO2 absorbing concrete block”, International
Journal of Technology Enhancements and Emerging Engineering Research (IJTEEE) Vol.2, No. 7, pp. 147-151
12. Syed Eashan A., Vasudev A., Kumar P.V., Reddy A.S., (2017) “Study on CO2 absorbing concrete”,
International Journal of Civil Engineering and Technology (IJCIET) Vol.8, No. 4 , pp. 1778-1784.
13. Siriruang C., Toochinda P., Julnipitawong P., Tangtermsirikul S., (2016) “CO2 capture using fly ash from coal
fired power plant and applications of CO2-captured fly ash as mineral admixture for concrete”, Journal of
Environmental Management, pp. 70-78, doi: 10.1016/j.jenvman.2016.01.010
14. Imbabi M. S., Carrigan C., McKenna S., (2013) “Trends and developments in green cement and concrete
technology”, International Journal of Sustainable Built Environment, Vol. 1, pp. 194-216, doi:
10.1016/j.ijsbe.2013.05.001
15. Bosoaga A. , Masek O., Oakey J. E., (2009) “CO2 capture technologies for cement industry”, Energy Procedia,
Vol.1, pp. 133-140, doi: 10.1016/j.egypro.2009.01.020
16. Hendriks C. A. , Worrell E. , De Jager D. , Block K. and Riemer P. (2004) “Emission Reduction of Greenhouse
Gases from the Cement Industry”
17. Qian C. , Ren L. , Xue B. , Cao T., (2016) “Bio-mineralisation on cement based materials consuming CO2 from
atmosphere”, Construction and Building Materials, Vol. 106, pp. 126-132, doi:
10.1016/j.conbuildmat.2015.10.105
18. https://en.wikipedia.org/wiki/Bacillus_cereus#/media/File:Bacillus_cereus_SEM-cr.jpg
19. https://www.researchgate.net/figure/SEM-image-for-Bacillus-Sphaericus_fig4_313029996
20. Alshalif A. F., Irwan J.M., Othman N., Zamer M.M. and Anneza L.H. (2017) “Carbon Dioxide (CO2)
Sequestration In Bio-Concrete”, An Overview. MATEC Web of Conferences 103, 05016 (2017)
doi:10.1051/matecconf/201710305016
21. Jang J.G., Kim G.M., Kim H.J., Lee H.K. (2016) “Review on recent advances in CO2 utilization and
sequestration technologies in cement based materials”, Construction and Building Materials, Vol. 127, pp. 762-
773, doi: 10.1016/j.conbuildmat.2016.10.017
236
Paper ID- 57
Development of Fragility Curves for Different Types of RC Frame Structures

Neeva Ahanthem1 and S.S. Ningthoukhongjam2
1
M.Tech. Student, Department of Civil Engineering, Manipur Institute of Technology, Imphal,
nivzthem@gmail.com
2
Department of Civil Engineering, Manipur Institute of Technology, Imphal, sukunao@gmail.com
Abstract. In this paper, seismic behaviours of different types of reinforced concrete (RC) frame (strong column-
weak beam and strong beam-weak column) have been investigated using SAP2000. Nonlinear static (Pushover)
analysis is employed to obtain the capacity curves of different RC frames. The fragility curves are then plotted from
the capacity curves obtained from pushover analysis. The guidelines given in HAZUS manual are referred for
developing fragility curves. The performance point of strong column–weak beam frame structure is obtained at B to
IO levels and that for strong beam-weak column frame structure at IO to LS levels. This indicates that the damages
of the strong beam-weak column frame structure will be severe than the strong column–weak beam frame structure.
Moreover, from the fragility curves, it is observed that the strong beam-weak column frame has higher probability of
damage than strong column–weak beam frame at particular value of spectral displacement.
Keywords: Pushover analysis • Capacity spectrum • Performance point • Fragility curve • HAZUS methodology •
Damage states
1 Introduction
The destruction caused by major earthquakes to mankind and its properties has been massive since time
immemorial. The losses that an earthquake has caused on framed structures have questioned the adequate strength of
the structures against strong ground motions. The scenario will be even more devastating if the building frame has
been constructed as strong beam-weak column type. Hence, the need for seismic vulnerability assessment of
structures is necessary for pre-earthquake disaster mitigation planning and post-earthquake recovery and retrofitting.
Fragility curve is one of such extensively used tool for seismic vulnerability assessment of the structures. It is a
statistical tool indicating the probability of surpassing a certain damage state as a function of an engineering demand
parameter (spectral displacement at a given frequency) that denotes the ground motion. Fragility curve is plotted
using the capacity curve of the structure. Whereas, the capacity curve is developed using Pushover (static-
nonlinear) analysis in which the structure is subjected to a monotonic displacement-controlled lateral load pattern of
constantly increasing magnitude until the structure become totally collapse.
Different authors have been working on different areas of investigating seismic behaviors of RC buildings.
Vasavada et al. [13] observed significant increase in strength and stiffness of RC building due to addition of infill
walls as compared to bare frame building (without infill wall). Kirchil et al. [14] investigated the effect due to the
changes in number of stories of the building on fragility parameters using regression analysis. It has been observed
that there was an effect on fragility parameters significantly due to increase in number of stories of the building.
Mwafy et al. [2] observed that for the twelve RC buildings taken into consideration, good correlation was obtained
between the calculated idealized envelopes of the dynamic analyses and the results obtained from static pushover
analysis. Shinozuka et al. [12] investigated that similar fragility curves are obtained by both Pushover and Time-
history analysis when minor damages are occurred in structure; however, similar fragility curves obtained by these
two methods is no longer applicable if major damages are occurred in structure where nonlinear effects play a
crucial role. From literature review, it has been observed that many research works have been carried out on
different areas of investigating the seismic behavior of RC buildings, fragility analysis and static nonlinear analysis.
However, as per the author knowledge, the comparative study on seismic behaviors of strong column–weak beam
and strong beam-weak column frames has not been performed so far by any author.
As per the code ACI-318 [1], the frame is considered to be strong-column weak-beam if the summation of
flexural strength of columns framing into a joint is greater than 1.2 times the summation of flexural strength of the
237
adjoining beams i.e, ΣMnc ≥ 1.2ΣMnb, where, ∑Mnc and ∑Mnb are the summation of nominal flexural strengths of the
columns and beams respectively. It is beneficial to disperse earthquake-induced energy by yielding of the beams
rather than the columns in multi-storied building frame structures, since columns provide total strength and stability
to the entire frame structure and their failure resulted into global collapse of the entire frame structure. Hence, it is
desirable to control formation of plastic hinges in columns to the possible extent while allowing dissipation of most
of the energy through the yielding of the beams. This is the concept of “strong-column weak-beam”.
In this paper, two different types of idealized reinforced concrete frame structures (Strong Column-Weak Beam
and Strong Beam-Weak Column) have been considered for investigation using Pushover analysis in SAP2000. First
a brief description of methodology used in the analysis is presented. Then details of the structural models are
discussed and the results of the analysis are presented in the form of capacity curves, performance point and fragility
curves.
2 Methodology
2.1 Pushover analysis
Pushover analysis is a static nonlinear analysis where a structure is subjected to gravity loading and a monotonic
lateral load pattern which is constantly increasing over elastic and inelastic behavior until an ultimate state is
reached. It can be executed as force-controlled (used when the load is known, for instance gravity loading) or
displacement-controlled (used when specified drifts are known). A curve is obtained between base shear and roof
displacement in pushover analysis which is known as the capacity curve.
Since pushover analysis gives only the capacity curve of the structure, however performance point of the
structure is determined by the intersection of the capacity curve and demand curve of the structure. Hence demand
curve of the structure has to be determined separately which is carried out in relations to the maximum displacement
induced by the earthquake either to the equivalent SDOF system or at the control node of the full structure and this
maximum displacement is known as the target displacement of the structure. The force-deformation curve for
pushover analysis as per ATC 40 [3] and FEMA 273 [6] is shown in Fig.1 where the points A, B, C, D, E indicate
the force deflection behaviour of the hinge and the points IO (Intermediate Occupancy), LS (Life Safety) and CP
(Collapse Prevention) indicate the acceptance criteria of the hinge.
Fig.1. Force-deformation for Pushover Analysis [8]
Performance point. Capacity Spectrum Method is one of the procedures ATC 40[3] prescribes for determining
performance point. By converting the base shears and roof displacements from a non-linear pushover to equivalent
238
spectral accelerations and displacements using the structure's originally elastic dynamic properties (participation
factor and modal mass coefficient) and superimposing an earthquake demand curve, the non-linear pushover
becomes a capacity spectrum. Spectra plotted in this format are known Acceleration-Displacement Response Spectra
(ADRS). The point at which the capacity curve intersects the demand curve represents the performance point, at
which capacity and demand are equal. By determining the performance point, engineers can develop an assessment
of the spectral acceleration, displacement, and damage that possibly will occur for particular structure responding to
a certain earthquake.
2.2 HAZUS methodology
The FEMA/NIBS earthquake loss estimation methodology, commonly known as HAZUS, is a complex collection of
components that work together to estimate casualties, loss of function and economic impacts on a region due to a
scenario earthquake. HAZUS technical manual offers the method for developing fragility curves for different kinds
of structures. Building fragility curves are lognormal functions that describe the probability of reaching, or
exceeding, structural and non-structural damage states, at given median estimates of spectral response, for
example spectral displacement. These curves take into account the variability and uncertainty associated with
capacity curve properties, damage states and ground shaking. Figure 2 provides an example of fragility curves for
the four damage states used in the FEMA/NIBS methodology and illustrates differences in damage-state
probabilities for three levels of spectral response corresponding to weak, medium, and strong earthquake ground
shaking respectively.
Fig.2. Example Fragility Curves for Slight, Moderate, Extensive and Complete Damage [9]
For a given damage state, a fragility curve is well described by the lognormal probability density function as
given below
P[ds|Sd] = ɸ[ ( )] (1)
Where,
Sd is the spectral displacement defining the threshold of a particular damage state,
239
Sd,ds is median value of spectral displacement at which the building reaches the threshold of damage states and can
be calculated by various damage state models,
βds is standard deviation of natural logarithm of spectral displacement for damage state, ds
ɸ is standard normal cumulative distribution function.
In this paper, fragility curves have been obtained on the basis of capacity spectra which are obtained from
pushover analysis. These capacity spectra are used to obtain the yield spectral displacement (Sdy) and ultimate
spectral displacement (Sdu). The values of the yield spectral displacement and ultimate spectral displacement are
used to obtain the values of medians at different damage states. Damage state model for median value of spectral
displacement (Sd,ds) proposed by Giovinazzi and Lagomarsino[10] has been used in this paper. The values of median
at the damage states are given in Table 1.
Table 1. Proposed damage state model
Spectral Displacement (m)

Damage States
(Sd,ds)
Slight 0.7 Sdy
Moderate 1.5 Sdy
Extensive 0.5 (Sdy+Sdu)
Complete Sdu
Development of damage state variability. Lognormal standard deviation or beta describes the total variability of
fragility curve damage states. The variability associated with capacity curve (βC), the variability associated with
demand spectrum (βD) which is a key parameter in calculating damage state variability and the variability
associated with the discrete threshold of each damage state (βTds ) are the three primary sources which contribute
to the total variability of any given state. Owing to the dependence of demand spectrum with the capacity spectrum,
their respective contributions to the total variability are combined by convolution process which is referred as
“CONV” in the equation (2). The third element βTds is assumed mutually independent of the first two variability
elements and is combined with the results of CONV process using SRSS (Square root of the sum of the squares)
method.
𝛽ds = √( [ ]) ( ) (2)
Where,
𝛽ds is the lognormal standard deviation parameter that describes the total variability of damage state (ds)
𝛽C is the lognormal standard deviation parameter that describes the variability of the capacity curve.
𝛽D is the lognormal standard deviation parameter that describes the variability of the demand spectrum (𝛽D = 0.45 at
short periods, 𝛽D = 0.5 at longer periods)
𝛽Tds is the lognormal standard deviation parameter that describes the variability of the threshold of damage state (ds)
240
Table 2. Variability Values used for two storey model
Degradation values for

Damage State Kappa Factor (к) Capacity Curve Total
Damage (βTds)
(βC) (βds)
Minor
Moderate Moderate
Slight Degradation 0.80
(0.4) (0.3)
(0.9)
Major
Moderate Moderate
Moderate Degradation 0.95
(0.4) (0.3)
(0.5)
Extreme
Extensive Moderate Moderate
Degradation 1.05
(0.4) (0.3)
(0.1)
Extreme
Moderate Moderate
Complete Degradation 1.05
(0.4) (0.3)
(0.1)
Sets of pre-calculated values of Damage-State Beta’s are mentioned in HAZUS technical Manual [9] (Tables 6.5
through 6.7). Hence, 𝛽ds values can be taken directly from the tables considering appropriate values of degradation
or Kappa factors (κ), 𝛽C and 𝛽D values for different types of structures. Table 2 shows the variability values which
are used for the two storeys considering the moderate cases of degradation.
3 Structural Modeling
A G+2 reinforced concrete frame is considered for the analysis of both the RC frames (strong column–weak beam
and strong beam-weak column). Geometry of both the RC frames are similar except different in member sizes.
Details of geometry for both the RC frames are given in Table 3. All the frames have common floor plan which is
shown in Fig. 3 and floor slabs are modeled as rigid diaphragm.
Fig. 3. Plan layout of both the structures
241
Table 3. Structural data
Items Dimensions
Column 0.42m × 0.42m

Strong column weak beam
Beam 0.35m × 0.38m
Column 0.3m × 0.3m

Strong beam weak column
Beam 0.3m × 0.36m
Size of the structure 13.11m × 10.06m
Floor height 3.04m
Thickness of slab 0.15m
Seismic zone V
Seismic zone factor 0.36
Importance factor 1
Response reduction factor 5
Type of soil Medium
M20 grade of concrete and Fe500 grade of reinforcing steel were used for the modeling of both the RC frames
(strong column–weak beam and strong beam-weak column). Mander model (1988) [11] was used for the modeling
of concrete and other basic material properties of the structure are as given in Table 4.
Table 4. Material properties
Modulus of elasticity of steel, Es 210000 MPa
Modulus of elasticity of concrete, Ec 22360.68 MPa
Characteristic strength of concrete, fck 20 MPa
Yield stress of steel, fy 500 MPa
3D modeling of both the RC frames was developed in SAP2000 software as shown in Fig. 4. Beams and columns
were modeled as nonlinear frame elements with lumped plasticity at the start and end of each member. P-M2-M3
hinges for columns and M3 hinges for beams were considered as per FEMA 356 [8]. Live load intensity of
2.5kN/m2 and floor finishes of 1kN/m2 were applied on each floors of both structures.
242
Fig. 4. 3D model of both the structures

4.1 Pushover curve
In this section, static nonlinear (pushover) analysis has been carried out on both the RC frame structures using
SAP2000 to develop capacity curve (pushover curve). Figs. 5a and 5b show the capacity curves of the strong
column-weak beam and strong beam-weak column frame structures respectively. It has been observed from the
capacity curves (Fig. 5) that strong column-weak beam frame can take more load than that of strong beam-weak
column frame since the slope of linear portion of capacity curve for strong column-weak beam frame (Fig. 5a) is
greater than that of strong beam-weak column frame (Fig. 5b). At first yield point B, strong column-weak beam
frame can take more load (~1300 KN) at less displacement (~0.03 m) whereas strong beam-weak column frame can
take less load (~800 KN) at more displacement (~0.045 m).
(a) (b)
Fig.5. Pushover curves for (a) strong column-weak beam and (b) strong beam-weak column frames
243
4.2 Performance point
The performance point in terms of base shear and displacement of the strong column-weak beam and strong
beam-weak column frame structures based on Capacity Spectrum Method are shown in Figs. 6a and 6b respectively.
(a) (b)
Fig. 6. Performance points for (a) strong column-weak beam and (b) strong beam-weak column frames
The performance point of the strong column-weak beam structure occurs at base shear of 1273.724kN and
displacement of 42mm (Fig. 6a), and that of strong beam-weak column structure occurs at base shear of 772.832kN
and displacement of 55mm (Fig. 6b). Occurrence of performance point at greater value of base shear and smaller
value of displacement for strong column-weak beam structure indicates good seismic performances.
4.3 Plastic hinge formation
Plastic hinges are formed at various levels of displacement for the two frame structures. The number and types of
hinges formation at different forcing steps for strong column-weak beam and strong beam-weak column frames are
shown in tables 5 and 6 respectively. The formation of hinges at the last step for the two RC frame structures are
shown in Figs. 7a and 7b.
In case of strong column-weak beam frame structure, the number of hinges formation in the range between B and
IO is 61 and that between IO and LS is 12. Hence a total of 73 hinges are formed in case of strong column-weak
beam frame structure. However, in strong beam-weak column frame structure, the number of hinges formation in the
range between B and IO is 38 and that between IO and LS is 22. And a total of 60 hinges are formed in case of
strong beam-weak column frame structure. Therefore, in strong column-weak beam frame structure, the damage of
the structure is relatively limited and the structure retains a significant portion of its original stiffness and strength,
while in strong beam-weak column frame structure, substantial damage has occurred to the structure as well as it
may have lost a significant amount of its original stiffness but a substantial margin remains for additional lateral
deformation before collapse would occur.
244
Table 5. Hinges result (strong column-weak beam)
LoadCase Step Displacement BaseForce AtoB BtoIO IOtoLS LStoCP CPtoC CtoD DtoE BeyondE Total
Text Unitless m KN Unitless Unitless Unitless Unitless Unitless Unitless Unitless Unitless Unitless
PUSH 0 0 0 242 2 0 0 0 0 0 0 244
PUSH 1 0.007923 416.902 242 2 0 0 0 0 0 0 244
PUSH 2 0.010518 538.51 241 3 0 0 0 0 0 0 244
PUSH 3 0.021024 922.752 223 21 0 0 0 0 0 0 244
PUSH 4 0.029669 1172.189 197 47 0 0 0 0 0 0 244
PUSH 5 0.030661 1189.637 191 53 0 0 0 0 0 0 244
PUSH 6 0.034345 1223.688 188 56 0 0 0 0 0 0 244
PUSH 7 0.035574 1230.816 188 56 0 0 0 0 0 0 244
PUSH 8 0.037159 1245.763 185 59 0 0 0 0 0 0 244
PUSH 9 0.039353 1255.503 185 59 0 0 0 0 0 0 244
PUSH 10 0.042223 1277.487 180 64 0 0 0 0 0 0 244
PUSH 11 0.051452 1329.047 177 67 0 0 0 0 0 0 244
PUSH 12 0.052614 1334.563 175 69 0 0 0 0 0 0 244
PUSH 13 0.055162 1337.505 175 63 6 0 0 0 0 0 244
PUSH 14 0.057249 1347.285 174 58 12 0 0 0 0 0 244
PUSH 15 0.060668 1350.135 172 60 12 0 0 0 0 0 244
PUSH 16 0.062652 1360.482 172 60 12 0 0 0 0 0 244
PUSH 17 0.063637 1363.496 172 60 12 0 0 0 0 0 244
PUSH 18 0.064569 1365.003 171 61 12 0 0 0 0 0 244
PUSH 19 0.0658 1367.816 171 61 12 0 0 0 0 0 244
PUSH 20 0.066292 1368.508 171 61 12 0 0 0 0 0 244
PUSH 21 0.066415 1368.851 171 61 12 0 0 0 0 0 244
PUSH 22 0.067058 1370.006 171 61 12 0 0 0 0 0 244
Table 6. Hinges Result (strong beam-weak column)
LoadCase Step Displacement BaseForce AtoB BtoIO IOtoLS LStoCP CPtoC CtoD DtoE BeyondE Total
Text Unitless m KN Unitless Unitless Unitless Unitless Unitless Unitless Unitless Unitless Unitless
PUSH 0 0 0 242 2 0 0 0 0 0 0 244
PUSH 1 0.007275 189.455 242 2 0 0 0 0 0 0 244
PUSH 2 0.010954 267.81 241 3 0 0 0 0 0 0 244
PUSH 3 0.020665 443.703 230 14 0 0 0 0 0 0 244
PUSH 4 0.03085 600.342 220 24 0 0 0 0 0 0 244
PUSH 5 0.040407 712.78 207 37 0 0 0 0 0 0 244
PUSH 6 0.04404 742.572 196 48 0 0 0 0 0 0 244
PUSH 7 0.045326 749.083 195 49 0 0 0 0 0 0 244
PUSH 8 0.046624 753.564 194 50 0 0 0 0 0 0 244
PUSH 9 0.046818 753.996 194 50 0 0 0 0 0 0 244
PUSH 10 0.0474 756.097 194 50 0 0 0 0 0 0 244
PUSH 11 0.04754 756.384 193 51 0 0 0 0 0 0 244
PUSH 12 0.050853 765.076 190 54 0 0 0 0 0 0 244
PUSH 13 0.053455 770.229 189 55 0 0 0 0 0 0 244
PUSH 14 0.053748 770.607 189 55 0 0 0 0 0 0 244
PUSH 15 0.054041 771.24 188 56 0 0 0 0 0 0 244
PUSH 16 0.062946 784.132 186 54 4 0 0 0 0 0 244
PUSH 17 0.070323 793.403 184 38 22 0 0 0 0 0 244
245
B-IO: 61 B-IO: 38
IO-LS: 12 IO-LS: 22
(a) (b)
Fig. 7. Hinge formation at last step for (a) strong column-weak beam and (b) strong beam-weak column frames
4.4 Fragility curve
The fragility curves for the two RC frame structures have also been plotted based on the spectral displacement
values that satisfy the damage states level (slight, moderate, extensive and complete) as per HAZUS methodology.
The fragility curves of strong column-weak beam frame and strong beam-weak column frame are shown in Figs. 8a
8b. It is evident from the two curves that both the structures have higher probability of slight damage than the other
forms of damages.
(a) (b)
Fig. 8. Fragility curves for (a) strong column-weak beam and (b) strong beam-weak column frames
246
The comparison of probability of damage is made at a spectral displacement of 5cm as the curves of the damages
do not overlap and are constant in both the fragility curves of the two structures. From Fig. 9, it has been observed
that the strong beam-weak column structure has higher probability of damage than the strong column-weak beam
structure.
Fig. 9. Probability of damage at a spectral displacement of 5cm
5 Conclusions
In this study, the performances of the two RC frame structures have been investigated using pushover analysis.
Since it is required that there is no formation of hinges above immediate occupancy level (IO) for the structure to be
safe, it can concluded that the seismic performance of the strong column-weak beam frame structure is better than
that of the strong beam-weak column structure. It has also been observed that the base shear and spectral
acceleration at performance point were found to increase in strong column–weak beam frame structure indicating
better performance and decrease in case of strong beam-weak column frame structure which means poor
performance of the structure. It can then be inferred that the results which are obtained on the basis of capacity,
demand and formation of plastic hinges of the structures give an insight to the actual performance of the structures.
Further, HAZUS methodology for the development of fragility curves has been discussed for the two RC frame
structures. It has been observed that the structure with strong beam-weak column case had higher probability of
damage at a spectral displacement of 5cm than the structure with strong column-weak beam case.
From the results obtained, it can be concluded that the pushover analysis is relatively simple in procedure and
less time consuming to explore the non-linear behavior of the structure and also HAZUS methodology gives an idea
to predict the damage level of the building corresponding to particular value of spectral displacement.
As a future scope for this study, the results generated using Pushover Analysis can be compared with the results
obtained from time history using HAZUS methodology.
References
1. Commentary on Building Code Requirements for Structural Concrete (ACI 318R-14), American Concrete
Institute, 2014
2. A.M. Mwafy, A.S. Elnashai (2001) Static pushover versus dynamic collapse analysis of RC buildings,
Engineering Structures 23, 407–424
3. ATC-40, Seismic Evaluation and Retrofit of Concrete Building, Volume 1and 2, Report NO.SSC 96-01,
Seismic Safety Commission, Redwood City, 1996
247
4. BIS, IS 1893 (Part 1): (2002): Criteria for Earthquake Resistant Design of Structures Part-1 General Provisions
and Buildings, Bureau of Indian Standards, Fifth revision
5. BIS, IS 456:2000, Plain and reinforced concrete code of practice, Bureau of Indian Standards, Fourth revision
6. FEMA 273, Federal Emergency Management Agency, NEHRP guidelines for the seismic rehabilitation of
buildings, Washington DC, U.S.A., October 1997
7. FEMA 440, Improvement of Nonlinear Static Seismic Analysis Procedures, Department of Homeland Security
Federal Emergency Management Agency, Washington, 2005
8. FEMA-356, Pre-standard and Commentary for Seismic Rehabilitation of Buildings, Federal Emergency
Management Council, Washington DC, USA, 2000
9. FEMA, HAZZUS-MH-MR 1 Technical Manual, Federal Emergency Management Agency, Building Seismic
Safety Council, Washington, DC, USA, 2003
10. Giovinazzi, S., Lagomarsino, S. (2006), “Macro-seismic and mechanical models for the vulnerability and
damage assessment of current building.” Bull Earthquake Engineering, 30, 7, September, pp. 415-443
11. J. B. Mander, M. J. N. Priestley, and R. Park (1998) Theoretical stress-strain model for confined concrete, J.
Struct. Eng., 114(8): 1804-1826
12. Masanobu Shinozuka, Maria Q. Feng, Ho-Kyung Kim, Sang-Hoon Kim (2000) Nonlinear Static Procedure for
Fragility Curve Development, J. Eng. Mech.126:1287-1295
13. Megha Vasavada, Dr. V R. Patel (2016) Development of Fragility Curves for RC Buildings using HAZUS
method, IRJET, e-ISSN: 2395 -0056, p-ISSN: 2395-0072, Volume: 03 Issue: 05th May
14. Murat Serdar Kircil, Zekeriya Polat (2006) Fragility analysis of mid-rise RC frame buildings, Engineering
Structures 28, 1335–1345
15. SAP User Manual, version 15, Berkeley (CA, USA): Computer and Structures, Inc., 2000
16. Wasekar S. B. and Chaudhari D. J. (2017) Development of Fragility and Vulnerability curves for RC
Buildings, 2nd InternationaL conference on Emerging trends in Engineering, Science and Management
ISBN:978:93-86171-45-0, 20th May.
248
Paper ID- 63
Chances of fall from height: Identification of root cause

in building construction projects
C.Vigneshkumar1, and Urmi R. Salve2

1
Research Scholar, 2 Assistant Professor
Department of Design,
Indian Institute of Technology Guwahati, Guwahati, India
vigneshkumarchellappa@gmail.com
Abstract. Construction industries have the highest number of labour injuries

and fatality rates with less number of workforces. Hazards associated with
construction projects cause the injuries and fatalities to labours in the
workplace. Hazard is defined as “an inherent property of a substance or
occurrence, which has the potential to cause loss of injury to property, person or
environment”. It is a great belief that, there is a chance to eliminate or minimize
the hazards in construction site, if root causes for hazards were identified. So
many researchers proposed in a different ways to find the root causes for
different hazards in construction projects.
This paper attempts to develop a fault tree model in order to find the root cause
for the hazards in building construction projects. This study limited to fault tree
model for labours fall from slopped roof. As a result, minimal cut sets were
analysed and 17 basic causes with 8 conditional causes were identified which
could trigger the labour to fall from roof in construction site. At the end, this
paper suggests that the fault tree model can be used systematically to identify
the root cause contributing the labour to falls and improve labours safety in
building construction projects.
Keywords: Hazards, Fault tree, Labour safety, Building projects.
1 Introduction
Construction industries are known to be one of the most hazardous industries all over
the world [1]. There is a popular belief that the labours working in construction sites
are facing risks associated with hazards. Hazard can cause harm or damage to human
or environment by any one of the three modes-dormant, armed and active [4].
Hazards associated with the building construction projects can be from work
materials, work equipment’s, work method and practices among others. Data
documented from the published reports shows that the fall from heights (from roof,
ladders, etc.), electrocution and struck by falling objects are the leading safety hazards
on site. In order to eliminate or reduce the hazards, the root causes for hazard should
be identified [7]. Many researchers proposed in a new different ways to find the root
cause for hazards in construction projects.
249
Fault tree analysis is a top-down, detective failure in which an undesired state of a

system is analysed using Boolean logic to combine a series of lower-level events
[3,6]. [2] Developed a fault tree model to identify multiple cause combinations for
fatalities in construction industry. These multiple cause combinations were connected
by AND gate (2 cause and 3 cause combinations) indicating that these 2 or 3 event
contributed simultaneously to these fatal falls. Further, [2] applied fundamental law of
Boolean algebra to reduce all possible cause combinations to the smallest cut sets.
Thus, this paper attempt to develop a fault tree model to identify multiple cause
combination for labour falls from roof in building construction projects. After
Boolean algebra and minimal cut sets (MCS) were analysed, the root causes were
identified which could trigger the labour to fall from roof. This study suggest that
fault tree model can establish all reasonable possible causes contributing to the falls,
in a systematic way and can improve labour safety in construction site.
2 Roof hazards and causes of fall
Roof is a part of the building envelope which covers the upper part of the building or
shelter which protects from animals, weathers, etc [8]. Roof work associated with
building construction projects include roof construction, roof dismantling and roof
maintenance. Statistics revealed that the fall from roof, is the leading hazard among
all other fall hazards (ladders, floor opening, scaffolding, etc.). Fall hazards associated
with roof work includes fall from roof opening, fall through unguarded rail, fall
through unprotected edge and fall through unstable roof material.
The causes for these hazards were classified as enabling causes, triggering causes
and support related causes. The enabling causes are the labour internal causes which
includes labours attitude, labours health and labours skill. The triggering causes are
the active external events indicating the failure of Personnel Protective system (PPE)
which causes the labours to fall. The support related causes are passive external event,
indicating the failure of components supporting the system. Here active devices
indicate the PPE (safety belts, lanyards, etc.) and passive devices indicates the
guardrails, safety nets, etc.
For this illustration, it is assumed that the labour falls from slopped roof during
roof work in construction site. Though it is a roof work, passive devices and active
devices are necessary for labours to protect them from fall. It is also assumed that the
labour worked without proper PPE. Further it is also assumed that no guardrails and
inadequate guardrails were properly installed. These trigger the labours to fall from
roof during construction. This illustration is developed by fault tree model and causes
for labours to fall from roof are identified in further study.
250
3 Fault tree model for fall from roof
In this study, LABOUR FALL FROM SLOPPED ROOF is considered as top event.
From the top event the reasonable possible causes are developed through logic gates.
The top event is further classified as LABOUR GENERAL CAUSES and LABOUR
SUPPORT 2. An OR gate represents the casual relationships and indicates that output
event occur if atleast one or more input occurs. Here LABOUR GENERAL CAUSES
represents the cause of fall and LABOUR SUPPORT 2 represents the roof collapse.
LABOUR GENERAL CAUSES (LGC) further expanded as LABOUR SPECIFIC
CAUSES and SAFETY CONDITIONAL CAUSES (SCC).
Table 1. Abbreviation of Boolean representation and Possible Causes of Labour fall from roof.
Causes Type Description Causes Type Description

LAL Basic Labour Alcohol RWI Basic Roof Weather Impact
Labour Equipment
LDR Basic Labour Drug LEI Basic
Impact
Labour Material
LPE Basic Labour Personality LMI Basic
Impact
LAC Basic Labour Acute OLI Basic Other Labour Impact
Labour Within Work
LCH Basic Labour Chronic LWZ Conditional
Zone
Labour Outside Work
LTR Basic Labour Training LOZ Conditional
Zone
LEX Basic Labour Experience IPPE Conditional Improper PPE
LAP Basic Labour Aptitude NGR Conditional No Guardrails
RDD Basic Roof Design Deficiency NSN Conditional No Safety Nets
RCD Basic Roof Deficiency IGR Conditional Inadequate Guardrails
RHC Basic Roof Hazardous Work IESN Conditional Ineffective Safety Nets
REI Basic Roof Equipment Impact NRC Conditional No Roof Cover
LGI Basic Labour Gust Impact
SCC represents the absence of conditions for safety like Improper PPE (IPPE), no
guardrails (NGR), no safety nets (NSN), etc, which was indicated by the oval shape.
The combination of conditional cause and the intermediate cause is represented by
INHIBIT gate. Fig 1 shows the illustration of fault tree model for labour fall from
roof. LABOUR GENERAL CAUSES further expanded into LABOUR ENABLING,
LABOUR TRIGGERING and LABOUR RELATED CAUSES. These causes are
further expanded to determine the basic causes of the top event. The cause that occurs
as a result of combination of other cause is called intermediate cause. A basic
initiating requiring no further development is called basic causes. Table 1 shows the
Boolean representation and possible causes of labour fall from roof.
251
Labour fall from slopped roof
OR
Labour general causes Labour support 2
A B
Labour specific causes Safety guard condition
OR
Labour Enabling Labour Triggering Labour Support 1
OR OR OR
Labour Attitude Labour Health Labour skill Roof Enabling Roof Triggering
OR OR OR OR OR
LA LP LC LA LT LA RD RH R
RE
L E H C R P D C WI
I
LD
R LE
RC
X
D
Labour Human Impact LG

I
A1
Labour specific impact Labour zone condition
OR OR
LEI LM OL LWZ LOZ

I I
IPPE NGR NSN IGR IESN NRC
Fig. 1. Fault tree illustration for labours falling from roof
252
4 Discussion
Fault tree can be analysed quantitatively and qualitatively. Fault tree qualitative
analysis leads to a determination of MCS [2], which contribute to identify all
combination of basic events that results in the occurrence of top event. MCS for fault
tree is expressed in terms of Boolean algebraic equation. Using this equation, the
MCS of cause and combination of these causes are found as shown in Table 2.
The result in the table shows that a total number of 120 MCS, which consists of 84
two-cause MCS and 36 three-cause MCS. There is no single causes MCS were
identified. As a result, INHIBIT gate represents the combination of basic causes and
conditional status of relation shown in fault tree model. For example, the causes
which trigger the labour to fall from roof are LAL x IPPE, which represents the basic
cause as LABOUR ALCOHOL and conditional cause as IMPROPER PPE. Another
example shows that LEI x LWZ x IPPE, which represents the basic cause as
LABOUR EQUIPMENT IMPACT with conditional causes as LABOUR WITHIN
WORK ZONE with IMPROPER PPE. From this study, it is identified that the two-
cause MCS is higher than three-cause MCS. Therefore there is much important to
concentrate on two-cause MCS.
Table 2. MCS of causes contributing to fall by applying Boolean algebra.
S No Two causes S No Three causes

1 WAL x IPPE 1 WEI x WWZ x IPPE
2 WDR x IPPE 2 WMI x WWZ x IPPE
3 WPE x IPPE 3 OWI x WWZ x IPPE
4 WAC x IPPE 4 WEI x WWZ x NGR
5 WCH x IPPE 5 WMI x WWZ x NGR
6 WTR x IPPE 6 OWI x WWZ x NGR
7 WEX x IPPE 7 WEI x WWZ x NSN
8 WAP x IPPE 8 WMI x WWZ x NSN
9 RDD x IPPE 9 OWI x WWZ x NSN
10 RCD x IPPE 10 WEI x WWZ x IGR
11 RHC x IPPE 11 WMI x WWZ x IGR
12 REI x IPPE 12 OWI x WWZ x IGR
13 RWI x IPPE 13 WEI x WWZ x IESN
14 WGI x IPPE 14 WMI x WWZ x IESN
15 WAL x NGR 15 OWI x WWZ x IESN
16 WDR x NGR 16 WEI x WWZ x NRC
17 WPE x NGR 17 WMI x WWZ x NRC
18 WAC x NGR 18 OWI x WWZ x NRC
19 WCH x NGR 19 WEI x OWZ x IPPE
20 WTR x NGR 20 WMI x OWZ x IPPE
21 WEX x NGR 21 OWI x OWZ x IPPE
22 WAP x NGR 22 WEI x OWZ x NGR
23 RDD x NGR 23 WMI x OWZ x NGR
24 RCD x NGR 24 OWI x OWZ x NGR
25 RHC x NGR 25 WEI x OWZ x NSN
26 REI x NGR 26 WMI x OWZ x NSN
253
27 RWI x NGR 27 OWI x OWZ x NSN

28 WGI x NGR 28 WEI x OWZ x IGR
29 WAL x NSN 29 WMI x OWZ x IGR
30 WDR x NSN 30 OWI x OWZ x IGR
31 WPE x NSN 31 WEI x OWZ x IESN
32 WAC x NSN 32 WMI x OWZ x IESN
33 WCH x NSN 33 OWI x OWZ x IESN
34 WTR x NSN 34 WEI x OWZ x NRC
35 WEX x NSN 35 WMI x OWZ x NRC
36 WAP x NSN 36 OWI x OWZ x NRC
37 RDD x NSN
38 RCD x NSN
39 RHC x NSN
40 REI x NSN
41 RWI x NSN
42 WGI x NSN
43 WAL x IGR
44 WDR x IGR
45 WPE x IGR
46 WAC x IGR
47 WCH x IGR
48 WTR x IGR
49 WEX x IGR
50 WAP x IGR
51 RDD x IGR
52 RCD x IGR
53 RHC x IGR
54 REI x IGR
55 RWI x IGR
56 WGI x IGR
57 WAL x IESN
58 WDR x IESN
59 WPE x IESN
60 WAC x IESN
61 WCH x IESN
62 WTR x IESN
63 WEX x IESN
64 WAP x IESN
65 RDD x IESN
66 RCD x IESN
67 RHC x IESN
68 REI x IESN
69 RWI x IESN
70 WGI x IESN
71 WAL x NRC
72 WDR x NRC
73 WPE x NGR
74 WAC x NGR
75 WCH x NGR
76 WTR x NGR
77 WEX x NGR
78 WAP x NGR
254
79 RDD x NGR
80 RCD x NGR
81 RHC x NGR
82 REI x NGR
83 RWI x NGR
84 WGI x NGR
5 Summary
In this study, fault tree model were developed and MCS were analysed, in which
17 basic cause and 8 conditional cause and 84 MCS (pair of causes) were identified
which could trigger the labour to fall from slopped roof during construction. This
paper suggests that fault tree model can establish all reasonable possible cause
contributing labours to fall in a systematic way and can improve labour safety in
construction site.
References
1. Abdul Rahim Abdul Hamid., Wan Zulkifli., Wan Yusuf and Bachan Singh.: Hazards at
construction site. Proceedings of the 5th Asia-Pacific Structural Engineering and
Construction Conference. 95-104 (2003).
2. Chia-Fen, Chi., Syuan-Zih, Lin., and Ratna Sari Dewi.: Graphical fault tree analysis for
fatal falls in the construction industry. Accident Analysis and Prevention. 359-369 (2004).
3. Fabian C Hadipriono.: Expert System for Construction safety I: Fault-Tree Models, J. of
Perf. of Constructed Facilities. 246-260 (1992).
4. Gregory Carter., and Simson D Smith.: Safety hazard Identification on Construction
Projects. J. of Cons. Engineering and Management. 197-205 (2006).
5. Managing the risk of fall falls while working on roofs in housing construction.:
Workplace Health and Safety Queensland.
https://www.worksafe.qld.gov.au/__data/assets/pdf_file/0003/101595/managing-fall-risks-
on-roofs-in-housing-construction.pdf
6. Pandey, M.: Fault Tree analysis, Engineering and Sustainable Development. 1-17.
7. Samaneh Zolfagharian., and Aziruddin Ressang.: Risk Assessment of Common
Construction Hazards among different Countries 6th International Conference on
Construction in the 21st Century. (2011).
8. The dangerous of working on roofs. Trilgate training trip sheet no-58. Gempler’s. 1-2.
255
Paper ID- 64
Smart light weight MR Damper for enhancement of Seismic mitigation
Daniel C1*, Hemalatha G1, Sarala L1, Tensing D1, Sundar Manoharan S2
1
Department of Civil Engineering, Karunya Institute of Technology and Sciences, Coimbatore,
India, danielckarunya@gmail.com
2
Sathyabama Institute of Science and Technology, Chennai, India.
Abstract
Magnetorheological (MR) fluids with exceptional rheological properties are skilled of exhibiting
quick performance to control vibrations during earthquake. The significant damping properties of
MR fluids was effectively controlled with the help of externally applied magnetic field and current,
Currently, MR fluids with nano Fe3O4 iron particles were used in the preparation of MR fluid to
reduce sedimentation. Fabrication of MR damper consist of nylon material to reduce the weight
and resist high temperature distortion. Weight of the proposed MR damper was of 445 grams with
the magnetic core. The MR fluid of proportion 30% (MRF30), 45% (MRF 45) and 60% (MRF 60)
of iron particle is prepared and cyclic load test with frequency 0.5 Hz and amplitude ± 5mm. The
maximum damping force was found to be 1.032 kN obtained for the MR fluid containing 60 %
Fe3O4 particles in magnetec oil.
Keywords: Damping force, MR Damper, Cyclic load test, Nano Fe3O4, Smart
Introduction
MR damper is semi active type of vibration mitigator and it is filled by Magnetorheological fluid
which can be precisely control by magnetic field [1-4]. MR fluid consist of micron size iron
particles with carrier oil and which can generate the chain formation in presence of magnetic field
when current is applied on it. In recent years, demand of MR damper devices has been intensifying
more and more. There are three types of MR fluid modes are available. Flow mode which is used
in vehicle suspension, prosthetic leg, washing machine as a damper application. Whatever the
passive damper used in present condition is not have this kind of features like change in damping
force with requirement as flow mode [5-8]. Based on the working modes, different MR damper
exhibit different hysteretic behavior. Aim of this paper is to design the light weight MR damper
for seismic application. Magnetic iron particle used in the proposed work is nano Fe3O4
synthesized from a waste product. The controllable damping force of the proposed MR dampers
possess light weight is higher than that of the traditional MR dampers. Additionally, for shocks
256
and vibration mitigation the flow mode MR Damper with various MR fluid is investigated
experimentally. Configuration of a flow mode MR damper for shock and vibration mitigation at
various MR fluid proportion is experimentally investigated.
Valve mode
In valve mode, the smart fluid was placed between the two stationary poles. However, the fluid
flow resistance is modified by controlling the magnetic field which is acting perpendicular to the
flow as shown in fig. 1. Some of the devices making use of the flow mode of operation are dampers,
shock absorbers, servo-valves and actuators [9].
Fig. 1. valve mode
MR fluid
Fig. 2. Schematic representation for the synthesis of MR fluid
MR fluid was successfully prepared by mixing the nano Fe3O4 of average particle size 12 nm in
the carrier oil [10-11]. In the present case, three different MR fluid was prepared by varying the
percentage of iron particles used, 30%, 45%, 60% by weight of nano Fe3O4 was mixed with 70%,
55%, 40% magnetec. The whole mixture was stirred at room temperature using overhead stirrer
for about 24 hours as depicts in fig. 2. The resultant MR fluid is further used for testing.
257
X ray diffraction studies
Fig.3. XRD pattern of as synthesized nano Fe3O4
Fig. 3 depicts the X-ray diffraction pattern of synthesized nano Fe3O4 measured using Schimadzu
equipment, Model XRD-6000, Tokyo, Japan. The particle size of the material was calculated using
the famous scherrer equation and was found to be 12 nm. The characteristic XRD plane (3 1 1)
for Fe3O4 was found at 35.44º. Since it was the most intense plane the average particle size was
calculated from the FWHM value of the same plane.
Scanning electron microscopic studies
Fig. 4. SEM image of nano Fe3O4
The surface morphological study was carried out using the SEM instrument JSM-6390, JEOL,
USA. The fig. 4 shows the SEM micrograph of synthesized nano Fe3O4. From the figure it is clear
that the material shows flake like structure.
258
Vibrating sample magnetometer
Fig. 5. Magnetic measurement of nano Fe3O4
Since the material synthesized is magnetic, magnetic measurement was done using vibrating
sample Magnetometer at room temperature. Fig. 5 depicts the magnetic hysteresis loop of nano
Fe3O4 particles. The sample showed magnetization of 59 emu/g and coercivity of 108G.
MR Damper
The schematic representation of the proposed MR damper and dimensions is depicted in fig. 6 a).
Fig. 6 b) shows the image of proposed MR damper fabricated by nylon material. The smart light
weight MR damper contains of the damper outer housing, inner piston with rod and multiple
annular gap, outer magnetic cores, seal ring and cylinder cap. The outer cylinder and piston pole
head, cylinder cap is fabricated by nylon rod material. The nylon material type 6 has high resistance
towards resistance, vibration, brittleness, deterioration, abrasion. Nylon is easy to machining and
high temperature distortion. The nylon material has a specific gravity of 1.15 cm3, tensile strength
of 93079.22 kN/m2, compressive strength of 110316.11 kN/m2 and deflection temperature range
starts from 3200 F. The conventional MR Damper has a weight of 2500 grams and the proposed
MR damper has a weight of 445 grams which is 82% reduction in weight. The MR damper is
featured by the four-shear forming layer in the piston pole head. The thickness of an annular inner
gap is 1mm. The magnetic field generation depends on the outer coil by which current is varied
from 0 to 1 A with the 0.25A interval. On applying current MR fluid flows inside is magnetized,
yield stress of fluid increases and the circuit is closed. In general, it is proposed that controlling
the current applied to the applied coil, the required damping force was generated by the MR
damper.
259
Fig. 6 a) Main structural dimension of the proposed MR damper b) fabricated light weight MR
Damper
Experimental investigation
The MR damper is tested by exciting the damper with the amplitude ±5 mm, current range from
0A, 0.25A, 0.5A, 0.75A and 1A and frequency 0.5 Hz using MTS Universal Testing machine as
shown in fig. 7. This vibration is similar to the cyclic load test. The resulted damping force of the
proposed MR damper is measured by a data acquisition system and accelerometer was used to
measure movement of the Magnetorheological damper. The outer coil is connected to the DC
power supply to produce power. The measured damping force with respect to displacement for
MRF 30, MRF 45 and MRF 60 is represented in fig. 8-10. The damping force obtained for MRF
30 at 0A was found to be 0.38kN whereas at 1 A it was raised to 0.552 kN. 31 % increase in
damping force was observed for MRF 30 at 1 A than 0A. similarly for MRF 45 and MRF 60, the
maximum damping force was observed at 1 A, and the resuly was found to be 0.8228 kN and 1.032
kN respectively. 44% of damping force increased for MRF 60. The damping force obtained in the
test is appropriate for vibration control.
260
Fig. 7. Experimental setup of light weight MR Damper
Fig. 8. Responses of MR Damper for MRF 30
261
Conclusion
In the present work the smart light weight MR Damper was successfully fabricated using light
weight nylon material. The conventional MR Damper has a weight of 2500 grams and the proposed
MR damper has a weight of 445 grams which is 82% reduction in weight. The nano Fe3O4 was
characterized by XRD and SEM analysis. The magnetic measurement of the smart material nano
Fe3O4 was obtained by VSM at room temperature. Three different MR fluid prepared by varying
the percentage of iron particle presented in namely MRF 30, MRF 35 and MRF 60 respectively.
Magnetec oil acts as the carrier fluid. Cyclic load test carried out with frequency of 0.5Hz and
amplitude ±5 mm for various input current range from 0-1A. The damping force was measure
using MTS Universal testing machine, the maximum damping force was observed for the fluid
MRF 60 at 1 A is 1.032 kN.
262
Acknowledgment
The authors thank Karunya Institute of Technology and Sciences, Coimbatore, Tamilnadu,
India for their constant support. We also extend our acknowledgement to the Department of
Science and Technology (Grant No: DST/TSG/STS/2015/30-G).
Reference
1. Chauhan, Nileshkumar D., and Dipal Patel. “Design Optimization of Flow Mode
Magnetorheological Damper.” Applied Mechanics and Materials 877 (February 2018):
403–408. doi: 10.4028/www.scientific.net/amm.877.403.
2. Goldasz, Janusz. “Study of a Magnetorheological Fluid Damper with Multiple Annular
Flow Gaps.” International Journal of Vehicle Design 62, no. 1 (2013): 21.
doi:10.1504/ijvd.2013.051601.
3. Liao, C R, D X Zhao, L Xie, and Q Liu. “A Design Methodology for a Magnetorheological
Fluid Damper Based on a Multi-Stage Radial Flow Mode.” Smart Materials and Structures
21, no. 8 (July 13, 2012): 085005. doi:10.1088/0964-1726/21/8/085005.
4. Jia, Yongshu. “Design and Experimental Research on the Vehicle Twin-Tube
Magnetorheological Fluids Damper Based on Pressure Driven Flow Mode.” Journal of
Mechanical Engineering 48, no. 10 (2012): 103. doi:10.3901/jme.2012.10.103.
5. Yu, Jianqiang, Xiaomin Dong, Shuaishuai Sun, and Weihua Li. “Hysteretic Model of a
Rotary Magnetorheological Damper in Helical Flow Mode.” Communications in
Computer and Information Science (2018): 15–24. doi:10.1007/978-981-13-2384-3_2.
6. Fu, Benyuan, Changrong Liao, Zhuqiang Li, Lei Xie, Peng Zhang, and Xiaochun Jian.
“Impact Behavior of a High Viscosity Magnetorheological Fluid-Based Energy Absorber
with a Radial Flow Mode.” Smart Materials and Structures 26, no. 2 (January 24, 2017):
025025. doi:10.1088/1361-665x/aa56f4.
7. Ciocanel, C., T. Nguyen, M. Elahinia, And N. G. Naganathan. “Squeeze-Flow Mode
Magnetorheological Fluid Mount.” Electrorheological Fluids and Magnetorheological
Suspensions (October 2007). doi:10.1142/9789812771209_0079.
8. Kim, Kyongsol, Zhaobo Chen, Dong Yu, and Changhyon Rim. “Design and Experiments
of a Novel Magneto-Rheological Damper Featuring Bifold Flow Mode.” Smart Materials
and Structures 25, no. 7 (May 24, 2016): 075004. doi:10.1088/0964-1726/25/7/075004.
9. Zeinali, Mohammadjavad, Saiful Amri Mazlan, Seung-Bok Choi, Fitrian Imaduddin, and
Lailatul Hamidah Hamdan. “Influence of Piston and Magnetic Coils on the Field-
Dependent Damping Performance of a Mixed-Mode Magnetorheological Damper.” Smart
Materials and Structures 25, no. 5 (March 30, 2016): 055010. doi:10.1088/0964-
1726/25/5/055010.
10. C, Daniel, Ajitha Magdalene, Hemalatha G, Tensing D, and Sundhar Manoharan S.
“Experimental Investigation on Magnetorheological Damper for Seismic Resistance of
Structures with Nano Fe3O4 MR Fluid.” International Journal on Applied Bio-Engineering
10, no. 2 (2016): 1–6. doi:10.18000/ijabeg.10140.
263
11. Cruze, Daniel, Hemalatha Gladston, Sophia Immanuel, Sarala Loganathan, Tensing
Dharmaraj, and Sundar Manoharan Solomon. “Experimental Investigation on
Magnetorheological Damper for RCC Frames Subjected to Cyclic Loading.” Advances in
Civil Engineering Materials 7, no. 3 (June 14, 2018): 20170112.
doi:10.1520/acem20170112.
264
Paper ID- 66
Rheological Behavior of Geopolymer Mortar with Fly ash, Slag and their
Blending
Biswajit Roy1* (https://orcid.org/0000-0002-2535-0686), Aminul Islam Laskar1
1
Department of Civil Engineering, National Institute of Technology Silchar, India, 788010
* Corresponding author: biswajit.1117@gmail.com
aminul.nits@gmail.com
Abstract. In the present paper attempt has been made to study the rheological behavior of
geopolymer mortar made with fly ash, ground granulated blast furnace slag (slag) and their
blending (1:1) as source materials. NaOH solution was used as alkaline activator. Molar
concentration of alkaline activator used in the present study were 4M, 8M, 12M and 14.5M. For
preparation of mortar, ratio of source material to sand was kept constant as 1:3. Three different
ratios of activator to binder were considered in the present experimental investigation. Experiments
were conducted by a rotational viscometer. It was observed that degree of thixotropy of slag based
geopolymer mortar reduces after addition of fly ash. Down curve of geopolymer mortar followed
Bingham model with good accuracy with and without blending. Rheological parameters of slag
based mortar reduced significantly after blending with fly ash.
Keywords: Rheology; Geopolymer; Fly ash; Slag; Yield stress; Plastic viscosity
265
1 Introduction
Single point workability tests are widely accepted tests and are empirical in nature that measure
time or distance in most of the times to determine the workability of fresh cementitious materials.
Two point tests, on the other hand, determine rheological parameters in terms of yield stress and
plastic viscosity which are considered as fundamental properties [12, 18, 30-31]. Fluid rheology
methods are used to describe the flow behavior of mortar and concrete under applied shear stress.
It is well established that non-Newtonian fluids such as cement mortar and cement concrete most
often behaves like a Bingham fluid with good accuracy [7, 12, 18, 27, 30-31]. The Bingham model
is represented by the following equation:
𝜏 = 𝜏0 + 𝜇𝛾̇ (1)
where, τ is the shear stress, 𝜏0 is the yield stress, µ is the plastic viscosity and 𝛾̇ is the shear strain
rate. Yield stress gives initial resistance to flow and plastic viscosity governs the flow after it is
initiated. Yield stress is the manifestation of friction among solid particles and plastic viscosity is
the contribution of suspending liquid that results from viscous dissipation due to the movement of
water and sheared material [12, 18-20].
Use of geopolymer binder in mortar and concrete is a field of recent research interest mainly due
to its eco-friendly nature. Geopolymer binder has some advantages over Portland cement such as
better resistance to chemical attack, better freeze-thaw resistance [1-4, 11, 26, 28], stronger
aggregate-matrix interface formation [29] and better fire resistance [32]. The mineral binders
containing amorphous siliceous material and alumina when activated with alkaline solutions
undergo reaction to form geopolymer binder. The alkali activation of alumino-silicate minerals is
a complex chemical process involving the dissolution of raw materials, transportation, orientation
and poly-condensation of the reaction products. Alkali activator used in geopolymerization
266
reaction is sodium hydroxide (NaOH), potassium hydroxide (KOH), mixture of sodium hydroxide
and sodium silicate or mixture of potassium hydroxide and potassium silicate [10, 19-20, 25].
Review of literature reveals that there had been a lot of research on geopolymer concrete with
different source materials and their blends for the determination of mechanical properties of
geopolymer concrete [1-4, 11, 26, 28-29, 32]. Attempt has also been made by researchers to
investigate rheological behavior of geopolymer concrete, mortar and paste; though such reported
literatures are not many. Palacios et al. [23] conducted rheological tests on slag based geopolymer
mortar and concluded that slag paste and mortar follow the Bingham model when NaOH solution
is used as an activator. Criado et al. [8] investigated the rheological behavior of fly ash based
geopolymer paste with NaOH as activator and observed that the Bingham model could be fitted to
those pastes. Laskar and Bhattacharjee [20] conducted rheological tests on fly ash based
geopolymer concrete and concluded that thixotropy is present in fly ash based geopolymer
concrete. They observed that fly ash based geopolymer concrete prepared with a mixture of sodium
silicate and sodium hydroxide solution follows the Bingham equation and yield stress and plastic
viscosity are affected by the molar concentration of alkaline activator. In another study, Laskar
and Bhattacharjee [19] observed that in presence of plasticizer and superplasticizer, rheological
parameters of fly ash geopolymer concrete are affected badly when molar strength of alkali is more
than 4.0 molar. Jang et al. [16] conducted flow test on slag-fly ash blend pastes at different slag-
fly ash ratio using a mixture of sodium hydroxide and sodium silicate as activator. Kashani et al.
[17] studied the interrelationship between surface chemistry and rheology in alkali activated slag
paste and concluded that alkali hydroxide activators cause significant increase in yield stress of an
activated slag paste in comparison to silicate activators at early stage and all activated slag exhibits
shear thinning and thixotropic behavior.
267
Investigation on rheological behavior and factors affecting yield stress and plastic viscosity of
geopolymer mortar incorporating mixture of fly ash and slag is yet to be explored. An attempt has
therefore, been made in the present study to investigate rheological behavior of geopolymer mortar
with fly ash and slag as source materials. Effect of molar strength, and activator to binder ratio (R)
on yield stress and plastic viscosity has also been studied.
2 Experimental Investigation
2.1 Mixture proportion and preparation
Mix designations and proportions of geopolymer mortars are shown in Table 1. In the table “M”
represents the molar strength of NaOH solution and “R” represents the activator to binder ratio by
weight. The dry mass of binder and sand was mixed thoroughly in a laboratory mortar mixer for
two minutes. Sodium hydroxide solution was added and mixed for three minutes. Required
quantity of mix was transferred in the cup of viscometer with the spatula for rheological testing.
Identical mixing procedure and mixing time was maintained for all the samples.
Fine aggregate. Locally available alluvial sand with water absorption of 1.5%, and specific
gravity of 2.6 was used in this study. The sieve analysis, specific gravity, moisture content, and
water absorption were determined as per IS: 2386 [14]. Sand was stored inside the laboratory
throughout the experimental investigation.
Fly ash. Fly ash conforming ASTM Class F collected from National Thermal Power plant at
Farakka (India) was used in this study. The specifications of the fly ash conform to IS: 3812 [15].
The chemical composition of fly ash used in this study is tabulated in Table 2. The specific gravity
and loss on ignition are 2.10 and 0.90 respectively.
268
Slag. Commercially available ground granulated blast furnace slag (slag) was used as another
source material. The specification of the slag conforms to IS: 12089 [13]. The chemical
composition of the slag is tabulated in Table 2. The specific gravity and loss on ignition are 2.80
and 1.41 respectively.
Table 1. Mix designations and proportions (by weight)
Mix No. Fly ash Slag Sand Molar strength R
1 1 - 3 4M 1.33
2 1 - 3 8M 1.33
3 1 - 3 8M 1.17
4 1 - 3 8M 1
5 1 - 3 12 M 1.33
6 1 - 3 14.5 M 1.33
7 - 1 3 4M 1.33
8 - 1 3 8M 1.33
9 - 1 3 8M 1.17
10 - 1 3 8M 1
11 - 1 3 12 M 1.33
12 - 1 3 14.5 M 1.33
13 0.5 0.5 3 4M 1.33
14 0.5 0.5 3 8M 1.33
269
15 0.5 0.5 3 8M 1.17
16 0.5 0.5 3 8M 1
17 0.5 0.5 3 12 M 1.33
18 0.5 0.5 3 14.5 1.33
Table 2. Chemical composition of binders
Chemical Constituents Fly ash (% by weight) Slag (% by weight)
CaO 5 37
SiO2 52 38
Al2O3 23 14
Fe2O3 11 1
MgO - 9
Alkaline activator. Sodium hydroxide palettes (98% purity and specific gravity= 2.13) was used
in the preparation of different concentration of activator solution. The alkali solution was prepared
24 hours prior to the start of experiment.
2.2 Rheological Measurement
Bohlin Visco 2000, a rate and temperature controlled viscometer was used to measure the
rheological properties of the mortar mixes. The viscometer is attached to a computer and operated
through software (Fig. 1). Samples are subjected to stepwise increasing shear rate sequence (15
per sec to 300 per sec in 1.5 minutes) followed by a stepwise decreasing shear rate sequence (300
270
per sec to 15 per sec in next 1.5 minutes) was used for all the mixtures. Stepwise decreasing shear
rate (300 per sec to 15 per sec 1.5 minutes) was used for determination of Bingham parameters.
The stepwise decreasing shear rate sequence is the most commonly used method for measuring
the rheological parameters of cement concrete [25]. Plug flow, if any, is corrected by point
elimination method [21]. Rheological parameters reported are the average of three respective
readings.
Fig. 1. Photograph of Visco 2000 used in the present study.
2.3 Thixotropy
271
An important phenomenon that affects the rheological measurements of cement-based materials is
thixotropy. Thixotropy will lead to error in test results in terms of yield stress and plastic viscosity
if it is not properly addressed. Mewis [22] used the term thixotropy to describe an isothermal gel-
sol transition due to mechanical agitation. Barnes et al. [5] described thixotropy as a decrease of
the apparent viscosity under constant shear stress or shear rate, followed by a gradual recovery
when the stress or shear rate is removed. The thixotropic sample experiences a reversible, time
dependent decrease in viscosity and represents a hysteresis loop when the torque is measured under
a linear increase and then a decrease in rotational frequency [9]. In cement paste, thixotropy is
governed by a combination of reversible coagulation, dispersion, and re-coagulation of cement
particles [6]. Researchers have suggested several approaches to investigate and measure
thixotropy. The simplest approach, developed by Mewis [22] is to measure torque under a linear
increase and then a decrease in rotational frequency. The hysteresis loop, if obtained, indicates the
presence of thixotropy, although the loops do not provide a good basis for quantitative assessment.
In this study, a stepwise increasing shear rate sequence (15 per sec to 300 per sec in 1.5 minutes)
followed by a stepwise decreasing shear rate sequence (300 per sec to 15 per sec in next
1.5miuntes) was used for all the mixtures. The reason for using an increase and then a stepwise
decrease in the shear rate and thus creating a hysteresis loop was mainly to investigate the degree
of thixotropic presence, if any, in the mortar mix.
3 Results and Discussions
3.1 Fly ash based mortar
Typical flow curve of fly ash based geopolymer mortar activated with different concentration of
NaOH solution (as per Table 1) is provided in Fig. 2. The down curve does not follow the up curve
and forms a loop. This is due to the thixotropic breakdown of the sheared material. The down curve
272
of the plot is a straight line which indicates that Bingham model can be fitted for fly ash based
geopolymer mortar.
Fig. 2. Typical flow curve of fly ash based geopolymer mortar (M= 8, R= 1)
Fig 3(a) & 3(b) shows the variation of rheological parameters with the variation of molar strength
at constant activator to binder ratio (R). Yield stress decreases continuously till 12 M solution and
then there is a slight increase at 14.5 M. This behavior is not in agreement with Laskar and
Bhattacharjee [20]. Plastic viscosity on the other hand, increases continuously with the increase in
molar strength. NaOH is a viscous liquid and any increase in NaOH content makes the mixture
more cohesive. Increase in plastic viscosity is also due to decrease in water content at higher molar
solution.
273
Fig. 3. Variation of (a) yield stress (b) plastic viscosity with molar strength for fly ash based mortar
Variation of yield stress and plastic viscosity with the variation of R at constant M is presented in
Fig. 4(a) & 4(b). There is continuous decrease in both yield stress and plastic viscosity with the
increase in R due to introduction of more and more water content at higher R thereby increasing
workability.
Fig. 4. Variation of (a) yield stress (b) plastic viscosity with alkali to binder ratio for fly ash based mortar
3.2 Slag based mortar
Typical flow curve of slag based geopolymer is provided in Fig. 5. It is evident that thixotropy is
present in slag based geopolymer mortar with NaOH as activator and Bingham model could be
10
274
fitted with good accuracy to the down curve. For constant activator to binder ratio (R), both yield
stress and plastic viscosity increases with the increase in molar strength (M) of NaOH solution
[Fig 6(a) & 6(b)]. This behavior is in good agreement with Kashani et al. [17]. Increase in yield
stress and plastic viscosity with increase in M is due to more cohesive mix at high molar strength
and decrease in water content. Like fly ash based geopolymer mortar, a decrease in yield stress
and plastic viscosity is observed with increase in R at constant M [Fig 7(a) & 7(b)].
Fig. 5. Typical flow curve of slag based geopolymer mortar (M= 8, R= 1.33)
11
275
Fig. 6. Variation of (a) yield stress (b) plastic viscosity with molar strength for slag based mortar
Fig. 7. Variation of (a) yield stress (b) plastic viscosity with alkali to binder ratio for slag based mortar
3.3 Blending of fly ash and slag
Fig. 8. Flow curve for blended mix (M=8, R=1.33)
Typical flow curve of blended mix is presented in Fig. 8. It may be observed that thixotropy is
present in all the mixes and the down curve obeys Bingham’s equation. In the present study, degree
12
276
of thixotrophy is assessed by evaluating the area enclosed under hysteretic loops of flow curve
[22]. The area of hysteretic loop represents the amount of work done in breaking down the
microstructure of mortar. Energy required for thixotropic breakdown of mortars having M= 4 and
R= 1.33 is presented in Table 3. A comparison of thixotropy among slag based mortar, fly ash
based mortar and mortar with blending of fly ash and slag shows that degree of thixotropy is the
highest in case of slag based mortar and is the lowest in fly ash based mortar.
Table 3. Thixotropic evaluation (M= 4, R= 1.33)
Mortar type Work done (Pa-s-1)
Fly ash 897
Slag 964
Blend (Fly ash + Slag) 922
Comparison of rheological parameters of blending with slag and fly ash based geopolymer mortar
is shown in Fig. 9-10. It may be observed both the rheological parameters of slag based
geopolymer mortar decrease to a large extent after blending with fly ash. Values of yield stress
and plastic viscosity of blended specimens are always greater than values of rheological parameters
with fly ash and less than values of rheological parameters with slag. It is to be noticed from Fig.
9 that values of yield stress for mortars incorporating fly ash and blended source materials are very
close and there are no significant differences.
13
277
Fig. 9. Comparison of Yield Stress with fly ash, slag and blending
Fig. 10. Variation of Plastic Viscosity with fly ash, slag and blending
4 Conclusion
14
278
In the present study, rheological behavior of fly ash and slag based mortar and their blending (1:1
by weight) was investigated at different molar concentration of NaOH activator and at different R-
values. It may be concluded that thixotropy is present in geopolymer mortar with fly ash, slag and
their blending when activated with NaOH solution. Degree of thixotropy is the highest with slag
and is the lowest with fly ash as the source material. The down curves in all the cases under
investigation follow Bingham equation with good accuracy. Values of rheological parameters such
as yield stress and plastic viscosity of slag based geopolymer mortar are maximum and are reduced
significantly after addition of 50% fly ash.
References:
1. Ariffin, M.A.M., Bhutta M.A.R., Hussin M.W., Tahir M.M., Aziah N.: Sulfuric acid
resistance of blended ash geopolymer concrete. Const. Build. Mater. 43, 80-86 (2013).
2. Bakharev, T., Sanjayan J.G., Cheng Y.-B.: Resistance of alkali-activated slag concrete to
acid attack. Cem. Conc. Res. 33, 1607-1611 (2003).
3. Bakharev, T., Sanjayan J.G., Cheng Y.-B.: Sulfate attack on alkali-activated slag concrete.
Cem. Conc. Res. 32, 211-216 (2002).
4. Bakharev, T., Sanjayan J.G., Cheng Y.-B.: Alkali activation of Australian slag cements.
Cem. Conc. Res. 29, 113–120 (1999).
5. Barnes, H.A., Hutton J.F., Walters K.: An Introduction to Rheology. Rheol. Ser. 3, Elsevier,
Amsterdam (1989).
6. Beaupre, B.: Rheology of High Performance Shotcrete, Ph.D thesis, University of British
Columbia, Vancouver, BC, Canada (1994).
15
279
7. Bouras, R., Kaci A., Chaouche M.: Influence of viscosity modifying admixtures on the
rheological behavior of cement and mortar pastes. Korea-Australia Rheol. J. 24, 35-44
(2012)
8. Criado, M., Palomo A., Fernandez-Jimenez A., Banfill P.F.G.: Alkali activated fly ash:
effect of admixtures on paste rheology. Rheol. Acta 48, 447-455 (2009).
9. de Larrard, F., Szitkar J.-C., Hu C., Joly M.: Design of a Rheometer for Fluid Concrete.
RILEM Workshop Special Concrete- Workability and Mixing, 201-208 (1993).
10. Duxson, P., Fernandez-Jimenez A., Provis J.L., Palomo A., van Deventer J.S.J.:
Geopolymer Technology: The current State of Art. J. Mater. Sci. 42, 2917-2933 (2007).
11. Fernandez-Jimenez, A., Palomo J.G., Puertas F.: Alkali-activated slag mortars: Mechanical
strength behavior. Cem. Conc. Res. 29, 1313-1321 (1999).
12. Ferraris, C.F. and Brower L.E.: Comparison of Concrete Rheometer: International Tests
at LCPC Nantes France, Report No. NISTIR 6819 (2001).
13. IS: 12089: Specification for Granulated Slag For the Manufacture of Portland Slag Cement.
Bureau of Indian Standards, Indian Standard Code of Practice, New Delhi, India (1999).
14. IS: 2386: Methods for Tests for Aggregates for Concrete. Bureau of Indian Standards,
Indian Standard Code of Practice, New Delhi, India (1999).
15. IS: 3812: Specification for Fly Ash for Use as Pozzolana and Admixture. Bureau of Indian
Standards, Indian Standard Code of Practice, New Delhi, India (1999).
16. Jang, J.G., Lee N.K., Lee H.K.: Fresh and hardened properties of alkali activated fly
ash/slag pastes with superplastizers. Const. Build. Mater. 50, 169-176 (2014).
16
280
17. Kashani, A., Provis J.L., Qiao G.G., van Deventer J.S.J.: The interrelationship between
surface chemistry and rheology in alkali activated slag paste. Const. Build. Mater. 65, 583-
591 (2014).
18. Koehler, E.P., Fowler D.W.: Development of a Portable Rheometer for Fresh Portland
Cement Concrete. Report No. ICAR 105-3F, International Center for Aggregate Research,
The University of Texas at Austin (2004).
19. Laskar, A.I., Bhattacharjee R.: Effect of Plasticizer and Super Plasticizer on Rheology of
Fly-Ash-Based Geopolymer Concrete. ACI Mater. J. 110, 513-518 (2013).
20. Laskar, A.I., Bhattacharjee R.: Rheology of fly ash based geopolymer concrete, ACI Mater.
J. 108, 536-542 (2011).
21. Laskar, A.I., Talukdar S.: Design of a new rheometer for concrete, J. ASTM Int. 5 (2008).
22. Mewis, J.: Thixotropy- a general review. J. Non-Newtonian Fluid Mech. 6, 1-20 (1979).
23. Palacios, M., Banfill P.F.G., Puertas F.: Rheology and Setting of Alkali Activated Slag
Pastes and Mortars: Effect of Organic Admixtures. ACI Mater. J. 105, 140-148 (2008).
24. Palomo, A., Santiago A., Fernandez-Jimenez A., Sobrados I., Sanz J.: Alkaline Activation
of Fly ashes: NMR Study of the Reaction Products. J. Am. Ceram Soc. 87, 1141-1145
(2004).
25. Palomo, A., Grutzeck M.W., Blanco M.T.: Alkali-activated fly ashes: A cement for the
future. Cem. Conc. Res. 29, 1323-1329 (1999).
26. Puertas, F., Amat T., Fernandez-Jimenez A., Vazquez T.: Mechanical and durable
behaviour of alkaline cement mortars reinforced with polypropylene fibres. Cem. Conc.
Res. 33, 2031-2036 (2003).
17
281
27. Sant, G., Ferraris C.F., Weiss J.: Rheological Properties of Cement Pastes: A Discussion
of Structure Formation and Mechanical Property Development. Cem. Conc. Res. 38, 1286-
1296 (2008).
28. Shi, C., Krivenko P.V., Roy D.: Alkali-activated cements and concretes. Taylor and
Francis, USA and Canada (2006).
29. Shi, C., Xie P.: Interface between cement paste and quartz sand in alkali-activated slag
mortars. Cem. Conc. Res. 28, 887–896 (1998).
30. Tattersall, G.H.: Workability and Quality Control of Concrete. E & FN Spon, London
(1991).
31. Tattersall, G.H., Banfill P.F.G.: The Rheology of Fresh Concrete. Pitman Publishing,
Marshfield, MA (1983).
32. Zhang, H.Y., Kodur V., Qi S.L., Cao L., Wu B.: Development of metakaoline-fly ash based
geopolymers for fire resistance application. Const. Build. Mater. 55, 38-45 (2014).
18
282
Paper ID- 67
EFFECT OF PILE SPACING AND RAFT THICKNESS ON

BEHAVIOUR OF PILED-RAFT FOUNDATION – A
PARAMETRIC STUDY USING FEM
Mukul Kalita1, Utpal Kumar Nath2, Palash Jyoti Hazarika3
1
Assistant Professor, Department of Civil Engineering, Assam Don Bosco University,
mukulkalita138@gmail.com
2
Associate Professor, Department of Civil Engineering, Assam Engineering College,
ukn.ce@aec.ac.in
3
Professor, Department of Civil Engineering, Assam Engineering College, pjhaz@rediffmail.com
ABSTRACT
This paper presents a parametric study of piled-raft foundation under static loading in
homogenous soft clay and loose sand. 3D finite element analyses have been carried out for
conventional un-piled raft as well as piled-rafts with 4, 9 and 16 number of piles underneath
the raft. A three dimensional finite element package for soil and foundation is used for plane
strain, linear elastic modeling of piled-raft system and Mohr-Coulomb yield criterion is used
to represent the two soil types as elastic-perfectly plastic material. The loading parameters,
length and diameter of circular piles, Poisson’s ratio and elastic moduli of raft, piles and soil
are kept constant throughout the analyses. Focus is thereby on two parameters - pile spacing
i.e. S/D ratio and raft thickness. Variation in the peak values of settlements, bending
moments, torsion, axial forces and shear forces are investigated in both raft and piles due to
varying pile spacing and raft thickness in clay and sandy soil. The general conclusion of this
study for providing a piled-raft foundation is to design a flexible/semi-flexible raft of
optimum thickness and S/D ratio so that the deflections and force parameters on the raft and
the piles are within permissible limits and the piles can effectively act as settlement reducers
in addition to sharing of loads.
Keywords – piled-raft, soft clay, loose sand, finite element analyses, S/D ratio, raft thickness
INTRODUCTION
In the past few decades, there has been an increasing recognition that the use of pile groups in
conjunction with the raft can lead to considerable economy without compromising the safety
and performance of the foundation. Such a foundation makes use of both the raft and the
piles, and is referred to here as a piled-raft. The use of piled-raft foundation has become
popular in recent years, as the combined action of the raft and the piles can increase the
bearing capacity, reduce settlements, and the piles can be arranged so as to reduce deflection
in the raft. A piled raft foundation is economical compared to the pile foundation because
piles in this case do not have to penetrate the full depth of soil stratum but it can be
terminated at higher elevations.
The main objective of the study is to investigate the influence of some of the factors
such as pile spacing and thickness of raft on the behavior of piled raft foundation system. The
scope of this work is to study the variation in the peak values of settlement, bending
moments, torsional moments, shear forces and axial forces for a piled-raft foundation due to
283
change in pile spacing and raft thickness in soft clay and loose sand, and to compare the
results for the two types of soil considered for arriving at some general hypothetical
conclusions.
METHODOLOGY
In this study a 16 m x 16 m raft with 0.6 m diameter massive circular piles were analyzed
using a finite element software package ‘Plaxis 3D Foundation’. A plane strain finite element
model was used to model the piled raft foundation. The raft and piles were assumed to be
linearly elastic. The Mohr-Coulomb yield criterion was used to represent both the soil types –
soft clay and loose sand, as elastic-perfectly plastic material. A single layer homogenous
isotropic soil with water table at a depth of 6 m from ground level was assumed for the study.
Here an undrained condition was assumed and total stress analysis was carried out using
plastic calculations. The soil was discretized as 15 noded triangular elements. The piles and
raft were modelled using a plate element. The side skin friction in piles was taken into
account by applying interface reduction factor Rinter. Sub-soil thickness is considered to a
depth of 30 m from ground surface for both the soil types. A non-uniform vertical loading in
the form of concentrated column loads has been imposed on the piled raft. Effect of lateral
loading is pseudo-statically considered by making one half of the piled raft heavier compared
to the other half. The various geometric and material properties of soil, raft, piles and
interfaces are tabulated below:
Table 1: Properties of the soil being used in the study
Soil type Soft Clay Loose Sand

Material model Mohr-Coulomb Mohr-Coulomb
Angle of friction (ϕu) 00 280
Stiffness (Eref) 1.5 x 104 kN/m2 6 x 103 kN/m2
Cohesion (Cu ) 30 kN/m2 1 kN/m2
Poisson’s ratio (ʋ) 0.35 0.30
Dilatancy angle (Ѱ) 00 10
Saturated unit weight (ϒsat) 17 kN/m3 18 kN/m3
Unsaturated unit weight (ϒd) 15 kN/m3 16 kN/m3
Drainage condition Undrained Undrained

Permeability 0.25 m/day 1 m/day
Depth of water table 6m 6m
284
Table 2: Properties of the materials being used in the study
Material Bored Pile Raft (Floor) Basement wall
Material model Linear elastic Linear isotropic Linear isotropic

Material type Non-porous Non-porous Non-porous
Stiffness (Eref) 2.15 x 107 kN/m2 2.15 x 107 kN/m2 2.15 x 107 kN/m2
Poisson’s ratio (ʋ) 0.15 0.15 0.15
Unit weight (ϒsat) 25 kN/m3 25 kN/m3 25 kN/m3
Permeability 0 m/day 0 m/day 0 m/day
Table 3: Properties of the soil-pile and soil-raft interface
Soil type Soft Clay Loose Sand

Rinter 0.8 0.9
Angle of friction (ϕi) 0 25.60
Cohesion (Ci) 24 kN/m2 0.8 kN/m2
Table 4: Dimensions of the materials being used in the study
Basement
Material Pile Raft
wall
Diameter,
0.6m - -
D
Thickness,
- 0.8m, 1.0m, 1.2m, 1.4m, 1.6m 0.5m
d
8m, 10m, 12m, 14m,
Length, L - -
16m
Number, N 0, 4, 9 - -
Spacing, S 6m - -
S/D ratio 10 - -
285
The loading and geometric details of the model have been depicted in the figure below:
Figure 1: Piled-raft configuration for analysis
RESULTS AND DISCUSSIONS
Effect of Pile Spacing
It is observed that the maximum positive moment in raft decreases non-linearly with increase
in pile spacing, for loose sand and soft clay, with pile diameter and raft thickness remaining
constant. For a pile length of 10 m, there are almost 21% and 20% reduction in positive raft
moment for clay and sand over a pile spacing increment from 4 m to 12 m.
286
Figure 2: Maximum positive raft moment versus pile spacing
Other parameters remaining same, the maximum negative moment in raft first decreases and
then increases with increase in pile spacing from 4 m to 12 m, for both the types of soil. For a
pile length of 10 m the maximum decrement in negative moments observed are almost 40%
and 47% for clay and sand respectively when the pile spacing increases from 4 m to 6 m.
Figure 3: Maximum negative raft moment versus pile spacing
The maximum torsional moment in raft first decreases and then increases with increase in
pile spacing from 4 m to 12 m, for both the types of soil. For a pile length of 10 m, the
maximum decrement in torsional moments observed are almost 39% and 34% for clay and
sand respectively when the pile spacing increases from 4 m to 6 m.
287
Figure 4: Maximum raft torsion versus pile spacing
The maximum shear force in raft exhibits a significantly decreasing trend with increase in
pile spacing, for both the types of soil. Over a pile spacing increment from 4 m to 12 m, there
are almost 51% and 48% reduction in raft shear for clay and sand respectively.
Figure 5: Maximum raft shear versus pile spacing
The maximum vertical deflection of raft varies directly with the spacing of piles for both the
types of soil. The maximum deflection in raft decreases by about 6% for clay and 11% for
sand when the pile spacing is decreased from 12 m to 4 m.
288
Figure 6: Maximum settlement of raft versus pile spacing
Maximum pile axial force is found to be increasing uniformly with increase in pile spacing.
For a pile spacing increment from 4 m to 12 m, increase in maximum pile axial values are:
Almost 10% for soft clay and
Almost 14% for loose sand
Figure 7: Maximum axial force in piles versus pile spacing
The maximum settlement of piles varies directly with the spacing of piles. For both the types
of soil, it is observed that the maximum vertical deflection of piles decreases with decrease in
pile spacing from 12 m to 4 m, the maximum reduction being 6% for clay and 11% for sand
over that range of pile spacing decrement.
289
Figure 8: Maximum settlemet of piles versus pile spacing
Effect of Raft Thickness
Pile diameter, pile spacing and pile length remaining constant, the maximum positive
moment in raft decreases significantly with increase in raft thickness, for loose sand and soft
clay with. The trend of decrease is almost uniform for both the types of soil. For a 9 pile
group, there are almost 52% and 35% reduction in positive raft moment for clay and sand
respectively when the raft thickness increases from 0.8 m to 1.6 m.
Figure 9: Maximum positive raft moment versus raft thickness for a 9-piled raft
Other parameters remaining same, the maximum negative bending moment in raft increases
directly with increase in raft thickness, for both the types of soil. Over a thickness increment
from 0.8 m to 1.6 m of raft, there are almost 40% and 36% increase in negative raft moment
for clay and sand respectively for a 9-piled raft.
290
Figure 10: Maximum negative raft moment versus raft thickness for a 9-piled raft
The maximum shear force in raft shows an increasing trend with increase in raft thickness,
for both the types of soil. As the raft thickness increases from 0.8 m to 1.6 m, the maximum
raft shear increases by about 10% and 12% in clay and sand respectively.
Figure 11: Maximum raft shear force versus raft thickness for a 9-piled raft
The maximum torsional moment in raft increases directly with increase in raft thickness, for
both the types of soil. Over a thickness increment from 0.8 m to 1.6 m of raft, there are
almost 39% and 36% increment in raft torsion for clay and sand respectively for a 9-piled
raft.
291
Figure 12: Maximum torsion in raft versus raft thickness for a 9-piled raft
The maximum settlement of raft varies directly with the thickness of raft. For both the types
of soil, the maximum deflection in raft increases almost uniformly with increase in raft
thickness from 0.8 m to 1.6 m, the maximum increment being 21% for soft clay and 28% for
loose sand.
Figure 13: Maximum settlement of raft versus raft thickness for a 9-piled raft
Significant effect of raft thickness on axial forces in piles have have observed for both the
types of soil. Maximum pile axial is seen to be increasing uniformly with with increase in raft
thickness. For a thickness increment from 0.8 m to 1.6 m of raft, increase in maximum pile
axial values are:
Almost 22% for soft clay and
Almost 24% for loose sand
292
Figure 14: Maximum axial force in piles versus raft thickness for a 9-piled raft
The maximum settlement of piles varies directly with the thickness of raft. For both the types
of soil, it is observed that the maximum vertical deflection of piles decreases significantly
with decrease in raft thickness from 1.6 m to 0.8 m, the maximum reduction being 17% for
soft clay and 23% for loose sand.
Figure 15: Maximum settlement of piles versus raft thickness for a 9-piled raft
CONCLUSION
The general observations of this study can be summerized as follows:

The force parameters (axial forces, bending moments, torsional moments, shear forces etc.) in
raft reach their minimum value for an optimum spacing of piles beyond which the moments
and forces in raft increase considerably. Increased pile spacing gives rise to increased vertical
deflection in both the raft and the piles. The force parameters are also considerably higher for
a raft which is rigid and stiff compared to a flexible raft. Again these forces acting on the raft
increase with increase in thickness of the raft. The piles supporting the raft basically reduce
the vertical deflection of the system. The force parameters in piles also exhibit an increasing
trend with increased raft thickness.
The most favourable solution to provide a piled raft system is to design a flexible/semi-
flexible raft of optimum thickness so that the forces coming on the raft and the piles are not
293
excessively high and the piles have adequate capacity to withstand the forces transferred to
them. To get the most serviceable and economic design of a piled raft, optimization analyses
have to be carried for determining the most favourable pile spacing that will give rise to
optimum forces in the raft and the piles.
REFERENCES
1. Maharaj, D. K., and Gandhi, S. R., Non-linear Finite element analysis of piled- raft
foundations, Proceedings of the Institution of Civil Engineers Geotechnical
Engineering 157, 2004 Issue GE3 pp. 107–113.
2. Novak, J., Reese, L. C., and Wang, S. T., Analysis of Pile-Raft Foundations with 3D
Finite-Element Method, Proceedings of the 2005 Structures congress and 2005
Forensic Engineering Symposium, New York.
3. Sanctis, L. D., Alessandro, and Mandolini, A., Bearing Capacity of Piled Rafts on
Soft Clay soils, Journal of Geotechnical and Geoenvironmental Engineering, Vol.
132, No.12, 2006, pp. 1600-1610.
4. Chow, H. S. W., Analysis of piled-raft foundation with piles of different length and
diameter, Ph.D. Thesis, 2007.
5. Oh, E. Y. N., Huang, M., Surarak, C., Adamec, R., and Balasubramaniam, A. S.,
Finite element modeling for piled raft foundation in sand, Eleventh East Asia-Pacific
Conference on Structural Engineering & Construction (EASEC-11), “Building a
Sustainable Environment”, November 19-21, 2008, Taipei, Taiwan.
6. Giretti, D., Modelling of piled raft foundations in sand, Ph.D. thesis (2009).
7. Oh, E. Y. N., Lin, D. G., Bui, Q. M., Huang, M., Surarak, C., and Balasubramaniam,
A. S., Numerical analysis of piled raft foundation in sandy and clayey soils,
Proceedings of the 17th International conference on Soil Mechanics and Geotechnical
Engineering (ICSMGE 2009): 1159-1162.
8. Rabiebi, M., Parametric Study for Piled Raft Foundations, Electronic journal of
Geotechnical Engineering Volume 14, Bundle A, 2009.
9. Lee, J., Kim, Y., and Sangseom, J., Three-dimensional analysis of bearing behaviour
of piled raft on soft clay, Computers and Geotechnics, Volume 37, Issue 1-2, pp. 103-
114, 2010.
294
Paper ID- 69
Removal of VOCs and Improvement of Indoor Air Quality Using Activated

Carbon Air Filter
Sujon Mondal1, Soham De1 and Dr. Purnachandra Saha2* (ORCID-0000-0002-2722-1669)

1
(Mail id: sujonmondal986@gmail.com, sohamde053@gmail.com)
2
Senior Associate Professor, School of Civil Engineering, KIIT Deemed to be University,
Bhubaneswar, India
(Mail id: dr.purnasaha@gmail.com) * Corresponding Author
Abstract
Volatile Organic Compound (VOC) is one of the most common air pollutants emitted from industries like Chemical,
Petrochemical industries, as well as when plastics are burned. It is very harmful to our environment which affects
climate change, the life cycle of plants and the health of all living beings. So it is necessary to control its emission
for improvement of air qualities which is beneficial to the indoor environment. The objective of the study is to
review the performance of different activated carbon-based air filters. Various activated carbon-based techniques are
the use of coconut shell, photocatalyst TiO2, Polystyrene foam, sorption-type, and granular activated carbon. Several
aldehydes and ketones have been removed effectively using activated coconut shell. Removal of nitrogen oxide was
possible by using TiO2 based activated carbon. It is observed from this study that activated carbon-based techniques
are effective for removal of VOCs and enhancing the indoor air quality.
Keywords: VOCs, IAQ, Activated carbon filter, Coconut shell, photocatalyst TiO 2, Polystyrene foam.
1. Introduction
Volatile organic component(VOC) is one the most harmful contaminant present in the air. Its boiling point lies
between 50oCto 260oC. It's consist of several components such as Acetaldehyde, Acetone, Benzene, Carbon
tetrachloride, Ethyl acetate, Ethylene glycol, Formaldehyde, Heptane, Hexane and several other [1]. There are
several detrimental effects that occur due to this such as serious health disorders which include irritation of nose and
eyes, damages liver, badly affect the nervous system [2]. It not only affects the human body but also affects the
change of climate, the temperature of the earth also reduces the plant growth. It is also responsible for the
photochemical smog which is hazardous in nature [3,2]. It reduces the indoor air quality (IAQ). IAQ is defined as
the air quality in and around the building. There are several factors that ./affect the IAQ such as temperature,
humidity, biological pollutants, air exchange rate, air movement, particle pollutants, and gaseous pollutants. One of
the major factors that affects the IAQ is the gaseous pollutants[4,5] For improvement of this IAQ and also outdoor
air quality there are several activated carbon-based filters are used which might be shell based such as coconut shell
based, pecan shell and almond shell based [6,7,8]or it might be activated charcoal, bamboo charcoal [9]. Another
type of activated carbon such as the granular activated carbon and the activated carbon fiber(AFC) [10,11] There are
several techniques that are used for the filtration process such as oxidation, catalysis, regeneration and reveres
reactor process [3,2,12].The main objective of the study is to study the different activated carbon-based filter and
comparing their
1) Removal Efficiency and Specific Surface area
2) Study the several filtration techniques that are available.
295
2. DIFFERENT TECHNIQUES OF ACTIVATED CARBON BASED AIR FILTER
Activated carbon-based air filter (ACBAF) are mainly used for the removal of different harmful contaminant and
impurities in the air which includes VOCs, CO2, NO2, SO2 etc. This filtration process mainly differs based on the
activated carbon used and process.[10] There are two types of techniques namely, adsorption and filter based are
used to improve the indoor air quality. In the adsorption process, increasing surface area with more pore
development should be used to improves the performance. [2]This types of techniques are mainly used for the
purification of the indoor as well as outdoor air. Filter based techniques mainly contain several activated carbon-
based filters which purify the air using oxidation, catalysis, regeneration and reveres reactor process[1].
2.1 ADSORPTION TYPE

Adsorption is a surface phenomenon in which solid surface attracts gas molecule or liquid solution. The adsorbing
solid is called adsorbent (activated carbon) and the adsorbed particle is called adsorbate (air contaminants).
When gas or vapor comes in contact with solid particles, a portion of the gas-composition is absorbed by the solid.
The gas The contact-solid surface is called adsorbate. Many organic and inorganic contaminants removed either
from the gaseous or liquid solutions by the natural process of adsorption on the very porous medium solid phase
with large internal surfaces. Adsorbates can attach themselves onto the surfaces in two ways a) physisorption b)
chemisorptions. In physisorption, adsorbents attach themselves to the solid medium by weak van der Waals
forces. But in case of chemisorptions, the absorbates sticks to the solid medium by forming (a) chemical bond
between the absorbates and the solid medium. The above-scripted steps of adsorption can be summarized as per
the following: solute diffuses near the solid surface, diffuses into the pores of the particle, then moves through the
pore wall and adsorbs to the pore wall surface. To increase the adsorption capacity of activated carbon a large
specific surface is favored [13].
2.2 ADSORPTION PROCESS

Activated carbons are mainly used as a filter material for removal of contaminant gases present in air, Porous
carbon-based materials have high thermal and chemical stability as well as good adsorption capabilities [14]
Coconut shell, Activated charcoals, Fiber rejects and commercial fibers of Polymetaphenylene is ophthalamide
,Fiber of Polyparaphenylene terephthalamide and bamboo carbon and anthracite activated carbon were used as raw
materials for activated carbon and Samples were milled inside planetary ball mill for 30 hours with the speed of 300
rpm to reduce the size, Hydrogen (H2) was used as adsorbate gas. The granular coconut shell charcoals used here in
the granulated form[15] The average diameter of the fibers was around 10 micrometer(pm)[11].
Different types of carbonaceous material for ex-plant, mineral origin which consists of carbon with more
concentration which can be converted to activated carbon using different types of a method such as chemical and
physical activation method, which affects the quality of the AC. For this reason, performed a pretreatment that
involved cutting, removing all of the pulp and some of the fibers, and air-drying the remaining shell. For uniform
quality in carbonization and activation, the water content of the shells was maintained at between 13% and 16%
[14]
Fig: 1 VOC removal process
296
2.2.1 Carbonization/ activation
Activated carbons can be produced by two different processes one is Physical or gas activation process and other
process is the Chemical activation. The choice of activation method solely depends upon the material used which
can have either a low or high-density carbon [13] After pretreatment the samples are placed in a high-temperature
kiln for carbonization by electric heating ,During carbonization, heating rate of 10 C/min and held for 2 hours [16]
Five activation temperatures 600, 700, 800, 900, and 1000∘ C [14] After carbonization, the samples were cooled to
room temperature under N2 flow (100cm3/min).[16] carbon is oxidized usually with the help of carbon dioxide or
steam at 800-1000 oC or and with low-temperature air flow, carbonization of primary raw material is followed by
the oxidation process [13].
2.2.1.1 Coconut shell, activated charcoal, Bamboo charcoal

The rapid development of industrial technologies increased the concentration of greenhouse gases such as VOCs,
CO2, NO2, N2, CH4, HFCs and because of this increasing global warming, raises ocean water levels,[14] Porous
carbon-based materials such as coconut shell, activated charcoal, bamboo charcoal, pecan shell and almond shell
based activated carbon has high thermal and chemical stability as well as good adsorption capabilities for such
harmful gases.
Phosphoric acid activated carbon almond shell having a surface area of 1340 m2/g and absorption of 347ppb.
coal based activated has the surface area 835m2/g and absorption is 476ppb whereas the steam-based activated
pecan shell has a surface of 917m2/g and absorption of 506ppb [7] coconut shell charcoal has least surface area
of about 36.5 m2/g. when chemically activated Coconut shell based activated carbon surface area increases up to
1768.8 m2/g [6] It can absorption is about 516ppb [7] but when modified or synthesis with tio2 nanoparticle
removal/absorption efficiency increase than the unmodified coconut shell based activated carbon [8] the removal
efficiency of the bamboo charcoal and activated carbon is 10% more than coconut shell based activated
carbon[9]
2.2.1.2 Granular activated carbon Fiber of polyparaphenylene terephthalamide and Fibers of

polymetaphenylene isophthalamide
Granular activated carbon is mainly used for the removal of harmful contaminants. It has a specific surface
area of 900 – 1200(m2/g) depending on the activation [10] In case of activated carbon fiber which may be of
different type of pitch-based activated, cellulose-based, pan-based ACF [17] The specific surface area varies
700-2500(m2/g) due to this the absorption of ACFs is more than that of the granular activated carbon. some
other derived of ACFs such as Nomex and Kevlar which have also specific area more than 1000(m2/g) [11] but
PAN has more surface area about 2400m2/g [17] also the so2 retention capacity of 266mgso2/gc which is more
than the Nomex (165.5 mgso2/g)and kevlar (192.5 mgso2/gc) making the PAN most efficient [11]
2.3 Filter based

Filter means to remove unwanted impurities from air or liquid. Filtration is the process by which the air or
liquid is getting filtered. Filtration can be done by several techniques such as using oxidation, catalysis,
regeneration and reveres reactor process. These techniques are mainly for the removal of impurities such as
VOCS , NO2 ,SO2, etc. filters mainly contain several activated carbons that is mention above such as ACF,
charcoal based activated carbon, granular activated carbon etc.
297
Fig 2 Simple process of filtration
3. Different types of filtering technique
Thermal oxidation is one of the techniques used for VOC removal and other contaminants in the air. Generally, the
temperature at which the oxidation takes place is 1300-1800ocF.the temperature also depends on the material used
and the destruction and removal efficiency.
Of higher removal efficiency value required more elevated temperature is required and also longer retention time
within the reactor.
The process mainly proceeds with the inflow of the air which is injected in the filter then it is heated to a certain
temperature. The thermal oxidizer is used for oxidation of the air so the VOC can be removed. Also, the heat is
recovered either in the cyclic process or it is stored in the heat storage system. The removal efficiency of this filter is
about 99% if the temperature lies between 1400o F to 2000o F [3]
Another major technique that is used is the catalysis for oxidation. The catalyst is mainly provided increases the
surface area for reaction and also for the reduction for the activation temperature. The are various process by which
the catalyst is used for the filtration process
The first common step is that the injection of impure air heated to a temperature of about 303 oF. Then its taken to
the heat exchanger for the for the further heating process it then is passed through the catalysis bed which is mainly
a honeycombed shaped which is coated with the catalyst this lead to the catalytic oxidation. another process where
the air is directly sent to the catalysis bed which is honeycombed shape and coated with tio2 then this is subjected to
UV light with help in photocatalysis and removal of harmful contaminant. The other process where the air and the
catalyst are directly injected into the filter where the photocatalysis takes place due to absorption of the light energy
of the catalyst particle where the temperature is also controlled according to the catalyst size. This reaction mainly
takes place in the solid-gas phase where the optimal particle size is about
12mm. The other techniques used is the where the catalyst is in a fluidized bed and there also photo catalysis takes
place at first the air is injected into the reactor where the fluidized base is present it is subjected to UV light for the
photocatalysis for the removal of the contaminant. The fluidized catalyst bed is mainly obtained by the impregnation
of the silica gel with sol-gel which makes the bed more fluidized which improves the removal efficiency [18]. There
are several another catalyst which is also used for the removal such as the Carbon Chem bituminous coal, envirotrol
bituminous coal, envirotrol coconut shell etc are some of the add-on the catalyst that is also used for the removal
contaminants from the air [12].
The removal efficiency of these photo catalysis techniques is about 95%. where the other techniques the efficiency
is about 90%. Another technique that is generally used is the reveres reactor. At first, the VOC leaden air enters the
filter then it is dried and is taken to a pressure regulated chamber then to adiabatically packed bed reactor where the
filtration process takes place. The feed flow direction is periodically changed. The main problem with the reactor is
that heat insulation which needs to be more than the packed bed. So in order to make it insulated it is vacuum
298
jacketed. There are several factors that affect the performance of RFR a)cycle period b)gas velocity c)adiabatic
temperature d)different component e)pressure etc. the main advantage of the reactor that no external heat/thermal
energy is required for the reactor to function.
Regeneration is another technique which is also used for the filtration process. This process is carried out by the
release of the N2 gas from the cylinder it gets purified by the gas purifier which contains silica gel and mol. Sieves
which is mainly used for moisture removal a d hydrocarbon. Then it is bubbled through VOC to bottle the gas is
then passed to the chamber containing the shell on one side and activated carbon fiber wrapped Teflon on other side
and a device was used to maintain the bed temperature also gas chromatography with FID was used for the test for
purification. Now this AFC can be can be regenerated completely by electrical heating in the presence of the
nitrogen gas which is the biggest advantage of this filter absorption of VOC or other contaminant doesn’t decree
With the regeneration process [2].
Table 1 Adsorption Efficiency of different activated carbon materials:
Material Adsorption Efficiency
Coconut shell charcoal 51.6%

Bamboo activated carbon 70.9%
Bamboo charcoal 72.3%
Phosphoric acid activated carbon 34.7%
Charcoal based 47.6%
Pecan shell 50.6%
Table 2 Effective size of carbon:
Material SURFACE AREA
Phosphoric acid activated carbon 1340 m2/g

Granular activated carbon 1050 m2/g
Charcoal based 835 m2/g
pecan shell 917 m2/g
activated carbon fiber 1600 m2/g
coconut shell charcoal 36.5 m2/g
299
80.00%
Adsorption efficiency(%)
70.00%
60.00%
50.00%
40.00%
30.00%
20.00%
10.00%
0.00%
Coconut shell Bamboo Bamboo charcoal Phosphoric acid Charcoal based Pecan shell
charcoal activated carbon activated carbon
Material
Fig 3 Adsorption efficiency (%)
1800
coconut shell charcoal
1600
1400 Phosphoric acid activated
Effective size (m2/g)
1200 carbon
1000 Granular activated carbon
800
Charcoal based
600
400 pecan shell
200
activated carbon fiber
0
Activated carbon materials
Fig 4 Effective surface area (m2/g)
300
CONCLUSION
Different Techniques of Activated Carbon Based Air Filters have been discussed in this study. Adsorption and
commercially available filters using activated carbon are efficient methods of air purification. Based on the study
following observations can be made-
1. Bamboo Based Activated Carbons Have Highest Adsorption Efficiency.

2. Activated Carbon Fibers Have Better Effective Surface Area Than Other Materials.
REFERENCES
[1] Gallego E., Roca F.J., Perales J.F., Guardino X., (2013) “Experimental evaluation of VOC removal efficiency
of a coconut shell activated carbon filter for indoor air quality enhancement” Building and Environment, 67 :14-25
[2] Das D., Gaur V., Verma N., (2004) “Removal of volatile organic compound by activated carbon fiber” Carbon
42: 2949–2962.
[3] Khan I.F., Ghoshal K.A., (2000) “Removal of Volatile Organic Compounds from polluted air” Journal of Loss
Prevention in the Process Industries 13: 527–545.
[4]. Shrimandilkar P.P., (March-April. 2013) “ Indoor Air Quality Monitoring For Human Health” International
Journal of Modern Engineering Research (IJMER) Vol.3, Issue.2, pp-891-897.
[5] Patnaik A., Kumar V., and Saha P., (2018). “Importance of Indoor Environmental Quality in Green Buildings",
Environmental Pollution, Water Science and Technology Library 77, doi.org/10.1007/978-981-10-5792-2_5
[6] Iqbaldin I.MN., Khudzir I.X., Azlan Mohd.MI., Zaidi AG., Surani, B., Zubri Z., (October 2013)
“PROPERTIES OF COCONUT SHELL ACTIVATED CARBON” Journal of Tropical Forest Science, Vol. 25,
No.4, pp. 497-503.
[7] Bansode R.R., Losso N.J., Marshall E.W., Rao R.M., Portier J.R., (2003) “Adsorption of volatile organic
compounds by pecan shelland almond shell-based granular activated carbons” Bioresource Technology 90 175–184.
[8] Hoang A.L., Le Thuy L.S.C., Jongsoo J., (2012) “Photocatalytic degradation of methylene blue by a combination
of TiO2-anatase and coconut shell activated carbon” Powder Technology 225 :167–175.
[9] Chuan C C., Ying P H., Wie C.W., Jun H. C., Yuan Y.W., (June 3, 2010 ) “Efficiency of Moso Bamboo
Charcoal and Activated Carbon for Adsorbing Radioactive Iodine” Clean – Soil, Air, Water 2011, 39 (2) :pp. 103–
108.
[10] Dwivedi P., Gaur V., Sharma A., Verma N., (2004) “Comparative study of removal of volatile organic
compounds by cryogenic condensation and adsorption by activated carbon fiber” Separation and Purification
Technology 39 : pp.23–37.
[11] J M., G M.C., and A B. F.,( 2000) “SOz Retention over Polyarybmide- Based Activated Carbon Fibers”
Environmental Progress (Vo1.19,N o.4) Winter.
301
[12] Dalai K.A., Cundall T.M and De M.,( AUGUST 2008) “Direct Oxidation of Hydrogen Sulphide to Sulphur
Using Impregnated Activated Carbon Catalysts” VOLUME 86.
[13] Mohammad K.A., and Ansari R.,( Oct-Dec 2009) “Activated Charcoal: Preparation, characterization and
Applications” Vol.1, No.4, pp 859-864.
[14] Pei H.H., Hao H.C., and Sheau H.L.,(5 February 2015) “Adsorption of Carbon Dioxide onto Activated Carbon
Prepared from Coconut Shells” Journal of Chemistry Volume 2015, Article ID 106590, 10 pages.
[15] Fatriansyah,F.J., Matari T., and Harjanto S.,( 2018-02-27) “The Preparation of Activated Carbon from Coconut
Shell Charcoal by Novel Mechano-Chemical Activation” Vol. 929, pp. 50-55.
[16] Yang K., Peng J., Srinivasakannan C., Zhang L., Xia H., Duan X.,( (2010) ) “Preparation of high surface area
activated carbon from coconut shells using microwave heating” Bioresource Technology 101: pp. 6163–6169.
[17] Youa Y.So., Parka H.Y., Parkb R.C., (2000) “Preparation and properties of activated carbon fabric from
acrylic fabric waste” Carbon 38 : pp. 1453–1460.
[18] Peral J,. Xavier D.n ., Ollis F.D.,( 1997) “Heterogeneous Photocatalysis for Purification, Decontamination and
Deodorization of Air” J. Chem. T echnol. Biotechnol., 70: pp. 117-140.
302
Paper ID- 73
A Comparative Study of Normal and Self Compacting Concrete

Deep Tripathi1, Rakesh Kumar2, P.K.Mehta3, Amrendra Singh4
1
deep@mnnit.ac.in, 2rkpat@mnnit.ac.in, 3pradeep11@mnnit.ac.in, 4amrendrasingh859@gmail.com,
CED, MNNIT Allahabad, Prayagraj, U. P., INDIA
ABSTRACT
In this research paper, the experimental results of both Normal Concrete (NC) and Self Compacting Concrete
(SCC) are presented. The fresh and hardened properties of both the concrete were investigated. The SCC was
developed by 20% incorporation of Fly ash (FA), so that the amount of Ordinary Portland Cement (OPC) is
almost same in both NC and SCC. The workability of the mix increases with the replacement of OPC by FA and
the mix becomes cost effective. For workability, the Slump test was conducted on NC while for SCC, Slump
flow, T50 time, V- funnel, L-box, U-box and J-ring tests were conducted. The water/binder (w/b) ratio was kept
constant for both the NC and SCC. The compressive, split tensile and flexural strengths of both NC and SCC
were found. For M25 grade concrete, cubes of size 100mm, cylinders of size 100mm×200mm, and beams of
size 100mm×100mm×500mm were cast to determine the compressive, split tensile and flexural strengths,
respectively. The samples were tested after curing in tap water at 7, 28, 56 and 90 days. The water absorption of
both NC and SCC were also studied up to 90 days. The XRD analysis of the both the concrete types were also
carried out for studying the micro structural changes.
Keywords: Self Compacting Concrete, Fly ash, Ordinary Portland Cement, X-Ray Diffraction
1. Introduction reinforcement with relative ease. In SCC, due

to the addition of pozzolanic materials the
All over the world, concrete plays a vital role cement consumption is less. This also helps in
in the field of construction. At present, NC is reducing the carbon-footprint because the
mostly used throughout the world; however, production of the cement is not environmental
some problems are faced during construction friendly and FA or some other pozzolanas are
viz., lack of skilled labour and concreting in industrial by-products. SCC can be produced
congested reinforcement areas etc. So to with easily / readily available materials and
speed up the construction and to get a durable superplasticiser. The other difference lies in
structure, a new type of, flowable concrete its design wherein the material proportion
was developed i.e., SCC. This is an varies [2]. FA is a waste material obtained
exceedingly flowable concrete and needs no from the hopper of chimney. Incorporation of
compaction. Generally, the SCC needs more FA in SCC enhances the flowability, strength
binder as compared to the NC. It is being used and durability with respect to control mix. It
successfully at construction sites. It differs is reported that with the inclusion of FA in the
from NC in material composition. It is SCC, its fresh properties were improved and
flowable under its own weight and able to it was found that there is less heat of
compact into congested reinforcement hydration due to which the cracking of
without the need of any additional vibration. concrete reduces [3]. Klug et al. [4] reported
SCC was firstly proposed by Prof. Okamura that for the same w/b ratio, the compressive
[1] in Japan, for construction to overcome the strength was same for both types of concrete,
shortage of skilled labour. The SCC offers and the split tensile strength of SCC was
many advantages over NC. The advantages better than NC however, the modulus of
include- increased speed of construction, cost elasticity of NC was higher than that of SCC.
savings due to lower labour requirement, At the same w/b ratio, the compressive
higher quality and aesthetically pleasing strength of SCC is at least 10% higher than
finishes, improved structural durability due to that of NC [5, 6]. Bertil Persson [7] reported
better compaction, low noise levels on site, that the creep, shrinkage and the modulus of
ability to pass through heavily congested
303
elasticity, of SCC was comparable with NC. 2.66 and 2.7, respectively and water
Ramanathan P et al. [8] reported that 30% absorption 1.0% and 0.9%, respectively,
replacement of OPC by FA in SSC showed conforming to IS 383-1987 were used [13].
improvement in properties as compared to the The bulk density of 10 mm and 20 mm
other replacement levels. It is also reported aggregates were 1590 kg/m3 and 1560 kg/m3,
that with increase in FA content the loss in respectively. The fineness modulus of fine
compressive strength was decreased and the aggregate was 2.7, and its value for the coarse
water absorption was decreased with aggregates of sizes 10 mm and 20 mm was 6.7
increasing percentage of FA. Gaywala et al. and 7.2, respectively. Class F- FA (procured
[9] reported that the compressive, split tensile, from NTPC Unchahar, U.P.) of grey colour
flexural and pull out strengths of SCC were having specific gravity 2.13, confirming to IS
maximum at 15% addition of FA in the mix. 3812-2000 [14] was used. A polycarboxylic
In SCC, with increase in FA content, the ether based Master Rheobuild 817RL
water absorption was increased and with superplasticizer having density approximately
curing period its value decreased [10]. It was 1.08 was used. Tap fresh potable water was
found that for the production of sustainable used in the mixture design and curing of the
SCC, FA is the most suitable pozzolana [11]. specimens. The chemical properties of the
X-Ray Diffraction (XRD) method is one of binders are presented in Table 1.
the techniques used for micro-structural
analysis. XRD is used to identify the Table 1. Chemical properties of the binders
compounds and minerals present in the
powdered specimen. The strength and S. Chemical OPC FA
permeability of both the SCC and NC No. Composition (%)
specimens were analysed by micro structural 1 Silicon dioxide 20.05 55.4
analysis. (SiO2)
2 Calcium oxide 61.95 4.19
In this study, M25 grade SCC and NC mix
(CaO)
were prepared using 43 grade OPC. Slump
3 Aluminum oxide 5.28 19.98
flow, T50 time, V-funnel, L- box, U- box and J-
(Al2O3)
ring experiments were performed on fresh
SCC. To check the workability of NC, Abrams 4 Iron oxide (Fe2O3) 4.12 6.82
cone apparatus was used. 100mm cubes, 5 Magnesium oxide 2.78 2.03
100mm×200mm cylinders and (MgO)
100mm×100mm×500mm beams of M25 grade 6 Potassium oxide 0.95 1.92
NC and SCC were cast to determine the (K2O)
compressive, tensile and flexure strengths, 7 Sodium oxide 0.24 0.61
respectively at 7, 28, 56 and 90 days curing. (Na2O)
8 Loss of ignition 3.12 2.41
Source : Manufactures’ data
2. Materials and their properties
3. Experimental Methodology
The OPC (Brand- M P Birla) of grade 43 was
used for this experimental programme. The 3.1 Mix Proportion
cement test results were: Normal consistency =
27%; Initial setting time = 45 minutes; Final M25 grade NC and SCC both were prepared,
setting time = 480 minutes; Specific gravity = using constant w/b ratio of 0.44. 20% of OPC
3.14; Compressive strength = 48.69 MPa (28 was replaced by FA for mix design of SCC.
days). The test values satisfy the provisions of The mix of NC was prepared as per the
IS: 8112-1989 [12]. The natural river sand, i.e provisions contained in IS: 10262-2009 [15].
fine aggregate of rounded shape was used The SCC mix was designed as per the
(conforms to zone II, as per prescribed in IS: EFNARC specifications [16]. The final mix
383-1987 [13]). It’s bulk density and specific proportions of both NC and SCC are presented
gravity were found to be 1680 kg/m3 and 2.65, in Table 2.
respectively. The Coarse aggregates (CA); 10
mm and 20 mm sizes, having specific gravity
304
Table 2. Mixture Proportioning of both NC Table 3. Test results of fresh SCC

and SCC
S. Tests Results EFNARC
3
S. Materials Quantity (kg/m ) No. limits [16]
No. NC SCC Min Max
1. Cement 345 356 1. Slump 690 mm 650 800
2. Fly ash -- 89 flow mm mm
3. Coarse Aggregate 1234 750 2. T50 time 4.0 sec 2 sec 5 sec
4. Fine Aggregate 686 890 3. V- 9.5 sec 6 sec 12 sec
5. Water 154 198 funnel
6. Water/binder 0.44 0.44 4. L-box 0.9 0.8 1.0
7. Superplastisizer -- 4.95 (h2/h1)
5 U-box 28 mm 0 30 mm
(h2-h1)
3.2 Water Absorption Test 6. J-ring 10 mm 0 10 mm
The water absorption test for all concrete
mixes was carried out to determine the change 4.2 Hardened Concrete Properties
in weight (%), as given by equation (1). If A is
the weight of concrete cube at 7, 28, 56 and 90 4.2.1 Compressive strength
days curing and B is the weight after
demoulding, The compressive strength of concrete cubes
𝐴−𝐵 were determined using compression testing
Water Absorption (%) = ×100 (1) machine (capacity 2000 kN) with loading rate
𝐵
of approximately 140 kg/cm2/min, as per the
provisions contained in IS: 516-1959 [17]. The
3.3 XRD Analysis
results of both NC and SCC are presented in
The behaviour of every material is related to Fig. 1.
its microstructure. Micro structure analysis is
the best way to decide the performance of
60 NC Compressive strength (MPa)
concrete because it controls the properties of
Compressive strength (MPa)
SCC Compressive strength (MPa)

concrete and its behaviour of concrete. XRD 50
analysis is one of the methods to find the
presence of mineral compounds in the concrete 40
specimen.
30
20
4. Results and Discussions 10
4.1 Fresh concrete Properties 0
7 days 28 days 56 days 90 days
The slump of 25 grade NC was 100mm. For
SCC, Slump flow, T50 time, L-box, U-box and Curing Period (MPa)
J- ring tests were performed. The Slump flow
test was performed to check the flowability of
SCC. The diameter of the SCC circle was Fig.1 Compressive strength of the
measured in two orthogonal directions as per specimens
the provisions contained in EFNARC [16]. T50 It was found that at early ages, the
time is the time taken by concrete to reach 500 compressive strength of SCC was lower in
mm diameter a diameter. To check flowability comparison to NC but at later ages it was
/ filling ability of the SCC, V-Funnel test was higher for SCC. At 7 days, the compressive
carried out. L-box, U-box and J-ring tests were strength of SCC is 9.9% lower than the NC.
performed to check passing ability of SCC. However, at 28, 56 and 90 days, the
The results are shown in Table 3. compressive strength of NC was lower by 6.0,
7.9 and 10.1%, respectively. Several other
researchers have also reported the same trend
305
[5, 6]. The strength gain may be attributed to

the formation of additional C-S-H gel in FA 8 NC Flexure strength (MPa)
mixed SCC and refined pore structure.
7 SCC Flexure strength (MPa)
Flexure strength (MPa)

6
4.2.2 Split tensile strength 5
4
Split tensile strength of both The NC and SCC
were determined by casting cylinders of size 3
100mm×200mm, as per the provisions of IS: 2
5816-1999 [18]. The results are presented in 1
Fig. 2.
0
3.5 NC Split tensile strength (MPa) Curing Period (days)
Split tensile Strength (MPa)
SCC Split tensile strength (MPa)

3
Fig. 3 Flexure strength of the specimens
2.5
2
The flexure strength of NC, at early ages was
more than the SCC; however, beyond that the
1.5 values increased. At 7 days, the flexure
1 strength of SCC was 5.4% lower than the NC.
However, it increased by 3.7, 2.0 and 1.7% at
0.5 28, 56 and 90 days, respectively.
0
Curing Priod (days) 4.3 Water Absorption
Durability of the concrete is affected by water
Fig. 2 Split tensile strength of the specimens
absorption because it is the indicator of
At 7 days, the Split tensile strength of SCC permeability of concrete. Thus, it is important
was found to be lower than the NC; however, to find the absorption of concrete specimens.
beyond that it was found to increase. At 7 The typical values of water absorption are
days, the split tensile strength of SCC was shown in Fig. 4.
8.5% lower than the NC. However, at 28, 56
and 90 days, the tensile strength was increased
3 NC water absorption (%)
by 2.8, 2.6 and 1.7%, respectively. The same
Water Absorption (%)
trend was observed by Klug et al. [4]. 2.5 SCC Water absorption (%)
2
1.5
4.2.3 Flexure strength
1
To determine the flexural strength of both NC
0.5
and SCC specimens, beams of size
100mm×100mm×500mm were cast. Beams 0
were tested, using two point method using 7 days 28 days 56 days 90 days
loading frame of capacity 500 kN at a loading Curing Period (days)
rate of 180 Kg/min, without any shock. The
cubes were tested after curing in tap water at 7, Fig. 4 Water absorption of NC and SCC
28, 56 and 90 days, as per the provisions of IS:
516-1959 [17]. The results are presented in It is evident from Fig. 4 that water absorption
Fig. 3. of NC was more than SCC and with time its
value decreases for both NC and SCC. FA
acts as a filler material and reduces the water
absorption. The water absorption of SCC was
306
21.2, 15.9, 13.5% and 10.8% lower than the More Ettringite is formed in NC in comparison
corresponding NC samples at 7, 28, 56 and 90 to SCC.
days, respectively. The same trend was also
observed by others [10].
5. Conclusion
4.4 XRD Analysis:
The following are concluded:
XRD analysis of both the samples NC and
SCC were carried out after 28 days curing.  The strength of SCC is better than the
The typical XRD traces are presented in Fig. 5 NC, except at the age of 7 days.
(a) and (b). Some major crystalline phases  At the same w/b ratio, the SCC is superior
were identified, i.e. Quartz (Q), Calcium and economical.
Silicate Hydrate (C-S-H), Portlandite (CH) and  Water absorption of SCC is lower.
Ettringite (e).  Microstructure is improved in SCC.
 Less Ettringite is formed in SCC.
 Higher amount of C-S-H gel is present in
SCC.
References
1. Okamura H., Maekawa K., Ozawa K.:
High-Performance Concrete. Gihodo
Publishing, (1993).
2. Okamura, H. and Ouchi, M.: Self-
Compacting Concrete. Journal of
Advanced Concrete Technology 1(1),
5-15, Japan (2003).
3. Kurita M. and Nomura T.: Highly-
flowable steel fiber-reinforced
a) NC concrete containing fly ash. Special
Publication 178, 159-176 (1998).
4. Klug, Y., Holschemacher, K.,
Wallevik, O. and Nielsson, I.:
Comparison of the Hardened
Properties of Self-Compacting and
Normal Vibrated Concrete. Paper
presented at 3rd RILEM Symposium
on Self Compacting Concrete,
Reykjavik (2003).
5. Bennek, W.: SCC Applied in the
Precast Industry (ICCX). :24-24,26,27
(2007)
6. Van Keulen, D. C, Onderzoek Naar
Eigenschappen Van Zelfverdichtend
Beton. Rapport TUE/BCO/00.07, TU
b) SCC Delft, April (2000).
7. Bertil Persson: A comparison between
mechanical properties of self-
Fig. 5 XRD traces after 28 days curing compacting concrete and the
corresponding properties of normal
(a) NC and (b) SCC concrete. Cement and Concrete
From the figures, it is clear that a higher peak Research, 31, pp.193-198 (2001).
of C-S-H is present in SCC than the NC. This 8. Dhiyaneshwaran S, Ramanathan P,
improves the durability and strength of SCC. Baskar I, Venkatasubramani R.: Study
307
on durability characteristics of self-

compacting concrete with fly ash.
Jordan Journal of Civil
Engineering.7(3), pp.342-52 (2013).
9. J.M. Khatib.: Performance of self-
compacting concrete containing fly
ash. Construction and Building
Materials, 22, pp. 1963–1971 (2008).
10. Gaywala NR, Raijiwala DB.: Self
compacting concrete: A concrete of
next decade. Journal of Engineering
Research & Studies. 2 (4), (2011).
11. Mohammed MK, Dawson AR, Thom
NH.: Production, microstructure and
hydration of sustainable self-
compacting concrete with different
types of filler. Construction and
Building Materials. 1 (49), 84-92
(2013).
12. IS: 8112-1989.: Specification for 43
grade ordinary Portland cement.
Bureau of Indian standards, New
Delhi, India.
13. IS: 383-1987.: Specification for coarse
and fine aggregate from natural
sources for concrete. Bureau of Indian
Standards, New Delhi, India.
14. IS: 3812- 2000.: Specification for
Pulverized Fuel Ash, Part 1: For Use
as Pozzolana in Cement. Cement
Mortar and Concrete.
15. IS: 10262-2009. Concrete mix
proportioning- Guidelines. Bureau of
Indian Standards, New Delhi, India.
16. EFNARC-2002. Guidelines for Self-
Compacting Concrete. London, UK:
Association House, 32-34.
17. IS: 516-1959. Methods of Tests for
Strength of Concrete.
18. IS: 5816-1999. Splitting Tensile
Strength of Concrete-Method of Test.
308
Paper ID- 74
Effect of Acidic Environment on Durability of RHA Concrete

Amrendra Singh1, Rakesh Kumar2, P.K.Mehta3, Deep Tripathi4
1
amrendrasingh859@gmail.com, 2rkpat@mnnit.ac.in, 3pradeep11@mnnit.ac.in, 4deep@mnnit.ac.in,
Department of Civil Engineering, MNNIT Allahabad, Prayagraj, U. P. India
ABSTRACT
Concrete is globally used construction material and its durability is the main concern. Acid is one of the
chemicals influencing the durability of concrete. Nitric acid (HNO3) is one of the acids influencing the
durability of concrete. Several mineral admixtures are used to improve the durability of concrete. Rice Husk
Ash (RHA) is one of such admixtures. This paper presents the results of an experimental programme to
investigate and compare the durability of Normal Concrete (NC) and RHA Aided Concrete (RHAC). M25 grade
of concrete using normal constituents was designed and used for referral. The RHA was used for partial
replacement of Ordinary Portland Cement (OPC). The different replacement levels considered were 10, 15, 20
and 25%. It is found that workability of concrete increases with increase in replacement level. The optimum
replacement level was 15%, with respect to the compressive strength. The cubes of both NC and RHAC were
cured in Tap water as well as in 5% Nitric acid solution till 90 days. The compressive strength of cubes was
determined after 7, 28, 56, and 90 days curing. The compressive strength of water cured RHAC is higher than
the NC. The compressive strength of both NC and RHAC decreases in acidic solution for all the exposure
periods; however, the decrease is lower in case of RHAC.
Keyword: Rice husk ash, Normal concrete, Ordinary Portland Cement
1 Introduction need to search for alternative materials which

may be used in place of cement or may be
Concrete is a man made construction material used to partially replace cement, without
used in construction industry. It is widely used adversely affecting the strength and durability.
for different types of construction works. It is reported that fly ash(FA), ground
Concrete is a appropriate mixture of granulated blast furnace slag(GGBS), RHA,
cementitious material, fine aggregate, coarse metakaolin(MK), silica fume(SF) etc are some
aggregate and water. Admixtures (mineral of the pozzolanic materials which could be
and/or chemical) are sometimes introduced in used for partial replacement of cement in
concrete to improve the characteristics of the concrete.
fresh concrete, the mixing process and/or the
final hardened material. The hardening is Durability of concrete depends on several
because of the chemical reaction between factors viz., the properties of its ingredients,
water and cementitious material which mix proportion, method of compaction and
continues for a long period and concrete other controls during placing and curing,
becomes stronger with age. But when concrete exposure to aggressive environment etc. If a
is exposed to aggressive environment, its concrete structure is exposed to acidic
durability may get adversely affected. Earlier, environment there is a high chance of damage
the strength was the main parameter of to the structure. Generally, concrete mix is
concrete without much concern for its designed to withstand such a harmful
durability. But at present, the durability a has aggressive exposure but if it is subjected to
become a major concern. such exposure for a long time, there may be a
Cement is the primary ingredient in concrete chance of failure of the structure.
which is produced artificially. During cement
production, large amount of Carbon dioxide is
emitted in the atmosphere. Cement industry Ismail et al. [1] investigated the usefulness of
accounts for around 5% of global carbon RHA (where in an appreciable amount of
dioxide (CO2) emissions. Due to higher cost of silica is present) as part re-placement of
cement and to reduce greenhouse gas cement and reported that it improved the
emissions from cement industry, there is a properties of RHAC. The concrete
309
incorporating 10% of the RHA as an OPC be 6.25 and 1.2%, while for 20mm aggregate
replacement showed some what higher these value were 7.27 and 1.1 %, respectively.
compressive strength and higher resistance The RHA was procured from KGR AGRO
against chloride-ion penetration, compared Private Ltd. Ludhiyana, Punjab. It was black
with the NC of the same water-to-cementitious coloured with specific gravity 2.15 and specific
materials ratio [2]. The finer fractions of RHA area 150000-180000 cm2/gm. The chemical
resulted in improvement of all properties of composition of RHA is included in Table-1 The
mixtures, with exception of final setting time HNO3 used for creating the acidic environment
[3]. The residual RHA provides a positive was of brand RENKEM and was procured from
effect on the compressive strength of concretes The Scientific Traders, Civil lines, Allahabad.
at early ages, but in the long term, the The specific gravity of nitric acid was 1.41 and
behaviour of the concrete containing RHA, normality was 72 %.
produced by controlled incineration, was more
significant [4]. A 30% replacement of OPC by Table 1. Chemical Compositions of OPC and
reburnt RHA leads to a substantial RHA
improvement in the permeability properties of Sl. Chemical compositions OPC RHA
blended concrete, compared to that of No (%)
unblended OPC concrete [5]. Ige et al[6] .
studied the effect of nitric acid on laterite 1. Silicon dioxide (SiO2) 19.45 96.7
concrete and reported that the compressive 2. Aluminum oxide (Al2O3) 4.64 1.01
strength reduces significantly with increase in 3. Iron oxide (Fe2O3) 3.23 0.05
acid concentration, immersion period and
4. Calcium oxide (CaO) 52.55 0.49
laterite content. The effect of richness of mix
5. Magnesium oxide (MgO) 2.15 0.19
on the resistance of laterized concrete against
6. Sodium oxide (Na2O) 0.22 0.16
acidic aggression becomes more pronounced
at the highest laterite content (50%). 7. Potassium oxide (K2O) 0.76 0.91
In this paper, the results of an experimental 8. Loss on ignition 2.76 4.81
programme aimed at investigating the effect of
5% HNO3 solution on RHA incorporated 3. METHODOLOGY
concrete in fresh and hardened states are
presented. 3.1 Mix proportioning
The concrete mix design (M25) was carried out
2. Materials and Their Properties
as per the procedure given in IS: 10262-2009
[9]. The mix proportion was-cement: FA:
OPC of Grade 43 (brand-jaypee) was used in
this experiment. The physical properties of CA::1: 1.31: 2.46 with water content of 186
litre/m3. The water/cement ratio was 0.40.
OPC was determined as per IS 8112-1989 [7].
Concrete cubes of size 150mm were cast to find
Normal consistency, initial setting time, final
the compressive strength of NC after at 7 and
setting time and compressive strength of cement
28 days of curing. 100mm cubes were cast for
were 31%, 92 minutes and 240 minutes and
52.50 N/mm2, respectively. The chemical both the NC and RHAC mix for further study.
These samples were submerged in tap water as
composition of OPC is presented in Table-
well as in 5% HNO3 solution for curing periods
1.The natural river sand, conforming to zone II
of 7, 28, 56 and 90 days.
as per IS-383-1987 [8], was used for making
normal concrete (NC). Its fineness modulus,
specific gravity and bulk density values were
found to be 2.492, 2.48 and 1680 kg/m3, 4. RESULTS AND DISCUSSIONS
respectively. Coarse aggregate (CA) of size 10
Cubes of size 150mm were cast for NC and
and 20 mm conforming to IS: 383-1987 [8] was
tested for compressive strength after curing of
used. The bulk density of 10 mm and 20 mm
7 and 28 days. After confirming the mix
aggregate was 1590 kg/m3 and 1560 kg/m3,
proportion of the NC, the optimum level of
respectively. The specific gravity of 10 and 20
OPC replacement by RHA was determined.
mm aggregates was found to be 2.67 and 2.7,
The OPC was replaced in part with RHA (10,
respectively. The fineness modulus and water
15, 20 and 25% on equal mass basis) to get the
absorption of 10 mm aggregates were found to
310
optimum replacement level with respect to composition of RHA. The maximum gain in
compressive strength. The optimum compressive strength is 9.37 and 6.44% at 7
replacement levels were determined at both 7 and 28 days, respectively. This gain in strength
and 28 days. After obtaining the optimum is primarily due to the formation of C-S-H gel
replacement level, cubes of size 100mm were in RHAC.
cast for both the NC and RHAC. After
Cmpressive Strength (MPa)

demoulding, the specimens were immersed 7 Day Compressive
separately in tap water and 5% HNO3 for the Strength
45 28 Day Compressive
period of 7, 28, 56 and 90 days.
40 Strength
4.1 Properties of Fresh Concrete
35
The results of the RHA inclusion on the 30
workability of concrete is shown in Fig-1.
25
The slump was recorded at different
percentages (0%, 10%, 15%, 20%, and 25%) 20
of RHA content. The workability of concrete 15
mixes is found to increase with RHA. The 0 10 15 20 25
slump of concrete mix was increased by 31 Rice Husk Ash (%)
mm when the replacement level was changed
from 0% to 25%.
Fig 2. Variation of 7 and 28 day compressive
strength at different replacement levels
110
4.2.2 Compressive Strength
Slump (mm)
100
90 The compressive strength of different types
80 of concrete specimens, cured at 7, 28, 56, 90
70 days were determined as per provisions
60 prescribed in IS: 516-1959 [10] and results
50 are shown in Fig-3. It is observed from this
figure that the concrete specimens cured in
5% Nitric Acid solution had lower
compressive strengths than those cured in tap
Rice Husk Ash(%)
water. It is also observed that the inclusion of
RHA in concrete enhanced the compressive
Fig 1. Variation in workability of Concrete at strength in comparison to the NC at all the
different replacement levels ages. The loss in compressive strength of NC
after 7, 28 ,56 and 90 days of Nitric acid
exposure is 15.61, 11.00, 9.23 and 10.94%,
4.2 Properties of Hardened Concrete respectively while for RHAC the above
change is 11.44, 12.92, 11.20 and 12.20%,
4.2.1 Optimization of RHA inclusion respectively.
Compressive strength test results for 7 and 28
days with varying percentages of RHA in
concrete are illustrated in Fig-2. In this study,
the replacement level of OPC with RHA was
varied from 0 to 25%, at an interval of 5%. It
is evident from this figure that the compressive
strength increases with the RHA content upto
15%, and after that it decreases. Thus the
optimum replacement level of OPC by RHA is
15%. The results support the findings of Ismail
and Waliuddint [1]; however, they reported
optimum replacement level as 10 %. The
difference may be due to different chemical
311
Cement and Concrete Research. Vol.

50 26, No. 6, pp. 963-977, (1996).
45 3. Harish Kizhakkumodom
Compressive Strength(MPa)
Venkatanarayanan, Prasada Rao

40 Rangaraju.:Effect of grinding of low-
35 carbon rice husk ash on the
30 microstructure and performance
25 in Tap Water(RC) properties of blended cement concrete.
in Nitric Acid(RC) Cement & Concrete Composites 55
20 348–363, (2015).
15 in Tap Water(RHAC)
in Nitric Acid(RHAC) 4. Kittipong Kunchariyakun , Suwimol
10 Asavapisit & Kwannate
5 Sombatsompop.: Properties of
autoclaved aerated concrete
0 incorporating rice husk ash as partial
7 Day 28 Day 56 Day 90 Day replacement for fine aggregate. Cement
& Concrete Composites 55, 11–16,
Age of Concrete
(2015).
Fig 3. Compressive Strength of different types 5. Bui D.D., Hu J. and Stroeven P.:
of concrete in Water & 5% Nitric acid. Particle size effect on the strength of
rice husk ash blended gap-graded
CONCLUSION Portland cement concrete. Cement &
Concrete Composites 27, 357–366,
 Workability of concrete increases with
(2005).
increase in replacement level of OPC by 6. Olubisi Ige, Stephanie Barnett, John
RHA. Chiverton, Ayman Nassif and John
 The compressive strength of water cured Williams.: Effects of steel fibre-
RHAC is higher than the NC. aggregate interaction on mechanical
 The optimum replacement level of OPC behaviour of steel fibre reinforced
by RHA in concrete, on equal weight concrete. Cement and Concrete Science,
basis, is 15%. Volume 116, Issue 4, (2017).
 At optimum replacement level, the 7. IS: 8112 (1989), “Specification for 43
compressive strength gain is about 7% grade ordinary Portland cement.”
at 28 days. Bureau of Indian standards, New Delhi,
 The compressive strength of all the India, (1989).
concrete mixtures reduces in acidic 8. IS: 383 (1987), “Specification for coarse
environment at all the exposure periods. and fine aggregate from natural sources
for concrete.” New Delhi: Bureau of
Indian Standards (1987).
9. IS: 10262-2009. Concrete mix
proportioning- Guidelines. Bureau of
References Indian Standards, New Delhi, India.
10. IS: 516–1959. Methods of Tests for
Strength of Concrete.
1. Muhammad Shoaib Ismail and
Waliuddint A. M..: Effect of rice husk
ash on high strength concrete.
Construction and Building Materials,
Vol. 10, No. I, pp. 521-526, (1996).
2. Zhang M. H., Lastra R., and Malhotra
V. M.: Rice-husk ash paste and
concrete: some aspects of hydration and
the microstructure of the interfacial zone
between the aggregate and paste.
312
Paper ID- 76
ALKALI-ACTIVATED CONCRETE – A REVIEW
Chitra Shijagurumayum1* and Thokchom Suresh2

1
Research Scholar and 2Associate Professor
Department of Civil Engineering, Manipur Institute of Technology, Imphal – 795001, Manipur, India
1,2
1
chitras10shij@gmail.com 2thoks1966@gmail.com
Abstract:
This review paper discusses the properties of alkali activated concrete (AAC) in which certain proportion
of fly ash is substituted by ground granulated blast furnace slag (GGBS), investigations to improve the
compressive strength of alkali activated concretes, morphology of structures when exposed to elevated
temperatures and the effect of GGBS on the setting time of such concrete.
AAC is manufactured by activating a source material rich in silica and alumina by strong alkaline liquid.
Fly ash is the most commonly used source material in production of geopolymer though other works such
as granulated blast furnace slag, silica fume, may also be successfully used as steering material. With the
production of alkali activated concretes, fly ash is no longer considered as a nuisance but a highly
promising raw material. Such alkali activated binders will be able to solve both the problems of
environmental threats and non-utilization of nonrenewable resources. They are also found to be more
durable under adverse conditions such as to exposure in acidic and sulphate environment.
Key words: Alkali activated concretes, Fly ash, Ground granulated blast furnace slag and Sulphate attack
1. Introduction
The production of Portland cement, a main component of making concrete, contributes
significant amount of greenhouse gas. The production of cement is increasing by about 3%
annually (McCaffrey 2002). The production of one ton of cement liberates about one ton of
CO2 to the atmosphere, as a result of de-carbonation of limestone in the kiln and the
combustion of fossil fuels (Roy 1999). The contribution of Portland cement production
worldwide to the greenhouse gas emission is estimated to be about 1.35 billion tons annually
or about 7% of the total greenhouse gas emissions to the earth’s atmosphere (Malhotra 2002).
Cement is also among the most energy-intensive construction materials, after aluminum and
steel. Furthermore, it has been reported that the durability of ordinary Portland cement (OPC)
concrete is under examination, as many concrete structures, especially those built in
corrosive environments, start to deteriorate after 20 to 30 years, even though they have been
designed for more than 50 years of service life (Mehta and Burrows2001).
In order to produce environmentally friendly concrete, Mehta (2002) suggested the use of
fewer natural resources, less energy, and minimize carbon dioxide emissions. He categorized
these short-term efforts as ‘industrial ecology’. The long-term goal of reducing the impact of
unwanted by-products of industry can be attained by lowering the rate of material
consumption. Above this, human activities on earth produce solid wastes in considerable
quantities of over 2500/ MT per year, including industrial wastes, agricultural wastes and
wastes from rural and urban societies. Recent technologies have shown that these materials
are valuable as inorganic and organic resources and can produce various useful products.
313
Amongst the solid wastes, the most prominent ones are fly ash, blast furnace slag, rice husk
(converted into ash), silica fume and demolished construction materials.
In 1978, Davidovits proposed that an alkaline liquid could be used to react with the silicon
(Si) and the aluminum (Al) in a source material of geological origin or in by-product
materials such as fly ash and rice husk ash to produce binders. As the chemical reaction that
takes place in this case is a polymerization process, Davidovits (1994, 1999) coined the term
‘Geopolymer’ to represent these binders. The estimated annual production of coal ash in
1998 was more than 390 million tons worldwide. The main contributors for this amount were
China and India. Only about 14 percent of this fly ash was utilized, while the rest was
disposed in landfills (Malhotra 1999). Prior to 1970, Rice husk ash (RHA) was usually
produced by uncontrolled combustion and the ash so produced was crystalline and possessed
poor pozzolanic properties. The utilization of RHA as a pozzolanic material in cement and
concrete provides several advantages, such as improved strength and durability properties,
reduced materials cost due to cement savings and environmental benefits related to the
disposal of waste materials.
2. Types of binders to produce AAC
The binders used to produced AAC may of the following types or a mixture of the following:
2.1. Fly Ash

Class F fly ash is activated using solutions of sodium silicate and caustic hydroxides.
Concretes made using fly ash as the pozzolanic material requires to be cured at a temperature
of 25 0C or more.
2.2. Ground Granulated Blast Furnace Slag
GGBS too just like fly ash is activated using a mixture of sodium hydroxide either in pellets
or flakes or powder form and sodium silicate solution. Water content is fixed as per the
desired water to binder ratio. The main advantage of using GGBS is that the concrete so
produced does not require thermal curing. So ambient curing in turn proves to be economical
too.
2.3 RICE HUSK ASH
Rice husk ash is proven to contain more amount of SiO2 which is as high as 85% and
comparatively lower percentage of Al2O3. In order to produce a binding agent, rice husk ash
is also activated using a mixture of sodium hydroxide, sodium silicate solution and water.
2.4 RED MUD
Red mud, a residue from Bayer’s process has also been successfully replaced certain
percentages of fly ash to produce binders. As in the previous cases, the activators are alkaline
solution of calcium silicate, sodium hydroxide and water.
2.5 SILLICA FUME
Addition of silica fume to calcined clay has also been able to successfully produce alkali
activated binders. Alkaline hydroxide pellets, potassium silicate and water were used as
activators.
3. AAC MANUFACTURING PROCESS
314
Activating solution is a mixture of sodium hydroxide, sodium silicate and water. The alkaline
hydroxide maybe in the form of flakes, powder or pellets. In certain studies, potassium or
calcium silicate is used in place of sodium silicate. The liquid to binder(solid) ratio is kept
between 0.3 to 0.45. Coarse aggregate passing 4.75mm sieve but retained on 19.00mm sieve are
used. For fine aggregates the sieves used were in the range of 150µm to 2.36mm. The coarse and
fine aggregates together took up 75% to 80% of the whole mixture in terms of mass just like the
case of OPC. To make the concrete more workable, superplasticizer such as naphthalene
sulphonate are used. In almost all the papers under review, the molarity of NaOH solution was in
the range of 8M to 16M.
It was also found out that the activating solution mix should be mixed one day prior to mixing
with the raw materials. After the concrete has been filled in the mould, it is preferable to cover
the moulds with plastic bags as most of geopolymer concrete specimens are to be cured in an
elevated temperature at around 800C. Thus, excessive evaporation could be prevented. However,
in the case of GGBS, elevated temperature curing is not required.
4.0 CHEMICAL STRUCTURE
The reaction of alkaline solution of sodium hydroxide and sodium silicate with fly ash is a
polymerisation process and gives a ring structure. (Davidovits 1999)
Mn [-(SiO2) z–AlO2] n . wH2O ; where: M = the alkaline element or cation such as potassium,
sodium or calcium; the symbol – indicates the presence of a bond, n is the degree of
polycondensation or polymerisation; z is1,2,3, or higher, up to 32.
The chemical reaction may comprise the following steps (Davidovits 1999; Xu and van Deventer
2000):
() Dissolution of Si and Al atoms from the source material through the action of hydroxide ions.
() Transportation or orientation or condensation of precursor ions into monomers.
() Setting or polycondensation/polymerisation of monomers into polymeric structures.
However, these three steps can overlap with each other and occur almost simultaneously, thus
making it difficult to isolate and examine each of them separately (Palomo et al. 1999).
Portland cement when used with GGBS, the hydration product of calcium hydroxide leads to the
formation of low calcium hydroxide to silicon dioxide ratio. In the case of using rice husk ash
burned in a controlled combustion, a strong pozzolanic material is formed which reacts quickly
with calcium silicate hydrate.
5.0 PROPERTIES AND PERFORMANCE
5.1 FRESH PROPERTIES
When fly ash is used as a base, the slump value increases with addition of extra water. To
achieve higher degree of workability, plasticizer based on naphthalene sulphonate has been
315
successfully used. However, it was observed that using plasticizer more than 2% of the binder,
leads to a decrease in compressive strength.
(D.D. Higgins, 2003) The AAC produced by replacing certain amounts of Portland cement with
GGBS were vibrated into cube moulds followed by a covering of the moulds by a damp cloth
(hessian and sheets of polythene) for 24 hours. It may be followed by water curing or air curing
at a relative humidity of 65%. Thus giving a rest period of 1 day. In some researches the rest
period up to 5 days was found to increase the compressive strength of fly ash based AAC.
Significant increase in compressive strength were observed for a rest period up to 3 days.
It was also seen that fly ash based AAC shows no sign of setting up to 120 minutes so it can be
handled effortlessly during the first 120 minutes.
5.1 HARDENED PROPERTIES OF AAC
5.1.1 DRYING SHRINKAGE
The shrinkage strain measurements started on the third day after casting the concrete. On the
third day after casting, the specimens were demoulded and the first measurement was taken.
Horizontal length comparator was used for length measurements. The next measurement was on
the fourth day of casting, considered as Day 1 for the drying shrinkage measurements. The
measurements then continued every day in the first week, once a week until the fourth week,
once in two weeks until the twelfth week, and then once in four weeks until one year. During the
drying shrinkage tests, the specimens were kept in a laboratory room where the temperature was
maintained at approximately at 230C. The relative humidity of the room varied between 40% and
60%. The creep coefficient, defined as the ratio of creep strain-to-instantaneous strain, after one
year for heat-cured geopolymer concrete with compressive strength of 40, 47 and 57 MPa is
around 0.6 to 0.7; for geopolymer concrete with compressive strength of 67 MPa this value is
around 0.4 to 0.5.
The heat-cured fly ash-based geopolymer concrete undergoes very little drying shrinkage in the
order of about 100 micro strains after one year. The drying shrinkage strain of ambient-cured
specimens is in the order of 1500 micro strains after three months. This value is many folds
larger than that of heat cured specimens, and the most part of that occurs during the first few
weeks.
5.1.2 MECHANICAL STRENGTH
For determination of compressive strength of AAC, cylinders of size 100mmx200mm and

150mmx300mm were used. There were no substantial gain in the compressive strength of heat-
cured fly ash based geopolymer concrete with age. In the case of cube specimens of size
20mmx20mmx20mm, hydrothermal curing yielded higher compressive strength compared to dry
curing. Longer mixing time of up to sixteen minutes increases the compressive strength of fly
ash-based geopolymer concrete.
316
In another compressive strength test on AAC using cylinders it was reported that 14 Molar
concentration of sodium hydroxide gives higher compressive strength values to those of 8 Molar
concentrations.
5.1.3 TEST FOR DURABILITY
Different types of solutions of concentrations usually ranging from 5% to 20% were used to test
for durability. Durability test is done was measuring the resistance of AAC to acids. The change
in mass, dimensions and compressive strength at saturated surface dry state were determined.
The test specimens of cubes and cylinders were immersed in desired acids at required
concentrations. On 7th and 28th day, the specimens were taken out and mass and dimensions were
measured. Lastly, the test for compressive strength were carried out.
For simulation of acid attack in sewer pipes, sulphuric acid is one of the most commonly used
acid. Some researches using sulphuric acid were reported to used acid concentration ranging
from 0.25%, 1% to 2%. For test against resistance to acid, the solution needs to be stirred every
week and also needs to be changed every month. It was found out that exposure to sulfuric acid
solution damages the surface of heat-cured geopolymer concrete test specimens and causes a
mass loss of about 3% after one year of exposure. The severity of the damage depends on the
acid concentration. The tests on heat-cured geopolymer mortar specimens indicate that the
degradation in the compressive strength due to sulfuric acid attack is mainly due to the
degradation in the geopolymer matrix rather than the aggregates. The degradation in compressive
strength of mortar specimens is larger than that of concrete specimens due to the larger
geopolymer matrix content by mass of mortar specimens.
7.0 RECENT RESEARCHES AND FUTURE SCOPE
Even though several studies have been carried out by many researchers in the field of alkali activated
binders, very few researches has been performed in the area of durability and blending of raw materials.
A lot more research in this useful field is required to fully understand the mechanism of strength
development and other properties.
8.0 APPLICATIONS OF AAC
Earlier in 1997, AAC were used to strengthened concrete structures and also as a coating to
transportation structures as a means of protection. It was reported in 2004 that railway concrete
sleepers made from fly ash based AAC and was reported to have very low drying shrinkage. The
sleepers also gave remarkable performance.
9.0 CONCLUSION
The review study bring forth the following outcomes with concern to AAC:
317
(i) Alkali activated fly ash specimen yield higher strength when it is covered and subjected to
thermal curing and direct curing is suggested only for NaOH based system.
(ii) A mixture of fly ash and slag at 50:50 with 10 molar NaOH and cured at 25degrees result in
compressive strength of 50 MPa.
(iii) Specimen prepared by blending 70% GGBS and 30% Portland cement gave high resistance to
sulphate attack as compared to suphate resisting portland cement.
(iv) Incorporating GGBS in the binder blend removes the process of thermal curing which solves
the problem of thermal curing for mass construction in the field.
(v) Increasing percentage of GGBS increases the compressive strength of alkali activated
concretes but it also leads to early setting which may not be desirable in certain situations.
(vi) Setting time is also affected by the molarity of sodium hydroxide. Higher molarity results in a
decrease of setting time of AAC.
(vii) Compressive strength is also directly proportional to molarity of sodium hydroxide and the
rate of gaining of strength is also directly proportional to the content of fly ash.
References:
1. Alaa M. Rashad and Dina M. Sadek “An investigation on Portland cement replaced by high-
volume GGBS pastes modified with micro-sized metakaolin subjected to elevated temperatures”,
International Journal of Sustainable built environment (2017) 6, pp.91-101
2. D.D Higgins and N.J. Crammond “Resistance of concrete containing ggbs to the thaumasite form
of sulphate attack”, Cement and concrete Composites 25(2003) pp.921-929
3. Rami A. Hawileh et.al “Performance of reinforced concrete beams cast with different percentages
of GGBS replacement to cement”, Archives of civil and mechanical engineering 17 (2017)
pp.511-519.
4. K. Ganesh Babu and V. Sree Rama Kumar “Efficiency of GGBS in concrete”, Cement and
Concrete Research 30 (2009) pp.1031-1036
5. Daniel L.Y.Kong and Jay G.Sanjayan, “Effect of elevated temperatures on geopolymer paste,
mortar and concrete”, Cement and Concrete Research 40, pp.334-336, 2010.
6. T. Bakharev, “Geopolymeric materials prepared using class F fly ash and elevated temperature
curing”, Cement and Concrete Research 35, pp.1224-1232, 2005.
7. M.S. Morsy, A.F. Galal and S.A. Abo-El-Enein, “Effect of temperature on phase composition
and microstructure of artificial pozzolana-cement pastes containing burnt kaolinite clay”, Cement
and Concrete Research, 28(8), pp.11571163, 1998.
8. Omer Arioz, “Effects of elevated temperatures on properties of concrete”, Fire Safety Journal 42,
pp. 516-522, 2007.
9. T.W. Cheng and J.P. Chiu, “Fire-resistant geopolymer produced by granulated blast furnace
slag”, Minerals Engineering 16, pp. 205-210, 2003.
10. K.D. Hertz, “Limits of spalling of fire-exposed concrete”, Fire Safety Journal 38, pp. 103-116,
2003.
11. J. Davidovits, “Geopolymers: Inorganic Polymeric New Materials”, Journal of Thermal analysis
37, pp. 1633-1656, 1991.
318
12. S.K. Handoo, S. Agarwal, S.K. Agarwal, “Physicochemical, mineralogical, and morphological
characteristics of concrete exposed to elevated temperatures”, Cement and Concrete Research 32,
pp. 1009-1018, 2002.
13. S. Songpiriyakji, “Engineering Properties of Mae Moh Fly Ash Geopolymer Concrete”,
International conference on Pozzolan, Concrete and Geopolymer, may 24-25, 2006
14. A. Rafeet, R. Vinai, M. Soutsos and W. Sha, “Guidelines for mix proportioning of fly ash /
GGBS based alkali activated concretes”, Construction and Building materials 147(2017), pp.
130-142
15. J. wongpa, K. kiattikomol and C. Jaturapitakkul, “New Geopolymer From Rice Husk-Bark
Ash”, International Conference on Pozzolan, Concrete and Geopolymer, May 24-25, 2006
16. D. Hardjito and B.V. Rangan, “Development and Properties of Low-Calcium Fly ash-Based
Geopolymer Concrete”, Research Report GC 1, Australia (2005)
17. Davidovits, J. (1994). Properties of Geopolymer Cements. First International Conference
on Alkaline Cements and Concretes, Kiev, Ukraine, 1994, SRIBM, Kiev State Technical
University.
18. Davidovits, J. (1999). Chemistry of Geopolymeric Systems, Terminology. Geopolymer
’99 International Conference, France.
19. Duan P. et.al (2017), “Compressive strength and microstructure of fly ash based
geopolymerblended with silica fume under thermal cycle”. Cement and concrete
composites 78: 108-119
20. McCaffrey R. (2002), "Climate Change and the Cement Industry", Global Cement and
Lime Magazine, (Environmental Special Issue)
21. Malhotra, V. M. (2002). "Introduction: Sustainable Development and Concrete
Technology." ACI Concrete International 24(7)
22. Mehta, P. K. and R. W. Burrows (2001). "Building Durable Structures in the 21st
Century." ACI Concrete International 23(03): 57-63.
23. Mehta, P. K. (2002). "Greening of the Concrete Industry for Sustainable Development."
ACI Concrete International 24(7): 23-28.
24. Mehta, P.K. (1973) “Siliceous Ashes and Hydraulic Cements Prepared Therefrom”,
Belgium Patent 802909.
25. Palomo, A., M. W. Grutzeck, M.T. Blanco (1999). "Alkali-Activated Fly Ashes, A
Cement for the Future." Cement and Concrete Research 29(8): 1323-1329.
26. Rafieizonooz M. et.al (2016), “Investigation of coal bottom ash and fly ash in concrete as
replacement for sand and cement” Construction and building materials 116: 15-24
27. Roy, D. M. (1999). "Alkali-Activated Cements, Opportunities and Challenges." Cement
and Concrete Research 29(2): 249-254.
28. Xu, H. and J. S. J. van Deventer (2000). "The Geopolymerisation of AluminoSilicate
Minerals." International Journal of Mineral Processing 59(3): 247- 266.
319
Paper ID- 80
Stability of Different shapes of Tunnels in Weathering Stages

of Basalt
1
Mohd. Faraz Athar (https://orcid.org/0000-0003-3409-567X) , 2Mohammad Zaid ( https://orcid.org/0000-0001-
6610-8960), 3Md. Rehan Sadique ( https://orcid.org/0000-0002-9570-6801)
1
U.G Student, Dept. of Civil Eng., ZHCET, Aligarh Muslim University, U.P, India ,
2
P.G Student, Dept. of Civil Eng., ZHCET, Aligarh Muslim University, U.P, India
3
Assistant Professor, Dept. of Civil Eng., ZHCET, Aligarh Muslim University, U.P, India
1faraz.athar12@gmail.com
Abstract. The stability analysis of rock tunnels of different shape and depth of
overburden, under static loading has been presented in this paper. The effect of
weathering of rock has also been taken into consideration. Tunnels has been
assumed in uniform rock mass of weathered Basalt rock. The finite element
model has been developed and analyzed using finite element code Abaqus. A
42m x 42m 2D rock model has been developed and 7m diameter of the tunnel
has been excavated though it. The depth of tunnel has been varied having
overburden as 5m, 10m, and 17.5m at the tunnel crown. The stability analysis
has been carried out by observing the deformation at the interface of lining and
rock mass. From the present study, it may be concluded that arch-shaped
tunnels are more stable in comparison to circular and horseshoe shape.
Moreover, as the depth of overburden increases on the tunnel stability has been
improved relatively.
Keywords: Weathering, Basalt, Tunnel, Arch, Horseshoe, Circular, Abaqus.

.
1 Introduction
Among the major civil engineering structures, underground tunnels are considered of
utmost importance due to its vast uses and strategic importance. Tunnels may serve
different purposes depending upon the requirements of the particular area. Tunnels
may be of different shapes ranging from circular to horseshoe tunnels. Due to rapid
urbanization in the Indian subcontinent and Government projects like “smart city” in
which cities are converted into metropolitans at very high pace. This process in old
cities where availability of land is a major problem can easily be sought out by
constructing underground tunnels for metros, subways, underpasses etc. The structure
with that much of importance requires high precision and accuracy while constructed,
the stability of tunnels when constructed in the rock mass is also very important aspect
before actual construction of the tunnel.
Finite element analysis plays important role for the study of tunnels and 3D FEM
prove to be an essential way of analysis [1]. H.Y. Liu et al [2] carried out the analysis
for the stability of existing tunnel-support system for a crossing tunnel and concluded
that local thickening is needed for the stability of existing tunnel [2]. Yamaguchi et al
described the influence of thrust of succeeding and preceding tunnels [3]. Zaid et al
(2018) carried out the static stability of tunnel in urban regions which has suggested
that shallow tunnels must be studied for the worst conditions including the effect of
weathering [4]. For the study of tunnels and deep excavations numerical analysis tools
are proved to be an effective way of analysis [5,6].
320
All these studies show the recent trend in this field, however, no one till now does not
encounter the problem of stability of different shapes of tunnels on account of rock
weathering our study is an extension to the on-going research on the topic. We aim to
find the static stability of different shapes of tunnels like circular, arc, and horseshoe
on weathering of basal [7].
2 Methodology
In the present study authors have opted finite element technique to solve out the
problem. A 2D Elastoplastic Numerical model of the tunnel is made through the finite
element software ABAQUS/Standard. We then assign the Loading and subsequent
properties and run the model for Static/general step in ABAQUS/Standard [8].
Results have been calculated for varying the cover depth and different stages of
weathering of rocks mass.
3 Geometry
A 2D tunnel geometry is considered in the present study. The rock mass is taken as a
square of the 42×42mcross-section, the diameter of the tunnel is taken as 7m in all the
cases. The thickness of the lining is taken as 0.12m [9]. Figure 1 shows the geometry
of the circular tunnel, horseshoe lining and arch-shaped lining. section, the diameter
of the tunnel is taken as 7m in all the cases. The thickness of the lining is taken as
0.12m [9]. Figure 1 shows the geometry of the circular tunnel, horseshoe lining and
arch-shaped lining.
(a) (b) (c)
Figure 1 Geometry of (a)Arc, (b)Circular, and (c)Horse Shoe tunnels
4 Property
The 2D plane strain elastoplastic model has been considered for the study of the
stability of differently shaped tunnels due to weathering. Mohr-Coulomb theory of
plasticity has been considered for the plastic behavior of tunnel. The basalt rock mass
has been considered for the analysis. the different stages of weathering considered for
the analysis are fresh basalt(W0), slightly weathered basalt(W1), medium weathered
basalt(W2) and highly weathered basalt(W3). The input parameters for the basalt rock
has been taken from available literature [4,10-13]. Table 1 shows the properties of
different stages of weathering of basalt rock.
321
Table 1. Input parameters of the numerical simulation
Weathering Density Young Modulus Poisson’s Friction Angle Cohesion

Stage (kg/m3) (GPa) Ratio (Degrees) (MPa)
W0 2960 46.5 0.186 63.38 26.25
W1 2740 20.6 0.260 53.71 18.50
W2 2470 2.80 0.272 33.33 8.08
W3 1820 0.6 0.272 43.87 1.64
Concrete 2400 31.6 0.150 - -
5 Analysis
The static analysis has been performed through the Static General step in Abaqus
Standard. The initial increment size of 0.001 has been provided and the minimum
increment size adopted by the computational 1E-05. The interaction was assigned
between the concrete lining and the inner surface of rock mass. The tangential
behavior has been assigned as 0.55 coefficient of friction and normal hard contact
behavior was assumed. A load of 500m overburden has been applied as the pressure
of 0.16MPa on the top surface and the gravity component of -1 has been applied [10].
The base of the model has been fixed in all directions and the sides have applied for
roller support by allowing deformation in the vertical direction. Figure 2 shows the
loading and boundary conditions applied to the model.
The meshing plays an important role in the finite element analysis of tunnels [7]. The
model has been meshed by element type CPE4R (A 4-node bilinear plane strain
quadrilateral, reduced integration, and hourglass control). The model has been meshed
by element size of 1 and lining has been meshed with an element size of 0.1 keeping
aspect ratio nearly 1. The shape of the element was quad-dominated with free meshing
type. The meshed model has been sown in figure 3
Figure 2 Load and boundary condition assembly. Figure 3 Mesh of the model.
6 Result & Discussion
The analysis for the static stability of shallow tunnels, intermediate depth tunnels, and
deep tunnels has been carried out in this study. The depth for the shallow tunnel was
5m, for intermediate depth tunnel was 10m and deep tunnels have 17.5m. The effect
of weathering has also been considered in the analysis. The results have been
observed at the crown (point 1), left spring (point 2), right spring (point 3), and invert
(point 4). The figure 4 shows the total maximum deformation at different locations as
mentioned above for arch shaped tunnels. Figure 4(a) shows results for 5m cover
322
depth of tunnel, figure 4(b) shows deformation results for 10m cover depth and figure
4(c) for 17.5m cover depth. Figure 4 shows significant change in deformation due to
weathering of rock mass. The figure 5 show deformation graphs for horse shoe
shaped tunnels for different depths and varying rock properties. The figure 6(a),(b),(c)
shows contour results for the deformation in circular shaped tunnels with varying
depth and weathering stage Figure 7 shows the comparison of stability of different
shape tunnels for highly weathered basalt at different locations in the inner face of
tunnel lining. Figure 8 shows the comparison of deformation for various overburden
for circular tunnel.
1.20E-02
1.00E-02
8.00E-03
6.00E-03
4.00E-03
2.00E-03
0.00E+00
CROWN LEFT RIGHT INVERT
SPRINGER SPRINGER
5M
w0 w1 w2 w3
(a)
1.20E-02
1.00E-02
8.00E-03
6.00E-03
4.00E-03
2.00E-03
0.00E+00
CROWN LEFT SPRINGER RIGHT INVERT
SPRINGER
10M
w0 w1 w2 w3
(b)
1.00E-02
8.00E-03
6.00E-03
4.00E-03
2.00E-03
0.00E+00
CROWN LEFT SPRINGER RIGHT INVERT
SPRINGER
17.5M
w0 w1 w2 w3
(c)
Figure 4 Deformation (in meters) at different location of tunnel for Arch shaped for overburden (a) 5m (b)
10m (c) 17.m
323
1.40E-02
1.20E-02
1.00E-02
8.00E-03
6.00E-03
4.00E-03
2.00E-03
0.00E+00
CROWN LEFT SPRINGER RIGHT SPRINGER INVERT
5M
w0 w1 w2 w3
(a)
1.20E-02
1.00E-02
8.00E-03
6.00E-03
4.00E-03
2.00E-03
0.00E+00
10M
w0 w1 w2 w3
(b)
1.00E-02
8.00E-03
6.00E-03
4.00E-03
2.00E-03
0.00E+00
17.5M
w0 w1 w2 w3
(c)
Figure 5 Deformation (in meters) at different location of tunnel for Horse shoe shape for overburden (a) 5m
(b) 10m (c) 17.m
324
W0 W0 W0
W1 W1 W1
W2 W2 W2
W3 W3 W3
Figure 6 (a) Deformation Countor for Figure 6 (b) Deformation Countor for Figure 6 (c) Deformation Countor for
overburden 5m of circular tunnel overburden 10m of circular tunnel 17.5m depth of circular tunnel
325
1.00E-02
Deformation (in meters)

9.00E-03
8.00E-03 Arched shape
7.00E-03
6.00E-03 Circular Shaped
5.00E-03
4.00E-03 Horse shoe
3.00E-03
2.00E-03
1.00E-03
0.00E+00
0 1 2 3 4 5
Loaction of Observation
Figure 7Comparison of Deformation for different shape tunnels
1.60E-04
1.40E-04
1.20E-04
Deformation
5m
(in meter)
1.00E-04
8.00E-05 10m
6.00E-05 17.5m
4.00E-05
2.00E-05
0.00E+00
0 1 2 3 4 5
Location of Observation
Figure 8 Comparison of Deformation with varying cover depth.
7 Conclusion
The stability analysis of rock tunnel has been carried out using Abaqus/Standard.
Two-dimensional finite element model concludes the following point:
1. The stability of different shapes of the tunnel has been investigated using
numerical simulation.
2. Plain Strain Elastoplastic 2D Model is Considered for the study.
3. The weathering has a significant effect on the stability of tunnels, as the
weathering of rock increases the instability of tunnel increases.
4. The comparison of the static behavior of different shape tunnel shows that arch-
shaped tunnel is relatively more stable as compared to circular and horseshoe
shaped tunnels.
5. The overburden depth of tunnel has also affects its stability significantly, as the
depth increases the stability of tunnel increases.
326
References
[1] V. PA, B. PG, and M. SC, “On a smart use of 3D-FEM in tunnelling,” in International symposium; 8th, Numerical
models in geomechanics; NUMOG VIII, 2002, pp. 361–366.
[2] H. Y. Liu, J. C. Small, J. P. Carter, and D. J. Williams, “Effects of tunnelling on existing support systems of
perpendicularly crossing tunnels,” Comput. Geotech., vol. 36, no. 5, pp. 880–894, 2009.
[3] I. Yamaguchi, I. Yamazaki, and Y. Kiritani, “Study of Ground-Tunnel Interactions of Four Shield,” vol. 7798, no.
3, pp. 289–304, 1999.
[4] M. Zaid, S. Mishra, and K. S. Rao, “Finite Element Analysis of Static Loading on Urban Tunnels,” in IGC-2018,
2018.
[5] C. D. L. Nogueira et al., Validation of Coupled Simulation of Excavations in Saturated Clay : Camboinhas Case
History, vol. 11, no. 3. 2011, p. 5622.
[6] F. Yang, J. Zhang, L. Zhao, and J. Yang, “Upper-bound Finite Element Analysis of Stability of Tunnel Face
Subjected to Surcharge Loading in Cohesive-frictional Soil,” vol. 00, no. 0000, pp. 1–10, 2015.
[7] J. N. Franzius and D. M. Potts, “Influence of Mesh Geometry on Three-Dimensional Finite-Element Analysis of
Tunnel Excavation,” Int. J. Geomech., vol. 5, no. 3, pp. 256–266, 2005.
[8] L. Z. Cai and C. L. Zhang, “Application of ABAQUS in Surrounding Rock Stability Analysis of Shallow Large
Cross-Section Tunnel,” Appl. Mech. Mater., vol. 777, pp. 8–12, 2015.
[9] M. W. Naqvi, M. Zaid, R. Sadique, and M. M. Alam, “Dynamic Analysis of Rock Tunnels Considering Joint Dip
Angle : a Finite,” in 13th International Conference on Vibration Problems, 2017, no. December.
[10] A. Gahoi, M. Zaid, S. Mishra, and K. S. Rao, “Numerical Analysis of the Tunnels Subjected to Impact Loading,” in
INDOROCK, 2017.
[11] S. Mishra, S. Rao, N. K. Gupta, and A. Kumar, “Damage to shallow tunnels in different geomaterials under static
and dynamic loading,” Thin-Walled Struct., vol. 126, 2017.
[12] S. Mishra, K. S. Rao, N. K. Gupta, and A. Kumar, “Damage to Shallow Tunnels under Static and Dynamic
Loading,” Procedia Eng., vol. 173, pp. 1322–1329, 2017.
[13] M. Zaid, A. Talib, and M. R. Sadique, “Effect of joint orientation on the seismic stability of rock slope with
transmission tower,” in IGC-2018, 2018, pp. 1–6.
327
Paper ID- 83
1 Evaluating toughness as a parameter to determine fatigue life of
2 Wollastonite micro fiber reinforced high flow pavement quality concrete
3 Shashi Kant Sharma1*, Mohit Kumar1
4 Deptt. of Civil Engineering, NIT Jalandhar, Jalandhar 144011, India.
5 Abstract
6 Rigid pavements are deigned on basis of flexural strength since concrete is brittle and supposed to show little
7 strains post peak stress. The stress ratio which is the ratio of flexural stresses to flexural strength generally
8 determines the fatigue life of the pavement. Under service pavements face a continuous reduction in flexural
9 strength on account of propagating cracks which cause fatigue damage. But the evolution of cracks in quasi
10 ductile concrete is different which is somehow anticipated to depend upon the toughness of concrete. Toughness
11 is the area under load deflection curve and indicates the energy possessed by concrete. Under repetitive loading
12 there is continuous reduction in strength of concrete but the total energy possessed by concrete should not vary.
13 This paper thus tries to find the relationship of fatigue life with stress ratio and, fatigue life with toughness &
14 stress ratio together to check for the validity of stress ratio-fatigue life equation in case of quasi ductile concrete.
15 Equations have been generated for normal pavement quality concrete (PQC), flyash admixed & wollastonite
16 micro fiber reinforced PQC. Results proved that though toughness affects the formation of flexural stresses in
17 any kind of concrete, but stress ratio is singularly effective for finding out the fatigue life since it reflects both
18 toughness (through flexural stresses) and flexural strength. Quantitatively, WMF reinforced concrete has very
19 high fatigue life than normal PQC and flyash admixed PQC respectively.
20 Keywords: Rigid pavements; Fatigue life; Toughness; Flyash; Wollastonite micro fiber
21 1 Introduction
22 Cement composites are mainly made up of three components; cement paste, cement mortar and cement
23 concrete. All these composites have same deficiencies; weak tensile strength (brittleness) and limited
24 deformation capacity in the presence of cracking (lesser toughening strength). Once cracks are originated these
25 composites fail in a brittle manner with little warning due to limited deformation capacity. The interface
26 between aggregates and cement paste called as interfacial transition zone (ITZ) acts as a major weak zone for
27 distribution of stresses. If weak, strains grow at a larger rate in this zone, which leads to crack formation and
28 finally failure of concrete mix. When load is first applied to the cement concrete, areas of stress concentration
29 develop at the tips of the initial flaws. With additional loading, microcracks form (or initially present) and grow
30 in the frontal process zones ahead of the flaw tips. It is assumed that the initial flaw will extend into the
31 microcracked region when the microcracking has reached a saturated level [18]. Afterwards, they coalesce and
32 turn into macro cracks. A reinforcing material is therefore required which may provide strength to the concrete
33 before cracking, and cause delay in the propagation of cracks once they are formed. The need of special
328
34 reinforcing is fulfilled by the introduction of fibers in concrete. With proper incorporation of fibers, the failure
35 mode of cement composites can change from brittle to quasi-ductile [13, 14, 15]. For a thin sheet fiber
36 reinforced cement concrete having initial flaw of approximately 60 μm, fibers with diameter <30 μm will be
37 deemed micro fibers. Generally microfibers are fine fibers with lengths less than 10 mm and diameters <40μm
38 [20, 1]. Examples of fibers are steel fibers, carbon fibers, PVA fibers, polypropylene fibers and wollastonite
39 micro fiber. Steel fibers have high yield strength (nearly 1100 MPa) and possess lesser bond strength with the
40 concrete [2]. Wollastonite, has found tremendous applicability in the improvement of repair works. Wollastonite
41 is a naturally occurring, acicular, inert, white mineral (calcium meta silicate [ß - CaO-SiO2]) of high modulus of
42 elasticity, which is less costly than steel and carbon micro-fibers [3]. Prices for wollastonite range from $0.2/kg
43 to $0.37/kg depending of its size, which is substantially cheaper than that of steel ($6.6/kg), carbon ($11/kg) or
44 glass ($2/kg) fibers [3]. It is formed due to interaction of lime-stone with silica in hot magmas. It is formed in
45 nature by the reaction of calcium carbonate CaCO3 with silica SiO2 in hot magmas. Wollastonite is mined
46 commercially for use in refractory ceramics and as a filler in paints. The processed granular material is of high
47 purity, inexpensive, and readily available.
48 Available in abundance along the Udaipur belt of Rajasthan state of the Indian union as a low cost
49 material, it may be ground to fine powder. Its composition of nearly equal proportions of lime and silica having
50 fine particle size are favourable indicators for its admixing in concrete by partial replacement of cement. In
51 itself, wollastonite does not possess binding properties like cement, but, in presence of microsilica, it improves
52 the properties of admixed concrete by fine packing of inert material [16, 17].
53 Soliman and Nehdi [19] investigated the effect of incorporating wollastonite micro-fibers in ultra-high
54 performance concrete (UHPC) on its early-age properties. Wollastonite micro-fibers were added at 0, 4, 8 and
55 12% as partial volume replacement for cement. Increasing the wollastonite micro-fibers content resulted in
56 compressive strength comparable to or higher than that of the control mixture without microfibers. Wollastonite
57 micro-fibers reduced shrinkage strains and increased cracking resistance compared to that of the control
58 mixture. Hence these fibers could be used in pavements to enhance the factors which increase the fatigue life.
59 But in the absence of proper quantitative results it becomes subtle to testify the role of fibers in enhancing the
60 fatigue life of pavements. Indian Roads Congress (IRC) 58 [7] codal provisions suggest stress ratio of concrete
61 as a parameter to determine the fatigue life of a normal PQC considering the fact that concrete is brittle. This
62 makes flexural strength as the only property to be tested. But in the light of fibers’s use in concrete present study
329
63 is envisaged to check the effect of both toughness and flexural strength in concrete. Toughness is the area under
64 load deflection curve and thus represents the energy possessed by the concrete which could be checked against
65 the total energy consumed by the concrete during loading cycles. Use of high flow concrete could expedite the
66 construction process and even find use in retrofitting concrete pavements. Hence high flow pavement quality
67 concrete is tested in this study. Literature studies are mostly concentrated on finding out fatigue life of pavement
68 with or without admixtures but toughness has not been compared with stress ratio.
69 2 Materials and Mixes
70 Ordinary Portland cement (OPC) 43 grade conforming to Indian standard code IS 8112 [11] was used. Graded
71 river sand conforming to Zone–II having fineness modulus of 3.23 and specific gravity of 2.58 was used as fine
72 aggregate. Crushed graded coarse aggregates of 20 mm and 10 mm conforming to IS 383 [8] were used in the
73 proportion of 60:40. The specific gravity of 20 mm and 10 mm aggregates were 2.62 and 2.58 respectively.
74 Fine amorphous wollastonite powder supplied by Rajwara Stonex Limited was used. It had a specific surface
75 and specific gravity values of 827m2/kg and 2.9 respectively. Densified 920D grade Microsilica supplied by
76 India Private Limited was used. It had a specific surface and specifc gravity value of 18000 m 2/kg and 2.05.
77 High water reducing poly carboxylate ether (PCE) based superplasticizer was used to introduce self compacting
78 workability conditions to the concrete.
79 3 Testing Program
80 3.1 Materials’ Testing
81 Powder materials were tested for physical properties such as surface area, particle size distribution, specific
82 gravity; and chemical composition. The mixes were classified into binary and ternary mixes. Binary mixes were
83 made by substituting cement with Wollastonite/flyash in intervals of 10% up to maximum 30%. For ternary
84 combinations, the mixes were prepared such that Wollastonite/flyash had same replacement levels as later and
85 Microsilica was added maximally up to 10% in addition for each mix, at intervals of 2.5% respectively, thereby
86 making 12 mixes. Pictorial (SEM image analysis) and quantitative analysis (XRD analysis shown in Table 1) of
87 hydrated compounds obtained from paste mixes was performed after 7 days and 14 days of hydration,
88 respectively.
89 Table 1. Mix composition and X Ray Diffraction Data of Pastes
330
Mix Percentage of constituting material % Ettringite % CH % CSH % C3A

C 100C 2.3 9.8 37.0 5.5
CW1 90C+10W 2.5 10.0 39.9 5.8
CW2 80C+20W 5.4 5.4 41.9 5.1
CW3 70C+30W 10.2 4.2 39.8 6.3
CWS1 87.5C+10W+2.5S 6.2 2.0 39.3 5.7
CWS2 85C+10W+5S 9.0 4.1 40.0 5.9
CWS3 82.5C+10W+7.5S 2.0 2.1 40.5 5.9
CWS4 80C+10W+10S 7.1 5.8 43.2 5.7
CWS5 77.5C+20W+2.5S 2.9 5.6 46.2 6.3
CWS6 75C+20W+5S 4.9 4.4 47.4 6.4
CWS7 72.5C+20W+7.5S 5.8 2.9 50.2 6.0
CWS8 70C+20W+10S 4.4 2.5 55.4 6.2
CWS9 67.5C+30W+2.5S 2.8 8.6 40.7 5.4
CWS10 65C+30W+5S 7.5 5.6 44.1 4.7
CWS11 62.5C+30W+7.5S 5.3 5.8 45.9 6.0
CWS12 60C+30W+10S 4.9 6.1 46.3 5.5
CF1 90C+10F 2.8 8.9 38.7 5.8
CF2 80C+20F 2.7 6.4 40.2 6.5
CF3 70C+30F 3.2 2.8 42.4 6.7
CFS1 87.5C+10F+2.5S 2.5 10 39.4 5.8
CFS2 85C+10F+5S 2.9 10.6 40.1 5.8
CFS3 82.5C+10F+7.5S 3.6 7.7 41.3 6.2
CFS4 80C+10F+10S 4.1 4.6 45.9 6.2
CFS5 77.5C+20F+2.5S 2.8 7.2 41.4 5.9
CFS6 75C+20F+5S 2.9 8.6 42.2 6.0
CFS7 70C+20F+7.5S 3.4 6.2 46.6 6.1
CFS8 72.5C+20F+10S 3.0 5.6 46.9 6.1
CFS9 67.5C+30F+2.5S 2.4 6.3 44 5.8
CFS10 65C+30F+5S 3.4 4.1 45.1 6.6
CFS11 60C+30F+7.5S 3.1 4.5 47.4 7.1
CFS12 62.5C+30F+10S 2.8 6.0 50.4 6.9
90 3.2 Wet Concrete Testing
91 The testing program of wet concrete aimed at sequentially finding out the workability related ambiguities with
92 the use of admixtures and wollastonite micro fiber in self-compacting concrete. For this one control concrete
93 mix of modulus of rupture 4.5 Kg/cm2 was designed as per IRC 44 [6] (shown in Table 2) and 15 cement
94 substituted trial mixes were sought from it. Flow trials were conducted to achieve SCC by changing the binder
95 content of the control concrete mix (by admixing); then correspondingly changing the fine aggregate to coarse
96 aggregate ratio along with superplasticizer content at a constant water to cementitious material ratio of 0.35.
97 Superplasticizer was added to fulfil the water demand for creating self-compacting conditions. Table 3 shows
98 the trial methodology to obtain SCC.
331
99 Table 2. Mix design of normal PQC

Mix design for Normal M-40 concrete used to derive SCC for retrofitting as per IRC 44
Cement Sand Coarse aggregates Water Nominal MSA=16mm
450 711 1057 165 Superplasticizer 0.3% ~1.23lt/cu m.
1 1.58 2.35 0.36
20mm 10mm CA:FA=60:40
740 317 C.A :- 20mm:10mm=70:30
Assuming that 20% cement reduction takes place on addition of superplasticizer
100 Abrams flow, V funnel, J ring and probe ring tests were conducted on the SCC trial mixes (Figure 1) as
101 per the guidelines of EFNARC [4, 5], whereas control concrete failed in these tests. Concrete volume of 6 litres
102 was used for Abrams flow test and J ring test, whereas 12 litres of concrete volume was used for V funnel test.
103 Probe ring test was conducted by pouring concrete in cylindrical moulds (150ɸ×300mm).
104 Table 3 Schematic method of yielding SCC from Normal PQC
Normal PQC concrete (W/C= 0.36)

105
106
1. C: WMF: S 2. CA: FA 3. Viscosity modifying agent (VMA)
(i) 100:0:0 (i) 60:40 (ii) 55:45 Superplasticizer
107 (ii) 90:10:0 If no (iii) 50:50 (iv) 45:55 If no (i) 0.15% (ii) 0.30% (iii) 0.45% (iv)
(iii) 80:20:0 0.60%
(iv) 70:30:0
108 (v) 87.5:10:2.5 + If yes
(vi) 85:10:5 Abram’s flow test, V funnel test, J Ring test,
109 (vii) --:--:-- If yes Probe Ring test If yes
110
SCC
111
112
113 (a) (b)
332
114
115 (c) (d)
116 Fig. 1 Workability tests of concrete mixes (a) Probe ring (b) V funnel (c) J Ring & (d) Abram’s flow test
117 3.3 Dry Concrete Testing
118 Flexural strength test was conducted on beams of size 500×100×100mm3 as per IS 516 [9] guidelines
119 respectively. Later on Plate load test was conducted on five pavement prototypes constructed with these mixes
120 i.e. CF3, CW3, CFS6 and CWS6. The prototypes were constructed with about 6 ton of concrete (DLC 100mm+
121 PQC 300mm) each, and contained a contraction joint, with dowel bars of 38 mm for load transference. Figure 2
122 shows the loading points as well as the locations where the dial gauges were setup to find the deflections in the
123 pavement. Location 1, 3, 2 & 4 correspond to corner, centre, left and right edge locations at which loads would
124 be applied. Dial gauges at location 1, 2, 3 & 4 were installed to observe the deflections on the loading side,
125 whereas dial gauges 5, 6 & 7 were installed to find out the deflections on the adjacent side of loading. The test
126 was performed under deflection controlled condition and performed as per IS: 1888 [10]. Also due to the
127 difference in stiffness of loading points different loading profile was obtained for each loading location. Plate
128 load test performed on normal concrete and CWS6 has been shown in Figure 3 (a & b).
129
130 Fig. 2 Schematic layout of Plate load test conducted in the laboratory
131 The model as well as prototype consists of two panels laid in the direction of traffic flow, having a contraction
132 joint in between. The panel size is 1800×1800×300 mm3, dowel bars of ɸ 35mm and 500mm length were
333
133 embedded as wire elements in the model, and a dry lean concrete (DLC) layer of 100 mm thickness was
134 sandwiched between PQC slab and properly compacted subgrade layer.
135
136 Fig. 3 Plate load test on (a) non-retrofitted and (b) FRSCC retrofitted PQC
137 Figure 4 (a) shows one panel of PQC wherein dowel bars are embedded within it and dry lean concrete layer has
138 been introduced beneath the PQC slab. The loading pattern at the edge of the slab is being shown in Figure 4 (b)
139 which is in the form of circular ring having diameter of 45 cm spaced at a distance of 30 cm from both the edges
140 of the panel. The loading was applied by means of circular bearing plate made of steel having 25.4 mm
141 thickness. The FEM model was designed having same geometric scale on ABAQUS and was subjected to
142 similar loading profile. Boundary conditions to the model were assigned on the basis of actual working of rigid
143 pavement. The top and bottom face of the slab were fixed against movement in two horizontal directions. The
144 bottom face of PQC was also restricted in the same manner to allow the bending of the slab. This restriction will
145 produce tensile stresses at critical zones and compressive stresses in their vicinity thereby producing cracks in
146 the pavement. The top and bottom face of the DLC was free to move in vertical direction and along the
147 transverse direction but restricted in the vehicular direction. The top face of the subgrade was free to move in all
148 directions but its bottom face was restricted in the vertical direction to simulate infinite subgrade behaviour. The
149 division lines in the panel show the area earmarked for different mesh spacing. The half area where load was
150 applied was provided with lesser mesh spacing (50mm inter nodal distance) for ease in converging to the
151 solution, whereas the half panel area away from load was given larger mesh spacing (100mm).
334
152
153 Fig.4 (a) Full model of PQC (b) Model showing the DLC, one panel of PQC and dowel bars
154 Validation of the model was done for all five types of concrete on the basis of Chi Square test which
155 analysed the goodness of fit between deflections of model and prototype at 5% significance level. It was
156 assumed that the model will get validated for other prototypes too taking into consideration similar properties of
157 constituents in the five prototype mixes. This also saved huge amount of time and cost to be incurred in
158 constructing all the prototypes. After validation, the model was run for 10000 loading cycles of 20KN at all the
159 loading locations so as to create a conditioned pavement under use. The model was then run with cyclic loading
160 with a frequency of 0.01hertz (864 axles load per day) with load-time variation using Fourier function at the
161 edges of the pavement near the transverse joint. Each concrete mix was checked for 1.5% areal strains (using
162 nodal displacement in both directions) at the critical node present on the bottom face of the slab. This value of
163 strain has been chosen on the basis that quasi ductile concrete shows upto 0.75% ultimate axial strains under
164 compressive loading and areal strains are the summation of axial strain in the longitudinal & transverse direction
165 respectively. It was assumed that cracks will fully propagate to the upper most region of slab when strains will
166 achieve this level in the lower layer. Load cycles of 20KN and 22.5KN were applied in such a way that strains
167 were checked each time after 50, 100, 200, 400, 800, 1600, 3200, 6400, 12800, 25600, 51200, 102400, 204800,
168 509600, 1019200, 3057600, 611520, 1223040 and 2446080 cycles until 1.5% strains were observed. Strains
169 beyond were calculated by extrapolating the curve.
170 4 Observations and Results
171 4.1 Powdery Material
172 4.1.1 Physical Composition of Powdery Materials. Particle size analysis has been shown in Figure 5.
173 Microsilica was finest among all considered powdery materials followed by WMF, fly ash and cement,
174 respectively. The peak of WMF lies exactly in between the peaks of microsilica and OPC. This interpretation
335
175 clearly infers that WMF used was median size to both microsilica and OPC and hence, an excellent interlocking
176 within these particles is anticipated physically. It is also clearly depicted that OPC used for the study, exhibits
177 particle sizes comparable to flyash as revealed by the presence of secondary peak in Figure 5. Peak patterns
178 analysis suggests that fly ash and OPC have nearly same size range but prolonged post peak profile of fly ash
179 indicates that there are numerous fractions of flyash which are even larger in size than OPC particles.
180
181 Fig. 5 Particle size variation for various materials
182 The degree of fineness of microsilica, WMF and fly ash with respect to OPC is: 60, 2.8 and 1.3,
183 respectively. Results strongly hinted that microsilica would be prominently proactive in comparison to the rest
184 as it has inherent ability to contribute strength development through its least surface area.
185 4.1.2 Chemical composition of powdery materials
186 Table 4 shows the quantitative results of the amount of oxides present in cement and other admixtures found
187 through X-Ray fluorescence spectrometer test conducted in accordance with IS: 12803 [12]. Following are the
188 few observations drawn from the study of chemical compositions:
189 1. Fly ash used in the present study contains appreciable amount of silica and highest amount of alumina
190 when compared to the rest of the materials.
191 2. WMF shares equal amounts of lime & silica, and traces of alumina. Their presence indicates that WMF
192 has tendency for self cementation & moderate rate of reaction. But by virtue of its crystalline nature,
193 WMF is more inert and serves as excellent pore filler.
194 3. Microsilica mainly consists of silica and rest other oxides are in meagre amount. In its amorphous state,
195 fine microsilica possesses higher surface areas because of which it has greater infilling capacity and
196 reactivity too. Thus, it is apt for reduction of CH and ettringite considerably.
336
197 Table 4 Chemical properties of cementitious materials including OPC

Compound Cement Flyash WMF Silica fume
SiO2 20.2 35 48 92.9
Al2O3 5.2 26 1.4 0.9
Fe2O3 3 8.7 0.6 0.72
MgO 1.51 5 0.2 0.57
SO3 2.2 3 - 0.16
Na2O 0.08 1.5 - 0.32
Chloride 0.014 0.005 - 0.037
Loss on ignition 4.3 5 4 2.6
CaO 62.9 15.3 45.9 1.4
K2O 0.6 0.5 - 0.4
198 4.2 Concrete
199 4.2.1 Workability of Concrete Mixes
200 Table 5 shows the achieved mix designs for different concrete mixes having different paste compositions. Some
201 mixes qualified for SCC whereas rest were just able to achieve high workability and lacked in one or other
202 property of a SCC. The table shows the minimum CA:FA ratio required for achieving the workability
203 parameters shown besides them along with the superplasticizer content needed in purview. Since the paper aims
204 to focus on fatigue life and its parameters so workability results have not been explained.
205 Table 5 Observations from the Workability Tests Conducted on trial concrete mixes
Mix CA: Abram’s V V Funnel J Ring Probe ring Super plasticizer

FA flow (600- Funnel after 5 min. difference (0- penetration (0-7) C.M.) % by weight
750) mm time (6- (9-15 sec) 10 mm) mm) of cement
12sec)
C 60:40 360 17 24 24 2 0.30
CW1 55:45 560 10 12 21 3 0.45
CW2 50:50 580 8 10 19 5 0.45
CW3 50:50 540 12 12 21 4 0.6
CWS1 55:45 580 9 11 15 3 0.45
CWS2 50:50 630 8 10 9 5 0.45
CWS3 50:50 620 9 10 10 4 0.45
CWS4 50:50 605 10 11 13 3 0.45
CWS5 50:50 646 7 11 11 5 0.45
CWS6 50:50 660 7 9 5 7 0.6
CWS7 50:50 620 8 11 8 5 0.6
CWS8 50:50 590 9 14 10 4 0.6
CWS9 50:50 570 10 12 14 5 0.6
CWS10 45:55 630 7 9 12 3 0.6
CWS11 45:55 575 9 13 17 2 0.45
CWS12 45:55 530 13 18 22 2 0.45
206
10
337
207 4.2.2 Flexural Strength
208 Flexural strength depends equally on mortar strength and mortar-aggregate interface strength. On admixing,
209 WMF improves the interfacial transition zone (ITZ) considerably. The flexural strength of concrete mixes after
210 28 & 90 days of curing has been shown in Figure 6 & 7 respectively. The rate of improvement is more profound
211 for higher part replacement of cement by WMF i.e. upto 30% (4.88 MPa against 4.87 MPa of reference concrete
212 at 28 days). Beyond this replacement level, no significant improvement was observed as more WMF did not
213 find more surface area of concrete matrix to reinforce. Surprisingly admixing of flyash reduced the flexural
214 strength of concrete mixes even after long durations. This may be due to the fact that bigger sized flyash
215 particles could not induce effective pore size refinement in the concrete. Even flyash has least tendency to
216 reduce ettringite in the interfacial transition zone. Though it was effective in reducing the calcium hydroxide
217 content in this zone but it could not fill up the deficiency in strength caused by replacement of cement with
218 flyash.
Flexural Strength at 28 days

Flexural strength N/sq mm.
5.5
F10
5
F20
4.5
F30
4 W10
3.5 W20
3 W30
0 2 4 6 8 10
Microsilica percentage in cementitious material
219
220 Fig. 6. Flexural tensile strength of concrete mixes after 28 days curing

7
Flexural strength N/sq
F10
6 F20
F30
mm.
5
W10
4 W20
W30
3
0 2 4 6 8 10
221
222 Fig. 7. Flexural tensile strength of concrete mixes after 90 days curing
223 Microsilica increases flexural strength in both flyash and WMF admixed concrete mixes (Figure 6 & 7) at a
224 much larger rate per unit mass than the other two admixtures. Prolonged curing shows significant improvement
11
338
225 in flexural strength of concrete but the concrete mix containing 30% flyash and 10% microsilica did not show
226 any improvement in the flexural strength with respect to corresponding mix at 7.5% microsilica (Figure 7). This
227 is attributed to the self-desiccation caused by presence of excessive amount of microsilica than that required.
228 Similar trend for strength gain with curing duration, due to microsilica has been observed in WMF ternary
229 mixes but it was found that percentage increase in flexural strength with microsilica addition is lesser in this
230 case. The percentage improvement in flexural strength with time (28-90 days) ranged from 10-20% generally in
231 all binary mixes in comparison to ternary mixes (10-25%). Ternary mixes of Wollastonite micro fiber showed
232 generally higher modulus of rupture value than normal concrete mix at all durations except CWS1 & CWS2.
233 This indicates that 20-30% WMF with any content range of microsilica will definitely improve flexural strength
234 of concrete. This proves that grain size refinement of WMF in the mortar as well as the mortar-aggregate
235 interface plays a major role than just pore size refinement of flyash in holding the concrete matrix together
236 under bending stresses. Here the complementary effect provided by microsilica further improves the interface
237 and mortar. As discussed in modulus of elasticity, even though stiffness at interface was higher for flyash-
238 microsilica mixes which improved stress distribution but the holding together of aggrgeate and mortar at
239 interface under bending tensile stresses is better for WMF-microsilica mixes.
240 Overall, the flexural strength decreased by 22.5% and 3.2% in binary flyash and WMF mixes
241 respectively with respect to reference concrete mix after 90 days curing. Whereas it decreased by 14.1% in
242 ternary flyash-microsilica mixes. An increase of 6.1% was observed in ternary WMF-microsilica mixes.
243 4.2.3 Compressıve toughness
244 Figure 8 shows that for binary mixes, as Flyash content increases the compressive toughness decreases
245 (toughness for CF10>CF20>CF30) whereas mix CW20 has highest toughness followed by CW30 and CW10.
246 Also if microsilica is added to flyash mixes then there is approximately linear increase in toughness. Same
247 results are not seen in WMF-microsilica ternary mixes. Ternary mix containing 10% WMF with microsilica
248 showed linear increase in toughness whereas that containing 30% WMF showed a linear decrease beyond 2.5%
249 microsilica addition. Only mix CWS5-CWS8 containing 20% WMF showed nonlinear relationship with
250 increasing amounts of microsilica such that toughness was highest for 7.5% microsilica and beyond that reduced
251 to large extent. The decrease in toughness due to flyash addition is obvious as flyash reduces modulus of
252 elasticity and even increase the brittleness of concrete post cracking. WMF addition in concrete though
253 decreased the modulus of elasticity to some extent but the stiffness provided by WMF due to its inherent nature
12
339
254 as well fiber reinforcement provided post crack ductility thereby bringing the toughness value to 47.15 MPa in
255 comparison to reference concrete’s 40.91 MPa. With addition of microsilica the ductility provided by WMF at
256 interfaces got reduced and the concrete became stiffer thereby reducing the toughness abruptly for higher WMF
257 contents i.e. >=20%. Hence there was reduction in compressive toughness right away from 2.5% microsilica
258 addition in CWS10 whereas it was found at 7.5% for mix CWS7. Addition of microsilica obviously increased
259 toughness for flyash-microsilica mixes on account of increase in stiffness or modulus of elasticity.
260 Overall, the compressive toughness factor decreased by 29.5% and 19.1% in binary flyash and ternary
261 flyash-microsilica mixes respectively with respect to reference concrete mix after 28 days curing. Whereas it
262 increased by 6.5% and 9.9% in binary WMF and ternary WMF-microsilica mixes respectively.
Compressive toughness at 28 days

60.00
Compressive toughness
F10
factor N/sq mm.
50.00
F20
40.00 F30
W10
30.00 W20
W30
20.00
0 2 4 6 8 10
Microsilica percentgae in cementitious material
263
264 Fig. 8. Compressive toughness factor of concrete mixes after 28 days curing
265 4.2.4 Fatigue life from pavement models
266 After application of load cycles, a curve was obtained in which areal strains regress upon the cycles. Using this
267 curve the value of cycles corresponding to 1.5% areal strains was evaluated. Figure 9 shows the extrapolated
268 regression curve of areal strains on cycles for normal concrete at 22.5KN. Similarly cycle values were obtained
269 for all concrete mixes.
13
340
270
271 Fig. 9. Extrapolated strains-cycles curve representing cycles at 1.5% areal strains
272 Table 6 shows the toughness and flexural strength of concrete mixes. It also shows the flexural stresses
273 produced just after the application of conditioning load of 10000 cycles on the concrete slab. All these three
274 parameters have been shown in conjunction with the fatigue life of the pavement corresponding to 20KN and
275 22.5KN loading.
276 Table 6. Toughness, flexural strength, flexural stresses and cycles for various PQC
Toughness Flexural Stress Stress Stress

Toughness Stress at
Mix factor strength ratio at Cycles at 22.5 ratio at Cycles
(Nmm) 20KN
(N/sq.mm.) 20KN KN 22.5KN
C 40.91 812828 4.866 2.079 0.427 54929243 2.285 0.470 5753000
CF1 32.91 653866 4.316 2.340 0.542 174922 2.602 0.603 48284
CF2 28.64 569086 3.728 2.774 0.744 2949 3.079 0.826 947
CF3 24.91 494903 3.061 3.035 0.991 249 3.219 1.052 131
CFS1 36.59 726989 4.316 2.063 0.478 2844939 2.300 0.533 294930
CFS2 36.59 726989 4.552 2.050 0.450 24720433 2.320 0.510 469392
CFS3 39.36 782096 4.787 1.868 0.390 125723225 2.116 0.442 35839292
CFS4 39.52 785275 4.787 1.854 0.387 139292249 2.180 0.455 21483225
CFS5 29.44 584982 3.846 2.674 0.695 6583 2.850 0.741 3531
CFS6 29.97 595580 4.042 2.682 0.664 10483 2.901 0.718 5839
CFS7 31.73 630551 4.277 2.440 0.570 89472 2.744 0.642 16949
CFS8 36.91 653347 4.356 2.426 0.557 100292 2.648 0.608 39021
CFS9 27.84 553190 3.139 2.853 0.909 684 3.116 0.993 195
CFS10 28.21 560608 3.375 2.810 0.833 929 3.145 0.932 530
CFS11 29.44 584982 3.571 2.621 0.734 4832 3.000 0.840 738
CFS12 31.31 622073 3.453 2.490 0.721 5473 2.696 0.781 1634
CW1 38.08 916662 4.513 2.180 0.483 2057282 2.496 0.553 113293
CW2 47.15 936819 4.807 2.140 0.445 27959305 2.393 0.498 893753
14
341
CW3 45.49 903967 4.876 2.200 0.451 24294902 2.479 0.509 583853
CWS1 38.77 895038 4.630 2.213 0.478 2843593 2.551 0.551 119382
CWS2 39.89 912693 4.748 2.194 0.462 11273939 2.550 0.537 268374
CWS3 40.43 883291 4.905 2.242 0.457 18254829 2.609 0.532 295382
CWS4 44.27 879593 4.944 2.250 0.455 21493005 2.551 0.516 424492
CWS5 48.48 963313 4.944 2.065 0.418 68382022 2.360 0.477 3184322
CWS6 49.39 981329 5.101 2.025 0.397 124949206 2.280 0.447 28429493
CWS7 52.85 1050212 5.258 1.900 0.361 170403026 2.130 0.405 96849335
CWS8 46.67 927281 5.18 2.150 0.415 74739202 2.413 0.466 8583929
CWS9 46.35 920923 4.983 2.159 0.433 40002949 2.433 0.488 1295939
CWS10 44.85 891250 5.14 2.227 0.433 38892042 2.519 0.490 1294943
CWS11 44.32 880652 5.297 2.250 0.425 55020482 2.545 0.480 2335949
CWS12 43.2 858398 5.219 2.313 0.443 31073994 2.611 0.500 759249
277 Statistical curve fitting analysis was performed at 95% confidence interval for all the concrete mixes to find five
278 types of relationships: between (a) stress ratio and toughness; (b) flexural stresses and toughness, (c) fatigue life
279 and both stress ratio & toughness; (d) fatigue life and toughness; and (e) fatigue life and stress ratio. Regression
280 equations provided in Table 7 serve as results obtained from the analysis. Corresponding curves are provided in
281 Figures 10-15. It was apparent from equations 5-12 that fatigue life depends mostly upon stress ratio irrespective
282 of toughness, load intensity or type of concrete mix. The overall relationship between fatigue life and stress ratio
283 in Equation 6 was not so appreciable and it was observed that the curve was having three zones corresponding to
284 three stress ratios: 0.35-0.5, 0.5-0.65 and >0.65. So relationships at these values were also checked. To verify
285 the role of toughness along with stress ratio, relationships were checked for them too. It was found that
286 toughness along with stress ratio has weaker correlation with fatigue life with respect to just stress ratio alone.
287 There was medium ranged correlation (0.6-0.8) between fatigue life and toughness which was discarded, yet it
288 indicated that toughness has some effect on fatigue life. This indication was proved true when a good correlation
289 between flexural stress and toughness was found at both load intensities and for all types of mixes in equations
290 1- 4. This is due to the fact that the cracks are formed only when the concrete elements are strained beyond a
291 certain limit. The higher the limit, the higher energy concrete will absorb to crack. When cracks are fully
292 developed at the bottom face of the PQC, they start propagating upwards. Cracks also widen during this time in
293 the bottom face in both longitudinal and transverse directions. The post peak stress energy held by concrete in
294 bottom layer enables it to get strained upto higher limits without widening the cracks further which hinders the
295 propagation of cracks vertically, thereby relieving the flexural stresses along the vertical direction. The stiffness
296 of concrete in upper layers which is also a parameter of toughness in turn relieves the stresses of bottom layer.
15
342
297 Highly stiff normal concrete thus has higher toughness and lesser flexural stresses in comparison to weakly stiff
298 flyash admixed concrete mixes (Table 6.). Hence, more is the toughness lesser will be the flexural stresses
299 produced in the pavement. This is also apparent from Table 6 which showed lesser flexural stresses associated
300 with highly Tough WMF reinforced mixes.
301 Medium ranged correlation (0.6-0.8) exists between stress ratio and toughness which explains the role
302 of flexural strength in stress ratio values. High flexural strength may or may not indicate a high toughness value
303 since it also depends upon stiffness as well as ductile nature of concrete post ultimate stress.
304
305 Fig.10. Flexural stresses at 20KN for (a) Normal PQC/Flyash admixed PQC (b) WMF reinforced PQC
306
307 Fig.11. Flexural stresses at 22.5KN for (a) Normal PQC/Flyash admixed PQC (b) WMF reinforced PQC
16
343
308
309 Fig.12. Fatigue life vs. (a) Stress ratio & Toughness (b) Stress ratio
310
311 Fig. 13. Fatigue life vs. (a) Stress ratio & Toughness (b) Stress ratio; at SR 0.35-0.5
312
313 Fig. 14. Fatigue life vs. (a) Stress ratio & Toughness (b) Stress ratio; at SR 0.5-0.65
17
344
314
315 Fig. 15. Fatigue life vs. (a) Stress ratio & Toughness (b) Stress ratio; at SR >0.65
316 Table 7. Regression equations of fatigue life on stress ratio and toughness
Eq. Correlating variables Coefficients (with 95% confidence bounds) Goodness of fit
No.
1. Flexural Stress vs. Toughness for a = 7.061e+06 (-2.397e+08, 2.539e+08) SSE: 0.2073
Normal/Flyash mixes at 20 KN b= -1.125 (-3.927, 1.677) R-square: 0.8919
f(x) = a*xb+ c c = 0.3432 (-4.848, 5.535) RMSE: 0.1263
2. Flexural Stress vs. Toughness for a = -0.00128 (-0.02975, 0.02719) SSE: 0.06187
Normal/Flyash mixes at 22.5 KN b = 0.6073 (-0.8712, 2.086) R-square: 0.9725
f(x) = a*xb+ c c= 7.001 (-3.594, 17.59) RMSE: 0.06899
3. Flexural Stress vs. Toughness for p1 = -2.134e-06 (-2.234e-06, -2.035e-06) SSE: 0.0009034
WMF mixes at 20KN p2 = 4.131 (4.04, 4.222) R-square: 0.994
f(x) = p1*x + p2 RMSE: 0.008336
4. Flexural Stress vs. Toughness for p1 = -2.625e-06 (-3.004e-06, -2.246e-06) SSE: 0.01321
WMF mixes at 22.5KN p2 = 4.876 (4.527, 5.226) R-square: 0.9451
f(x) = p1*x + p2 RMSE: 0.03188
5. Fatigue life vs. Stress ratio & p00 = 3.481e+08 (-5.765e+08, 1.273e+09) SSE: 1.165e+16
Toughness (overall) p10 = -1.661e+09 (-5.435e+09, 2.113e+09) R-square: 0.8695
f(x, y) = p00 + p10*x + p01*y + p01 = 943.1 (132.3, 1754) RMSE: 1.455e+07
p20*x2 + p11*x*y + p30*x3 + p20 = 3.129e+09 (-1.488e+09, 7.747e+09)
p21*x2*y p11 = -3325 (-6257, -391.6)
p30 = -1.939e+09 (-3.507e+09, -3.698e+08)
p21 = 2618 (-30.11, 5265)
6. Fatigue life vs. Stress ratio (overall) a = 5.506e+04 (5488, 1.046e+05) SSE: 6.483e+15
f(x) = a*xb+ c b= -8.104 (-9.027, -7.182) R-square: 0.9274
c = -6.362e+06 (-1.086e+07, -1.869e+06) RMSE: 1.048e+07
7. Fatigue life vs. Stress ratio & p00 = 2.711e+09 (2.225e+09, 3.197e+09) SSE: 8.904e+14
Toughness at SR 0.35-0.5 p10 = -1.013e+10 (-1.192e+10, -8.35e+09) R-square: 0.986
f(x, y) = p00 + p10*x + p01*y + p01 = -376.2 (-692.4, -59.9) RMSE: 5.852e+06
p20*x2 + p11*x*y p20 = 9.24e+09 (7.491e+09, 1.099e+10)
p11 = 847.3 (116.7, 1578)
8. Fatigue life vs. Stress ratio for SR a1 = 1.715e+08 (1.636e+08, 1.794e+08) SSE: 2.936e+14
0.35-0.5 b1 = 0.3561 (0.3511, 0.3611) R-square: 0.9954
{
( )⁄
} c1 = 0.06573 (0.06208, 0.06938) RMSE: 3.238e+06
f(x) = a1*
9. Fatigue life vs. Stress ratio & p00 = 7.507e+06 (-7.081e+06, 2.209e+07) SSE: 2.051e+10
Toughness at SR 0.5-0.65 p10 = -3.386e+07 (-7.562e+07, 7.891e+06) R-square: 0.97
f(x,y) = p00 + p10*x + p01*y + p01 = 10.41 (1.497, 19.33) RMSE: 4.529e+04
p20*x2 + p11*x*y p20 = 3.759e+07 (9.588e+06, 6.558e+07)
p11 = -19.18 (-35.58, -2.778)
10. Fatigue life vs. Stress ratio for SR a= 7.359 (-1.954, 16.67) SSE: 1.266e+10
0.5-0.65 b= -16.63 (-18.52, -14.73) R-square: 0.9815
f(x) = a*xb RMSE: 3.12e+04
11. Fatigue life vs. Stress ratio & p00 = 1.728e+05 (2.777e+04, 3.178e+05) SSE: 8.567e+06
Toughness at SR>0.65 p10 = -3.351e+05 (-5.628e+05, -1.074e+05) R-square: 0.9374
18
345
f(x, y) = p00 + p10*x + p01*y + p01 = -0.0782 (-0.2628, 0.1064) RMSE: 882.5
p20*x2 + p11*x*y p20 = 1.534e+05 (8.104e+04, 2.258e+05)
p11 = 0.09553 (-0.1202, 0.3112)
12. Fatigue life vs. Stress ratio for SR a1 = 1.012e+05 (-4.936e+05, 6.96e+05) SSE: 2.449e+06
0.65-1.5 b1 = 0.2834 (-0.5933, 1.16) R-square: 0.9821
{
( )⁄
} c1 = 0.2527 (-0.004975, 0.5103) RMSE: 434
f(x) = a1*
317 Hence whilst calculating the fatigue life on basis of stress ratio i.e. flexural stresses upon flexural
318 strength, the toughness factor is introduced indirectly by the flexural stresses. Thus the stress ratio is singularly
319 capable enough to regress the value of fatigue life. Regarding the life of concrete, it could be said that overall
320 WMF reinforced concrete has very high fatigue life in comparison to both normal and flyash admixed concrete.
321 At 20 KN maximum fatigue life values of WMF reinforced, flyash admixed and Normal concrete mixes are 170,
322 139.3 and 54.9 million, respectively whereas these are 96.8, 57.8 and 5.75 million, respectively at 22.5KN load.
323 Though overall flyash admixed concrete is a poor performer and showed very low fatigue life on account of
324 high flexural stresses and low flexural strength.
325 5 Conclusion
326 Following points could be read as the conclusion on the basis of this study:
327 1. On an average fatigue life of WMF reinforced rigid pavements is very high. Its peak fatigue life value
328 is 3.1 times at 20KN to 16.8 times at 22.5KN in comparison to normal PQC. Similarly it varies like
329 1.22 times at 20KN to 1.67 times at 22.5KN in comparison to flyash admixed PQC. Wollastonite micro
330 fiber reinforced concrete is beneficial not only because of the fact that they exhibit higher flexural
331 strength but also higher toughness.
332 2. Since WMF reinforced concrete exhibited high flow, hence it would not be wrong to say that high
333 fatigue life could be achieved with the use of high flow fiber reinforced concretes.
334 3. Flyash admixed concrete pavements have on an average very low fatigue life than normal PQC.
335 4. Stresses developing in lower most layer of concrete are relieved either by the stiffness of layers above
336 or due to the post peak stress energy possessed by the concrete which strains the concrete without
337 fracture. In either way it’s the toughness that plays the role because both stiffness and post peak stress
338 energy are components of toughness. Yet, it is obvious that higher flexural strength will ensure high
339 fatigue life as more cycles will be consumed to fracture the concrete. Higher toughness reduces the
340 crack propagation in concrete by relieving the stresses and thus indirectly affects the stress ratio of
341 concrete.
19
346
342 5. Stress ratio remains a sufficient parameter in order to evaluate fatigue life of a pavement.
343 6. Fatigue life equation could be a consolidated equation for all types of loads and for all types of
344 concrete since it depends only upon stress ratio. For a given value of stress ratio it could be similar for
345 different concrete types or different load types.
346 References
347 1. Banthia, N, and Sheng, J. (1996). “Fracture toughness of microfiber reinforced cement composites.”
348 Cement and Concrete Composites, Elsevier, 18(4), 251-269.
349 2. Banthia, N., and Trottier, J. F. (1994). “Concrete reinforced with deformed steel fibers, Part I: Bond-
350 slip mechanisms. ACI Mat. J., 91(5), 435–446.
351 3. Clark, P. (1998). “Future of automotive body materials: Steel, Aluminum & Polymer Corporation.”
352 Massachusetts Institute of Technology, < http://readpdf.net/file/future-steel-vehicle.html>.
353 4. EFNARC. (2002). “Specification and guidelines for self-compacting concrete.”
354 555 EFNARC, Association House, 99 West Street, Farnham, Surrey, UK.
355 5. EFNARC. (2005). “The European guidelines for self-compacting concrete: specification, 557
356 production and use”, EFNARC, www.efca.info or www.efnarc.org.
357 6. IRC 44. (2008). “Tentative guidelines for cement concrete mix design for pavements,” Indian Roads
358 Congress, New Delhi, India.
359 7. IRC 58. (2015). “Guidelines for the design of plain jointed rigid pavements for highways,” Indian
360 Roads Congress, New Delhi, India.
361 8. IS 383. (1970). “Indian standard methods of tests for gradation of coarse aggregates.”
362 Bureau of Indian Standards, New Delhi, India.
363 9. IS 516. (1959). “Indian standard methods of tests for strength of concrete.” Bureau of
364 Indian Standards, New Delhi, India.
365 10. IS: 1888. (1982). “Indian standard methods of load tests on soils.” Bureau of Indian
366 Standards, New Delhi, India.
367 11. IS 8112. (2013). “Ordinary Portland cement 43 grade specification.” Bureau of Indian
368 Standards, New Delhi, India.
369 12. IS 12803. (1989). “Methods of analysis of hydraulic cement by X-ray fluorescence
370 spectrometer.” Bureau of Indian Standards, New Delhi, India.
371 13. Leung, C. K. Y. (1992). “Fracture-based two-way debonding model for discontinuous fibers in elastic
372 matrix.” J. Engg. Mech., ASCE, 118(11), 2298-2318.
373 14. Li, V. C., and Leung, C. K. Y. (1992). “Tensile failure modes of random discontinuous fiber reinforced
374 brittle matrix composites.” J. Engg. Mech., ASCE, 118(11), 2246–2264.
375 15. Li, V. C., and Wu, H. C. (1992). “Conditions for pseudo strain-hardening in fiber reinforced brittle
376 matrix composites.” Appl. Mech. Rev., 45(8), 390–398.
377 16. Low, N.M.P., and Beaudoin, J. J. (1992). “Mechanical properties of high performance cement binders
378 reinforced with Wollastonite micro-fibres.” Cement and Concrete Research, 22(5), 981-989.
20
347
379 17. Low, N.M.P., and Beaudoin, J.J. (1993). “The effect of Wollastonite micro-fibre aspect ratio on
380 reinforcement of portland cement-based binders.” Cement and Concrete Research, 23, 1467-1479.
381 18. Nelson, P. K., Li, V. C., and Kamada, T. (2002). “Fracture toughness of microfiber reinforced cement
382 composites.” Journal of Materials in Civil Engineering, 14(5), 384–391.
383 19. Soliman, A.M., and Nehdi, M.L. (2012). “Effect of natural Wollastonite microfibers on early-age
384 behavior of UHPC”, Journal of Materials in Civil Engineering, 24(7), 816–824.
385 20. Wu, H.C. (2001). “Discussion on mechanical properties of steel microfiber reinforced cement pastes
386 and mortars.” Journal of Materials in Civil Engineering, 13(3), 240 -241.
387
21
348
Paper ID- 85
High Flow Pavement Quality Concrete Reinforced with Wollastonite
Microfiber for Rigid Pavements
Shashi Kant Sharma1*, Mohit Kumar1
Deptt. of Civil Engineering, NIT Jalandhar, Jalandhar 144011, India.
Abstract
Fibers have always been needed for obtaining concrete mixes with higher strength and lower shrinkage. But the
cost of fibers, loss in workability and increased abrasion makes it imperative to put a restriction on their
quantity. This study tests the pavement quality concrete (PQC) reinforced with wollastonite micro fiber (WMF)
a cheap abundant mineral fiber for high flow i.e. approaching self compacting concrete (SCC), strength,
toughness, modulus of elasticity, shrinkage and abrasion resistance. A PQSCC would be an asset in speedy and
economic construction of durable rigid pavements for sustainable transport.
Results proved that SCC could successfully be obtained from the replacement of cement with microfiber
materials. Flexural strength is improved by 10% whereas compressive strength approaches equal value to
normal concrete after 90 days of curing. Modulus of elasticity though reduced with the increase in volume of
mortar but WMF and microsilica compensated the loss by lending their own stiffness to the concrete.
Compressive toughness also improved by 10% thereby indicating higher fatigue life ensuring least early
maintenance. Even though a rise in abrasion was obvious due to the acicular microfiber but the dense
reinforcement by fiber improved the bonding in concrete interphases and thus normalized the abrasion.
Keywords: Microfiber; Fatigue life; Toughness; Shrinkage; Abrasion resistance
1 Introduction
A concrete having high flexural strength (>4.5MPa) and high abrasion resistance has the capability to serve as a
pavement quality concrete. Durability of concrete is always better than bituminous pavements provided it has
controlled amount of shrinkage. Shrinkage stresses results in minute crack formation which accumulate under
repetitive axle loads to cause bigger cracks, thereby fracture of concrete.
Fibers have been used to control shrinkage in concrete successfully and almost all types of fibers play this role
[22,20,1,10,25,30,31,4]. They have additional benefits of increasing compressive strength and flexural strength
349
of concrete [17,18,16,24,8]. WMF is one such type of microfiber which is found abundantly in Rajasthan state
of India. It has no processing cost and bears only packing and transportation cost (INR 2/Kg). The calcium meta
silicate composition enables it to act as a pozzolan [12,23,14] whereas its acicular nature does the fiber effect
[16,29]. Since it is acicular, there are issues involved with its interaction in the concrete mix which may affect
the homogeneity of concrete further affecting workability and abrasion resistance. Compressive toughness is a
measure of area under the load deflection curve which depends upon three factors: modulus of elasticity,
ultimate strength and post fracture strains development in concrete. All these factors depend upon the
substitution levels of cement as well as the intensity of working of WMF either as a pozzolan or a fiber. In order
to evaluate the role of a fiber in developing fatigue strength of concrete, compressive toughness has a lot of
significance, since it is a measure of total energy absorbed by concrete under static loading, and thus more or
less indicates the energy absorbed by concrete under discontinuous dynamic loading.
Abrasion resistance of concrete is higher because of its high strength as well as smoothness imparted by cement
mortar. Various types of fiber alter concrete matrix and cause either balling effect [11,7,21] due to increased
cohesive forces between fibers which reduces workability and strength, and brings higher abrasion due to the
texture of fiber. WMF has acicular nature which could increase the abrasion of mortar with tyres of vehicles.
Under such circumstances only an increase in strength may not totally compensate the increase in abrasion [5].
There is a need of grain size refinement at fiber-mortar and fiber-coarse aggregate interface which could
smoothen the acicular effect of fiber. Hence in this study efforts have been made to improve these interfaces
with the use of admixtures.
Summarizing the objectives, it could be said that the study aimed at obtaining shrinkage resisting high flow
pavement quality concrete, possibly achieving self-compacting conditions, with improved fatigue life and
without compensating abrasion resistance. Since mostly flyash has been used in literature [3,15,19,27,13,26,28]
to provide bulk to mortar for obtaining high flow and bring comparable strength to normal concrete, comparison
has been made with flyash admixed high flow concrete.
2 Materials
Ordinary Portland cement 43 grade conforming to Indian standard code IS 8112 (1995) was used as the
principal binder in all mixes which was substituted with either class F flyash & microsilica, or WMF &
microsilica. Graded river sand conforming to Zone–II having fineness modulus of 3.23 and specific gravity of
350
2.58 was used as fine aggregate. Crushed graded coarse aggregates of 20 mm and 10 mm having specific gravity
values of 2.62 and 2.58 respectively and conforming to IS 383 (1970) were used in the proportion of 60:40. Fine
amorphous wollastonite powder supplied by Rajwara Stonex Limited was used. It had a specific surface and
specific gravity values of 827m2/kg and 2.9 respectively. Densified 920D grade Microsilica having specific
surface and specifc gravity value of 18000 m2/kg and 2.05 was used. High water reducing poly carboxylate ether
(PCE) based superplasticizer was used to introduce high flow to the concrete. Chemical properties of these
materials have been shown in Table 1. The particle size analysis of these materials has been shown if Figure 1.
TABLE 1. Chemical properties of cementitious materials including OPC

Compound Cement Flyash WMF Silica fume
SiO2 20.2 35 48 92.9
Al2O3 5.2 26 1.4 0.9
Fe2O3 3 8.7 0.6 0.72
MgO 1.51 5 0.2 0.57
SO3 2.2 3 - 0.16
Na2O 0.08 1.5 - 0.32
Chloride 0.014 0.005 - 0.037
CaO 62.9 15.3 45.9 1.4
K2O 0.6 0.5 - 0.4
Microsilica Cement Flyash WMF

10
9
8
7
Quantity (%)
6
5
4
3
2
1
0
0.01
0.02
0.03
0.04
0.06
0.09
0.13
0.20
0.30
0.45
0.67
1.01
1.51
2.27
3.41
5.12
7.70
133.10
200.00
300.52
11.57
17.38
26.11
39.23
58.95
88.58
Particle size (microns)
Figure 1. Particle size variation for various materials
3 Mix Proportions
A normal high flow SCC having flexural strength of 4.5MPa was designed as per specifications laid down in
IRC 44. The proportions of various constituents have been shown in Table 2.
Table 2. Mix design of normal PQC
351
Mix design for Normal M-40 concrete used to derive SCC for retrofitting as per IRC 44
1 1.58 2.35 0.36
20mm 10mm CA:FA=60:40
740 317 C.A :- 20mm:10mm=70:30
Efforts were directed towards yielding a SCC from normal concrete, but all the high flow concrete
mixes were also evaluated. Three mixes were constituted for each flyash and WMF by substituting cement @
10-30%. Furthermore twelve mixes each for WMF and flyash were achieved by substituting cement with
WMF/flyash & micro silica @ 10-30% & 2.5-10%, respectively. Thus in all thirty mixes were constituted for
trial program. These are shown in Table 3 along with X Ray diffraction analysis of their constituted pastes after
14 days of hydration. Also, Table 4 shows the schematic diagram of workability trials. The control concrete was
designed as a semi flowable concrete. Abrams flow (flowability), V funnel (flowability), J ring (passability) and
probe ring tests (segregation resistance) were conducted on the trial mixes (Fig. 2) as per the guidelines of
EFNARC (2002, 2005).
(a) (b) (c) (d)

Fig. 2. Workability tests of concrete mixes (a) Probe ring (b) V funnel (c) J Ring & (d) Abram’s flow test
TABLE 3. Mix composition and X ray diffraction data of pastes

Mix Percentage of constituting material % Ettringite % CH % CSH % C3A
C 100C 2.3 9.8 37 5.5
CF1 90C+10F 2.8 8.9 38.7 5.8
CF2 80C+20F 2.7 6.4 40.2 6.5
CF3 70C+30F 3.2 2.8 42.4 6.7
CFS1 87.5C+10F+2.5M 2.5 10 39.4 5.8
CFS2 85C+10F+5M 2.9 10.6 40.1 5.8
CFS3 82.5C+10F+7.5M 3.6 7.7 41.3 6.2
CFS4 80C+10F+10M 4.1 4.6 45.9 6.2
CFS5 77.5C+20F+2.5M 2.8 7.2 41.4 5.9
CFS6 75C+20F+5M 2.9 8.6 42.2 6
CFS7 70C+20F+7.5M 3.4 6.2 46.6 6.1
CFS8 72.5C+20F+10M 3 5.6 46.9 6.1
352
CFS9 67.5C+30F+2.5M 2.4 6.3 44 5.8

CFS10 65C+30F+5M 3.4 4.1 45.1 6.6
CFS11 60C+30F+7.5M 3.1 4.5 47.4 7.1
CFS12 62.5C+30F+10M 2.8 6 50.4 6.9
CW1 90C+10W 2.5 10 39.9 5.8
CW2 80C+20W 5.4 5.4 41.9 5.1
CW3 70C+30W 10.2 4.2 39.8 6.3
CWS1 87.5C+10W+2.5M 6.2 2 39.3 5.7
CWS2 85C+10W+5M 9 4.1 40 5.9
CWS3 82.5C+10W+7.5M 2 2.1 40.5 5.9
CWS4 80C+10W+10M 7.1 5.8 43.2 5.7
CWS5 77.5C+20W+2.5M 2.9 5.6 46.2 6.3
CWS6 75C+20W+5M 4.9 4.4 47.4 6.4
CWS7 72.5C+20W+7.5M 5.8 2.9 50.2 6
CWS8 70C+20W+10M 4.4 2.5 55.4 6.2
CWS9 67.5C+30W+2.5M 2.8 8.6 40.7 5.4
CWS10 65C+30W+5M 7.5 5.6 44.1 4.7
CWS11 62.5C+30W+7.5M 5.3 5.8 45.9 6
CWS12 60C+30W+10M 4.9 6.1 46.3 5.5
Table 4 Schematic method of yielding SCC from Normal PQC
Normal PQC concrete (W/C= 0.36)
1. C: WMF: S 2. CA: FA 3. Viscosity modifying agent (VMA)

(i) 100:0:0 If no (i) 60:40 (ii) 55:45 If no Superplasticizer
(ii) 90:10:0 (iii) 50:50 (iv) 45:55 (i) 0.15% (ii) 0.30% (iii) 0.45% (iv)
(iii) 80:20:0 If yes 0.60%
(iv) 70:30:0
(v) 87.5:10:2.5
(vi) 85:10:5 Abram’s flow test, V funnel test, J Ring test,
(vii) --:--:-- If yes Probe Ring test If yes
SCC
4 Experimental Program
Flexural strength test was carried out in the laboratory on prismatic specimens of 100×100×500 cu.mm. size, at
400 mm c/c span on two steel rollers of 38 mm diameter via third point loading method as per the specifications
of IS: 516:1959.
4.2 Compressive strength, Toughness and Modulus of Elasticity
353
Cylindrical concrete specimens were tested under compressive loads as per JSCE-SF5. The code specifies
guidelines for finding compressive strength, toughness and modulus of elasticity of steel fiber reinforced
concrete. Fig. 3 (a & b) show the assembly of LVDT applied to the concrete specimen for measuring
deformation. The stress was increased at a rate of 0.25 N/mm2/second. The compressive toughness is evaluated
as compressive toughness factor and is given by:
= ⁄ ) (1)
Where, is compressive toughness factor (N/mm2)
Tc is compressive toughness (J) measured from area under load deflection curve (Fig. 3(c)) and is
deformation corresponding to 0.75% strain or 1.125mm deformation for cylinder of size ϕ150mm × 300 mm. If
specimen fails before reaching , then the deformation shall be corresponding to 0.75% strain.
(a) (b) (c)
Figure 3. (a) Assembly for performing compression test (b) Concrete specimen showing irregular cracks due to fiber
reinforcement (c) Area under load deformation curve for concrete
4.3 Drying shrinkage
This test was carried out according to the specifications of ASTM C1581, using a 50 mm thick concrete
specimen, 150 mm in height, cast around a steel ring of 25 mm thickness and 300 mm outside diameter as
shown in Fig.4. The compressive strain developed in the steel ring caused by shrinkage of the mortar or concrete
specimen is measured from the time of casting. Cracking of the test specimen is indicated by a sudden decrease
in the steel ring strain. The age at cracking and the rate of tensile stress development in the test specimen are
indicators of the material’s resistance to cracking under restrained shrinkage. From the strain- time variation
average strain rate is recoded which yields average stress rate by the equation 1 as follows:
 net    t  k (2)
354
 net = net strain, m/m, α = strain rate factor for each strain gage on the test specimen (m/m)/day1/2, t = elapsed
time, days, and k = regression constant
The stress rate is calculated from the strain rate by the equation 2 as follows:
G   avg
q (3)
2  tr
Where, q = stress rate in each test specimen, MPa/day, G = 72.2 GPa, |  avg | = absolute value of the average
strain rate factor for each test specimen, (m/m)/day1/2 and, t r = elapsed time at cracking or elapsed time when
the test is terminated for each test specimen, in days.
Figure 4. Apparatus for measuring compressive strains in restrained shrinkage test
4.4 Abrasion resistance
Abrasion test was performed as per IS: 9284-1979. In this test an abrasive sand charge (4000g), passing 1mm
and retained on 0.5 mm sieve was stroked with a 10 cm a cement composite cube specimen. A pneumatic
hollow needle with an arrangement for striking charge from a height of 50 mm, with an air pressure of 0.14
N/mm2 was used. Two impressions (at 2.5 cm each from edge) are made on a single face by rotating the sample
at 180°. The abrasion resistance is measured as the inverse of loss in weight of sample. Fig. 5 shows the various
specimens tested in the laboratory for abrasion loss.
Figure 5. Abrasion test conducted on cubical mortar specimens
5 Results & Discussion
355
5.1 Workability
Following observations were made whilst conducting workability tests for SCC on trial mixes:
(i) It was found that for a given binder content, there is an optimum ratio of CA: FA. If the value of this
ratio is higher than the optimum value, then the flow of the mix is reduced apparently. If it is lower
than it induces laitance. It was supposed that segregation too depends on the ratio of CA: FA, such that
at higher ratio, segregation would be higher. But the assumption was wrong because at higher coarse
aggregate content only flowability and passability were maximally affected. The possible reason is that
the binder material was sufficient for holding either the coarse or fine aggregates, provided the total
volume of aggregates remains the same. The ratio has to be decreased to induce maximum flow.
There are two forms of segregation; one form in which the separation of coarse aggregates occurs in
very lean mixes containing higher specific gravity due to higher volume of aggregates, and other form
in which there is reduction of cohesion on account of reduced surface tension caused by higher
volumes of water, owing to separation of cement paste from the aggregates mass. This is called as
laitance formation and is different from bleeding. With higher volume of fine aggregate this type of
segregation occurs. Also flowability of the mix reduces due to low cohesion in the mix. In order to
increase the flow, if further superplasticizer is added, then the increased effective water content reduced
the viscosity of the paste to such an extent that laitance formation takes place. It is known from
literature that high viscosity of fresh cement paste component mitigate the downward movement of the
heavier aggregate particles, mix with low water/cement ratios are less prone to segregation [2,6].
Hence, while addition of superplasticizers for a given binder mass, if segregation occurs, then fine
aggregates need to be reduced or the superplasticizer content has exceeded. In present study the last
possible superplasticizer content has been reported for a given CA:FA and the corresponding flow
parameters have been noted.
(ii) A check on the average surface areas of the binder content and that of the fine aggregates made it
coherent that the maximum possible content of fine aggregates for a given binder content is directly
proportional to the average surface area of the binder. This means that the increment in fine aggregate
content over and above that of the mix design value is directly proportional to the increment in surface
area caused by the fraction of admixture in the paste.
356
(iii) Binder content, CA/FA ratio and surface morphology; all effect the flow of SCC to big or small extent
depending upon the property. Flowability and passability were found to be dependent more on the
binder content and CA/FA ratio, whereas the segregation resistance depended more on the surface
morphology of the binder material.
(iv) A maximum of 27.5% cement substitution with WMF & microsilica could be achieved such that no
more than 20% WMF could be added in any case so as to maintain segregation resistance. Also, atleast
5% microsilica is needed to maintain homogeneity and cohesiveness in the concrete mix.
Table 5 shows the observations made from the test. Following results are suggested on the basis of role of
each of the admixture in imparting flow to the SCC:
(i) WMF addition
Cohesion in small amounts is necessary for a good flow of mix because the mix flows homogeneously,
together with minimal opposition caused by viscous forces. Passability also decreases with increased cohesion,
because the mix try to remain altogether when the obstruction confronts it, but for a mix to show segregation
resistance it should have higher cohesion. Apart from viscous forces in the pore solution, the inter-particle
frictional forces tend to reduce the flowability and passability but enhance the segregation resistance of concrete
mix. WMF is acicular and it has high adsorption tendency as well as the tendency to be a part of pore solution at
its higher contents. Hence it increases the inter-particle friction by producing friction between cement particles
on which it get adsorbed, and it also increases the viscosity of pore solution at its higher contents. Therefore,
with WMF addition the flowability and passability of mix reduced, though up to 20% WMF its rate of reduction
was less. Segregation resistance increased with increment in WMF content.
(ii) Flyash addition
Flyash at all contents improved the flowability and passability of mixes, but it decreased the
segregation resistance of self compacting concrete. Replacement of cement with high volume flyash reduced the
viscosity of the mix. The smooth flyash particles did not stick to other materials in the mortar, and being having
lower specific gravity made the paste lighter which favoured good flow under impact. Hence flyash at its higher
content improved the flow of mortar. At lower content it reduced the cohesiveness of the mix and thus reduced
the flow of mortar.
(iii) Microsilica addition
357
With microsilica addition at lower amount 0-5%, the flowability, passability and segregation resistance
increased because microsilica has smooth texture and spherical shape, which induces a ball bearing effect
between cement particles on whom it get adsorbed. The ball bearing effect does not allow the cement particles
to move away though it allows them to roll or slide over each other on account of the sticky nature of
microsilica. Khayat & Aitcin [9] also suggested that presence of microsilica affects the properties of fresh
concrete by inducing cohesivity and thus reducing the bleeding of concrete. Both of these factors work
contradictory as far as shrinkage is concerned; on one hand cohesivity reduces shrinkage, whereas on other the
reduced bleeding increases it. The increased cohesivity also requires more slump for a given flow, with respect
to a normal concrete. But it has one advantage which is homogeneity, and thus enables good passability and
flowability, thereby enabling microsilica admixed concrete as a pumpable concrete. Hence, at higher amount of
microsilica, the flowability and passability decreased but the segregation resistance increased, because the pore
solution between the cement particles get thicker and more viscous due to increased microsilica content.
Table 5. Observations from the Workability Tests Conducted on trial concrete mixes
Mix CA: Abram’s V V Funnel J Ring Probe ring Super plasticizer

FA flow (600- Funnel after 5 min. difference (0- penetration (0-7) C.M.) % by weight
750) mm time (6- (9-15 sec) 10 mm) mm) of cement
12sec)
C 60:40 360 17 24 24 2 0.30

CW1 55:45 560 10 12 21 3 0.45
CW2 50:50 580 8 10 19 5 0.45
CW3 50:50 540 12 12 21 4 0.6
CWS1 55:45 580 9 11 15 3 0.45
CWS2 50:50 630 8 10 9 5 0.45
CWS3 50:50 620 9 10 10 4 0.45
CWS4 50:50 605 10 11 13 3 0.45
CWS5 50:50 646 7 11 11 5 0.45
CWS6 50:50 660 7 9 5 7 0.6
CWS7 50:50 620 8 11 8 5 0.6
CWS8 50:50 590 9 14 10 4 0.6
CWS9 50:50 570 10 12 14 5 0.6
CWS10 45:55 630 7 9 12 3 0.6
CWS11 45:55 575 9 13 17 2 0.45
CWS12 45:55 530 13 18 22 2 0.45
CF1 55:45 570 10 12 14 5 0.3
CF2 55:45 581 9 17 13 7 0.3
CF3 55:45 604 7 18 11 9 0.3
CFS1 55:45 585 9 16 12 5 0.45
CFS2 50:50 615 8 14 11 8 0.45
CFS3 50:50 600 9 10 15 6 0.45
CFS4 50:50 618 10 12 18 3 0.45
10
358
CFS5 50:50 587 7 9 9 7 0.45

CFS6 50:50 633 6 9 8 6 0.45
CFS7 50:50 663 8 9 10 6 0.45
CFS8 50:50 623 10 11 13 6 0.45
CFS9 50:50 633 6 15 11 10 0.45
CFS10 50:50 663 5 17 8 12 0.3
CFS11 45:55 623 7 17 13 9 0.3
CFS12 45:55 594 8 21 15 8 0.3
5.2 Drying shrinkage resistance of concrete mixes
Table 6 shows the cracking time and final stress rate for different mixes. While testing at a laboratory, shrinkage
is measured by two parameters: stress rate due to shrinkage and time for crack development. Both of these
parameters vary according to paste composition (fineness), which affects the continuity of pores in hardened
paste as well as the rate of hydration. It is not sufficient enough to say that the shrinkage of a concrete would be
higher, if it contains larger fraction of fine materials. This would be true to say only, if the water adsorbed by the
particle at the capillary voids is not held strongly, if the particle is not able to convert the capillary voids into gel
pores, and if the pores are continuous. All of these factors are dependent upon the size and reactivity of the
particle.
(i) Results on WMF with microsilica
Initially when WMF is added in lower quantity up to 20%, the cement particles are not far apart in the solution,
and the voids in between them are easily filled or blocked by WMF due to its pore size or grain size refinement.
Though the water is released when WMF enters into the cement voids, but this water only is used in the
accelerated hydration done by cement particles. As a result of which, the water is not able to escape from the
pores (mainly gel pores) and shrinkage is reduced. With the increment in the content of WMF in the mix, the
cement particles increase their distance, and more and more WMF particles enter into the pore solution present
in between these voids. As a result of which the number of capillary spaces increase whereas the gel pores
reduce. Since the void space is larger, the initial hydration, as well as insufficient capacity to block the voids,
provided by WMF, is not sufficient to stop the water from escaping from these capillary voids. Therefore, an
increased shrinkage stress was reported for WMF content greater than 20%. As a reinforcer, WMF will increase
the tensile strength and toughness of mortar, which would resist the crack development and also will delay the
propagation of cracks by distributing the stresses over a local region. Thus, the overall effect observed for WMF
addition, was the delay in crack development in spite of high stress rate.
(ii) Results on Flyash with micro silica
11
359
For flyash mixes, the hydration rate is higher than normal concrete uptil 7 days. The effect of this acceleration
increases its crack development time in comparison to normal cement concrete. After 14 days the effect of
acceleration diminishes and the samples start cracking. Though flyash contain alumina and calcium oxide,
which promote ettringite formation in presence of sulphates, but the consumption of lime due to flyash
hydration may hinder the formation of ettringite. Also, the reactivity of alumina depends upon, whether it is
available in glass part of the compound or not. If it is present in glass part than it provides a long term source of
ettringite formation in presence of sulphates. The reactivity of flyash as noticed from the XRD test suggests, that
ettringite formation is hindered in presence of flyash. This suggests that flyash does not contain reactive alumina
and highly neutralizes CH; if it did, then ettringite formation would have been promoted rather than hindered.
Thus, the stress rate is equal to or a bit higher than normal concrete because effect of pore refinement is nullified
by presence of fine material (fraction of flyash much finer than cement) verified by particle size analysis. Hence
with the increment in flyash content both the stress rate and the crack time increase to a small extent.
Table 6. Results obtained from the restrained shrinkage test
Mix Max.μ Initial Age at Strain rate (α) at Avg. Elapsed Stress Potential
Strain age max. elapsed time T μstrain time at rate (q) for
(Days) strain (ε=α×T½+K) rate max. MPa at cracking
(Days) strain (Tr Tr
C -96 0.24 12.3 Y=10.48X-58.35 10.48 Days)
12.06 0.109 Mod High
CW1 -70 0.03 18.8 Y= 6.872X-37.00 6.87 18.82 0.057 Mod Low
CW2 -55 0.25 116.8 Y= 4.836X-35.12 4.83 16.56 0.043 Mod Low
CW3 -96 0.24 523.6 Y= 9.069X-54.14 9.07 23.45 0.068 Mod Low
CWS1 -56 0.47 128.0 Y= 10.95X+0.280 10.95 27.52 0.075 Low
CWS2 -64 0.34 919.5 Y= 5.266X-42.20 5.27 19.19 0.043 Mod Low
CWS3 -68 0.34 022.5 Y= 9.956X-17.54 9.96 22.17 0.076 Mod Low
CWS4 -67 0.50 427.4 Y=11.13X-2.352 11.13 26.89 0.077 Low
CWS5 -72 0.26 118.7 Y= 4.316X-51.34 4.32 18.51 0.036 Mod Low
CWS6 -70 0.55 026.5 Y= 2.317X-56.68 2.32 26.03 0.016 Low
CWS7 -78 0.47 823.6 Y= 4.856X-51.09 4.86 23.21 0.036 Mod Low
CWS8 -76 0.27 920.5 Y= 5.390X-51.34 5.39 20.23 0.043 Mod Low
CWS9 -39 0.40 928.0 Y= 8.081X+6.033 8.08 27.63 0.056 Low
CWS10 -40 0.55 028.0 Y= 6.505X-3.328 6.51 27.47 0.045 Low
CWS11 -54 0.34 224.1 Y= 10.69X-1.725 10.69 23.79 0.079 Mod Low
CWS12 -55 0.34 228.0 Y= 12.36X+3.933 12.36 27.68 0.085 Low
CF1 - 0.32 314.4 Y= 12.05X-68.02 12.05 14.13 0.116 Mod Low
CF2 -96 0.44 218.9 Y= 15.07X-22.94 15.07 18.49 0.127 Mod Low
CF3 120- 0.42 416.0 Y= 15.59X-52.78 15.59 15.59 0.143 Mod Low
CFS1 - 0.37 219.1 Y= 12.09X-58.50 12.09 18.80 0.101 Mod Low
CFS2 113- 0.45 120.5 Y= 10.28X-66.31 10.28 20.08 0.083 Mod Low
CFS3 117- 0.29 716.6 Y= 12.05X-69.14 12.05 16.35 0.108 Mod Low
CFS4 116- 0.27 315.0 Y= 11.82X-67.93 11.82 14.77 0.111 Mod Low
CFS5 119
-84 0.55 428.0 Y= 16.22X+4.654 16.22 27.47 0.112 Mod Low
CFS6 116
-89 0.34 428.0 Y= 16.03X-2.588 16.03 27.68 0.110 Mod Low
CFS7 -69 0.53 228.0 Y= 13.61X+12.02 13.61 27.49 0.094 Low
CFS8 -95 0.34 228.0 Y= 18.55X+5.899 18.55 27.68 0.127 Mod Low
CFS9 -96 0.50 228.0 Y= 16.42X+0.420 16.42 27.52 0.113 Mod Low
CFS10 - 0.29 218.0 Y= 11.88X-54.38 11.88 17.73 0.102 Mod Low
2
105 2 12
360
CFS11 - 0.29 21.8 Y= 16.85X-22.19 16.85 21.52 0.131 Mod Low

CFS12 - 0.45 26.9 Y= 18.88X-2.764 18.88 26.50 0.132 Mod Low
108 1
5.3 110 resistance of mortars
Abrasion 5
Fig. 6 shows the results for abrasion resistance of mortar mixes observed after 28 days of curing which could be
summarized as following:
(i) WMF addition
The abrasion resistance studied after 28 days showed that flyash decreased abrasion resistance whereas WMF
increased it. Excess of WMF imparts more packing in a given volume of paste by grain size rather than pore size
refinement. Even though it densifies the mortar, which should increase the strength of mortar and its resistance
against abrasion, but the dilution of cement in mortar, causes reduction in C-S-H formation too. Apart from this
WMF reinforces the mortar too, due to its bonding with it. The bonding results on account of WMF getting
adsorbed on cement particles and afterwards being hydrated there itself. It was found, that up to twenty percent
content of WMF, the strength loss due to loss in C-S-H formation caused by dilution, is nullified by packing and
reinforcing effect. After 20% WMF content, the packing starts hindering further C-S-H formation, as a result of
which the water content in the pore solution between cement particles diminishes and the availability of
nucleating surfaces for hydration products too decreases. This also affects the strength of interfacial transition
zone where the mortar containing WMF binds aggregates. The overall effect was reduction in abrasion strength
at WMF content greater than 20%. Figs 4.85 show clearly, the effect produced by admixtures on the abrasion
resistance of mortars.
(ii) Flyash addition
Flyash addition was found to reduce the 28-day abrasion resistance of mortar mixes. With an increment in the
flyash content up to 30%, there was a corresponding uniform increment in weight loss of specimens due to
abrasion. The possible reasons for loss in abrasion resistance due to flyash are; loss in strength due to dilution,
lesser packing effect of flyash, less strength enhancement due to lower rate of C-S-H formation by flyash. Even
though the water liberation was more in case of flyash, but the dilution of cement and lesser hydration rate of
flyash caused reduction in strength of mortar at 28 days.
(iii) Microsilica addition
13
361
Even in case of microsilica inclusion, WMF also interferes in the pore size refinement and grain size refinement
done by microsilica. Though it is not able to get adsorbed on the cement particles’ surface because of lower
surface area but it totally affects the filling of spaces in cement particles’ voids. This does not affect strength to a
higher extent which seems to increase with the WMF content initially because of its reinforcing effect. But the
reduced pore size refinement caused by filling of voids by acicular WMF has a straightaway effect on the
density of hydration products and thus abrasion resistance. The density increased with the age and it is expected
that after 90 days WMF too should give same abrasion resistance as that of WMF-microsilica mixes.
Hence for 28 days, it could be said that microsilica addition in WMF mixes only led to moderate strength
increment in comparison to flyash-microsilica mixes. Flyash-microsilica mixes have extra advantage of both
strength increment and pore refinement, due to increment in hydration caused by microsilica in bigger sized
flyash particles (approx same size as cement particles) which provided more nucleating surfaces. Hence the rate
of abrasion resistance is more in microsilica-flyash mixes in comparison to WMF-microsilica mixes, though
overall WMF-microsilica mixes have highest abrasion resistance values.
1.000
0.900
0.800
0.700
Percent Weight Loss
0.600
0.500
0.400
0.300
0.200
0.100
0.000
CFS2
CF1
CF2
CF3
CFS1
CFS3
CFS4
CFS5
CFS6
CFS7
CFS8
CFS9
CWS10
CWS11
CWS12
C
CFS10
CFS11
CFS12
CW1
CW2
CW3
CWS3
CWS1
CWS2
CWS4
CWS5
CWS6
CWS7
CWS8
CWS9
Fig. 6. Abrasion resistance of binary and ternary mortar mixes
Overall stating abrasion loss increased by 172.6% and 79.9% in binary flyash and ternary flyash-microsilica
mixes respectively with respect to reference concrete mix after 28 days curing. Whereas, it increased by 22.4%
and 2.5% in binary flyash and ternary flyash-microsilica mixes, respectively.
5.4 COMPRESSIVE STRENGTH
For showing the strength of mixes, curves have been plotted taking into consideration variation of microsilica
content since representation of mixes as a whole would have been cumbersome otherwise. As it is apparent from
Fig. 7 that compressive strength of mix CF10 (53.32 MPa) approaches that of reference concrete (53.94 MPa)
14
362
after 28 days. This means dilution of cement has negligible effect on this mix. With an increase in flyash content
the compressive strength has decreased which is shown for 0% microsilica for mix CF20 and CF30. Also the
curve profile for these mixes clearly depict that maximum strength is achievable for 5% microsilica having
range from 40-45MPa. Higher microsilica content in these mixes reduces the compressive strength whereas in
mix containing 10% flyash the strength continues to increase. This is mainly due to filling of capillary voids of
concrete matrix with upto 5% microsilica and for flyash content >=20% beyond which extra flyash remained
like a useless material outside cement voids. Incase of WMF, mix CW20 has higher compressive strength in
comparison to CW10 and CW20 for 0% microsilica. Higher grain size refinement apart from infilling of
capillary voids due to orientation of lime and ettringite crystals by acicular crystals of WMF may have
compensated the loss due to dilution of cement. As the micosilica content further increases mix with 30% WMF
behave differently than other two mixes and shows approximately mildly sloped linear variation upto 7.5%
microsilica and later decreases. Whereas the optimum content of microsilica corresponding to 10-20% WMF is
near to 5% with compressive strength varying from 45-50 MPa. This indicates that WMF and microsilica
together could fill the capillary voids for above mentioned contents optimally above which they tend to remain
outside of cement voids. For 30% WMF, the compressive strength did not increase because of heavy dilution of
cement just compensated by grain size refinement. Pore size refinement is not expected to a high level for both
WMF and bigger sized flyash particles upto 28 days.
Compressive strength at 28 days

compressive strength N/sq
55 F10
50 F20
45 F30
mm
40
W10
35
W20
30
W30
25
0 2 4 6 8 10
Microsilica percentage in cementititous material
Fig.7. Compressive strength of concrete mixes after 28 days curing
15
363
Compressive strength at 90 days

90
compressive strength N/sq

F10
80
F20
70
F30
mm
60
W10
50 W20
40 W30
0 2 4 6 8 10
Microsilica percentage in cementititous material
Fig.8. Compressive strength of concrete mixes after 90 days curing
As shown in Fig. 8, all the mixes showed approximately linear variation for compressive strength with
increasing levels of microsilica at 90 days, due to appreciable amount of pore size refinement. Flyash mixes
showed nearly comparative strength with WMF mixes. Mix CWS12 with 10% microsilica showed highest
compressive strength of 80.34MPa whereas normal concrete mix had 63.66 MPa strength. Also at 0%
microsilica content all these mixes showed approximately equal strength which proves the pozzolanic nature of
these materials.
Overall stating compressive strength decreased by 23.4% and 20.1% in binary flyash and WMF mixes
respectively with respect to reference concrete mix after 90 days curing. Whereas it decreased by 6.4% in
ternary flyash-microsilica mixes. An increase of 1.6% was observed in ternary WMF-microsilica mixes.
Flexural strength depends equally on mortar strength and mortar-aggregate interface strength. On admixing,
WMF improves the interfacial transition zone (ITZ) considerably. The flexural strength of concrete mixes after
28 & 90 days of curing has been shown in Fig. 9 & 10 respectively. The rate of improvement is more profound
for higher part replacement of cement by WMF i.e. upto 30% (4.88 MPa against 4.87 MPa of reference concrete
at 28 days). Beyond this replacement level, no significant improvement was observed as more WMF did not
find more surface area of concrete matrix to reinforce. Surprisingly admixing of flyash reduced the flexural
strength of concrete mixes even after long durations. This may be due to the fact that bigger sized flyash
particles could not induce effective pore size refinement in the concrete. Even flyash has least tendency to
reduce ettringite in the interfacial transition zone. Though it was effective in reducing the calcium hydroxide
16
364
content in this zone but it could not fill up the deficiency in strength caused by replacement of cement with
flyash.

Flexural strength N/sq mm. 5.5
F10
5
F20
4.5
F30
4
W10
3.5 W20
3 W30
0 2 4 6 8 10
Fig. 9. Flexural tensile strength of concrete mixes after 28 days curing

7
6.5 F10
6
F20
5.5
5 F30
4.5 W10
4 W20
3.5
W30
3
0 2 4 6 8 10
Microsilica increases flexural strength in both flyash and WMF admixed concrete mixes (Fig. 9 & 10) at a much
larger rate per unit mass than the other two admixtures. Prolonged curing shows significant improvement in
flexural strength of concrete but the concrete mix containing 30% flyash and 10% microsilica did not show any
improvement in the flexural strength with respect to corresponding mix at 7.5% microsilica (Fig.10). This is
attributed to the self-desiccation caused by presence of excessive amount of microsilica than that required.
Similar trend for strength gain with curing duration, due to microsilica has been observed in WMF ternary
mixes but it was found that percentage increase in flexural strength with microsilica addition is lesser in this
case. The percentage improvement in flexural strength with time (28 -90 days) ranged from 10-20% generally in
all binary mixes in comparison to ternary mixes (10-25%). Ternary mixes of Wollastonite micro fiber showed
generally higher modulus of rupture value than normal concrete mix at all durations except CWS1 & CWS2.
This indicates that 20-30% WMF with any content range of microsilica will definitely improve flexural strength
of concrete. This proves that grain size refinement of WMF in the mortar as well as the mortar-aggregate
17
365
interface plays a major role than just pore size refinement of flyash in holding the concrete matrix together
under bending stresses. Here the complementary effect provided by microsilica further improves the interface
and mortar. As discussed in modulus of elasticity, even though stiffness at interface was higher for flyash-
microsilica mixes which improved stress distribution but the holding together of aggrgeate and mortar at
interface under bending tensile stresses is better for WMF-microsilica mixes.
Overall stating flexural strength decreased by 22.5% and 3.2% in binary flyash and WMF mixes respectively
with respect to reference concrete mix after 90 days curing. Whereas it decreased by 14.1% in ternary flyash-
microsilica mixes. An increase of 6.1% was observed in ternary WMF-microsilica mixes.
5.6 Modulus of Elasticity
As shown in Fig. 11 modulus of elasticity of all mixes at 28 days was lower than the normal concrete mix (30.3
GPa) due to increase in mortar volume having lesser stiff component. E value variation trend of all the mixes at
0% microsilica content was similar to that of compressive strength such that there was decrease in E values with
increase in flyash content, and mix CW20 showed higher E value than mixes CW10 and just higher value than
CW30.
Modulus of Elasticity at 28 days

30
F10
25
F20
20 F30
W10
15
W20
10 W30
0 2 4 6 8 10
But later on the behaviour is quite different. Flyash mixes showed linear behaviour as is apparent from the
figure, such that mix having 30% flyash has highest E value (28 GPa) followed by those with 20% and 10%. All
these values are higher than those of WMF mixes corresponding to 10% microsilica. WMF mixes also showed
an increase in E value with an increase in microsilica content, but the variation was not linear, it increased for
upto 7.5% (all corresponding values in range of 23-25GPa) and decreased for 10% such that mix with 30%
WMF has higher E value followed by those with 20% and 10%. The reason behind continuous linear rise in E
18
366
values for flyash mixes is only attributed to microsilica which stiffened the matrix at the interfaces even though
if not present in cement voids at higher levels. Whereas in WMF mixes, there was self- desiccation of
microsilica as fine WMF (fineer than flyash) also adsorbed water over its surfaces which resulted in scarcity of
water at higher contents of microsilica thereby resulting in weak interfaces and mortar itself. Since E value
depends more on stress distribution which further depends upon interface strength, and rather than strength of
mortar, therefore here the variation trend of mixes is different from that of compressive strength.
Overall stating E value decreased by 52.2% and 39.4% in binary flyash and WMF mixes respectively with
respect to reference concrete mix after 28 days curing. Whereas it decreased by 26.6% and 26.7% in ternary
flyash-microsilica and WMF-microsilica mixes respectively.
5.7 Compressive Toughness
Fig.12 shows that for binary mixes, as Flyash content increases the compressive toughness decreases (toughness
for CF10>CF20>CF30) whereas mix CW20 has highest toughness followed by CW30 and CW10. Also if
microsilica is added to flyash mixes then there is approximately linear increase in toughness. Same results are
not seen in WMF-microsilica ternary mixes. Ternary mix containing 10% WMF with microsilica showed linear
increase in toughness whereas that containing 30% WMF showed a linear decrease beyond 2.5% microsilica
addition. Only mix CWS5-CWS8 containing 20% WMF showed nonlinear relationship with increasing amounts
of microsilica such that toughness was highest for 7.5% microsilica and beyond that reduced to large extent. The
decrease in toughness due to flyash addition is obvious as flyash reduces modulus of elasticity and even increase
the brittleness of concrete post cracking. WMF addition in concrete though decreased the modulus of elasticity
to some extent but the stiffness provided by WMF due to its inherent nature as well fiber reinforcement
provided post crack ductility thereby bringing the toughness value to 47.15 MPa in comparison to reference
concrete’s 40.91 MPa. With addition of microsilica the ductility provided by WMF at interfaces got reduced and
the concrete became stiffer thereby reducing the toughness abruptly for higher WMF contents i.e. >=20%.
Hence there was reduction in compressive toughness right away from 2.5% microsilica addition in CWS10
whereas it was found at 7.5% for mix CWS7. Addition of microsilica obviously increased toughness for flyash-
microsilica mixes on account of increase in stiffness or modulus of elasticity.
Overall stating compressive toughness factor decreased by 29.5% and 19.1% in binary flyash and ternary flyash-
microsilica mixes respectively with respect to reference concrete mix after 28 days curing. Whereas it increased
by 6.5% and 9.9% in binary WMF and ternary WMF-microsilica mixes respectively.
19
367
Compressive toughness at 28 days

60.00
Compressive toughness
55.00
factor N/sq mm.

F10
50.00
45.00 F20
40.00 F30
35.00 W10
30.00 W20
25.00
W30
20.00
0 2 4 6 8 10
Microsilica percentgae in cementitious material
Fig. 12. Compressive toughness factor of concrete mixes after 28 days curing
6 Conclusions
It is always beneficial to add microfibers in concrete to enhance the flexural tensile strength and reduce the
shrinkage of concrete rather than mineral admixtures irrespective of the type of microfiber used. This study
validates this statement, but there are always limitations in regards to the optimum quantity of microfiber to be
used. This is because of the fact that parameters like workability, strength, modulus of elasticity and abrasion
resistance of concrete have correspondingly different optimum quantities for a certain type of microfiber.
Following conclusions from this study verifies this fact:
1. Ternary WMF-microsilica mortar mixes have same abrasion resistance (just 2.5% more) alike reference
mortar of PQC and are thus suitable for rigid pavements. Even binary WMF mortar mixes could be used for
yielding abrasion resistant high flow concrete.
2. Drying shrinkage of ternary WMF-microsilica mixes lies between low to moderately low rate as per
restrained shrinkage test results. Binary WMF mixes show moderately low shrinkage. Hence WMF is a
suitable fiber for reducing the shrinkage rate of concrete. Flyash on the other hand gave moderately low
shrinkage rates is not sufficiently capable to reduce drying shrinkage rate of concrete with or without the
presence of microsilica. This means that even microsilica cannot complement flyash in controlling drying
shrinkage.
3. Fatigue life of pavement which depends upon compressive toughness could be enhanced by nearly 10% by
using ternary WMF-microsilica mixes. Though alone WMF could also bring an increase of nearly 7% in the
fatigue life. Flyash cannot increase the fatigue life of pavement even under prolonged curing as there was a
big difference in compressive toughness of all flyash mixes with respect to reference concrete after 28 days
curing.
20
368
4. Flyash alone reduces E value by half times that of reference concrete. Its only along with microsilica that
flyash is able to compensate the loss in E value and bring the overall loss to 27%, which is similar to that of
ternary WMF-microsilica mixes.
5. WMF is capable enough to play the role of a microfiber. This is apparent from the fact that even after
substituting cement it reduced the flexural strength of concrete by only 3%. When complemented by
microsilica it increased the flexural strength by 6% with respect to normal concrete after 90 days curing.
6. WMF acts both as a microfiber and a pozzolan. The increment in compressive strength of concrete by 1.6%
after 90 days of curing proves this fact. Had it been acting alone as a microfiber, it could not have been able
to increase the strength with time. Also flyash-microsilica combination is capable enough of giving good
strength (just 6.4% reduction).
7. SCC is achievable for WMF reinforced concrete mixes too. Surprisingly flyash admixed mixes could not
yield more combinations than CFS5 & CFS6 which have 20% flyash and 2.5-5% microsilica. WMF alone
can’t yield SCC passing of mixes CWS2, CWS3, CWS6, CWS7 prove that 10-20% WMF along with 2.5-
7.5% microsilica could yield a better SCC.
Overall, it could be concluded that 20% WMF along with 5-7.5% microsilica could be used to yield a
shrinkage mitigating high strength concrete. Though, both compressive and flexural increase but flexural
strength increases more in comparison to compressive strength. Also it provides stiffness to concrete and
does not allow anticipated heavy reduction of modulus of elasticity on account of increase in volume of
mortar. Furthermore abrasion resistance of WMF-microsilica mixes is nearly similar to that of normal PQC.
Hence its use as PQC is recommended.
References
1. Banthia, N., and Gupta, R. (2006). “Influence of polypropylene fiber geometry on plastic shrinkage
cracking in concrete.” J. of Cement and Concrete Research, 36(7), 1263-1267.
2. Betancourt, G.H. (1988). “Admixtures, workability, vibration and segregation.” Journal of Materials
and Structures, 21 (124), 286-288.
3. Bouzoubaa, N., and Lachemi, M. (2001). “Self-compacting concrete incorporating high volumes of
Class F fly ash preliminary results.” Cement and Concrete Research, 31, 413-420.
4. Clark, P. (1998). “Future of automotive body materials: Steel, Aluminum & Polymer Corporation.”
Massachusetts Institute of Technology, < http://readpdf.net/file/future-steel-vehicle.html>.
5. Fwa, T. F., and Paramasivam, P. (1990). “Thin steel fibre cement mortar overlay for concrete
pavement.” Cement & Concrete Composites, 12, 175-184.
21
369
6. Grunewald, S., and Walraven, J. C. (2001). “Rheological study on the workability of fiber reinforced
mortar.” Proceedings of the Second International Symposium on Self Compacting Concrete, Tokyo,
Japan, 127-136.
7. Kaushik, S.K., Gupta, V.K., and Gupta, S. (1994). “Properties of fresh and hardened SFRC.
Proceedings of the National Seminar on Fiber reinforced Comentitious Products, Roorkee, India,
January, 247-257.
8. Khan, T.A.H., Laid, S.M., and Ramakrishna, B. (1987). “Experimental study of SFRC under
compression, indirect and pure bending.” Proceedings of the international symposium on fiber
reinforced concrete, Madras, India, 2.39-2.48.
9. Khayat, H.K., and Aitcin, P.C. (1992). “Silica fume in concrete- an overview.” Flyash, Silica fume,
Slag and Natural Pozzolans in Concrete, Ed. V.M. Malhotra, ACI SP-132, Detroit, Michigan, 2, 835-
72.
10. Kovler, K., Sikuler, J., and Bentur, A. (1992). “Free and restrained shrinkage of fibre reinforced
concrete with low polypropylene fibre content at early age.” Fourth Rilem Int. Symp. on Fibre
Reinforced Cement and Concrete, International Union of Laboratories and Experts in Construction
Materials, Systems and Structures, Sheffield, UK, 91–101.
11. Krishnamoorthy, T.S., and Kumar, B.H. (1994). “Influence of steel fibers on the compression and
flexural behavior of concrete.” Proceedings of the National Seminar on fiber Reinforced Cementitious
Products, Roorkee, India, Januray, 25-32.
12. Kukreja, C.B., and Chawla, S. (1989). “Flexural characteristics of steel fiber reinforced concrete.” The
Indian Concrete Journal, March, 154-157.
13. Langan, B.W., Weng, K., and Ward, M.A. (2002). “Effect of silica fume and Fly ash on heat of
hydration of portland cement.” Cement and Concrete Research, 32(7), 1045–1051.
14. Li, V. C., Wu, H. C., and Chan, Y. W. (1996). “Effect of plasma treatment of polyethylene fibers on
interface and cementitious composites properties.” J. American Ceramic Soc., 79(3), 700–704.
15. Lomboy, G., Wang, K., and Ouyang, C. (2011). “Shrinkage and Fracture Properties of Semiflowable
Self-Consolidating Concrete.” J. Mater. Civil Engineering- Energy Efficient and Environmentally
Friendly Paving Materials, 23, Special Issue, 1514-1524.
16. Low, N.M.P., and Beaudoin, J. J. (1992). “Mechanical properties of high performance cement binders
reinforced with Wollastonite micro-fibres.” Cement and Concrete Research, 22(5), 981-989.
17. Naaman, A. E., and Shah, S. P. (1976). “Pull-out mechanism in steel fiber-reinforced concrete.” J.
Struct. Div., ASCE, 102(8), 1537–1548.
18. Otter, D.E., and Naaman, A.E. (1987). “Strain rate effects on the compressive properties of fiber
reinforced concrete.” Proceedings of the international symposium on fiber reinforced concrete, Madras,
India, 2.225-2.235.
19. Parra, C., Valcuende, M., and Gomez, F. (2011). “Splitting tensile strength and modulus of elasticity of
self-compacting concrete.” Construction and Building Materials, 25(1), 201-207.
20. Pfyl, T., and Marti, P. (2001). “Versuche an stahlfaserverstärkten Stahlbetonelementen.” Institute for
Structural Engineering, ETH Zürich, Zürich, Switzerland.
22
370
21. Ramakrishnan, R. (1993). “Recent advancement in concrete fiber composites.” Proceedings of the
International Symposium on Innovative World of Concrete (ICI-IWC-93), Madras, India, 163-192.
22. Saje, D., Bandelj, B., Sustersic, J., Lopatic, J. and Saje, F. (2011). “Shrinkage of polypropylene fiber-
reinforced high-performance concrete.” Journal of Materials in Civil Engineering, 23(7), 941–952.
23. Sasturkar, P.J., and Kaushik, S.K. (1994). “Flexural behavior of fiber reinforced concrete beams.”
Proceedings of the National Seminar on Fiber Reinforced Cementitious Products, 177-183.
24. Soliman, A.M., and Nehdi, M.L. (2012). “Effect of natural Wollastonite microfibers on early-age
behavior of UHPC”, Journal of Materials in Civil Engineering, 24(7), 816–824.
25. Swamy, R. N., and Stavrides, H. (1979). “Influence of fibre reinforcement on restrained shrinkage and
cracking.” J. of American Concrete Inst., 76(3), 443–460.
26. Thomas, M.D.A., Shehata, M.H., Shashiprakash, S.G., Hopkins, D.S. and Cailb, K. (1999). “Use of
ternary cementitious systems containing silica fume and Flyash in concrete.” Cement and Concrete
Research, 29, 1207–1214.
27. Turcry, P., Loukili, A., Haidar, K., Pijaudier-Cabot, G., and Belarbi, A. (2006). “Cracking tendency of
self-compacting concrete subjected to restrained shrinkage: experimental study and modeling.” Journal
of Materials in Civil Engineering, 18(1), 46–54.
28. Wittman, F.H. (1976) “On the action of capillary pressure”, Cement and Concrete Research, 6(1), 49-
56.
29. Zhu, W., and Bartos, P.J.M. (2003). “Permeation properties of self-compacting concrete.” Cement and
Concrete Research, 33(6), 921–926.
30. Zollo, R. F. (1984). “Collated fibrillated polypropylene fibres in Fibre Reinforced Concrete.”,
American Concrete Institute, Detroit, ACI SP-81, 397–409.
31. Zollo, R. F., Ilter, J. A., and Bouchacourt, G. B. (1986). “Plastic and drying shrinkage in concrete
containing collated fibrillated polypropylene fibres.” 3rd Int. Symp. on Developments in Fibre, 86(1),
RILEM Technical Committee 49-TFR, Cachan, France.
23
371
Paper ID- 86
Wet State Analysis of WMF Reinforced & Flyash Admixed Cement
Composites
Shashi Kant Sharma1*, Mohit Kumar1
Deptt. of Civil Engineering, NIT Jalandhar, Jalandhar 144011, India.
Abstract
Present study deals with the comparison of cement composites admixed with naturally available cheap microfiber-
wollastonite micro fiber (WMF), and abundantly used industrial waste material- flyash. Microsilica has also been
utilised in traces to check the mixes for any improvement. Cement composites have been checked for water
demand, setting times and compatibility with polycarboxylate ether (PCE) based superplasticizer, workability and
shrinkage resistance. These parameters determine the feasibility of producing flowable or self compacting concrete
from the admixed binder material.
The results obtained from the tests show, that inspite of having high water demand (17% higher than flyash’s
mixes), WMF admixed pastes have comparable setting times to flyash’s mixes. At their normal consistencies (low
water contents), flyash admixed pastes show higher compatibility with superplasticizer, whereas for higher water
contents both flyash and WMF admixed pastes show equal compatibility. Workability tests to check the self
compacting behaviour of concrete mixes show that WMF is highly efficient in achieving flowable homogeneous
concrete. This shows that, even though flyash admixed simple concrete is better, but WMF (upto 20% WMF)
admixed concrete has a good potential to serve as a self compacting concrete, with only a small increment in water
demand. Drying shrinkage is very low (reduced by 49%) for WMF mixes in comparison to normal concrete mix.
Keywords: Cheap materials; Pozzolans; Fibers; Compatibility; Flowable; Self compacting concrete
1 Introduction
The use of fine materials in cement composites for the development of high flowing durable mix is performed for
a long time. The fine materials which have been used are mainly pozzolanic materials [3]. Flyash, microsilica and
slag are some famous artificial pozzolanic materials which are obtained as a by product of industries. Their use in
concrete production reduce pollution, apart from improving the strength and durability. Flyash is obtained as a
waste material from the thermal power plants. ASTM C 618 categorizes it into two types: Class C (high calcium)
and Class F (low calcium) flyash. Flyash, as a cement substitute, converts portlandite produced by the hydration of
calcium silicates present in cement grains, into calcium silicate hydrate C-S-H. Thus it substitutes cement, and on
372
the other hand converts the expansive and less durable portlandite into strength producing product. Microsilica is a
by product from the reduction of high purity quartz with coal in electric arc furnaces during the manufacture of
ferrosilicon alloys and silicon metal [14]. Qing et al. [15] found that microsilica decreases the permeability of
concrete and thus, also increases the corrosion resistance or durability of concrete. Kocak [9] investigated the
effect of flyash and microsilica substitution in cement paste and mortars. It was found, that setting time of cement
paste increases linearly with flyash and microsilica substitution. It was concluded, that flyash and microsilica slow
down hydration speed, decrease hydration heat and temperature, thus, are effective against contraction by
increasing the setting time. Low specific weights of microsilica and flyash causes a reduction in the specific
weight of flyash and microsilica substituted cements.
Bouzoubaa and Lachemi [1] investigated the feasibility of producing SCC with high volumes of Class F flyash
(40-60%). Nine SCC mixtures and one control concrete were studied at constant cement content (400 kg/m3) and
the water/cementitious material ratios ranged from 0.35 to 0.45. The 28 days compressive strength values ranged
from 26-48 MPa which were quite lesser than normal concrete but it indicated that economical medium strength
SCC could be made by substituting cement with high volumes of flyash. Lomboy et al. [11] investigated the
differences in the strength and shrinkage properties of semi flowable self compacting concrete (SFSCC) and
ordinary pavement concrete containing 20% Class F flyash. Compressive strength, splitting tensile strength,
modulus of elasticity and fracture strength of both the concretes were evaluated at 1,3,7,14 and 28 days and were
found to have negligible differences except modulus of elasticity, which reduced for SFSCC. Also shrinkage
induced cracking was found higher in SCC. SCC mixtures shrink on water loss either due to surface evaporation
or temperature gradient due to hydration (temperature difference in top and middle layers of mixes). Therefore
fiber addition is brought into SCC mixes. Hossain et al. [7] studied the influences of fiber types/size/dosages and
fiber combinations on workability and rheological properties. The maximum dosage of PVA is limited to 0.125%
compared to 0.3% of metallic ones in developed FRSCC mixtures due to PVA’s higher workability
reduction/viscosity increase capability. Study indicated that a homogenous fiber distribution can be obtained up to
a critical fiber content and once that is surpassed, a stiff structure of the granular skeleton make self compaction
impossible as has been proved in the study by Grunewald & Walraven [6]. The optimization of fresh and
rheological properties of FRSCC is essential in order to ensure better mechanical and durability characteristics in
the hardened state. It is important for FRSCC to have low yield stress and optimal viscosity to ensure required
373
flowability and to prevent segregation. This study makes an initial effort to find out the effect of Wollastonite
micro fiber in wet state of cement composites. Also restrained shrinkage test was performed to check the
improvement, if any brought by these fibers in the concrete since pavements are prone to cracking at initial stages
of curing.
2 Materials and Methods
2.1 Materials Used
Ordinary Portland cement (OPC) 43 grade conforming to Indian standard code IS 8112–1995 was used. Graded
river sand conforming to Zone–II having fineness modulus of 3.23 and specific gravity of 2.58 was used as fine
aggregate. Crushed graded coarse aggregates of 20 mm and 10 mm conforming to IS 383-1970 were used in the
proportion of 60:40. The specific gravity of 20 mm and 10 mm aggregates were 2.62 and 2.58 respectively.
Fine amorphous wollastonite powder supplied by Rajwara Stonex Limited was used. It had a specific surface and
specific gravity values of 827m2/kg and 2.9 respectively. Densified 920D grade Microsilica supplied by India
Private Limited was used. It had a specific surface and specifc gravity value of 18000 m2/kg and 2.05. Medium
lime Class F Flyash was obtained from National Thermal Power Plant Ghaziabad and was found to have specific
surface and specific gravity values of 350 m2/kg and 2.52. High water reducing poly carboxylate ether (PCE)
based superplasticizer was used to introduce self compacting workability conditions to the concrete.
Particle size analysis was performed for WMF, cement, fly ash and microsilica using Ankersmid laser
based analyzer. Figure 1 shows graphically, the particle size distribution of various materials. Specific surface of
OPC was found by Blaine’s method, and that of WMF, flyash and microsilica was found by BET method.
Chemical composition of OPC, WMF, microsilica and flyash was analyzed by X-ray fluorescence spectrometer in
accordance with IS 12803. Table 1 represents the results obtained from this test. Further the mixes were subjected
to SEM image analysis.
374
Fig. 1. Percentage Sizes of various particles
Table 1. Chemical property of Cementitious materials including OPC
Compound Cement Flyash WMF Microsilica

SiO2 20.2 35 48 92.9
Al2O3 5.2 26 1.4 0.9
Fe2O3 3 8.7 0.6 0.72
MgO 1.51 5 0.2 0.57
SO3 2.2 3 - 0.16
Na2O 0.08 1.5 - 0.32
Chloride 0.014 0.005 - 0.037
CaO 62.9 15.3 45.9 1.4
K2O 0.6 0.5 - 0.4
2.2 Mixture Proportions
Cement has been substituted with either WMF or flyash @ 10-30%, and with microsilica @ 2.5-10%.
Binary mixes have been designated as CW1-CW3 and CF1-CF3 for three percentage substitution (10%-30%) with
wollastonite, and flyash respectively. Ternary mixes have been designated as CWS1-CWS12 and CFS1-CFS12
with three sets of four microsilica substitutions each as shown in Table 2.
Table 2. Mixes chosen for the tests and their material composition
Mix Designation Percentage of material in mix Paste Composition (%age)
Cement Flyash Wollastonite micro Microsilica Silica C+F/WMF+S
C 100 - - - 100
W - 100 - 100
F - 100 - - 100
M0 - - - 100 100
CW1 90 - 10 - 90+10
CW2 80 - 20 - 80+20
CW3 70 - 30 - 70+30
375
CWS1 87.5 - 10 2.5 87.5+10+2.5

CWS2 85 - 10 5 85+10+5
CWS3 82.5 - 10 7.5 82.5+10+7.5
CWS4 80 - 10 10 80+10+10
CWS5 77.5 - 20 2.5 77.5+20+2.5
CWS6 75 - 20 5 75+20+5
CWS7 72.5 - 20 7.5 72.5+20+7.5
CWS8 70 - 20 10 70+20+10
CWS9 67.5 - 30 2.5 67.5+30+2.5
CWS10 65 - 30 5 65+30+5
CWS11 62.5 - 30 7.5 62.5+30+7.5
CWS12 60 - 30 10 60+30+10
CF1 90 10 - - 90+10
CF2 80 20 - - 80+20
CF3 70 30 - - 70+30
CFS1 87.5 10 - 2.5 87.5+10+2.5
CFS2 85 10 - 5 85+10+5
CFS3 82.5 10 - 7.5 82.5+10+7.5
CFS4 80 10 - 10 80+10+10
CFS5 77.5 20 - 2.5 77.5+20+2.5
CFS6 75 20 - 5 75+20+5
CFS7 72.5 20 - 7.5 72.5+20+7.5
CFS8 70 20 - 10 70+20+10
CFS9 67.5 30 - 2.5 67.5+30+2.5
CFS10 65 30 - 5 65+30+5
CFS11 62.5 30 - 7.5 62.5+30+7.5
CFS12 60 30 - 10 60+30+10
Table 3 shows the mix proportions of the normal concrete mix of 4.5 MPa prepared in accordance with the specifications laid
down in IRC 44-2008. Subsequent self compacting concrete mixes have been prepared by sequentially performing the
following steps: substituting the binder material, addition of superplasticizer and changing the coarse aggregate to fine
aggregate content of the normal concrete mix
Table 3. Mix design of normal PQC
Mix design for Normal M-40 concrete used to derive SCC for rehabilitation as per IRC 44
1 1.58 2.35 0.36
20mm 10mm CA:FA=60:40
740 317 C.A 20mm:10mm=70:30
2.3 Tests on Pastes
Normal consistency test was performed on the paste mixes, which provides a measure of water demand of the
mixes to resist a specified pressure. IS: 4031 (Part 4) guidelines were followed for determining the normal
376
consistency of cement. Initial setting time and final setting time were determined as per IS 4031 (Part 5). To
evaluate the compatibility of cement & admixed slurries with superplazticizers, Marsh cone test was performed
according to ASTM D 6910 at three water to cementititous material ratios 0.3, 0.35 and 0.4. The test measures the
viscosity of slurry mix (946 ml) in the form of time (seconds) taken to flow from Marsh funnel into a graduated
cup. This test enables to ascertain the optimum dosage of a superplasticizer for cement slurry. Since a
superplasticizer works mainly on the cement slurry, the predominance of supplementary cementitious materials on
slurry flow at their higher content was expected and had to be checked.
2.4 Tests on Concrete Mixes
Fresh state tests i.e. Abram’s flow test, V funnel test, J ring test and Probe ring test were performed on concrete to
check its workability as a self compacting concrete. Abrams flow and V Funnel test measured the flowability; J
Ring test measured the passability of SCC, and Probe Ring test was performed to find the segregation resistance of
fresh concrete. In the present study, only J ring test has been performed to measure the passability of concrete.
EFNARC [4] guidelines have been followed to check the properties of SCC as per the test requirements. Further
drying shrinkage was tested as per ASTM C1581 in a hollow cylindrical steel ring apparatus.
3.1 Physical and Chemical Properties of Materials
Figure 1 clearly illustrates that microsilica is finest among all considered powdery materials followed by WMF,
fly ash and cement respectively. The peak of WMF lies exactly in between the peaks of microsilica and OPC. This
interpretation clearly infers that WMF used was median size to both microsilica and OPC, and hence, an excellent
interlocking within these particles is anticipated physically. It is also clearly depicted that OPC used for the study,
exhibits particle sizes comparable to flyash as revealed by the presence of secondary peak in Figure1. Peak
patterns analysis suggests that fly ash and OPC have nearly same size range, but from the prolonged post peak
profile of fly ash it is learnt, that there are numerous fraction of flyash which are even larger in size than OPC
particles. From this analysis, it is learnt that microsilica has particle size ranging from 0.01-0.5 microns and about
80 percent of WMF particles were in the range of 0.5-4.47 microns. About 80 percent of OPC and flyash particles
are greater than 4.47 microns but if one precisely looks at particular size range i.e. upto 0.9 microns, OPC particle
377
is much finer than flyash. The findings strongly suggest that larger fraction of flyash having size greater than OPC
may retard the reaction rate as well as may lower density of mix. However, a fair chance for improvement in
workability is speculated due to higher rate of dispersion of cement particles on account of high negative charge
carried by their larger sizes. On other hand, ultra-fine nature of microsilica, WMF and fly ash would enhance the
pore infilling capacity significantly.
Results obtained from Blaine’s air permeability test on cement, and BET permeability test on WMF,
flyash and microsilica prove, that microsilica is the finest among all, followed by WMF, flyash and OPC
respectively. If one compares the degree of fineness of microsilica, WMF and fly ash with OPC, they are in the
order; 60, 2.8 and 1.3 times finer than OPC respectively. Results strongly hint that microsilica would be
prominently proactive in comparison to the rest as it has inherent ability to contribute strength development
through its pozzolanic reactivity. BET’s study result strongly suggests that WMF is fine enough to modify the
properties of concrete in its fresh or hardened states. The inclusion of WMF would have significant physical effect
of modifying the flocculation of cement and the median particle size being smaller, presence of WMF would infill
micro pores and voids of the cement system better as a result of which coefficient of permeability of cement
matrix would reduce significantly. Fibrous mineral wollastonite micro-fiber has been used for development of
high performance cement composites offering economic benefits over steel or carbon micro-fibers [12, 13]. The
fly ash used in the present study has specific surface area very close to OPC but the influence of fly ash on the
properties of fresh concrete is linked to the spherical shape of the fly ash particles which provide lubrication and
deflocculate cement particles due to negative charge over them; this being called as “ball-bearing effect” [2].
The specific gravity results clearly show, that Microsilica offers the lowest value of specific gravity
followed by fly ash, wollastonite micro-fiber and OPC respectively. The specific gravity of admixtures affects the
flowability of fresh concrete, along with their viscosity in the paste form. Hence, it could be said that mixes
prepared with microsilica would have tendency to achieve better flow, followed by flyash, WMF and plain OPC
respectively.
Table 1 shows the quantitative results of the amount of oxides present in cement and other admixtures, as
has been found through X ray fluorescence spectrometer test when conducted in accordance with IS: 12803.
The following are the few observations drawn from the study of chemical compositions:
378
(i) Fly ash used in the present study contains appreciable amount of silica and highest amount of alumina
when compare to the rest of the materials. By the presence of substantial amount of lime in it, the
aforesaid lime can be said to belong to the category of medium lime fly ash and would liberate moderate
heat of hydration. Higher content of silica strongly indicates good potential to reduce formations of CH
and ettringite.
(ii) WMF shares equal amounts of lime & silica, and traces of alumina. Their presence indicates that WMF
has tendency for self cementation & moderate rate of reaction. Silica in WMF also makes it possible for
the later to reduce ettringite and CH content of the mix. By virtue of its crystalline nature, WMF is more
inert than been reactive and served as excellent pore filler.
(iii) Microsilica mainly consists of silica and rest other oxides are in meagre amount. In its amorphous state,
microsilica is very fine particle and possesses higher surface areas. By virtue of its microfine nature,
microsilica is a promising material for infilling micro pores and apt for reduction of hydrated cement
compounds like CH and ettringite considerably.
3.2 Normal Consistency
Normal consistency is related with water requirement which could make the paste workable. For pure cement it
depends upon the quantities of compounds responsible for the hydration, as well as the fineness of cement
particles. Higher the rate of hydration of C3A in cement, higher would be the formation of ettringite crystals; thus
causing reduction in workability. With the introduction of admixtures, the fineness of binder powder further
increases. This increase depends upon the particle size (surface area) as well as quantity of admixtures. The
surface area of admixture particle, affects the water required for lubrication of mixes due to adsorption at their
surfaces’, which consumes water, and also filling of pores between cement particles, which releases water. The
release of water is more if the admixture is inert and it would be less if it is reactive enough to cause instantaneous
hydration whilst filling the pores. As has been observed during the laboratory study, flyash has been found to
achieve normal consistency at a water content of 30.5% and 30.7%, with and without microsilica respectively.
WMF has been found to achieve the same at 36.9% and 34.7% respectively, whereas in case of combined mixes
the water demand is more i.e. 39.1% and 37.1% respectively. This shows that flyash is both coarser and inert
enough to cause enough water release at lower water contents, to form a paste of normal consistency. Overall there
379
is no major effect of addition of flyash on flyash admixed pastes, but as has been observed singularly, minute
increment in water demand has been observed with microsilica substitution for each flyash content in these mixes.
Microsilica, increase the hydration rate due to its ultra fine glassy silicious particles which penetrate into the
minute capillary voids very easily, release water and consume it instantaneously to cause hydration, thereby
increasing the stiffness of the paste. Hence additional water is required to provide normal consistency to the stiffer
pastes. WMF has led to an increment in water demand due to three possible reasons: (i) due to pozzolanic action
on account of possible glassy fine silicious material in it, (ii) on account of reactivity due to calcium oxide present
in it and (iii) highly viscous nature of WMF due to it its acicular structure. Flyash further increase water demand in
WMF admixed mixes which has been seen on an overall & singular scale. In comparison to flyash, both WMF and
microsilica makes the paste stiffer by reacting at first instance.
Fig. 2 shows the general trend of normal consistencies of cement pastes with or without mineral
admixtures for singular, binary and ternary mixes.
50.0
45.0
40.0
35.0
% Water content
30.0
25.0
20.0
15.0
10.0
5.0
0.0
CF1
CF2
CF3
CFS1
CFS2
CFS3
CFS4
CFS5
CFS6
CFS7
CFS8
CFS9
C
W
CWS10
CWS11
CWS12
F
CFS10
CFS11
CFS12
M0
CW1
CW2
CW3
CWS1
CWS2
CWS3
CWS4
CWS5
CWS6
CWS7
CWS8
CWS9
Fig. 2. Normal consistency values for singular, binary and ternary mixes
WMF though, is a fine material and releases water by grain size refinement, but by virtue of its acicular structure,
it showed tendency to increase the friction to flow. Water releasing tendency is more pronounce when cement was
partly replaced by wollastonite in the range of 10-20% by weight of cement. This tendency is more active in
comparison to friction inducing tendency. Beyond 20% replacement level, water demand is substantial for
achieving normal consistency. This is attributed to extra water absorbed by the WMF particles coating around the
sphere of the cement grains. Figs. 2 indicates that incorporation of fly ash upto 20% by weight produced higher
normal consistency values (32 & 31) in comparison to cement’s consistency but at 30% part replacement level,
this value (29) found to be even less than that of cement’s consistency (30.5). The trend of reduction in normal
380
consistency is approximately linear with the percentage flyash replacement level as can be seen from Figure 2. The
reduction in normal consistency is mainly contributed by the presence of bigger particle size of flyash and by
virtue of its less moisture attraction characteristics.
For microsilica addition in the range of 0-5%, the ball bearing effect of microsilica small spherical
particles tend to increase the workability of the mix by reducing the friction between cement particles. Also, the
tendency for water release due to pore size and grain size refinement caused by microsilica, releases more water
into the mix and reduces the friction respectively. But due to its larger surface area, microsilica beyond 5% starts
making a thicker paste around cement particles and increases the viscosity at a higher rate.
3.3 Initial Setting Time and Final Setting Time
Present study clearly revealed that incorporation of wollastonite microfiber into cement paste increases both initial
and final setting times apart from increase in consistency value. With substitution of WMF, flyash and microsilica,
initial & final setting times of the pastes increased. This is due to the increment in the dormant period with respect
to the plain cement, brought by all admixtures. Overall, the initial setting time values increased by 21.1% &
109.4%, 18.7% & 104.6%, and 10.8% & 110.2%, when WMF and flyash substituted cement, with or without
microsilica. Similarly, on an average, the final setting times increased by 25.5% & 84.3%, 27.1% & 95.9% and
16.3% & 92.3% when WMF and flyash substituted cement, with or without microsilica.
Though, with admixtures, there was an increment in setting times, but if the effect of percentage change
in setting times with percentage change in admixtures’ quantities is observed, it was found that there a relative
increment in initial setting time values was brought by all admixtures with percent increment in their quantities.
But, only flyash increased the final setting time, with an increment in its quantity. These findings are in the line
with study conducted by Soliman and Nehdi [16]; they found that incorporation of WMF into cement pastes
increase both initial and final setting times by reducing the number of active sites for hydration products to form
along with reducing the cement content by its replacement, which complements the reduction in hydration rate.
For flyash, at lower water/cement ratio, the retardation is less and vice versa. Microsilica according to
literature prolongs hydration at both dormant and acceleration period, at lower water/cement ratios [10]. At higher
ratios it is found to decrease the dormant period and initiate the acceleration period at early stage, thereby reducing
10
381
the time when acceleration period ends. Hence it could be assumed that microsilica reduces both of the setting
times i.e. initial and final. Even though it increases the release of water but loss of hydration is large enough to
reduce the initial setting times and final setting times of the paste. On contrary inclusion of flyash into the cement
paste reduced final setting time but increases initial setting times. The results have been shown in Figure 3.
Initial Setting Time Final Setting Time

500
450
Setting time in minutes
400
350
300
250
200
150
100
50
0
Fig. 3. Setting time of pastes
Microsilica is the finest, and thus it gets adsorbed more easily and completely over the cement particles,
therefore microsilica reduces the hydration rate to the highest extent, and this trend is followed by WMF and
flyash whose surface areas vary accordingly. All of the admixtures increase the dormant period because dormant
period is influenced by the breaking of initially hydrated layer from the surface of cement particles, and if the
adsorbed admixtures on the surface hinder it, then cement particles will not hydrate even if there is sufficient
amount of water available for hydration to take place. When the content of microsilica is higher than 7.5%, there is
a reduction in the increment rate of initial setting time. This is because, microsilica up to 7.5% by mass of mix is
sufficient to cover the cement particles by adsorption. Over and above this percentage, microsilica would only
increase the thickness of inter-crystal paste between cement particles which has minimal effect on the dissolution
of cement crystals form its surface, though they indirectly reduce the spaces for hydration products to form and
affects the final setting time to a higher extent. Hence after 7.5% of microsilica, there is reduction in the increment
rate of initial setting time.
The final setting time corresponds to the time duration between the instant, the water was added and
initial hydration started to the time when the acceleration period ended. Literature studies state that microsilica in
presence of enough water reduces its retarding behaviour and the acceleration period ends early, whereas flyash
11
382
increase its retardation at higher water contents. At the end of dormant period the water level of the mix is highest
because of delayed hydration, and when the film of hydrates break on the cement particles then the admixtures
start reacting with the lime so formed after hydration of new cement compounds. Microsilica has the highest
reactivity followed by WMF and flyash because of their surface areas. Hence it is obvious that the acceleration
period of microsilica would end earlier than WMF, followed by flyash. More is the admixture present more would
be the reaction rate, but is has been observed that flyash increases the final setting times with the rise in its content.
This is because flyash because of low surface area and lower potential to reduce lime did not enter into the voids
between cement particles fully. Though with the increment in its content the fraction lower than 5 microns
increased inside the cement voids but a simultaneous larger increment in fraction greater than 5 micron takes place
in the paste which causes dilution of the cement. The increased secondary hydration by flyash due to its increased
content in cement voids is much lesser than the reduction in hydration caused by the relative reduction of cement.
For WMF and microsilica this is not so, because of their higher reaction rates. Microsilica and WMF have nearly
47.3 and 2.2 times more surface area than flyash which explains their higher hydration rate than the dilution
caused by them.
3.4 Compatibility with Superplasticizer
Due to the necessary use of superplasticizer for making a medium strength self compacting concrete, it was
necessary to check the compatibility of superplasticizer with cement pastes containing admixtures. The admixtures
present in the paste led to a change in proportions of C3A in the powder material.
It was found, that whether PCE based superplasticizer is mixed initially or after mixing of water, there is
no significant difference in the behaviour of PCE, which suggests, that cement contains good ratio of aluminates
and sulphates in it. This is proved by the X ray fluorescence test, which gives the amount of aluminates and
sulphates in the cement as 5.2 and 2.2 percent respectively. The flyash introduction, which contained 26 percent
aluminates and 3% sulphates, also did not alter the superplasticizer behaviour. This suggests that the aluminates
present in flyash are not highly reactive and have a higher inert crystalline content. Tables 4 shows the data
obtained from testing of the pastes at different W/C ratios (0.3, 0.35 & 0.4), with different contents of
superplasticizer added sequentially to the mixes, to check their flow and also indicates the optimum flow time in
red colour.
12
383
Following paragraphs entail the results obtained, when water was added along with superplasticizer into the
powder material. The analysis has been done by observing the effect of admixture type and content on the flow
time of pastes at different W/C ratios and superplasticizer content.
(i) At W/C=0.3
The water content is low and therefore the activity of microsilica and WMF is retarded. This retardation is not
similar to the retardation in hydration caused by the admixtures at dormant period; it is related to time period just
after the addition of water. This retardation effect is lower for lower contents of these admixtures and gets
amplified at their larger contents. Therefore both microsilica and WMF at their lower content did not get adsorb on
the cement particles, thereby showing a raised water demand. This demand kept on increasing with their higher
contents due to retardation, and owing to more water consumption by their particles’ surfaces. Hence, flow time
shows a small increment with microsilica and WMF addition at their lower content, and at their higher contents,
flow time shows a high increment.
Flyash mixes have normal consistency in this range because of the workability created by the repulsive forces
of adsorbed particles of flyash carrying negatively charged sulphate ions, which repel the cement particles far
apart, thus releasing more water [3]. Therefore no retardation was observed in the effects of flyash, which kept on
reducing the water demand at its increased content, by filling the voids in cement particles. The reason for
continuous reduction in water demand by flyash is the low reaction tendency of flyash [10]. The higher fraction of
large sized flyash particles in the paste, further reduced the water demand due to dilution. The aluminates present
in the flyash could have done two functions depending upon their reactivity: firstly they could have consumed the
superplasticizer, and secondly, converted the ettringite and lime into monosulphates and C-S-H respectively. This
somehow should have reduced the water content in pore solution. It is not sure that up to what extent the
aluminates were reactive, but overall, there is a decrement in the flow time by the use of flyash.
As far as increment in superplasticizer is concerned, it has been observed that fine admixtures start behaving
normally with an increment in effective water content caused by superplasticizer. Thus, for WMF and microsilica,
flow time reduced up to 20% WMF and 5% microsilica content and thereafter it increases due to the water demand
created by excess WMF and microsilica, as their further adsorption was not possible on cement particles beyond
13
384
that content. No behaviour change was observed in the flyash mixes with an increment in superplasticizer because
the behaviour of flyash did not vary much with a small increment in the effective water content of mix.
Initially whatever may be the effect of admixture but at higher superplasticizer contents there is enough water
in the paste and the paste behaviour becomes independent of the behaviour of its constituents. This generally
happens at superplasticizer contents above 1.5%, 1.2%, and 0.9% for water /cementitious material ratio of 0.3,
0.35, and 0.4, for both pure cement paste and admixed pastes.
(ii) At W/C=0.35
The water content is high for flyash admixed pastes, whereas it approaches normal consistency for pastes
containing lower contents of WMF and microsilica.
Therefore flyash was observed to release entrapped water at a higher rate, at its lower contents and as the
flyash content increased, the rate decreased abruptly due to no further increment in adsorption of flyash at cement
particles’ surfaces. Hence, at medium W/C ratios, flyash admixed pastes were observed to increase flow up to 20%
content and then it remained constant.
WMF and microsilica acceleration effect was enhanced at this stage. WMF and microsilica mixes released
more water at their lower content in comparison to their water consumption by adsorption of water at their
respective surfaces. Hence they gave lower flow times. At higher content, WMF and microsilica were found to
increase the flow time as their further adsorption on cement particles reduced. This increment in flow time was
lower in comparison to same mixes at lower W/C ratio (0.3%), which is obvious. Hence, for pastes having higher
content of WMF and microsilica, the W/C ratio of 0.35 is low, and they behave as being retarded.
(iii) At W/C=0.4
The water content is high enough for flyash admixed pastes, whereas it approaches normal consistency for
WMF and microsilica admixed pastes, especially those containing higher content of WMF and microsilica. Same
effect for flyash was observed at this W/C ratio as was observed at 0.35% W/C. For WMF-microsilica mixes the
effect of WMF and microsilica was unhindered up to 20% content of WMF and up to 5% content of microsilica,
after which it registered a small decrement. This happened because, even though the amount of water in the pore
14
385
solution is higher for the free movement of WMF and microsilica particles, but there is a limit beyond which all
the surfaces of cement particles get engulfed with admixtures and no further adsorption is possible.
Also, the average flow times of flyash and WMF admixed pastes corresponding to superplasticizer contents of
1.5% (at W/CM=0.3), 1.2% (at W/CM=0.35) & 1.2% (at W/CM=0.4) are 82.36%, 109.94% & 46% and 166.38%,
122.72% & 81.26% respectively, of plain cement’s flow times at low, medium and high water contents.
Hence a water content of 0.35 % is favourable in this case for achieving flowable pastes. Since for a SCC, the
volume of mortar is nearly two third of the volume of concrete, therefore a water content of nearly 0.35% would
be better for achieving a SCC.
Table 4. Flow time (seconds) variations of ternary slurry mixes

0.6 0.9 1.2 1.5 1.8 0.6 0.9 1.2 1.5 0.3 0.6 0.9 1.2 1.5
Mix Flow at W/C=0.3 Flow at W/C=0.35 Flow at W/C =0.4
C 938 358 235 200 184 900 400 228 218 850 310 213 196 190
CW1 373 280 218 188 175 280 266 215 210 229 201 192 189 186
CW2 704 595 525 452 410 540 195 150 135 210 165 95 80 77
CW3 878 736 611 531 485 588 525 441 433 270 210 189 183 180
CF1 316 220 186 160 147 344 254 216 208 190 120 93 90 89
CF2 248 189 149 125 114 264 228 180 172 180 110 87 83 80
CF3 222 165 127 114 107 278 233 197 188 176 105 85 80 78
CWS1 334 236 187 159 149 270 240 210 203 211 160 89 69 65
CWS2 351 240 185 158 149 532 182 144 137 201 183 185 181 173
CWS3 466 299 233 198 184 592 346 328 318 309 240 128 100 92
CWS4 500 298 231 195 182 628 328 313 300 351 264 148 106 99
CWS5 549 413 359 307 290 231 189 152 148 196 181 184 175 168
CWS6 605 414 362 310 288 327 240 160 146 186 148 81 65 62
CWS7 830 514 455 386 353 328 238 193 189 287 275 265 258 242
CWS8 859 514 438 370 342 470 305 250 246 309 274 289 269 249
CWS9 806 612 511 441 406 525 415 354 345 250 177 152 136 130
CWS10 872 625 516 447 404 550 403 341 336 207 150 131 122 119
CWS11 931 612 509 438 395 725 473 399 389 360 248 207 189 183
CWS12 944 581 485 410 380 767 452 391 374 412 279 232 203 193
CFS1 288 202 170 144 132 254 197 188 185 173 149 124 113 108
CFS2 313 219 187 158 147 229 189 171 169 143 120 93 83 77
CFS3 416 285 245 210 193 429 397 317 314 196 128 96 88 78
CFS4 458 309 264 229 215 476 384 353 346 258 192 162 80 69
CFS5 243 182 149 122 114 300 221 195 186 156 114 94 86 82
CFS6 274 208 167 139 128 267 196 171 164 129 104 90 84 80
CFS7 434 311 251 208 195 522 380 342 333 193 153 129 116 107
CFS8 535 381 307 257 237 597 438 394 387 230 143 95 79 75
CFS9 187 122 108 96 92 272 222 190 181 142 92 78 73 72
CFS10 206 133 116 103 98 246 203 175 163 119 89 83 79 77
CFS11 369 234 208 184 170 348 277 245 233 165 131 102 90 87
CFS12 406 263 224 195 184 405 322 286 278 188 145 110 93 85
3.5 Workability of Wet Concrete
15
386
Table 5 provides the results obtained from performing a series of workability tests on fresh concrete mixes to
check their flowability, passability and segregation resistance. Following results have been suggested on the basis
of role of each of the admixture in imparting flow to the SCC:
(i) WMF addition
Cohesion in small amounts is necessary for a good flow of mix because the mix flows homogeneously, together
with minimal opposition caused by viscous forces. Passability also decreases with increased cohesion, because the
mix try to remain altogether when the obstruction confronts it, but for a mix to show segregation resistance it
should have higher cohesion. Apart from viscous forces in the pore solution, the inter-particle frictional forces tend
to reduce the flowability and passability but enhance the segregation resistance of concrete mix.
WMF is acicular and it has high adsorption tendency as well as the tendency to be a part of pore solution at its
higher contents. Hence it increases the inter-particle friction by producing friction between cement particles on
which it get adsorbed, and it also increases the viscosity of pore solution at its higher contents.
Therefore it was observed that with WMF addition the flowability and passability of mix reduced, though up to
20% WMF its rate of reduction was less. Segregation resistance increased with increment in WMF content.
(ii) Microsilica addition
With microsilica addition at lower amount 0-5%, the flowability, passability and segregation resistance increased
because microsilica has smooth texture and spherical shape, which induces a ball bearing effect between cement
particles on whom it get adsorbed. The ball bearing effect does not allow the cement particles to move away
though it allows them to roll or slide over each other on account of the sticky nature of microsilica. Khayat &
Aitcin [8] also suggest, that presence of microsilica affects the properties of fresh concrete by inducing cohesivity
and thus reducing the bleeding of concrete. Both of these factors work contradictory as far as shrinkage is
concerned; on one hand cohesivity reduces shrinkage, whereas on other the reduced bleeding increases it. The
increased cohesivity also requires more slump for a given flow, with respect to a normal concrete. But it has one
advantage which is homogeneity, and thus enables good passability and flowability, thereby enabling microsilica
admixed concrete as a pumpable concrete [5].
16
387
Hence at higher amount of microsilica, the flowability and passability decreased but the segregation resistance
increased, because the pore solution between the cement particles get thicker and more viscous due to increased
microsilica content.
(iii) Flyash addition
Flyash at all contents improved the flowability and passability of mixes, but it decreased the segregation resistance
of self compacting concrete.
Table 5. Observations from the Workability Tests Conducted on trial concrete mixes
Mix CA: FA Abram’s V Funnel V Funnel after 5 J Ring Probe ring Super plasticizer(%
flow (600-750) time (6- min. (9-15 sec) difference (0- penetration (0- by weight of
mm 12sec) 10 mm) 7) mm) cement. mat.)
C 60:40 360 17 24 24 2 0.30
CW1 55:45 560 10 12 21 3 0.45
CW2 50:50 580 8 10 19 5 0.45
CW3 50:50 540 12 12 21 4 0.6
CWS1 55.45 580 9 11 15 3 0.45
CWS2 50:50
55:45 630 8 10 9 5 0.45
CWS3 50:50 620 9 10 10 4 0.45
CWS4 50:50 605 10 11 13 3 0.45
CWS5 50:50 646 7 11 11 5 0.45
CWS6 50:50 660 7 9 5 7 0.6
CWS7 50:50 620 8 11 8 5 0.6
CWS8 50:50 590 9 14 10 4 0.6
CWS9 50:50 570 10 12 14 5 0.6
CWS10 45:55 630 7 9 12 3 0.6
CWS11 45:55 575 9 13 17 2 0.45
CWS12 45:55 530 13 18 22 2 0.45
CF1 55:45 570 10 16 15 5 0.3
CF2 55:45 581 9 17 13 7 0.3
CF3 55:45 604 7 18 11 9 0.3
CFS1 55:45 585 9 16 12 5 0.45
CFS2 50:50 615 8 14 11 8 0.45
CFS3 50:50 600 9 10 15 6 0.45
CFS4 50:50 573 10 12 18 3 0.45
CFS5 50:50 625 7 9 9 7 0.45
CFS6 50:50 650 6 9 8 9 0.45
CFS7 50:50 618 8 9 10 6 0.45
CFS8 50:50 587 10 11 13 6 0.45
CFS9 50:50 633 6 15 11 10 0.45
CFS10 50:50 663 5 17 8 12 0.3
17
388
CFS11 45:55 623 7 17 13 9 0.3

CFS12 45:55 594 8 21 15 7 0.3
3.6 Drying shrinkage resistance of concrete mixes
This test corresponds to the behaviour of concrete in the period before 28 days, and thus the behaviour of concrete
in the first seven days has more prominent effect on the drying shrinkage of concrete. For small, unrestrained
concrete specimens (prisms), a low ultimate shrinkage (strain) is considered to be less than 520 millionths (at 50%
relative humidity and 73° Fahrenheit). Typical concrete shrinkage has been measured at 520 to 780 millionths.
Table 6 shows the cracking time and final stress rate for different mixes.
While testing at a laboratory, shrinkage is measured by two parameters: stress rate due to shrinkage and time for
crack development. Both of these parameters vary according to paste composition (fineness), which affects the
continuity of pores in hardened paste as well as the rate of hydration. It is not sufficient enough to say that the
shrinkage of a concrete would be higher, if it contains larger fraction of fine materials. This would be true to say
only, if the water adsorbed by the particle at the capillary voids is not held strongly, if the particle is not able to
convert the capillary voids into gel pores, and if the pores are continuous. All of these factors are dependent upon
the size and reactivity of the particle.
(i) Results on WMF with microsilica
Soliman and Nehdi [16] have also suggested that there is a reduction in total shrinkage with increasing content and
aspect ratio of WMF. It was suggested that as the WMF content increased from 4 to 12%, the reduction in total
shrinkage at 7 days increased from 11 to 16% for very fine microfibers and from 2 to 9% for medium fine
microfibers, with respect to that of the control mixture. Microfibers delay the coalescence and propagation of
cracks at early age through better stress transfer at microcracks. On the other hand, the increase in the cracking age
with increasing microfiber content implies a higher crack-bridging efficiency and ability of larger size WMF
microfibers to overcome the reduction in matrix strength induced by dilution effect.
Initially when WMF is added in lower quantity up to 20%, the cement particles are not far apart in the solution,
and the voids in between them are easily filled or blocked by WMF due to its pore size or grain size refinement,
represented by dense zigzag hydrated WMF particles in the SEM image shown in Figure 4(a). Though the water is
18
389
released when WMF enters into the cement voids, but this water only is used in the accelerated hydration done by
cement particles. As a result of which, the water is not able to escape from the pores (mainly gel pores) and
shrinkage is reduced. With the increment in the content of WMF in the mix, the cement particles increase their
distance, and more and more WMF particles enter into the pore solution present in between these voids. As a
result of which the number of capillary spaces increase whereas the gel pores reduce. This is verified from the
SEM images of mixes containing WMF, as has been shown for CW3 in Figure 4(b). Since the void space is larger,
the initial hydration, as well as insufficient capacity to block the voids, provided by WMF, is not sufficient to stop
the water from escaping from these capillary voids. Therefore, an increased shrinkage stress was reported for
WMF content greater than 20%. As a reinforcer, WMF will increase the tensile strength and toughness of mortar,
which would resist the crack development and also will delay the propagation of cracks by distributing the stresses
over a local region. Thus, the overall effect observed for WMF addition, was the delay in crack development in
spite of high stress rate. Since the time period of cracking also depends upon the stress rate development, therefore
the variation trend for time period of cracking is similar to the stress development.
(ii) Results on Flyash with micro silica
Coarser flyash particles act as micro aggregates which reduce the shrinkage, increase the packing of aggregates
with the paste and thus increase the tensile strength, increase the density of the concrete, thereby leading to
reduction in crack propagation and porosity, causing a direct increase in compressive strength and stiffness. Also
this portion of flyash, absorbs water in capillary pores due to surface tension and holds it for long time hydration.
Hence, when microsilica is present, the unreacted coarser flyash at later ages avail the water to the microsilica, and
cause its hydration, thereby increasing the strength of the concrete. An excessive flyash beyond 30% of total
cementitious material has been found to reduce strength due to dilution of cement and increment in porosity cause
by its coarser particles.
For flyash mixes, the hydration rate is higher than normal concrete uptil 7 days. The effect of this acceleration
increases its crack development time in comparison to normal cement concrete. After 14 days the effect of
acceleration diminishes and the samples start cracking (Figure 5 showing minute cracks formation in mix CFS4).
Though flyash contain alumina and calcium oxide, which promote ettringite formation in presence of sulphates,
but the consumption of lime due to flyash hydration may hinder the formation of ettringite. Also, the reactivity of
19
390
alumina depends upon, whether it is available in glass part of the compound or not. If it is present in glass part
than it provides a long term source of ettringite formation in presence of sulphates. The reactivity of flyash as
noticed from the XRD test suggests, that ettringite formation is hindered in presence of flyash. This suggests that
flyash does not contain reactive alumina and highly neutralizes CH; if it did, then ettringite formation would have
been promoted rather than hindered. Thus, the stress rate is equal to or a bit higher than normal concrete because
effect of pore refinement is nullified by presence of fine material (fraction of flyash much finer than cement)
verified by particle size analysis. Hence with the increment in flyash content both the stress rate and the crack time
increase to a small extent. Form the results obtained it has been found that WMF mixes has the highest stress rate
increment capacity and its crack development time is followed by microsilica. Microsilica has maximum stress
rate reduction capacity due to higher paste strength imparted by high rate of hydration. Flyash has lowest crack
time increment capacity.
Langan et al. [10] studied the influence of microsilica on bulk paste modification and pressed, that excessive
fine cement promotes rapid hydrations, which is further complemented by microsilica, and thus cause expansion
of concrete. On loss of water, this expanded concrete starts shrinking, but the restraint offered by concrete on
account of high strength provided by microsilica, as well as the pore discontinuity provided through filler effect &
pore refinement, controls the shrinkage. The results of this study are thus, in agreement with their study.
Figure 4. (a) SEM image of CW1 & (b) CW3
20
391
Figure 5. Crack development in Mix CFS4 after 14 days during Ring Shrinkage test
TABLE 6. Results obtained from the restrained shrinkage test
Mix Max. Initial Age at Strain rate (α) at Avg. Elapsed Stress Potential
age max. elapsed time T microstr time at rate at for
Micro (Days) strain ½
(ε=α×T +K) ain rate ) max. cracking cracking
strain (Days) strain (MPa)
C -96 0.24 12.3 Y=10.48X-58.35 10.48 (Days)
12.06 0.109 Mod High
CW1 -70 0.03 18.8 Y= 6.872X-37.00 6.87 18.82 0.057 Mod Low
CW2 -55 0.25 116.8 Y= 4.836X-35.12 4.83 16.56 0.043 Mod Low
CW3 -96 0.24 523.6 Y= 9.069X-54.14 9.07 23.45 0.068 Mod Low
CWS1 -56 0.47 128.0 Y= 10.95X+0.280 10.95 27.52 0.075 Low
CWS2 -64 0.34 919.5 Y= 5.266X-42.20 5.27 19.19 0.043 Mod Low
CWS3 -68 0.34 022.5 Y= 9.956X-17.54 9.96 22.17 0.076 Mod Low
CWS4 -67 0.50 427.4 Y=11.13X-2.352 11.13 26.89 0.077 Low
CWS5 -72 0.26 118.7 Y= 4.316X-51.34 4.32 18.51 0.036 Mod Low
CWS6 -70 0.55 026.5 Y= 2.317X-56.68 2.32 26.03 0.016 Low
CWS7 -78 0.47 823.6 Y= 4.856X-51.09 4.86 23.21 0.036 Mod Low
CWS8 -76 0.27 920.5 Y= 5.390X-51.34 5.39 20.23 0.043 Mod Low
CWS9 -39 0.40 928.0 Y= 8.081X+6.033 8.08 27.63 0.056 Low
CWS10 -40 0.55 028.0 Y= 6.505X-3.328 6.51 27.47 0.045 Low
CWS11 -54 0.34 224.1 Y= 10.69X-1.725 10.69 23.79 0.079 Mod Low
CWS12 -55 0.34 228.0 Y= 12.36X+3.933 12.36 27.68 0.085 Low
CF1 - 0.32 314.4 Y= 12.05X-68.02 12.05 14.13 0.116 Mod Low
CF2 -96 0.44 218.9 Y= 15.07X-22.94 15.07 18.49 0.127 Mod Low
CF3 120- 0.42 416.0 Y= 15.59X-52.78 15.59 15.59 0.143 Mod Low
CFS1 - 0.37 219.1 Y= 12.09X-58.50 12.09 18.80 0.101 Mod Low
CFS2 113- 0.45 120.5 Y= 10.28X-66.31 10.28 20.08 0.083 Mod Low
CFS3 117- 0.29 716.6 Y= 12.05X-69.14 12.05 16.35 0.108 Mod Low
CFS4 116- 0.27 315.0 Y= 11.82X-67.93 11.82 14.77 0.111 Mod Low
CFS5 -84
119 0.55 428.0 Y= 16.22X+4.654 16.22 27.47 0.112 Mod Low
CFS6 -89
116 0.34 428.0 Y= 16.03X-2.588 16.03 27.68 0.110 Mod Low
CFS7 -69 0.53 228.0 Y= 13.61X+12.02 13.61 27.49 0.094 Low
CFS8 -95 0.34 228.0 Y= 18.55X+5.899 18.55 27.68 0.127 Mod Low
CFS9 -96 0.50 228.0 Y= 16.42X+0.420 16.42 27.52 0.113 Mod Low
CFS10 - 0.29 218.0 Y= 11.88X-54.38 11.88 17.73 0.102 Mod Low
CFS11 - 0.29 221.8 Y= 16.85X-22.19 16.85 21.52 0.131 Mod Low
CFS12 105- 0.45 226.9 Y= 18.88X-2.764 18.88 26.50 0.132 Mod Low
108 1
110 5
4 CONCLUSIONS
Following conclusions could be drawn for combined effect of WMF and flyash on concrete mixes:
21
392
(i) Water demand, setting times and compatilibilty
The rise in water demand of a paste as shown by normal consistency test, generally indicate towards increased
initial and final setting times, as the excessive water adsorbed on fine binder material decrease the stiffness of
paste. This is generally applicable for pastes containing lesser reactive coarser pozzolans, but the behaviour of
pozzolans vary with fineness and reactivity. Finer pozzolans ensure high water demand but do not guarantee lower
setting times, whereas higher fineness, complemented with high reactivity results in both high water demand and
lower setting times with respect to pozzolan admixed pastes. WMF when substitutes cement by 20% with 5%
silica fume show comparable water demand and normal consistency to cement. Its compatibility with PCE based
superplasticizer is also good and achieves comparable flow to flyash admixed pastes at higher water to
cementitious material ratio, for a similar superplasticizer content as in flyash admixed pastes.
(ii) Workability
As far as rheological properties of concrete is concerned, flyash though impart high flowability due to better
lubrication by free water, ball bearing effect and lesser density, but is not able to provide viscosity to the concrete
mix, which is necessary for making a mix homogeneous. Hence, flyash admixed concrete mixes show better
flowability but have poor passability and segregation resistance as the mix does not remain as a single unit. On the
other hand WMF admixed mixes show medium flow owing to being lighter than cement grains and high water
releasing capacity at 20% content in binder; good passability and better segregation resistance at 20% content in
binder along with 5% microsilica. This all happens inspite of it being reckoned as a flow inhibitor fiber.
(iii) Drying shrinkage
Shrinkage is generally low for mixes containing 20% of pozzolanic admixture and 5% of microsilica. In general,
drying shrinkage reduced by 49% for WMF-microsilica mixes, whereas for flyash-microsilica mixes, an increment
of 1.25% was observed.
References
1. Bouzoubaa, N., and Lachemi, M. (2001). “Self-compacting concrete incorporating high volumes of Class
F fly ash preliminary results.” Cement and Concrete Research, 31, 413-420.
2. CUR Report (1991). “Fly ash as addition to concrete.” Centre for Civil Engineering Research and Codes,
Report 144, Netherlands, 99.
22
393
3. Drazan J, Zelic J (2006). The effect of fly ash on cement hydratıon in aqueous suspensions.
Ceramics−Silikaty. 50(2): 98-105.
4. EFNARC (2005). “The European guidelines for self-compacting concrete: Specification, Production and
Use”, EFNARC, www.efca.info or www.efnarc.org.
5. FIP 1998. “Condensed silica fume in concrete, state of the art report.” FIP Commission on Concrete,
Thomas Telford, London, 37.
6. Grunewald, S., and Walraven, J. C. (2001). “Rheological study on the workability of fiber reinforced
mortar.” Proceedings of the Second International Symposium on Self Compacting Concrete, Tokyo,
Japan, 127-136.
7. Hossain, K.M.A., Lachemi, M., Sammour, M., and Sonebi, M. (2012). “Influence of polyvinyl alcohol,
steel and hybrid fibers on fresh and rheological properties of self-consolidating concrete.” Journal of
Materials in Civil Engineering, 24(9), 1211–1220.
8. Khayat, H.K., and Aitcin, P.C. (1992). “Silica fume in concrete- an overview.” Flyash, Silica fume, Slag
and Natural Pozzolans in Concrete, Ed. V.M. Malhotra, ACI SP-132, Detroit, Michigan, 2, 835-72.
9. Kocak, Yilmaz (2010). “A study on the effect of fly ash and silica fume substituted cement paste and
mortars.” Scientific Research and Essays, 5(9), 990-998.
10. Langan, B.W., Weng, K., and Ward, M.A. (2002). “Effect of silica fume and Fly ash on heat of hydration
of portland cement.” Cement and Concrete Research, 32(7), 1045–1051.
11. Lomboy, G., Wang, K., and Ouyang, C. (2011). “Shrinkage and Fracture Properties of Semiflowable
Self-Consolidating Concrete.” J. Mater. Civil Engineering- Energy Efficient and Environmentally
Friendly Paving Materials, 23, Special Issue, 1514-1524
12. Low, N.M.P., and Beaudoin, J. J. (1992). “Mechanical properties of high performance cement binders
reinforced with Wollastonite micro-fibres.” Cement and Concrete Research, 22(5), 981-989.
13. Low, N.M.P., and Beaudoin, J.J. (1993). “The effect of Wollastonite micro-fibre aspect ratio on
reinforcement of portland cement-based binders.” Cement and Concrete Research, 23, 1467-1479.
14. Neville AM (2006). “Proporties of concrete. Pearson Education limited.” England.
15. Qing Y, Zenan Z, Deyu K, Rongshen C (2007). Influence of nano-SiO2 addition on properties of
hardened cement paste as compared with silica fume. Constr. Build Mater. 21: 539-545.
16. Soliman, A.M., and Nehdi, M.L. (2012). “Effect of natural Wollastonite microfibers on early-age
behavior of UHPC”, Journal of Materials in Civil Engineering, 24(7), 816–824.
23
394
Paper ID- 88
Comparison of the behaviour of deep beams

designed using IS Code and STM method
Sivananthan V. (0000-0001-7727-9343), Vikalp Gupta (0000-0001-8622-
378X), Sanket Rawat (0000-0002-1985-7579), and G. Muthukumar*
(0000-0002-4836-0975)
Department of Civil Engineering, BITS Pilani, Pilani Campus, India –

333031
*muthug@pilani.bits-pilani.ac.in
Abstract: The present study mainly deals with the investigation of flexural
behavior of deep beams designed using Indian Standard and Strut and Tie
Modelling (STM) method. The comparison is mainly made through the
load-deflection behavior obtained using extensive experimental studies.
The ductile behavior and crack formation are also analysed in detail for the
deep beams designed using current Indian standard and recommendations
are made to adopt a more reliable and effective method, STM.
Keywords: Deep beams; STM; IS Code; Crack Pattern; Load-deflection.
1 Introduction
Deep beams commonly fall in the category of beams in which the span to depth ratio is
relatively small and this ratio is considered in Indian Standard (IS) 456 [1] and ACI 318 [2]
with the boundary limit of 2.5 & 4 respectively. ACI 318 [3] also emphasizes on the position
of the concentrated load for classification of the beam i.e. if the distance of the applied
concentrated load is less than two times the depth of the member from the face of support,
then it should be classified as deep beam. Deep beams are mainly characterized by the
predominant existence of arch action rather than the bending action. Therefore, the theory of
elastic analysis is normally valid only prior to cracking in these cases. IS 456 [1] prescribes
certain guidelines to be followed for the design of deep beams in Indian Scenario, which is
mainly intended to control crack width rather than the ultimate strength of the beam. Strut
and Tie Modelling (STM) method, on the other hand, can be applied wherever discontinuity
exist in any part of structural member, irrespective of span to depth ratio. The discontinuity
occurs in the regions where concentrated load acts, openings and changes in the cross section
[4]. In order to depict actual state of stress when the concrete is cracked, the dependence on
finite element analysis may not be found to be convenient and STM is found to be best
alternative for the regions where assumption of linear strain distribution is not valid. STM
gives the lower bound estimates of the capacity of concrete structures thus it is always
conservative [5]. In STM, several iterations are required to produce an efficient model and it
does not provide unique solution resulting in giving freedom to the designers. Basically, STM
method consists of three major components such as Strut, Ties & Nodal Zone [5]. The Strut
is defined as compression members which represents concrete stress fields whose principal
compressive stresses are predominantly acting along the center line of the strut and the
dimension of Strut depends upon the contact area between Strut and Nodal Zone. The tie is
defined as the tension member which consists of reinforcing steel or prestressing steel that
includes surrounding concrete lying concentrically with the axis of the tie, however
surrounding concrete is assumed to resist no axial tension in the model. The Nodal Zone is
the intersection point of either of the Struts, Tie or concentrated load and it may also define
as the zone in which the volume of concrete is responsible for transfer of the forces. STM
covers the detailing of structural members consistently with the design and results in larger
amount of reinforcement under the same loading thus incorporating more ductility and safety..
Efforts are made in the present study to compare both the methods and evaluate the possibility
of the later for Indian scenario. This study would definitely be effective in providing
justifiable comparison to improve the presently followed approach for design and detailing
of deep beams, and thus ensuring enhanced performance.
395
2 Test Specimens
The experimental program consisted of casting of two sets of specimens of deep beams: One
designed using IS 456:2000 [1] method and the other designed using STM method and one
set of specimens of slender beam using IS 456:2000 [1] method. Deep beam designed using
STM method conformed to the guidelines of ACI 318-14 [2]. Hereafter the terms IS code
method and STM method should be treated as deep beam designed as per IS 456:2000 [1]
and ACI 318-14 [2] respectively. Fig.1 shows the cross-section and elevation of the deep &
slender beam specimens. The materials used in preparation of samples include OPC 43 grade
cement, Yamuna river sand as fine aggregate and locally available coarse aggregate. M30
grade concrete and Fe 500D steel were used to cast both the sets of reinforced concrete deep
beams. Furthermore, several trails of mix proportioning had been conducted & consequently
it was finalized on the basis of 7 days and 28 days compressive strength. Weigh batching was
used for quantifying the ingredients of each mix. The properties of sand such as water content,
gradation and specific gravity were studied in detail. The samples were removed from the
moulds after 24 hours and wet curing of samples was done in locally available clean water.
Fig. 1. Elevation and cross section of a) Deep Beam & b) Slender Beam
All the specimens were designed preliminarily using the same factored point load of 134 kN
(≈ factored bending moment of 20 kN-m) using both STM method & IS code method for
the beam dimension as shown in Fig. 1. The summary of design & detailing of beams is
mentioned in Table 1.
Table 1: Design Parameters obtained from IS 456 Code [1] & STM Conforming to ACI 318-
14 [2]
Design Parameters IS 456:2000 Method STM Method IS 456:2000

(Deep beam) Conforming to ACI Method (Slender
318M-14 (Deep beam)
beam)
Beam Size 700 x 150 x 300 mm 700 x 150 x 300 mm 700 x 150 x 200 mm
Clear cover 20 mm 20 mm 20 mm
396
Tension 2-12φ bars 2-12φ bars 3-12φ bars

reinforcement
Holding bars at top 2-8φ bars 2-8φ bars 2-8φ bars
Side Face Vertical: 8 mm 2- Vertical: 8 mm 2- Vertical: 8 mm 2-

reinforcement legged stirrups at 300 legged stirrups at 150 legged stirrups at
mm c/c mm c/c 150 mm c/c
Horizontal: 8 mm bars Horizontal: 8 mm bars
parallel to longitudinal parallel to longitudinal
reinforcement at 150 reinforcement at 150
mm c/c on both faces mm c/c on both faces
Bearing -No special At supports: 50 mm -No special

calculation required- ISF 12 mm thickness. calculation required-
At Loading: 75 mm
ISF 12 mm thickness.
The detailing of both sets of deep beam specimens is shown in Figure 2. While designing
deep beam using STM method as per ACI code [5] recommendations, the angle between
Strut and Tie was kept lower than 25 degrees in order to control cracks and to avoid the
problems of shortening of struts and elongation of ties occurring in almost same directions.
It was also ensured that truss selected was determinant as indeterminant truss may indulge
complexity in the calculations. The proportioning of struts depends upon the requirement of
compressive resisting forces. The size of nodal zones is increased or either compression
reinforcement is provided in case if the struts do not have sufficient capacity. However, this
revision may lead to change in the size of bearing plate or column.
Fig. 2. Reinforcement detailing in deep beams designed using a) IS code b) STM method
Fig. 3. Reinforcement detailing in slender beam designed using IS code method

The test setup consisted of metal bearings and bearing plates which were modified to act as
flexure test fixtures at the support is shown in Figure 4. Concentrated load was applied at the
midspan using a 1000 kN universal testing machine. Each test was fully video recorded to
capture the behavior in detail. The midspan deflection was read from the digital meter. Also,
in order to analyze the exact pattern of the crack formation, the complete beam was divided
into the squares of size 50 mm each as shown in Figure 4. During test, the load at which the
flexural and diagonal cracks initiated was noted from the video & written on the tested
specimen, nearby the corresponding crack. The markings should not be mistaken as
proportional to the actual crack width. The numbers next to each crack shall denote the load
at which the particular crack initiated, and the upward or downward arrow shall denote
whether the crack initiated before attaining peak load or beyond peak load, respectively.
397
Fig. 4. Failure pattern in deep beam using IS 456 Code & STM Method respectively
Fig. 5. Failure pattern in slender beam using IS 456 Code Method

a. Load-deflection Curve
Through the comparison of the load deflection curves (Fig. 6), it can be analyzed that both
IS code and STM designed deep beams follow a similar profile up to a certain point of loading.
Three specimens were tested and results of all three specimens are shown for the purpose of
comparison. Beyond that point, the IS deep beam reached the ultimate load and failed
suddenly, exhibiting a steeper slope on either side of the peak. Whereas, the STM deep beam
continued to take greater loads, exhibiting a definite yielding before it reached the ultimate
load and then it failed suddenly, with steep descending post the peak load. The higher mid
span deflection in the deep beam designed using STM method is due to the gentle slope of
the curve while reaching the peak ultimate load and thereafter sudden failure takes place. In
deep beam designed using STM method, the sufficient amount of strain softening zone is
formed as compared to deep beam designed using IS code method. Thus, it can be inferred
from the load-deflection curve that failure in deep beam designed using IS code method is
more brittle in comparison with the beam designed using STM method. Also, the load taken
by the deep beam designed using STM method is quite high in comparison with the beam
designed using IS code method resulting into high mid span deflection and higher strength.
From the load-deflection curve of slender beam, it can be inferred that the load increases at
a faster rate and reached the ultimate load. However, it exhibited full yielding beyond the
peak load and sufficient amount of strain softening zone was formed. The slender beam has
capability to take much higher mid-span deflection and the reason behind such behaviour is
the large softening zone. Thus, inspite of a lower ultimate strength, the mid span deflections
in slender beam were significantly higher than that of deep beam designed using IS code &
STM method.
b. Crack Pattern
As expected, the first flexure crack developed in deep beam designed using IS code method
at a load which was approximately 60% of ultimate load whereas first diagonal cracks
appeared at 86% of ultimate load. However, in STM deep beams, the first flexural cracks
appeared at 70% of ultimate load and the first diagonal crack formed at 74% of ultimate load.
The crack pattern formation of deep beam initially started from the flexure crack which was
developed from the soffit of the beam and soon after that, the diagonal crack initiated from
the inner edge of support to the point of application of loading. The higher amount of shear
reinforcement led to change in diagonal cracking and thus reduced the inter-spacing of cracks.
As load increased progressively, the number of cracks originating from the support to the
398
point of application of load also increased. After the ultimate load was reached and as load
started to decrease, the vertical and horizontal crack patterns got deepened at both the ends
of support in beams designed using IS code method. However, in STM method, with decrease
of load the cracks at the ends of support were not deepened.
It should be noted that in STM deep beams, diagonal cracks were developed fully from both
the supports and which is why, it sustained additional loads before reaching the ultimate load,
in the yield region. This was achieved by the provision of crack control reinforcement across
the strut elements in STM deep beams. From Table 1, it can be well established that the
amount of reinforcement is quite more in deep beam designed using STM method and due to
reason that the first flexural crack developed in beam using STM method is delayed. Also,
from Table 1 it is found that on increasing the number of vertical shear reinforcement i.e.
decreasing the spacing between them, the beam is capable of taking additional loads and
hence better crack control could be achieved. From the study conducted by Zhang & Tan [6],
it was found that after the formation of diagonal crack, the ultimate shear strength of deep
beams is dependent on arch capacity which in turn depends upon the boundary condition and
geometry of the diagonal strut. Thus, it can be inferred that the geometry of truss selected by
the designer plays a significant role in deciding the shear strength of deep beam and it needs
to be done with utmost care.
However, in case of slender beam, the first flexure crack developed at 60% of ultimate load
whereas the first diagonal crack developed at 78% of ultimate load. All the samples have
failed in flexure and the diagonal cracks were formed after particular duration of loading. For
slender beams designed using IS code method, the crack formation pattern similar to STM
method deep beam was observed.
350 350
1
300 300
1 2
250 250
3
2
Load, kN
Load, kN
200 200
3
150 150
100 100
50 50
0 0
0 2 4 6 8 10 12 14 0 5 10 15
Mid span deflection, mm Mid span deflection, mm
Fig. 6. Load vs displacement curve for deep beam designed using IS 456 Code & STM
Method respectively
250
1
200 2
3
Load, kN
150
100
50
0
0 5 10 15 20
Mid Span Deflection, mm
Fig. 7. Load vs displacement curve for slender beam designed using IS 456 Code Method
399
c. Ductility:
Ductility signifies the deformation capacity of member after yielding and thus helps the
structure to get rid of that energy and to dissipate it out. The brittleness increases when either
the depth of beam increases or the area of reinforcement decreases. Also, the rotation capacity
of deep beams is quite small. The ductility of reinforced concrete members depends upon the
two parameters, the size of the member and concrete strength. The study conducted by the
Rao et al. [7] suggested that ductility number which is a measure of brittleness of beam
increases with the size of beam at given flexural reinforcement and increases with increase
in beam depth. During uniaxial test on concrete, from the stress-strain curve it is observed
that the higher the concrete grade the steeper is the portion of stress-strain curve, the sharper
the peak of the curve & less the failure strain. For low grade concrete, the curve has relatively
flat top & a high failure strain, thus with increasing grade of concrete the ductility decreases.
Also, study conducted by Cheng [8], Tan et al. [9] suggested that the loading dimension,
support plates and distribution of reinforcement influence the size effect in beam. Here, the
depth of beam must be considered in the size effect of the beam, and as proved by Kani [10]
that width of beam has negligible effect on the structural behavior of deep beams as failure
shear stress is not affected by the width of beam.
The deflection ductility index is defined as the ratio of deflection at ultimate state to the
deflection at yield state. The higher the deflection ductility index, better the performance of
structural member to sustain higher deformations before failure and thus, it confirms the
applicability in seismic design and redistribution of moments. From the curve, the deflection
ductility index is found to be in the order of 1.3-1.7 for deep beam designed using IS code
method, while in deep beam designed using STM method it is of order of 1.5-2, suggesting
a lower ductility in case of IS code deep beams. Similarly, for the slender beam the ductility
is in order of 1.9-2.5 from the load-deflection curve which is higher as compared to deep
beam. Thus, it can be said that the ductility of the deep beam is lower as compared to slender
beam.
4 Conclusion
The present study clearly highlights the advantage of designing deep beams using STM
method over IS code method. In STM method, the design of beam may not have unique
solution because it is an iterative process and thus most efficient solution with high
performance can be adapted in the design. The selection of the type of truss primarily depends
upon the loading pattern and finally on designer’s choice thus providing flexibility and
freedom to the designers. The proper modelling of load path taken by the forces in the beam
and detailing of key elements is responsible for this high performance. STM deep beams were
found to perform better than IS deep beams. Following conclusions can be drawn from the
present experimental study:
 From the load deflection curve, it is assessed that the beam designed using STM
method is able to take sufficiently larger load & deflection (at mid span) as compared
to the beam design using IS code method.
 Also, the provisions related to vertical shear reinforcement for deep beams in IS code
may be relooked.
 It is found that deep beam designed using STM method possesses sufficient ductility
in comparison with the beam designed using IS code method which can also be
calculated by the deflection ductility index.
 From the experimental study for a slender beam, it can be concluded that slender beams
have greater ductility due to significant amount of softening zone and due to that reason,
it is capable of taking higher mid span deflections. The crack pattern formation in
slender beam is similar to that of deep beam designed using STM method. Also, the
deflection ductility index of slender beam ranges in the order of 1.9-2.5 ensuring
sufficient ductility of beam.
400
5 References
1. IS 456: 2000, Indian Standard Plain and Reinforced Concrete - Code of Practice, Fourth
revision, Bureau of Indian Standards, New Delhi, 2000.
2. ACI 318M-14, Building Code Requirements for Structural Concrete and Commentary,
American Concrete Institute, Farmington Hills, 2014, p. 519.
3. ACI Committee 318, Building Code Requirements for Structural Concrete (ACI 318-05)
and Commentary (318R-05), American Concrete Institute, Farmington Hills, MI, 2005,
430 pp.
4. Schlaich & Schafer (1991) Design and detailing of structural concrete using strut-and-
tie models. The Structural Engineer/Volume 69/No.6, pp.113-125.
5. Subramanian, N., Design of Reinforced Concrete Structures, Oxford University Press,
New Delhi, 2013.
6. Zhang N. & Tan K (2007) Size effect in RC deep beams: Experimental investigation
and STM verification. Elsevier, Engineering Structures, pp.3241-3254.
7. Rao et al. (2007) Experimental Studies on Ductility of RC beams in Flexure and Size
Effect”, FraMCOS-6, 17-22, June, 2007, Catania, Italy, pp. 671-675.
8. Tan KH & Cheng GH. (2006) Size effect on shear strength of deep beams: Investigating
with strut-and-tie model, J Struct Eng., pp.673–685.
9. Tan KH, Cheng GH, Zhang N. Experiment to mitigate size effect on deep beams. Mag
Concrete Res (in press).
10. Kani GNJ (1967) How safe are our large reinforced concrete beams, ACI
1967;64(3):128-41.
401
Paper ID- 89
Efficient Static Analysis Of Laminated Composite And Sandwich Shell

Structures Using ABAQUS
Aman Garg1, H.D Chalak2
1,2
Civil Engineering Department, National Institute of Technology, Kurukshetra, Haryana, India-
136119
1
E-mail: amang321@gmail.com
Abstract
In the present study, an attempt has been made to carry out efficient analytical analysis of laminated composite and
sandwich shell structures under static conditions. The models were made using ABAQUS/CAE. Effect of different
types of finite elements on predicting the behaviour of laminated composite and sandwich shell structures has been
carried out. Influence of material properties, geometrical properties, boundary conditions, lamination schemes,
stacking sequences etc. were carried out on different types of shell structures (spherical, cylindrical, hypar shaped
shells). The obtained results were compared with the results those available in literature using different types of
theories. At last, also an attempt has been made to carry out the study on the order of suitability of various theories
and proposed model in ABAQUS. Comparative studies were also on different types of shells under different loading
conditions was also concluded.
Keywords: Laminated composite shells, sandwich shells, static analysis, spherical shells.
1 Introduction
Laminated composite and sandwich shell structures are gaining popularity in field of civil, aeronautical, automobile,
defence, naval industry, nuclear containers, pressure vessels etc. due to their excellent properties such as high
strength to weight/stiffness ratio, lesser self-weight etc. Main problem with laminated composite and sandwich
structures is sudden change in material properties at interface and weak in shear as compared to extensional rigidity.
Number of theories are available for the analysis of laminated composite and sandwich shell structures. The simplest
type of theory available is called as classical laminated theory (CLT). The main problem with this theory is that it
ignores transverse shear stresses during analysis. CLT was used by [1-3] for analysis of laminated composite and
sandwich shell structures. Hence, this theory predicts satisfactory behaviour only for thin shells but not for
moderately thick and thick shell structures. This is because CLT assumes plane section remains plane after bending.
To incorporate effects of transverse shear stresses, first-order shear deformation shell theories (FSDT) are proposed
which assumes constant transverse shear stresses across the thickness of shell. This theory is employed by [5-7]. It is
seen that inclusion of stretching-bending effects in case of FSDT improves the accuracy in predicting the behaviour.
In actual, the shear stress distribution across the thickness of laminated composite and sandwich shell structures is
not constant. Therefore, this theory requires shear correction factor which depends upon number of conditions such
as material, geometrical properties; end conditions; laminated scheme etc. [8, 9].
Defects presented in FSDT are solved by higher order shear deformation shell theories (HSDT). These theories
assume parabolic variation of transverse shear stresses across the thickness. Extensive work is available on analysis
of laminated composite and sandwich structures using HSDT [10-15]. HSDT gives constant distribution of transvers
shear strains across the thickness of shell with discontinuous transverse shear stresses across thickness of shell but in
402
actual the behaviour of laminated composites are opposite to it [16]. Even some HSDTs are not able to give stress
free boundary conditions at top and bottom of shells.
All the above stated theories are called as equivalent single layer (ESL) theories because deformations/displacement
functions are assumed as unknown with respect to mid plane or reference layer. The disparities posed by ESL are
solved by layer wise theories (LWT). These theories are of two types: discrete layer wise theories and refined
theories. In discrete theories, analysis is carried out for each separately and then are integrated over the whole
domain/thickness [17, 18]. These theories are able to predict the behaviour of laminates more accurately as
compared to above discussed theories but are computationally costlier because of increase in number of unknown
with increase in number of layers. Refined theories (also known as zig zag theories (ZZT)), in which number of
unknowns are not dependent upon number of layers by choosing in-plane displacement function having piecewise
variation across the thickness of plate and at same time satisfying transverse shear stress continuity at interfaces at
same time. ZZTs are used by [19-24] for static and free vibration analysis of laminated composite and sandwich
shells.
Carrera proposed a combination of all the above discussed models into a single equation called as Carrera unified
formulation (CUF) and used by [25-27] for analysis of laminated composites. Some literature is also available on
analysis of laminated composites using 3D models [28-30]. These models predict behaviour of laminates with great
accuracy but are computationally costlier. Some review literature is also available on analysis of laminated
composite and sandwich structures [31-43].
During review work, it is found that the application of software packages for the analysis of laminated composite
and sandwich shell structures is not fully explored. In this paper, an attempt has been made on the analysis of
laminated composite and sandwich shell structures using ABAQUS/CAE. The efficiency of the model is checked by
comparing the obtained results with those already available in literature.
2 Modelling
Finite element (FE) is one of the most widely used method for analysing laminated composite and sandwich shell
structures under different conditions. FE based commercial software package ABAQUS/CAE is employed to study
the behaviour of laminated composite and sandwich structures under static and free vibration conditions. Effects of
material properties, boundary conditions, geometric properties, lamination scheme etc. on behaviour of laminated
composite and sandwich shells is carried out in detail. The obtained results are compared with the already published
results available in literature.
The material properties used during the analysis are reported in table 1.
Table 1. Material properties used during analysis
Material Material Properties

E1 E2 E3 G12 G13 G23 Poisson’s Ratio
A 25E2 Open E2 0.5E2 0.5E2 0.2E2 0.25
B 0.8E2 Open E2 3.2E2 1.2E2 1.2E2 0.25
Preliminary studies: Firstly, preliminary studies are carried out in order to choose the element and mesh size in order
to predict more accurate behaviour. For preliminary study, simply supported laminated cylindrical panel made up of
material A as in Table 1 is used. Three different types of elements are chosen available in element library of
ABAQUS. FE S8R5, S3R and S8R are used. For details of these element, one may refer the ABAQUS manual [44].
The obtained results are compared with those already available in literature. Results are reported in table 2.
403
Table 2. Convergence and element suitability studies for simply supported laminated composite cylindrical
shell panel (0/90/0) made of material A.
Element Name a/h=5 a/h=10 a/h=100
S3R (16x16) 2.078 0.9018 0.4498
S8R (16x16) 2.072 0.9001 0.4468
S8R5 (4x4) 2.069 0.9023 0.4583
S8R5 (8x8) 2.070 0.9025 0.4588
S8R5 (10x10) 2.075 0.9028 0.4589
S8R5 (12x12) 2.078 0.9031 0.4592
S8R5 (14x14) 2.080 0.9031 0.4592
S8R5 (16x16) 2.080 0.9301 0.4592
Shu [45] 2.081 0.9292 -
Elasticity [9] 2.118 0.9396 -
HOZT [21] 2.081 0.9301 0.4592
Form the results available in table 2, it can be seen that the results predicted by using FE S8R5 are nearest to the
results available in literature. Also, the mesh convergence rate is performed for the same element. It can be seen that
the results converge at mesh size of 14x14. Hence, in all further studies, mesh size of 14x14 is taken.
Cross-ply laminated composite cylindrical shells: Effects of lamination schemes, and geometric properties are
studied for simply supported laminated composite cylindrical shells made up of material A subjected to sinusoidal
loading. The similar problem is already solved by Shu [45] and Huang [9]. The results are reported in table 3. The
obtained results are much closer to results predicted by Huang using elasticity theory [9]. It can be seen that the
number of layers greatly affects the behaviour of laminated composite shell. With increase in a/h ratio, deflection at
centre of shell shows decreasing trend for all cases of lamination schemes. However, for five layered laminated
shell, deflections are less as compared to four layered laminated shell but higher than that of three-layered laminated
shell.
Analysis of laminated simply supported spherical shells: Under this section, simply supported laminated composite
cross-ply and angle-ply laminated spherical shells are studied subjected to sinusoidal loading conditions. The
obtained results are compared by those available in [21] by using HOZT. The results are reported in table 4. It can
be seen that angle-ply laminated shells performed better as compared to cross-ply shells in terms of both deflection
and stresses under present conditions.
404
Table 3. Non-dimensional deflection and stress for simply supported laminated cylindrical panel subjected to
sinusoidal loading
Lamination a/h Source 𝑤

̅ 𝜎̅1
0/90/0 5 Present 2.082 -0.9913
HOZT [21] 2.081 -0.9751
Shu [45] 2.081 -0.1040
Elasticity [9] 2.118 -1.022
10 Present 0.9300 -0.7401
HOZT [21] 0.9301 -0.7327
Shu [45] 0.9292 -0.7369
Elasticity [9] 0.9396 -0.7463
0/90/0/90 5 Present 2.996 -1.405
HOZT [21] 2.944 -0.401
Shu [45] 2.981 -1.409
Elasticity [9] 3.042 -1.388
10 Present 1.598 -1.139
HOZT [21] 1.580 -1.139
Shu [45] 1.579 -1.129
Elasticity [9] 1.609 -1.137
0/90/0/90/0 5 Present 2.198 -0.1008
HOZT [21] 2.165 -0.9974
Shu [45] 2.187 -1.036
Elasticity [9] 2.205 -1.040
10 Present 1.015 -0.8269
HOZT [21] 1.009 -0.8195
Shu [45] 1.008 -0.8223
Elasticity [9] 1.020 -0.8340
405
Table 4. Non-dimensional displacement and stress for simply supported laminated spherical shells subjected
to sinusoidal loading
Lamination Scheme Source R/a a/h 𝑤

̅ 𝜎̅1
0/90/0 Present 20 10 0.7601 -0.6001
HOZT [21] 0.7524 -0.5852
Present 100 0.7628 -0.6107
HOZT [21] 0.7546 -0.5908
Present 20 100 0.3697 -0.4307
HOZT [21] 0.3621 -0.4263
Present 100 0.4589 -0.5389
HOZT [21] 0.4312 -0.5300
0/90/0/90 Present 20 10 0.7631 -0.5408
HOZT [21] 0.7535 -0.5336
Present 100 0.7608 -0.5428
HOZT [21] 0.7557 -0.5396
Present 20 100 0.4201 -0.3608
HOZT [21] 0.4117 -0.3590
Present 100 0.5089 -0.4792
HOZT [21] 0.5042 -0.4745
45/-45/45 Present 20 10 0.6489 -0.2007
HOZT [21] 0.6395 -0.1918
Present 100 0.6589 -0.2217
HOZT [21] 0.6520 -0.2127
Present 20 100 0.1800 -0.0400
HOZT [21] 0.1704 -0.0317
Present 100 0.3789 -0.1748
406
HOZT [21] 0.3677 -0.1624
45/-45/45/-45 Present 20 10 0.5896 -0.2008
HOZT [21] 0.5432 -0.1645
Present 100 -0.5781 -0.1998
HOZT [21] 0.5613 -0.1838
Present 20 100 0.1509 -0.0408
HOZT [21] 0.1459 -0.0305
Present 100 0.2699 -0.1300
HOZT [21] 0.2617 -0.1234
Analysis of laminated composite hypar shells: Two different types of laminated hypar shells: simply supported and
fully clamped made of material A subjected to sinusoidal loading. Predicted results are reported in table 5 (for
simply supported) and table 6 (for clamped). Same shells subjected to uniformly distributed loading are also studied
and results are reported in Table 7 (for simply supported) and Table 8 (for clamped). The results obtained are near to
those predicted by HSDT [10] using HSDT along with FE method. However, as thickness of shell increases, the
results are not close as compared to thin shells. The reason for this can be ignorance of transverse displacement
function by HSDT [10] during formulations.
Table 5. Non-dimensional central deflection for simply supported hypar shell subjected to sinusoidal loading
(c/a=1)
Source c/a (0/90) (0/90/0) (0/90/90/0) (0/90/0/90)
Present 0 10.7025 4.4982 4.4089 5.1929
HSDT [10] 10.6524 4.3430 4.3441 5.0857
Present 0.05 1.7018 1.5690 1.4026 1.4982
HSDT [10] 1.6785 1.3973 1.3511 1.3866
Present 0.10 0.6897 0.6392 0.5072 0.5210
HSDT [10] 0.5672 0.5390 0.4966 0.4781
Present 0.15 0.3182 0.3001 0.2890 0.2896
HSDT [10] 0.3167 0.2966 0.2716 0.2545
Present 0.20 0.2189 0.2010 0.1982 0.1762
HSDT [10] 0.2180 0.1974 0.1800 0.1680
407
Table 6. Non-dimensional central deflection for clamped hypar shell subjected to sinusoidal loading (c/a=1)
Source c/a (0/90) (0/90/0) (0/90/90/0) (0/90/0/90)
Present 0 2.9018 1.1056 1.1982 1.3269
HSDT [10] 2.8703 1.0783 1.0967 1.2944
Present 0.05 1.0985 0.7025 0.7860 0.8069
HSDT [10] 1.0168 0.6331 0.6440 0.7138
Present 0.10 0.3895 0.2900 0.3001 0.3015
HSDT [10] 0.3335 0.2783 0.2820 0.2988
Present 0.15 0.1610 0.1524 0.1590 0.1500
HSDT [10] 0.1524 0.1420 0.1433 0.1484
Present 0.20 0.1089 0.0998 0.0982 0.0931
HSDT [10] 0.0852 0.0836 0.0837 0.0836
Table 7. Non-dimensional central deflection for simply supported hypar shell subjected to uniformly distributed
loading (c/a=1)
Source c/a (0/90) (0/90/0) (0/90/90/0) (0/90/0/90)
Present 0 17.0895 6.9875 7.0045 9.0236
HSDT [10] 16.9763 6.7055 6.8436 8.1137
Present 0.05 2.4802 1.9826 2.0861 2.1945
HSDT [10] 2.3774 1.8988 1.9629 2.0922
Present 0.10 0.5998 0.5729 0.6098 0.7059
HSDT [10] 0.6193 0.5680 0.5972 0.6252
Present 0.15 0.2578 0.2525 0.2478 0.1952
HSDT [10] 0.2610 0.2493 0.2638 0.1767
Present 0.20 0.1484 0.1462 0.1583 0.1599
HSDT [10] 0.1388 0.1360 0.1434 0.1504
408
Table 8. Non-dimensional central deflection for clamped hypar shell subjected to uniformly distributed
loading (c/a=1)
Source c/a (0/90) (0/90/0) (0/90/90/0) (0/90/0/90)
Present 0 4.0583 1.6025 1.6902 1.8999
HSDT [10] 3.9672 1.4189 1.4859 1.7894
Present 0.05 1.4897 0.9872 0.9979 1.0089
HSDT [10] 1.3371 0.8008 0.8459 0.9640
Present 0.10 0.4038 0.3891 0.4028 0.4105
HSDT [10] 0.3901 0.3265 0.3451 0.3791
Present 0.15 0.1897 0.1984 0.1990 0.1999
HSDT [10] 0.1576 0.1564 0.1613 0.1732
Present 0.20 0.0812 0.0970 0.0989 0.1015
HSDT [10] 0.0805 0.0880 0.0879 0.0917
Analysis of sandwich cylindrical and spherical shell panels: Very few works are available on the analysis of
laminated sandwich shell structures. Under this section, simply supported sandwich cylindrical and spherical panels
are analysed in which face layers are made up of material A, while core is made up of material B (thickness of core
=0.8h and remaining layers are of equal thickness).
Table 9. Non-dimensional deflections for simply supported sandwich spherical panels subjected to sinusoidal
loading
Source a/h R/a (0/90/core/0/90) (45/-45/core/45/-45)
Present 10 5 1.700 1.175
HOZT [21] 1.688 1.171
Present 10 1.785 1.287
HOZT [21] 1.715 1.283
Present 100 1.728 1.324
HOZT [21] 1.724 1.325
Present 100 5 0.3607 0.0800
HOZT [21] 0.3593 0.0795
409
Present 10 0.6500 0.2209
HOZT [21] 0.6495 0.2142
Present 100 0.8897 0.4894
HOZT [21] 0.8854 0.4855
Table 10. Non-dimensional deflections for simply supported sandwich cylindrical panels subjected to
sinusoidal loading
Source a/h R/a (0/90/core/0/90) (45/-45/core/45/-45)
Present 10 5 3.680 3.904
HOZT [21] 3.501 3.855
Present 10 3.498 3.892
HOZT [21] 3.484 3.836
Present 100 3.482 3.834
HOZT [21] 3.479 3.829
Present 100 5 1.86 2.206
HOZT [21] 1.77 2.119
Present 10 1.892 2.201
HOZT [21] 1.858 2.191
Present 100 1.905 2.221
HOZT [21] 1.886 2.216
2.3 Discussion
Number of studies are reported above for laminated composite and sandwich shells. It can be seen that spherical
shells gives less deflection at mid-point when compared to cylindrical sandwich panel keeping concerned factors
same. Even the deflection decreases with increase in a/h value thus thin shells are more effective. Symmetric shells
are found to be performed in a better way as compared to corresponding unsymmetrical shells. Behaviour of shell
panel is greatly influenced by end conditions and type of loading acting upon the structure. This can be reflected in
case of hypar shells (tables 5, 6, 7 & 8). The proposed model is also capable to predict the behaviour of shell more
accurately when compared with those available in literature. This highlights the application of FE software packages
for analysis of laminated composite and sandwich shell structures without any large computational efforts.
410
2.4 Conclusion
Following points can be concluded after performing static studies on laminated composite and sandwich shell
panels:
1. The present model holds fairly good results when compared with those available in literature.
2. Three layered shells are found to be stiffer than subsequent number of laminated shells.
3. Boundary conditions, geometrical parameters and loading conditions significantly affects the behaviour of
laminated structures.
4. All sides clamped shell are stiffer than simply supported shells.
References
1. Reissner E, Stavsky Y (1961) Bending and Stretching of Certain Types of Heterogeneous Aeolotropic
Elastic Plates. J. Appl. Mech. 28:402-408. doi:10.1115/1.3641719.
2. Jones RM (1975) Mechanics of Composite Materials, Mc Graw–Hill, New York.
3. Reddy JN (1945) Mechanics of Laminated Composite Plates and Shells : theory and analysis, Second Edi,
CRC Press, Florida.
4. Schmidt R, Reddy JN (1988) A Refined Small Strain and Moderate Rotation Theory of Elastic Anisotropic
Shells. J. Appl. Mech. 55:611-617. doi:10.1115/1.3125837.
5. Palmerio AF, Reddy JN, Schmidt R (1990) On a moderate rotation theory of laminated anisotropic shells—
Part 1. Theory. Int. J. Non. Linear. Mech. 25:687–700. doi:10.1016/0020-7462(90)90007-V.
6. Palmerio AF, Reddy JN, Schmidt R (1990) On a moderate rotation theory of laminated anisotropic shells-
Part 2. Finite-element analysis. Int. J. Non. Linear. Mech. 25:701–714. doi:10.1016/0020-7462(90)90008-
W.
7. Kapania RK, Mohan P (1996) Static, free vibration and thermal analysis of composite plates and shells
using a flat triangular shell element. Comput. Mech. 17:343–357. doi:10.1007/BF00368557.
8. Pai PF (1995) A new look at shear correction factors and warping functions of anisotropic laminates. Int. J.
Solids Struct. 32:2295–2313. doi:10.1016/0020-7683(94)00258-X.
9. Huang NN (1994) Influence of shear correction factors in the higher order shear deformation laminated
shell theory. Int. J. Solids Struct. 31:1263–1277. doi:10.1016/0020-7683(94)90120-1.
10. Pradyumna S, Bandyopadhyay JN (2008) Static and Free Vibration Analyses of Laminated Shells using a
Higher-order Theory. J. Reinf. Plast. Compos. 27:167–186. doi:10.1177/0731684407081385.
11. Balah M, Al-ghamedy HN (2002) Finite element formulation of a third order laminated nite rotation shell
element. Civ. Eng. 80:1975–1990. doi:10.1016/S0045-7949(02)00222-.
12. Mantari JL, Oktem AS, Guedes Soares C, Static and dynamic analysis of laminated composite and
sandwich plates and shells by using a new higher-order shear deformation theory. Compos. Struct. 94:37–
49. doi:10.1016/j.compstruct.2011.07.020.
13. Mantari JL, Guedes Soares C (2012) Analysis of isotropic and multilayered plates and shells by using a
generalized higher-order shear deformation theory. Compos. Struct. 94:2640–2656.
doi:10.1016/j.compstruct.2012.03.018.
14. Mantari JL, Oktem AS, Guedes Soares C (2012) Bending and free vibration analysis of isotropic and
multilayered plates and shells by using a new accurate higher-order shear deformation theory. Compos.
Part B Eng. 43:3348–3360. doi:10.1016/j.compositesb.2012.01.062.
15. Fard KM, Livani M, Veisi A, Gholami M (2014) Improved high-order bending analysis of double curved
sandwich panels subjected to multiple loading conditions. Lat. Am. J. Solids Struct. 11:1591–1614.
doi:10.1590/S1679-78252014000900006.
16. Chalak HD, Chakrabarti A, Iqbal MA, Hamid Sheikh A (2012) An improved C0 FE model for the analysis
of laminated sandwich plate with soft core. Finite Elem. Anal. Des. 56:20–31.
doi:10.1016/j.finel.2012.02.005.
17. Guo Y, Ruess M (2015) A layerwise isogeometric approach for NURBS-derived laminate composite shells.
Compos. Struct. 124:300–309. doi:10.1016/j.compstruct.2015.01.012.
411
18. Maturi DA, Ferreira AJM, Zenkour AM, Mashat DS (2015) Analysis of Three-Layer Composite Shells by
a New Layerwise Theory and Radial Basis Functions Collocation, Accounting for Through-the-Thickness
Deformations. Mech. Adv. Mater. Struct. 22:722–730. doi:10.1080/15376494.2013.846444.
19. Icardi U, Ferrero L (2011) Multilayered shell model with variable representation of displacements across
the thickness. Compos. Part B Eng. 42:18–26. doi:10.1016/j.compositesb.2010.09.022.
20. Yaqoob Yasin M, Kapuria S (2013) An efficient layerwise finite element for shallow composite and
sandwich shells. Compos. Struct. 98:202–214. doi:10.1016/j.compstruct.2012.10.048.
21. Kumar A, Chakrabarti A, Bhargava P (2013) Finite element analysis of laminated composite and sandwich
shells using higher order zigzag theory. Compos. Struct. 106:270–281.
doi:10.1016/j.compstruct.2013.06.021.
22. Maturi DA, Ferreira AJM, Zenkour AM, Mashat DS (2013) Analysis of Laminated Shells by Murakami’s
Zig-Zag Theory and Radial Basis Functions Collocation, J. Appl. Math. 2013:1–14.
doi:10.1155/2013/123465.
23. Versino D, Gherlone M, Di Sciuva M (2014) Four-node shell element for doubly curved multilayered
composites based on the Refined Zigzag Theory, Compos. Struct. 118:392–402.
doi:10.1016/j.compstruct.2014.08.018.
24. Kumar A, Chakrabarti A, Bhargava P (2013) Vibration of laminated composites and sandwich shells based
on higher order zigzag theory. Engg. Struct. 56:880-888. doi:10.1016/j.engstruct.2013.06.014
25. Cinefra M, Carrera E (2013) Shell finite elements with different through-the-thickness kinematics for the
linear analysis of cylindrical multilayered structures. Int. J. Numer. Methods Eng. 93:160–182.
doi:10.1002/nme.4377.
26. Carrera E, Cinefra M, Li G, Kulikov GM (2016) MITC9 shell finite elements with miscellaneous through-
the-thickness functions for the analysis of laminated structures. Compos. Struct. 154:360–373.
doi:10.1016/j.compstruct.2016.07.032.
27. Carrera E, Pagani A, Valvano S (2017) Shell elements with through-the-thickness variable kinematics for
the analysis of laminated composite and sandwich structures. Compos. Part B Eng. 111:294–314.
doi:10.1016/j.compositesb.2016.12.001.
28. Kwon YW (2013) Analysis of laminated and sandwich composite structures using solid-like shell elements.
Appl. Compos. Mater. 20:355–373. doi:10.1007/s10443-012-9273-8.
29. Brischetto S (2017) Exact three-dimensional static analysis of single- and multi-layered plates and shells.
Compos. Part B Eng. 119:230–252. doi:10.1016/j.compositesb.2017.03.010.
30. Tornabene F, Brischetto S (2018) 3D capability of refined GDQ models for the bending analysis of
composite and sandwich plates, spherical and doubly-curved shells, Thin-Walled Struct. 129:94–124.
doi:10.1016/j.tws.2018.03.021.
31. Reddy JN (1989) On refined computational models of composite laminates. Int. J. Numer. Methods Eng.
27:361–382. doi:10.1002/nme.1620270210.
32. Noor AK, Burton WS (1990) Assessment of Computational Models forMultilayered Composite Shells.
Appl. Mech. Rev. 43:67–97. doi:10.1115/1.3119162.
33. Kant T, Swaminathan K (2000) Estimation of transverse/interlaminar stresses in laminated composites - a
selective review and survey of current developments, Compos. Struct. 49:65–75. doi:10.1016/S0263-
8223(99)00126-9.
34. Abrate S, Di Sciuva M (2017) Equivalent single layer theories for composite and sandwich structures: A
review. Compos. Struct. 179:482–494. doi:10.1016/j.compstruct.2017.07.090.
35. Kapania RK (2013) A Review on the Analysis of Laminated Shells, J. Eng. Mech. 111:88–96.
doi:10.15898/j.cnki.11-2131/td.2011.01.021.
36. Noor AK, Burton WS (1989) Assessment of Shear Deformation Theories for Multilayered Composite
Plates, Appl. Mech. Rev. 42:1. doi:10.1115/1.3152418.
37. Carrera E (2002) Theories and finite elements for multilayered, anisotropic, composite plates and shells.
Arch. Comput. MethodsEngng.9(2):87-140. doi:10.1007/BF02736649.
38. Mackerle J (2002) Finite element analyses of sandwich structures: a bibliography (1980–2001). Eng.
Comput. 9(2):206-245doi:10.1108/02644400210419067.
39. Carrera E (2003) Historical review of Zig-Zag theories for multilayered plates and shells, Appl. Mech. Rev.
56:287-308. doi:10.1115/1.1557614.
40. Zhang YX, Yang CH (2009) Recent developments in finite element analysis for laminated composite
plates. Compos. Struct. 88:147–157. doi:10.1016/j.compstruct.2008.02.014.
41. Kreja I (2011) A literature review on computational models for laminated composite and sandwich panels.
412
Cent. Eur. J. Eng. 1:59–80. doi:10.2478/s13531-011-0005-x.

42. Qatu MS, Asadi E, Wang W (2012) Review of Recent Literature on Static Analyses of Composite Shells:
2000-2010. Open J. Compos. Mater. 02:61–86. doi:10.4236/ojcm.2012.23009.
43. Caliri MF, Ferreira AJM, Tita V (2016) A review on plate and shell theories for laminated and sandwich
structures highlighting the Finite Element Method. Compos. Struct. 156:63–77.
doi:10.1016/j.compstruct.2016.02.036.
44. ABAQUS manual.
45. Xiao-ping S (1996) An improved simple higher-order theory for laminated composite shells. Comput.
Struct. 60(3):343–50. doi:10.1016/0045-7949(95)00406-8
413
Paper ID- 93
DYNAMIC RESPONSE OF WEATHERED JOINTED ROCK SLOPE

HAVING THE TRANSMISSION TOWER
Zaid Mohammad1*( https://orcid.org/0000-0001-6610-8960) Khan Mudassir Ali1 (https://orcid.org/0000-0001-
7824-189X) Sadique M. R1 (https://orcid.org/0000-0002-9570-6801)
mohammadzaid1@zhcet.ac.in, mudassir290@gmail.com, rehan.sadique@gmail.com
1 Department of Civil Engineering, ZHCET, Aligarh Muslim University, Aligarh, U.P., India-Pin: 202002
*Mohammad Zaid (mohammadzaid1@zhcet.ac.in)
Abstract. Risk assessment and stability of slope having any superstructure on its surface is necessary aspect
to observe the vulnerability of slope. The paper deals with stability of the rock slope having the transmission
tower on its top near Karcham Wangtoo hydroelectric power house under the seismic loading. 2D Finite element
analysis using ABAQUS has been carried in this study to calculate the settlement in various phase. The different
weathering stages of basalt rock have been adopted for the in this study. The four different seismic loading has
been considered to observe the effect of light, moderate, strong, and major earthquake having magnitude of
4.6M, 5.6M, 6.6M, and 7.6M respectively on rock slope. It has been observed that Seismic loading with the
joint orientation, weathered rock, has much influenced the stability of slope. It has also been found that critical
angle of joint orientation is 30 degrees with the slope at which slope was unstable having transmission tower at
1d distance from the edge of the slope.
Keywords: Finite element Analysis, Elastoplastic, Mohr Coulomb model, Slope stability, Weathered rock.
1. Introduction
Electricity is the form of energy used by the mankind for myriads of purpose to make the life easier and
comfortable. Demand of electricity has been increased rapidly during recent decades due to population increase,
development of high voltage electric appliances etc. However, in India 13.5% of electricity is produced from the
water by constructing the hydroelectric power plant near the different hydraulic structures [1]. A hydroelectric
power plant consists of a high dam that is built across a large river to create a reservoir, and a station where the
process of energy conversion to electricity takes place and transmission tower to convey the electricity up to the
required place. Stability of any rock slope having the transmission tower on its surface in seismic affected
region should be analyse for safe and efficient use of tower for long time. Stability of slope is influenced by the
various factor like geological features, joint orientation, bedding plane, foliation [2]. Furthermore, soil structure
interaction and seismic effect should also be considered while computing the stability of rock slope [3,4]. There
have been various methods developed to analyse the slope, like probabilistic approach [ 6,7], deterministic
approach using reliability theory [8] stochastics approach with finite element analysis [9], however in his study
2D finite element technique is used to observe the behaviour of transmission tower on the slope stability in
seismic region.
2. Problem Outline
A 2D elastoplastic plain strain seismic analysis has been carried out to analyse the stability of rock slope having
the transmission tower on its top surface. The slope is located near the Karcham Wangtoo hydroelectric power
house along the Satluj river at Wangtu in Kinnaur district of Himachal Pradesh in seismic zone V with stepped
topography as shown in Figure 1 [4]. Cumulative response of joint orientation, weathering class of basalt rock
and position of transmission tower has observed on stability of rock slope. The orientation of joint has been
varied from 0-degree to 90-degrees with an interval of 15 degrees as considered by [5]. To observe the effect of
transmission, tower the position of footing is varied from 0 to 5d from the edge of slope where d is the size of
footing equal to 4.5m. The weathered class of basalt is considered, that is, Highly Weathered Basalt (W3) as
shown in Table 1. Mohr coulomb constitutive model have adopted to simulate the plastic behaviour of rock in
all the analyses.
The effect of light, moderate, strong, and major earthquake having magnitude of 4.6M, 5.6M, 6.6M, and 7.6M
respectively on rock slope used in this study has magnitude on Richter scale which had occur in this region. The
effect of horizontal component of earthquake has been taken into consideration
414
The geometry shown in Figure 1 [4] has three stepped topography having footing of transmission tower on its
top surface and on middle step road of 15 m wide passing near the Karcham Wangtoo Hydropower house, and
last step shows the bed of Satluj river.
Fig. 1. Geometry of Karcham Wangtoo Slope
3. Material Properties
The highly weathered basalt rock mass has been input in the model. The property of rock mass and footing are
shown in Table 1[10].
Table 1. Material Properties of the Rock mass and Footing
Property Values
Basalt Mass Density 1820 kg/m3

Young's Modulus 600 MPa
Poisson Ratio 0.272
Friction Angle 43.87
Cohesion 1.64Mpa
Concrete Mass Density 2400 kg/m3

Young's Modulus 25 GPa
Poisson Ratio 0.15
4. Finite Element Analysis

2D plain strain finite element stability analysis has been performed on the basalt rock slope for gravity loading
in seismic condition followed by the static condition. Interaction of footing with the slope has been assigned as
general surface to contact with coefficient of friction of 0.85 [4]. Normal behavior property of contact has also
been applied for pressure overclosure of hard contact type. The load of transmission tower applied on the
footing was a line load of 28340N/m.
The bottom of slope has been taken as fixed to simulate the rigid base and rock mass extend to a large depth and
roller support was adopted for the sides of the model in which horizontal deformation restrained in x-direction
and vertical settlement allowed in y-direction. The loading and boundary conditions has shown in Figure 2.
415
(a)
(b)
Fig.2. Load and Boundary conditions for Static Loading (a) and Seismic Loading (b).
5. Earthquake loading
This paper compares the effect of change in earthquake intensity. There are four different magnitude
earthquakes have been considered, that is, 4.6M, 5.6M, 6.6M, and 7.6M. The acceleration-time history of all the
mentioned earthquakes have been applied in the load section in Abaqus/Explicit. These earthquakes are
considered based on different classes of earthquake, light earthquake-4.6M, moderate earthquake 5.6M, strong
earthquake 6.6M, major earthquake 7.6M.
A 4.6M light earthquake struck in the Gulf of Mexico at southeast of New Orleans, Louisiana, 2018. The
acceleration-time history is shown in figure 3(a). A 5.6M moderate earthquake has struck at Republic of
Guatemala in South America, 2018 and its acceleration time history shown in figure 3(b). A 6.6M strong
earthquake has struck in Andrean of Islands, Aleutian Island, 2018, shown in figure 3(c). And a 7.6M major
earthquake struck at Caribbean Sea between the coast of Honduras and the Cayman Islands and is shown in
figure 3(d).
416
2
Acceleration (in 1
m/s^2) 1
0
-1 0 2 4 6 8 10 12
-1
Time (in seconds)
(a)
1
Acceleration (in
0.5
m/s^2)
0
0 5 10 15 20 25 30
-0.5
-1
Time (in seconds)
(b)
2
Acceleration
1
(m/s^2)
0
0 5 10 15 20 25
-1
-2
Time (in seconds)
(c)
20
Acceleration
10
(m/s^2)
0
0 50 100 150 200 250
-10
Time (in seconds)
(d)
Fig. 3. Acceleration-Time history of different earthquakes used in the study (a) 4.6M Light Earthquake of
Louisiana, 2018, (b) 5.6M Moderate Earthquake of Guatemala, 2018, (c) 6.6M Strong Earthquake of Aleutian
Island, 2018, and (d) 7.6M Major Earthquake of Caribbean Sea, 2018.
6. Mesh and Element Type

The slope has been discretized with 3-node linear plane strain triangular element (CPE3 element type). The
scaling factors were defined as linear bulk viscosity parameter is 0.06 and quadratic bulk viscosity parameter
equals to 1. The global seed size was 0.8 for the footing and the transmission tower. The triangular shaped free
meshed was assigned to the modal geometry. The Figure 3 shows the type of mesh adopted in this study.
For earthquake loading, infinite boundary condition has been applied on both the sides of the slope model. For
the infinite boundary condition, CINPE4-a 4-node linear 2-D quadrilateral element type has been adopted.
417
Fig. 4. Meshing of the model
7. Results and Discussion

A two dimensional numerical simulation has been carried out to analyse the stability of rock-slope having
transmission tower at its top. The location of transmission tower has been varied from 0d (placed near the edge
of slope) to 5d (where d is the width of footing = 4.5m). The effect of discontinuity has also been considered by
varying orientation of joint set with the slope. The orientation has been varied from 0-degree (i.e. parallel to
slope) up to 90-degrees (i.e. normal) with the slope. Effect of weathering on the strength of rock mass has been
by considered. The properties of highly weathered basalt rock has been incorporated in the study. To seek the
dynamic behaviour of slope, acceleration-time history of four different earthquake of different magnitude has
been incorporated explicitly.
0.07
0.06
Total Maximum Deformation (in meter)
0.05
0d
0.04
1d
2d
0.03
3d
4d
0.02 5d
0.01
0
0 10 20 30 40 50 60 70 80 90
Angle of Orientation of Joint with Slope (in Degrees)
(a)
418
0.4
0.35
Total Maximum Deformation (in meters)
0.3
0.25
0d
0.2 1d
2d
0.15 3d
4d
0.1 5d
0.05
0
0 10 20 30 40 50 60 70 80 90
Angle of Orientation of Joint with Slope (in Degrees)
(b)
0.7
0.6
Total Maximum Deformation (in meter)
0.5
0d
0.4
1d
2d
0.3
3d
4d
0.2 5d
0.1
0
0 10 20 30 40 50 60 70 80 90
Angle of Orientation of Joint with Slope (in degrees)
(c)
419
7
Total Maximum Deformation (in meters)
0d
5
1d
4 2d
3d
3
4d
2 5d
0
0 10 20 30 40 50 60 70 80 90
Angle of Orientation of Joint (in Degrees)
(d)
Fig.5. Total Maximum Deformation for varying angle of orientation of joint with slope for different location of footing under
earthquake loading(a) 4.6M Light Earthquake of Louisiana, 2018, (b) 5.6M Moderate Earthquake of Guatemala, 2018, (c)
6.6M Strong Earthquake of Aleutian Island, 2018, and (d) 7.6M Major Earthquake of Caribbean Sea, 2018.
Figure 5 shows variation of total maximum deformation with change in orientation of joint for different location
of transmission tower footing for earthquakes of different magnitude. The variation of deformation at the edge
of slope has been reported in all cases. A significant increase in the magnitude of total maximum deformation
has observed from 15-degrees to 30-degrees of angle of orientation with slope. The magnitude of maximum
deformation decreases from 30-degrees angle of orientations with slope to 90-degrees of joint orientation for all
cases of different location of footing of transmission tower and magnitudes of earthquake.
7.6M 6.6M 5.6M 4.6M
2.5
Deforamation at the Edge of Slope (in
1.5
meter)
0.5
0
0d 1d 2d 3d 4d 5d
Position of Transmission Tower Footing
Fig. 6. Comparison of Total Maximum Deformation at the edge of slope for different magnitude of Strong Ground Motion
for 0d location of footing.
420
The figure 6 shows comparison of effect of different magnitude of earthquake on the stability of slope for 0d
location of transmission tower footing. The point 0, 1, 2, 3, 4, and 5 on x-axis represents 0d,1d, 2d, 3d, 4d, and
5d location of footing respectively. The Figure 5 shows that 1d location of footing has maximum deformation
in comparison to other cases.
8. Conclusion
The stability analysis of a highly weathered jointed rock slope having transmission tower footing has been
checked. Finite element analysis has been performed for seismic analysis of slope under different earthquake
using ABAQUS/ Explicit. Following point may be concluded from the present study:
1. For 15-degrees of joint orientation with the slope and 2d location of transmission tower footing, the
minimum increase in deformation has been observed by an amount of 2% of 4.6M earthquake
deformation at the location and maximum increase in deformation has been observed by 18% of 4.6M
earthquake deformation at the location, for 0-degree of joint orientation with slope at 0d location of
footing when earthquake magnitude increases from 4.6M to 5.6M.
earthquake deformation at the location, for 90-degrees of joint orientation with slope at 2d location of
earthquake deformation at the location, for 15-degree of joint orientation with slope at 0d location of
4. The above comparison of deformation percentage shows that 6.6M to 7.6M or more has much serious
effect on the stability of transmission tower lying in seismically active area.
5. The angle of orientation of joint with slope must have to be greater than or less than 30-degrees for
slope to be stable.
6. The earthquake of magnitude greater than 7M on Ritcher scale has much significant effect on the slope
as compared to lower magnitude earthquakes for highly weathered basalt rock.
7. 1d location of transmission tower footing (4.5m distance between the edge of slope and edge of
footing) is most critical as compared to different locations.
8. Total maximum deformation proved to be more time consuming observation to decide stability of slope
as compared to calculation of factor of safety.
9. The method adopted proved to be universal as it can be applied for variety of geometry, geology, etc.
10. Strain gauges can be used to note direct observation of deformation on site which can be utilised to find
the insitu stability of slope before going for construction at site.
9. References
[1] http://www.cea.nic.in/reports/monthly/executivesummary/2017/exe_summary-06.pdf, October 12th ,2018
[2] L. R. Alejano, A. Rodriguez-Dono, E. Alonso, and G. Fdez.-Manín, “Ground reaction curves for tunnels
excavated in different quality rock masses showing several types of post-failure behaviour,” Tunn. Undergr. Sp.
Technol., vol. 24, no. 6, pp. 689–705, 2009.
[3] D. V. V Griffiths and P. a. A. Lane, “Slope stability analysis by finite elements,” Geotechnique, vol. 49, no.
3, pp. 387–403, 1999.
[4] M. R. Sadique, “Analysis of Bridge Pier Foundation On Jointed Rocks,” M. Tech Dissertation, Indian
Institute of Technology Roorkee, 2010.
[5] M. W. Naqvi, M. Zaid, M. R. Sadique, and M. M. Alam, “Dynamic Analysis of Rock Tunnels Considering
Joint Dip Angle: A Finite Element Approach,” 13th International Conference on Vibration Problems, IIT
Guwahati, 2017
421
[6] B. Xu, & B. Low, 2006. Probabilistic stability analyses of embankments based on finite-element method.
Journal of Geotechnical and Geoenvironmental, 2006
[6] K. Gavin, & J. Xue, Design Charts for the Stability Analysis of Unsaturated Soil Slopes. Geotechnical and
Geological Engineering, 28(1), pp. 79–90. 2010.
[7] Cormac Reale, Jianfeng Xue, Zhangming Pan, and Kenneth Gavin ,Deterministic and probabilistic multi-
modal analysis of slope stability. Computers and Geotechnics, 66((5)), pp. 172–179,2015.
[8] Maria Lia Napoli, Monica Barbero, Elena Ravera, Claudio Scavia A stochastic approach to slope stability
analysis in bimrocks International Journal of Rock Mechanics and Mining Sciences journal homepage:
www.elsevier.com/locate/ijrmms, 101 2018,41-49
[9] A. Gahoi, M. Zaid, S. Mishra, and K. S. Rao, “Numerical Analysis of the Tunnels Subjected to Impact
Loading,” in INDOROCK, New Delhi, 2017.
422
Paper ID- 98
Response of Single and Multilayered Flexible base for Static and

Earthquake loading under Framed RC Structure
Gaurav D. Dhadse1 (https://orcid.org/0000-0002-8857-9509),

Dr. G.D. Ramtekkar2 and
Dr. Govardhan Bhat3 (https://orcid.org/0000-0002-8683-1918)
1
National Institute of
Technology, Raipur, C.G., India
gdhadse@yahoo.com
Abstract. As it is well known that, if the effect of Soil Structure Interaction

(SSI) is to be assessed, then structure must be placed on physical soil mass with
actual properties to get the perfect behavior of soil and structure rather than
Winkler’s Spring Model. In case of non interaction analysis problem, the
structure is to be placed on non yielding support; hence it is assume that there
will be zero displacement beneath the footing or displacement is considered less
than the permissible limits as stated by IS codes. But the actual condition is
totally different and settlement due to soil properties can be seen by naked eyes.
In non interaction analysis the footings are design based on Safe Bearing
Capacity (SBC) of Soil, but as per literature review and SSI analysis done in this
paper it is observed that the footing displacement is more than permissible value
and stress value just below the footing is also higher than SBC.
To demonstrate the actual nature of ground response, a multi-storey frame is

analyzed with 4 different cases. The SSI System modeling is done using Finite
Element Method ANSYS Software program. Portal frame is modeled as elastic
(linear), whereas the ground is modeled as both elastic-plastic (linear elastic and
non-linear). The study gives insight into variation of displacement and stress
intensities in soil mass while considering linear and non-linear behavior of
ground.
In analysis of Soil, the vertical displacements in soil mass are plotted along
depth and discussed whereas stresses are plotted along the width with variation
for different types of soil and different loadings. Thus after complete analysis, it
is observed that, both the stresses and displacements are out of permissible limit
after considering actual stiffness of soil.
Keywords: SSI, Finite Element Method, Stress, Displacement, Linear and non
Linear Soil Mass, etc.
1. Introduction
In order to understand the effect of Soil Structure Interaction over general Fix support condition,
the structure must be placed on soil mass and then allow the structure to behave accordingly to the support
condition of soil mass. The amount of soil mass required to be taken while modeling is decided by Shear
Wave Velocity criteria and direct method is used for analysis the SSI problem. Also, about 90%
researchers did their work considering soil mass as linear and single layered and none of them provide
solid reason for providing extent of soil mass in modeling. So here in this paper, the above mentioned
gaps in study is fulfilled with the help of analysis of four models considering both single layered and multi
layered as well as linear and non linear soil mass as given below. Output of all analysis shows amount of
displacement as well as stress intensity in soil. Paper further converges on importance of stiffness of soil
over safe bearing capacity while designing structure. In this paper four models are considered which are as
follows,
1. Single layered soil + frame structure model (Gravity loading)
2. Single layered soil + frame structure model (Gravity loading + Lateral loading)
3. Multi layered soil+ frame structure model (Gravity loading)
4. Multi layered soil+ frame structure model (Gravity 423 loading + Lateral loading)
Thus SSI analysis is carried out for all above mentioned models considering Elastic and Elasto-
plastic soil mass. Also for the analysis, Finite Element Modeling of frame structure, isolated footing and
semi-infinite soil mass is done in ANSYS 11.0. Basic aim of this paper is to check the behavior of soil
mass and understanding the settlements and stress distribution pattern in soil mass.
2. Models under Investigations

In present problem a 3 bay x 3 storey RCC space frame resting on homogeneous soil mass with
elastic and elasto-plastic properties and subjected to gravity and combined gravity + lateral loading with
single layered and multilayered soil mass is analyzed. The problem under consideration is symmetric
about both axes in terms of material, geometry and loading. The superstructure of proposed model is
depicted in Figure 1.
Fig. 1: Plan showing columns and footings

To investigate the interaction behavior, the interaction analyses are carried out for the following
two cases considering all frames of structure as plane frame.
Case-1: The interaction analysis of plane frame isolated footing-soil system considering the columns
supported on individual column footings and resting on soil media (Elastic Analysis). (IAE)
Case-2: The interaction analysis of plane frame isolated footing-soil system considering the columns
supported on individual column footings and resting on soil media (Elasto-plastic Analysis). (IAEP)
where,
IAE - Interaction Analysis Elastic
IAEP - Interaction Analysis Elasto-Plastic
The frame and foundation is considered to be elastic and soil mass is considered as both elastic and elasto-
plastic. The geometric and material properties of proposed models are given in Table 1(Single layered)
and Table 2 (Multi layered). And elevation of Frame 1, Frame 2, Frame 3, Frame 4 and plan of all beams
are shown in figure 2 (a), 2(b) and 2(c), 2(d) respectively.
Fig. 2 (a): Elevation of Frame 1 and Fig. 2 (b): Elevation of Frame 2 and
Frame 3 Frame 4
Fig. 2(c): Plan showing Ground beams 424 Fig. 2(d): Plan showing Slab
and Plinth beams beam for first, second and terrace floor
Table 1: Geometric and material properties of frame, footing and soil mass (Single Layered)
COMPONENT DESCRIPTION DATA
Number of storey 3
Number of bays in X direction 3
Number of bays in Y direction 3
Floor to floor height 3.0 m
Plinth height 1.0 m
Bay width in X direction 4.0 m

Frame
Bay width in Y direction 4.0 m
Beam dimensions (0.23 x 0.3) m
Columns C1,C13,C4,C16(A) (0.23 x 0.3) m
Columns C2,C14,C3,C15,C5,C9,C8,C12(B) (0.38 x 0.23) m
Columns C6,C10,C7,C11(C) (0.23 x 0.45) m

Thickness of all slabs 0.15 m
Footing under Columns C1,C13,C4,C16(F1) 1.5m x 1.5m x 0.35m
Isolated Footing Footing under Columns C2,C14,C3,C15,C5,C9,C8,C12(F2) 1.8m x 1.8m x 0.4m
Footing under Columns C6,C10,C7,C11(F3) 2.2m x 2.2m x 0.5m
Modulus of elasticity of concrete for(M20) 22360 x 106 N/m2

Concrete
Poisson’s ratio of concrete 0.15
Extent of Soil Mass* 30.0m x 15.0m
Modulus of elasticity of soil 100 x 106 N/m2

Soil Mass
(Medium hard clay) Poisson’s ratio of soil 0.2
Cohesion 50 x 103 N/m2
Table 2: Geometric and material properties of frame, footing and soil mass (Multilayered)
COMPONENT DESCRIPTION DATA
Frame Same data for Frame as given in Table 1

Isolated Footing Same data for Isolated Footings as given in Table 1
Concrete Same data for Concrete as given in Table 1
Soil Mass Total Extent* 60.0 m x 58.34 m

Depth 13.65 m
First layer (Firm Modulus of elasticity of soil 100 x 106 N/m2

Clays or Stiff Clays
or Medium hard Poisson’s ratio of soil 0.30
clay) Cohesion 50 x 103 N/m2
Frictional angle 190
Depth 27.20 m
Second Layer (Silty Poisson’s ratio of soil 0.25
Sand)

425
Depth 17.49 m

Third Layer (Dense
Poisson’s ratio of soil 0.20
Sand and Gravel)
* = Extent of soil mass depend upon Significant Depth, which is defined as the depth up to which the
stress increment due to applied loads can produce significant settlement and shear stress.
Thus significant depth is generally taken as the depth at which the vertical stress is 1/5 th of load
intensity. In this case vertical stress is calculated from Bousinesq’s Equation. Also depth of soil mass is
calculated from shear wave velocity criteria.
3. Loads
3.1 Gravity Loads
Loads are calculated by usual way i.e. for slab, Live Load = 3kN/m 2, Floor Finish = 1.25 kN/m2,
and after this factored it and distributed on beams as per IS 456-2000 [1]. For beams, the loads of walls
are calculated and factored it.
As far as dead load is concerned, Ansys’11 calculated dead weight of members from mass
density and acceleration due to gravity hence these properties are required while assigning. Following
Table 4 shows factored UDL intensities on all beams [1].
Table 4: Factored UDL Intensities on Beams
STRUCTURAL COMPONENT U.D.L. INTENSITIES (N/m)
Ground Beams 25530
Plinth beams 9315
First and Second floor
Outer slab beams 34630
Inner slab beams 41315
Terrace floor
Outer slab beams 22900
Inner slab beams 38900
3.2 Lateral Loads

Gravity loading is same as section 3.1 of first model and Lateral loading is calculated according
to IS 1893-2002 [2] for Seismic Zone III and Ordinary Moment Resting Frame. Thus distribution of Base
Shear i.e. Lateral Load Intensity on each floor of frame is given in Table 5 and Gravity loading is given in
Table 5 respectively.
Table 5: Lateral Load Intensity on Each Floor
FLOOR LOAD INTENSITIES (N)/FRAME
Terrace floor 60170
Second floor 46095
First floor 20870
Ground level floor 525
4. Finite Element Modeling

The interaction analysis of the problem is carried out using ANSYS software (Version 11.0). The
finite element discretization of the problem is shown in Figure 3.
426
Fig. 3: Finite element discretization of frame-footing-soil system (Single layered and Multi layered)
The frame structure, footing and soil mass is discretized with 8 noded plane stress element
(PLANE 82) for case 1 and case 2 with two degree of freedom per node (Ux and Uy). It provides more
accurate results for mixed automatic meshes and can tolerate irregular shapes without as much loss of
accuracy. The 8-node elements have compatible displacement shapes and are well suited to model curved
boundaries. The element may be used as a plane element or as an axisymmetric element. The element has
plasticity, creep, swelling, stress stiffening, large deflection, and large strain capabilities [3].
Fig. 4: PLANE 82 element geometry

It is assumed that various member joints are perfectly rigid. The vertical displacements in soil mass are
restrained at the bottom boundary whereas horizontal displacements are restrained at vertical boundaries.
The soil mass is idealized as isotropic, homogeneous, half-space model. For Case 2, elastic-plastic
behavior of the ground is modeled using Drucker-Prager criterion in ANSYS’11. The input consists of
only three constants:
• The cohesion value (must be > 0)
• The angle of internal friction
• The dilatancy angle.
The interface characteristics between the isolated footing and soil are represented by TARGE169 and
CONTA172 elements. The element size for frame and footings are taken as 100 mm x 100 mm. The soil
mass is discretized with 1000 mm x 1000 mm finer meshes in close vicinity of footing where stresses are
of higher order [3].
5. Analysis of Soil Mass

In Soil mass analysis, the displacement as well as stress approach is used. And this part of study discussed
about the response of soil mass for elastic and plastic soil properties, due to frame structure over it. Thus
soil mass for the same models mentioned is analyzed and discussed.
5.1 Displacement Approach

Here the vertical displacement of soil mass in all models for IAE and IAEP analysis is presented
and discussed. Also two types of geometry are adopted, while modeling soil mass in mentioned models.
So, vertical displacement is plotted along the depth of soil mass to study the variation of displacements.
Also it is not possible to check displacement at every meter width of soil mass so unique approach is used,
in which displacement is plotted along the depth for mid width of soil mass as shown in figure 5
427
STARTING
STARTING
POINT
POINT
END POINT END POINT

Fig. 5: Single Layered and Multi layered Soil mass height consider for comparing vertical Displacements
• As depth of soil mass is taken according to literature available i.e. 15 m (assuming there is bed rock
below this depth) so bottom boundary is restrained against translations in both direction and hence at
15 m depth displacement must be zero regardless of soil material stiffness.
• Also main reason behind switching 15 m depth in models 3 and model 4 is that, in general there
should not be bed rock at such shallow depth and 15 m decided on the basis of significant depth
criteria in which depth is taken up to which the stress increment due to applied loads can produce
significant settlement and shear stress. Hence it is assume that there should not be significant
settlement and shear stress below this depth which is practically wrong. Hence in order to produce
exact volume of soil mass required to model SSI problem the shear wave velocity concept in
Eorocode 8 is used to model layered soil mass. Thus more exact results are found for these three
layered soil mass with increasing stiffness than single layered soil mass with same stiffness as top
layer of multilayered.
As in Gravity load models the displacement is maximum at central width of soil mass and in
Gravity + Lateral load model, due uplifting the maximum displacement of soil mass shifted towards last
footing. Hence in order to show the variation displacement in soil mass for both these loading, this path is
adopted. As discussed earlier frame 2 and frame 4 shows worst results so comparison is also done for
these two frames only. Following figure 6, 7, 8, and 9 shows the variation of vertical displacement along
above mentioned path for single layered and multilayered soil mass.
SINGLE LAYERED SOIL MASS
(GRAVITY & GRAVITY+LATERAL)
SECOND FRAME
20
IAEP SINGLE
Vertical Dispt. of Soil Mass (mm)
LAYERED(GRAVITY)
16
IAEP SINGLE LAYERED
12 (GRAVITY+LATERAL)
IAE SINGLE LAYERED

8 (GRAVITY)
4 IAE SINGLE LAYERED

(GRAVITY+LATERAL)
0
0 5 10 15 20
Depth of Soil Mass (m)
Fig. 4.70: Variation of Vertical Displacement along mid width depth of soil mass for single layered soil mass
(Second Frame)
428
18 SINGLE LAYERED SOIL MASS

IAEP SINGLE

16 LAYERED
FOURTH FRAME (GRAVITY)
14
IAEP SINGLE
12 LAYERED
10 (GRAVITY+L
ATERAL)
8 IAE SINGLE
LAYERED
6
(GRAVITY)
4
IAE SINGLE
2 LAYERED
(GRAVITY+L
0
ATERAL)
0 5 10 15 20
Fig. 4.71: Variation of Vertical Displacement along mid width depth of soil mass for single layered soil mass
(Fourth Frame)
MULTI LAYERED SOIL MASS

70 SECOND FRAME IAEP MULTI
LAYERED
60 (GRAVITY)
50 IAEP MULTI
LAYERED
(GRAVITY+
40
LATERAL)
IAE MULTI
30 LAYERED
(GRAVITY)
20
IAE MULTI
10 LAYERED
(GRAVITY+
0 LATERAL)
0 10 20 30 40 50 60 70
Fig. 4.72: Variation of Vertical Displacement along mid width depth of soil mass for multi layered soil mass
(Second Frame)
MULTI LAYERED SOIL MASS

FOURTH FRAME IAEP MULTI
70 LAYERED
(GRAVITY)
60
IAEP MULTI
50 LAYERED
(GRAVITY+L
40 ATERAL)
IAE MULTI
30 LAYERED
(GRAVITY)
20
IAE MULTI
LAYERED
10
(GRAVITY+L
ATERAL)
0
0 10 20 30 40 50 60 70
Fig. 4.73: Variation of Vertical Displacement along mid width depth of soil mass for multi layered soil mass
(Fourth Frame)
429
Thus from the above curves followings points are concluded,

• In single layer soil mass, depth is 15 m and elastic and elasto-plastic soil properties are used. Thus
figure 6 and 7 shows that there is gradual decrease in displacement for elastic and elasto-plastic soil
mass.
• In single layered soil mass IAEP (gravity + lateral) shows maximum displacement than IAEP
(gravity) hence it concluded that after applying lateral loading, mid width displacement increases than
gravity loading.
• For IAE also the same variation of results is found in single layered.
• In multi layered IAEP soil mass displacement suddenly decreases as layer changes and become
stiffer. But for IAE variation of displacement is again gradual and displacements are less than IAEP.
• As discussed earlier in horizontal displacement of frame and vertical displacement of footing, that
footings get uplifted after applying lateral load so at mid width the displacement in IAEP due to
gravity loading is more than displacement due to gravity + lateral loading. Hence figure 8 and figure
9 show the same.
• In single layered, the same frame is come out safe but in multilayer the frame tilted. Hence, if more
practical approach is used more practical results are found.
5.2 Stress Approach

In orders to show the distribution of stress in soil mass, the graphs are plotted between width of
soil mass and stress intensity. The width of soil mass at mid depth of each layer is consider for multilayer
model and width at 1/3 rd depth and 2/3 rd depth is considered for single layered soil mass model. The
exact location of layers along which the stress is plot is shown in figure 10(a) and 10(b)
STARTING
END
POINT
POINT
Fig. 10 (a): Location width at 1/3rd and 2/3rd depth for single layered soil
STARTING
POINT
END
POINT
Fig. 10 (b): Location width at mid depth of each layer for multi layered soil
Thus stress is plot at above mentioned location for IAE and IAEP in single and multilayered (Gravity and
Gravity+ lateral loading) models and presentation of 430
stress is done in following figures 11and 12.
1. Single layered (Gravity + Lateral)

FRAME 2 IAE FRAME 2 IAEP
200000 200000 FIRST
FIRST
STRESS INTENSITY
STRESS INTENSITY
LAYER1/3
LAYER 1/3
150000 150000 DEPTH
DEPTH
(N/SQ.M)
(N/SQ.M)
FIRST LAYER
FIRST 2/3 DEPTH
100000 LAYER 2/3 100000
DEPTH
50000 50000
0 0
0 20 40 0 10 20 30 40
WIDTH OF SOIL MASS (M) WIDTH OF SOIL MASS (M)
FRAME 4 IAE FRAME 4 IAEP

STRESS INTENSITY (N/SQ.M)
FIRST FIRST

180000 LAYER 1/3 200000 LAYER1/3
160000 DEPTH DEPTH
140000 150000
120000 FIRST FIRST
100000 LAYER 2/3 LAYER 2/3
80000 100000
DEPTH DEPTH
60000
40000 50000
20000
0
0
0 20 40
0 10 20 30 40
WIDTH OF SOIL MASS (M)
Fig. 11: Distribution of stress along width for soil mass of single layered (gravity + lateral) IAE and IAEP
model
2. Multi layered (Gravity + Lateral)

FIRST LAYER
FRAME 2 IAE FIRST LAYER FRAME 2 IAEP

MIDDLE
MIDDLE
140000 140000
120000 120000 SECOND
SECOND LAYER
100000 LAYER 100000
80000 MIDDLE
MIDDLE 80000
60000 THIRD LAYER
THIRD LAYER 60000
40000 MIDDLE
MIDDLE 40000
20000
0 20000
0 20 40 60 80 0
0 20 40 60 80
FRAME 4 IAE FIRST FRAME 4 IAEP FIRST

LAYER LAYER
140000 MIDDLE 140000 MIDDLE
120000 SECOND 120000

SECOND
100000 LAYER 100000 LAYER
MIDDLE
80000 80000 MIDDLE
60000 THIRD
LAYER 60000 THIRD
40000 MIDDLE 40000 LAYER
20000 MIDDLE
20000
0 0
0 50 100 0 50 100
WIDTH OF SOIL MASS (M) WIDTH OF SOIL MASS (M)
Fig. 12: Distribution of stress along width for soil mass of multi layered (gravity + lateral) IAE and IAEP
model
431
Thus from above curves, the following points are concluded,

• From all the curves, it shows that stress is distributing on soil mass considered. Also as we go dipper
in soil mass the distribution curve become more flat thus it shows that stress is distributing and get
vanished after suitable depth, of course such depth is adopted in all models. Hence SSI model must be
correct if this stress distribution follows soil mass in correct way, as it is observed that if depth is
15 m then stress intensity increase than 58.34 m depth hence it is necessary to take proper depth for
analysis.
• As in single layered soil mass the depth is less than multi layered soil mass, thus for this shallow
depth in single layered, stress induced due to structural load also more so distribution space required
for distribution of stress is also limited hence if we consider actual depth regardless significant depth
then stress distribution also shows best result and value of stress in multilayered model also less than
15.0 m single layered model.
• In single layered soil mass, Frame 2 and Frame 4 shows maximum stress but in multi layered soil
mass it doesn’t shows result like this.
• If we compare stress in IAE and IAEP side by side then it shows that the stress value is almost same
but the response of soil changes from IAE to IAEP for upper layer of all models. Thus soil behave as
plastic keeping stress value nearly same but response after loading majorly changes. Hence plasticity
of soil must be considered for accurate analysis.
• In multi layered models stress in IAEP under exterior footings increases slightly than IAE and for
inner footings the stress values are slightly decrease than IAE. Thus exact response can be found out.
• Also in lateral + gravity loading the peak value of stress slightly shifted in opposite direction of
application of lateral load. Of course layer nearer to footings shows drastic effect but dipper layer
shows slight change in behavior but shifting of peak can easily be seen.
• Also after checking the upward soil pressure on footing, the maximum value in single layered soil
mass is found to be 3295 kN/m2 and for multi layered soil mass is 2607 kN/m2; both values are
greater than 200 kN/m2 (SBC of soil is same for single layer and top layer of multi layered model).
Hence footings are in tension and structure may fail.
• Thus designers may switch to other type of footing but they have to check design again for SSI effect
or they may increase the stiffness of soil mass by any soil stabilization technique.
• As from the stress analysis, the stress just below the footings is so high that soil fails its strength and
that is the basic reason of increasing the vertical displacement.
• Therefore both the stress and displacement are out of limit after considering actual stiffness hence
while designing, in addition to SBC, the stiffness of soil mass must be considered.
6. Conclusion
Soil analysis is done and vertical displacement in soil mass are plotted along depth and discussed. After
this, distribution of stress in soil mass is plotted along width. Thus vertical displacement and stress
intensity are out of limit for frame and soil whereas horizontal displacement is out of limit for frame only,
hence this provides solid reason for saying that there is importance of Soil Structure Interaction of Framed
Structure.
1. Stress in soil mass is distributing in nature as go dipper stresses try to vanish but intensity of stress at
just below footing is so high that soil fails but while modeling soil mass those stress are most
important so that displacement or stress should not be affected by boundary conditions.
2. Maximum stress value in 15 m depth soil mass is greater than 58.34 m depth soil mass hence it can be
concluded that for proper distribution of stress proper depth must be considered.
3. Also after checking the upward soil pressure on footing, the maximum value in single layered soil
mass is found to be 3295 kN/m2 and for multi layered soil mass is 2607 kN/m2; both values are
greater than 200 kN/m2 (SBC of soil is same for single layer and top layer of multi layered model).
Hence footings are in tension and structure may fail.
4. Thus designers may switch to other type of footing but they have to check design again for SSI effect
or they may increase the stiffness of soil mass by any soil stabilization technique.
5. As from the stress analysis, the stresses just below the footings are so high that soil fails its strength
and that is the basic reason of increasing the vertical displacement.
6. Therefore both the stress and displacement are out of limit after considering actual stiffness hence
while designing, in addition to SBC, the stiffness of soil mass must be considered.
7. Finite element method has found to be very useful method to study the soil structure interaction effect
with rigor. In fact, the technique becomes useful to incorporate the effect of material nonlinearity,
non-homogeneity and interface modeling of soil and foundation.
432
References:
1. IS 456 – 2000, ‘Indian standard Criteria for Plain and Reinforce Structure’, ICS 91.100.30 New Delhi.
2. IS 1893 – (Part 1)-2002 ‘Indian Standard Citeria for Earthquake Resistant Design of Structures’ (ICS
91.120.25).
3. C. S. Krishnamoorty, ‘Finite Element Analysis’, 2nd edition, Mc Graw Hill, New Dehli (2010).
4. Siddharth G. Shah, Solanki C.H., Desai J.A., ‘Soil structure interaction analysis methods - State of art-
Review’, International Journal of Civil and Structural Engineering, ISSN 0976 – 4399, Volume 2, No 1
(2011).
5. F. Jancy, A. Rajagopal, B. Umashankar, M. R. Madhav, ‘Finite Element Modeling of Ground - Structure
Interaction Considering Non-Linear Response of the Ground’, Proceedings of Indian Geotechnical Conference
Kochi (Paper No.N-281) (2011).
6. M.Garevski et.al. ‘Effect on Soil Medium on Response of Base Isolated Multistory Frame Structures’, 15 th
WCEE Conference Lisboa (2012).
7. H.M. Rajashekhar Swamy, ‘Non-linear Dynamic analysis of Soil Structure Interaction of Three Dimensional
Structure for Varied Soil conditions’, Doctor of Philosophy In Civil Engineering Thesis, Department of Civil
Engineering, Manipal Institute of Technology Manipal (2012).
8. Mohamed Saad Eldin, Arafa El-Helloty, ‘Effect of Opening on Behavior of Raft Foundations Resting on
Different Types of Sand Soil’, International Journal of Computer Applications (0975 – 8887), Volume 94 –
No.7 (2014).
9. D. K. Jain and M. S. Hora, ‘Interaction Analysis of Space Frame-Shear Wall-Soil System To Investigate
Foundation Forces Under Seismic Loading’, ARPN Journal of Engineering and Applied Sciences, ISSN 1819-
6608, Vol. 9, No. 8 (2014).
10. Vivek Garg and M.S. Hora , ‘Interaction Effect of Space Frame-Strap Footing-Soil System on Forces in
Superstructure’, ARPN Journal of Engineering and Applied Sciences, ISSN 1819-6608, VOL. 7, NO. 11
(2012).
11. Gaurav D. Dhadse, Milind V. Mohod, ‘Importance of Soil Structure Interaction for Framed Structure’,
Proceedings of International Conference on Advances in Civil and Mechanical Engineering Systems(ACMES
2014) , 23-24 Dec.2014, Government College of Engineering, Amravati. (C-25) (pp 18-27) (2014).
12. Gaurav D. Dhadse, , ‘Effect of Soil Structure Interaction for Framed Structure Resting on Multilayered Soil
Mass’, International Journal of Engineering Research, Volume No.5, Issue Special 1, pp : 99-105, January,
2016, ISSN:2319-6890(online),2347-5013(print) (2016).
13. Gaurav Dhadse, ‘Effect of Soil Structure Interaction on Vertical Irregularity in Framed RC Structure’,
Proceedings of “The Sixth International Congress on Computational Mechanics and Simulation” at IIT
Mumbai on 27 June to 01July (2016).
433
Paper ID- 100
An Approach for Studying Dynamics of A Rotor Bearing System using

ANSYS
Kedar Kishor Patil1, Sanjay Hari Sawant2
1
Research Scholar, Department of Mechanical Engineering, Shivaji Universty, Kolhapur, 416013, India.
kedarkishorpatil@gmail.com
2
Research Guide, Department of Mechanical Engineering, Shivaji Universty, Kolhapur, 416013, India
sanjaysawant2010@gmail.com
Abstract
An approach for performing rotor dynamic analysis using ANSYS as a tool is presented. In this paper, a simple
offset disk rotor model is studied. Critical speeds are calculated by analytical way and Campbell diagram is
plotted in MS Excel. For checking the approach for rotor dynamic analysis using ANSYS software, same
problem of an offset disk rotor model is solved and corresponding Critical speeds and Campbell diagram is
plotted. Furthermore rotor bearing system proposed by H. D. Nelson has studied and solved in ANSYS for
different bearing conditions. The numerical results, based on the present approach, stand with very good
agreement with results is stated. Hence it may be a more effective way to foresee the opportunity of using
ANSYS as a tool for rotor dynamic calculations.
Keywords:Rotor bearing system, Critical speeds, Campbell diagram, finite element model
1. Introduction
Rotor dynamics deals with the vibration of rotating elements. While designing rotors, it is of prime
importance to consider rotor dynamics characteristics such as rotors unbalance excitation, critical speeds,
gyroscopic effect, natural frequencies and amplitudes of synchronous vibration etc. Considering these
characteristics at the design phase may prevent the system from severe catastrophic failures. These rotors
dynamic characteristics can be determined with the help of Finite Element Methods. ANSYS is widely used for
finite element analyses of rotating machinery.
In this paper it is proposed to study the basics of rotor dynamics, solve rotor-bearing system problems using
hand/ MS-excel, review finite element method for vibration problems, understand usage of ANSYS and solve
example problems.
Our primary interest relative to this paper, is a contribution by Nelson and McVaugh [4], who presented
procedure for dynamic modeling of rotor-bearing systems which consist of rigid disks, distributed parameter
finite rotor elements, and discrete bearings. A finite element model including the effects of rotary inertia,
gyroscopic moments, and axial load has developed. They have also found out critical speeds for different types
of flexible isotropic and flexible orthotropic bearing conditions. In addition the natural frequencies of rotor
bearing system between different support conditions are compared and the Campbell diagram is plotted for
different boundary conditions.
2. Rotordynamic Analysis in Ansys

ANSYS is a computer-aided engineering (CAE) program tool. During this work only ANSYS Classic
version 13.0 has been used. To analyze rotating models two different coordinate system can be used in ANSYS,
stationary and rotating. In this work only stationary reference frame is used because it gives the possibilities to
plot Campbell diagram and to do modal analysis. When using a stationary reference frame, the reference
analysis system is attributed to the global coordinate system, which is a fixed one. In such analysis system, the
gyroscopic moments due to nodal rotations are included in the damping matrix and the equation of motion
becomes,
[M ]{u}  ([C]  [Cgyro]){u}  [ K ]{u}  { f }
Above equation is used by ANSYS when the displacements are calculated in a stationary reference frame [1].
2.1 Elements used in ANSYS

In ANSYS for critical speed calculations of rotor-bearing systems, MASS21, BEAM188 and COMBI214
elements are adopted [2].
434
Mass 21 is a point element and it is defined by a single node. The degrees of freedom of the element can be
extended up to six directions: translations in the nodal x, y, and z directions and rotations about the nodal x, y,
and z axes. With the KEYOPT (3) option, the rotary inertia effects to the element can be included or excluded
and also the element can be reduced to a 2D capability. If the element has only one mass input, it is assumed
that mass acts in all coordinate directions.
BEAM188 is a two-node beam element in 3-D with tension, compression, torsion, and bending capabilities.
It is developed based upon Timoshenko beam theory. Hence the element includes shear deformation effects.
When the KEYOPT (1) = 0 (default) the element has six degrees of freedom at each node: translations in the
nodal x, y, and z directions and rotations about the nodal x, y, and z axes. Also, this beam element is associated
with sectional library which consists of different section shapes. So that, a BEAM188 element may be modeled
with the desired section shapes and thereby real constants for the chosen section are automatically included.
COMBI214 is a spring-damper element developed for rotordynamics applications. The element has springs
and dampers in four directions; one horizontal, one vertical and two cross coupled springs.
2.2 Analysis commands in ANSYS

The most important commands when rotor dynamics calculations should be done are [2]
Coriolis, on Coriolis,on is used on rotating components in a rotation reference frame. Ansys is when taking
in account Coriolis matrix and spin softening matrix automatically.Coriolis, on,,, on Coriolis, on,,, on is used
for rotating components in a stationary reference frame. Ansys is taking in account the gyroscopic matrix in the
calculation. Important is that the component is axis-symmetric around the axis of rotation.
Omega Omega makes the whole model rotate around the global coordinate system.
Omega Omega makes the whole model rotate around the global coordinate system.
CmomegaCmomega is the same as omega but the difference is that cmomega is pointing at a component not
the whole model and is only rotating that component around a defined axis. Thanks to this the model can have
many different components rotating with different rotational speed around different axis.
Campbell,onIs used when pre stressed should be account in the model. The command is saying to the
program that the eigenvalues should be saved for all the sub steps.
Plcamp,, 1, rpm Plcamp is used in the post processing to show the Campbell diagram. “1” is the slope of a
line of how many excitations per revolution, thus “1” means one excitation per revolution. Rpm is the unit of the
x-axis of the diagram. Rpm means revolution/min and there is also rds which means radians/second. Then
cmomega is used the component name has to be written in the command also.
Prcamp,, 1, rpm Prcamp is writing out the critical speeds from the Campbell diagram. The critical speed are
there they excitation line meets any mode lines. Rpm is the unit which the rotational speed will be written out in.
“1” is the slope of the excitation line which is drawn in the Campbell diagram. Then cmomega is used the
component name has to be written in the command also.
Antype, Modal Define that a modal analysis should be done.
MODOPT, Method, NMODE, FREQB, FREQE, Cpxmode, Nrmkey
Method is what kind of method the program should use. NMODE is the number of modes the program
should calculate. FREQB is the start frequency. FREQE is the end frequency. Cpxmode is the choice if the
program should calculate the complex eigenvalues MXPAND, NMODE, FREQB, FREQE, Elcalc, SIGNIF
defines how many modes should be expanded. NMODE is the number of modes which should be expanded.
FREQB is the start frequency. FREQE is the end frequency. Elcalc is if the program should calculate the
element reaction and reactions forces in the model during the modal analysis. Either ON or OFF. SIGNIF
defines threshold which tells the program to only expand modes over this threshold.
3 Numerical Examples
For checking the approach of solving rotordynamics analysis in ANSYS following examples have studied.
3.1 An offset disk rotor system
Figure 3.1 Offset disk system, courtesy of [3]

An offset disk rotor system; with various parameters like mass of the disk m= 10 kg, the diametral mass moment
of inertia, Id= 0.02 kg-m2 and polar mass moment of inertia Ip = 0.04 kg-m2. The disk is placed at b= 0.25 m
from right support. The shaft has the diameter of 10 mm, the total span length of 1m and the Young’s modulus
of 2.1*1011 N/M2 The shaft is assumed to be massless and consider gyroscopic effect [3].
Solution: m=10kg, d = 10 mm, L = 1 M, a=0.75 M, b=0.25M, Id=0.02 kg-m2 ,Ip=0.04 kg-m2 , E= 2.1*1011 N/M2

I  d4
64 =4.9087 *10-10 m4
435
Firstly defining influence coefficients for offset disk problem,

a 2b 2
11 
3EIl = 1.137*10-4 m/N
(3a 2l  2a3  al 2 )
12   21   3.031*104
3EIl 1/N
(3al  2a  l )
2 2
 22    1.4146*103
3EIL 1/N-m
y   a11 a12   f 
    
x   a21 a22  
M 

Solving above equations, we get Frequency Equation for above problem is,
( )4  2()( )3  95.99( )2  4.664()( )  93.658  0
Condition 1
  
4 2
  91.33  93.658  0
2
1,2  -92.35 and 1.015 (negative roots are neglected)
   f crf 1 * a11m
= 1.015
 f
crf 1
= 29.88 rad/sec
Condition 2
  
4 2
  33.55  31.22  0
2
1,2  0.98 and 5.71
1,2   B cr1,2 * a11m
 B cr1  0.98* 29.66 = 28.47 rad/sec
 B cr 2  5.71* 29.66 = 169.34 rad/sec
For above problem only one forward critical speed is obtained and two backward critical speeds are
obtained. For the 1st mode very less divergence is occur and for 2 nd gyroscopic effect is clearly seen; the
divergence between forward and backward whirling goes on increasing.
An analysis was performed using ANSYS to determine the Critical speeds. BEAM188, Mass21 and
Combi214 elements are adopted. The results are compared to that of analytical ones and very good agreement is
obtained. This close agreement demonstrates the approach of using ANSYS for rotordynamic calculations is
correct.
Figure 3.2 Campbell diagram plotted in MS Excel Figure 3.3 Campbell diagram plotted in ANSYS
436
3.2 Rotor –bearing system proposed Nelson and McVaugh [4]
In order to illustrate the application of finite element model, a typical rotor –bearing system which was first used
by Nelson and McVaugh [4] is analyzed to determine its critical speeds in which a density of 7806 kg /m3 and
elastic modulus 2.078*1011N /m2 were used for the distributed rotor and a concentrated disk with a mass of
1.401kg , polar inertia 0.002 kg-m2 and diametral inertia 0.00136 kg-m2 was located at station five. The
distributed rotor was modeled as eighteen elements shown in Figure (3.29).The data of these elements is listed
in Table(1). Two identical bearings, idealized as undamped and linear ,were located at stations eleven and
fifteen.The following two cases of bearing stiffness were analyzed:
Table 3.1 Geometric Data, courtesy of [4]
Node Inner Outer

Element Bearing
Location Diameter Diameter
Node No. & Disc
(cm) (cm) (cm)
1 0 0 0.51
2 1.27 0 1.02
3 5.08 0 0.76
4 7.62 Disc 0 2.03
5 8.89 0 2.03
6 10.16 0 3.3
7 10.67 1.52 3.3
8 11.43 1.78 2.54
9 12.7 0 2.54
10 13.46 Bearing 0 1.27
11 16.51 0 1.27
12 19.05 0 1.52
13 22.86 0 1.52
14 26.67 Bearing 0 1.27
15 28.7 0 1.27
16 30.48 0 3.81
17 31.5 0 2.03
18 34.54 1.52 2.03
Isotropic Bearings
The shaft is supported by identical bearings with Kyy = Kzz = 4.378 *107 N/m; Kyz = Kzy = 0
An analysis was performed to determine the critical speeds from Campbell diagram for isotropic bearing
condition
Figure 3.4 Rotor bearing configuration in ANSYS Figure 3.4 ANSYS Campbell Diagram for isotropic Bearings condition
Orthotropic Bearings
The shaft is supported by identical bearings with the following data:
Kyy = Kzz = 3.503 *107 N/m
Kyz = Kzy = -8.756 *106 N/m
An analysis was performed to determine the critical speeds from Campbell diagram for orthotropic bearings.
437
Figure 3.5 ANSYS Campbell Diagram for Orthotropic Bearings condition
4 Ansys Results
4.1 Results for offset disc rotor system
In this study; an offset disk problem is solved by both analytical and numerical methods. By plotting the
Campbell diagram in ANSYS following results are obtained.
Table 4.1 comparison of results for offset disc
Critical speeds Analytical Results Ansys Results
Forward critical speed 1 29.88 Rad/sec 29.127 Rad/sec
Backward critical speed 1 28.47 Rad/sec 28.334 Rad/sec
Backward critical speed 2 169.34 Rad/sec 168.899 Rad/sec
Percentage of error for the Result obtained from ANSYS:

(1) For forward critical speed 1 = (29.88-29.127)*100/29.88 = 2.52 %
(2) For backward critical speed 1 = (28.47-28.334)*100/28.47 = 0.477 %
(3) For backward critical speed 2 = (169.34-168.89jh9)*100/169.34 = 0.260%
From this it is found that the percentage of error for the ANSYS results is very less. These tasks are performed
for acquiring the basic knowledge and skill in the usage of ANSYS software for the given scope.Hence it is
found that the procedure of using ANSYS software for performing rotor dynamic analysis is acceptable.
4.2 Results for Rotor bearing configuration of Nelson and McVaugh
4.2.1 Isotropic bearing condition
Table 4.2 Comparison of the results for isotropic bearing condition
Mode No. Paper Results (RPM) Ansys Results (RPM)

First backward whirling 15470 15292
First forward whirling 17159 16609
Second backward whirling 46612 49522
Second forward whirling 49983 49563
Third backward whirling 64752 64815
Third forward whirling 96457 83548
Percentage of error for the Result obtained from ANSYS

(1) For First forward critical speed = (15470-15292)*100/15470 = 1.12%
(2) For Second backward critical speed = (49522-46612)*100/49522 = 5.87 %
(3) For Third forward critical speed = (96457-83548)*100/96457 = 13.38%
It has been obtained that percentage of error in ANSYS results is very less and varies between 1 to 13.38
percent.
438
4.2.2 Orthotropic Bearing condition

Table 4.8 Comparison of the results for orthotropic bearing condition
Mode No. Paper Results (RPM) Ansys Results (RPM)
First Backward Whirling 13063 13713
First Forward Whirling 17310 16091
Second Backward Whirling 39889 40614
Second Forward Whirling 48293 49540
Third Backward Whirling 62414 59888
Third Forward Whirling 81917 80536
Percentage of error for the Result obtained from ANSYS

(1) For First forward critical speed = (13713-13063)*100/13713 = 4.74%
(2) For Second backward critical speed = (40614-39889)*100/40614 =1.78 %
(3) For Third forward critical speed = (81917-80536)*100/81917 = 1.68%
It has been obtained that percentage of error in ANSYS results is very less and varies between 1 to 5 percent.
5. Conclusion
The fundamentals of rotordynamics have been understood. Critical speeds and Campbell diagram are
plotted by using analytical method (hand/MS-Excel) and ANSYS software. A finite element model of an offset
disk system and multi bearing rotor system using ANSYS is presented. The results from analytical method and
ANSYS are correlated. Effect of isotropic bearing condition and orthotropic bearing condition on critical speeds
of a rotor bearing systems has been studied. Approach regarding rotordynamic analysis in ANSYS is
understood.
References
[1] Giancarlo Genta (2013), “Dynamics of rotating systems” Springer-Verlag London publication
[2] ANSYS Classic version 13.0, Help Topic (2013)
[3] Rajiv Tiwari (2012) Tutorials of rotordynamics, www.nptel.iitm.ac.in
[4] Nelson, H.D. and McVaugh, J. M. (1976), “The dynamics of rotor bearings ystems, using finite elements”,
Trans. ASME, J. Eng. Ind., Vol. 98, No. 2, pp.593-600
Nomenclature
{f} External force vector
{X} Generalized displacement vector
{u} Generalized displacement vector
Velocity vector
{u }
Velocity vector
{ X }
Acceleration vector
{u }
Acceleration vector
{X }
[M] Mass matrix
[C] Damping matrix
[K] Stiffness matrix
[Cgyro ] Gyroscopic matrix
Id Diametral inertia
Ip Polar inertia
Kx Bearing stiffness in x direction
Ky Bearing stiffness in y direction
E Young’s modulus
m Mass of offset
 Frequency
 Rotational velocity
 Influence coefficient
439
Paper ID- 102
Construction of Overhead water tank

- an unconventional approach
Mainak Mallik1
1
National Institute of Technology, Arunachal Pradesh, India
mainaknitap @gmail.com
Abstract. Construction of an overhead water tank leads to extensive

scaffolding and formwork for the full height of the tank. A tank has
been constructed with “Slip forming and Heavy Lifting Technique” to
meet the requirement of Steel Melt Shop and Rolling Mill furnace of a
steel plant located at Sadat City, Egypt with the total capacity of
275cu.m at a staging height of 36.75 meter above ground level. It
consists of twin chambers, each having capacity of 75cu.m and
200cu.m supported on a circular shaft. A comparison with the method
of “Slip forming and Heavy Lifting Technique” and “Conventional
method of construction” had been presented. “Slip forming and Heavy
Lifting Technique” is need of the hour to provide an economical
solution towards the requirement of time bound finish of any project.
Keywords: Overhead water tank, Heavy lifting technique, Slip form
shuttering
1 Introduction
In industries, whether it is a steel plant or power plant, Liquid Storage Tanks are
commonly used for storing water, chemicals, petroleum products, etc. and also for
storing water in public water distribution systems. The elevated tanks may be
rectangular or circular in shape. These tanks are elevated so that once they are filled,
the water flows under gravity to the area it is to serve. The tanks are supported on
staging which may be simple slab or a spherical dome with or without opening in the
centre, latter being mostly used in practice. Where the size of the tank is large, a
network of beams is provided over columns to support the base slab. If columns are
provided in staging, these are braced by ring beams not only at top and bottom but
also at a number of places along the height. In such a case, the effective length of
columns is taken as the distance between centres of adjacent bracings. The staging
and the tank is subjected to wind pressure and seismic forces depending upon the
location of the tank. In some cases staging is also provided in the form of circular
shaft. All over the world Intze tank is one of the most popular form of construction for
Overhead water tank.
The conventional approach for the construction of such type of overhead water tank is
to construct the foundation, followed by the construction of the supporting structure
440
of the tank bowl either in the form of circular shaft or column/beam upto the staging
height and
nd at the end the construction of tank bowls. This will lead to extensive
scaffolding and formwork for the full height of the tank. It takes almost 10 months for
the completion of an OH tank.
The construction of the tank bowl, however, is the most critical one, because of the
height of the clearance between ground level and the cantilevered slab of the tank
bowl. Conventional construction means was to use temporary H-shapeH shape steel members
supported vertically upon the ground to support the cantilevered portion
portion of the tank
bowl under construction, as shown in Figure 1 [1]. Generally the height of staging of
an OH tank varies from 20m to 30m for domestic purpose but it may goes upto 40m
for industrial purpose. As a result of that the unsupported length being large ge and the
temporary joints of these members being likely to be far from rigid, the stability of the
scaffolding system turned out to be a critical one. It has been observed that the cost of
scaffolding system was inevitably high and found to be as high as one fifth of the total
construction cost, as was unreasonably high. Moreover, since the stability problem of
the old system was so critical, the potential risk of its failure might be a matter to be
worried about. In order to overcome this an alternate scaffolding
sc system [1
[1] (Fig.2)
was suggested but it also has some drawbacks like to use the central shaft as
supporting member there will be a waiting period till it attains its full strength.
Fig.1- Conventional Scaffolding System Fig.2- Conventional Scaffolding System

(Support from ground) (Support from the shaft)
In order to overcome all these problems a method has been adopted for the
construction of an Overhead
Overhea tank to cater the water requirement of Steel Melt shop,
located at Sadat City, Egypt which involved construction of Tank by a combined
technique - Slip forming and Heavy Lifting Technique. In this method, Slip forming
Technique has been adopted for the construction of the shaft after the footing of the
tank is cast in conventional method.
method. Later the tank bowl is cast encircling the shaft at
ground level and then same is lifted along the shaft by Heavy lifting technique and is
placed finally on the shaft by some special arrangement. It is obvious that the original
diameter inside the tank bowl has been designed more than the shaft diameter so that
441
it can be lifted along the shaft keeping in mind that the capacity of the tank bowl is as
per requirement. Inn view of this
th innovative approach,, which is not only cost effective
but also time saving thehe client and the
t architect accepted these challenges as well as
changes without any reservation.
reservation
2. Method of Construction
There are six major steps starting from the construction of footing to the casting of the
top slab of the tank bowl,
bowl based on which total construction of the Overhead water
tank had been carried out.
Stage-I
Step-I-1:
1: As per normal practice the first step is the construction of the raft ffoundation
followed by circular shaft construction upto 2.0 above ground level.
Fig.3 Step-I-1- Construction of footing and tank shaft

Fig.3-
Step-I- 2: In the year 1942 Swedish company Byggförbättring [2] [3] had invented the
SLIP FORMING Technique. In this this technique a system of mechanical lock nut
hydraulic jacks has been connected to a hydraulic power unit to and all the vertical
walls of large-scale
scale structures are allowed to be built simultaneously. Scaffolding
caffolding is
not required. The sequence of work is as follows-
follows Positioning of the formwork along
the shaft wall, followed by pouring of concrete, and afterwards allowing the initial
setting time of concrete moving the same formwork upwardly to pour freshly mixed
concrete on top of the first concrete.
concrete. Till the staging height is achieved tthese
sequences are kept on repeating. For carrying out the construction work two options
are there- either without any break the full construction of the shaft will be done by
virtue of day and night shift work or with the break during night time intermittently
continuous work. Known slip-forming
slip apparatus comprised of a self elf-climbing
formwork and reinforcing steel is placed as the forms move upwardly. Three
platforms in the form of top deck, working deck and lower deck are arranged as
442
shown in fig.4. It takes only 7days to construct a staging height of 40m with the slid
speed of formwork as 15 to 25 cm per hour, in conjunction with continuous
concreting. are used The
T slid movement of the formwork has been carried by means
of hydraulic jacks.. The formwork,
formwork, which are generally 1.20 m high, travel 4 m in 24
hours, so that the concrete emerging at the bottom is 4 to 6 hours old. It is firm
enough to be able to hold without forms but must not have set sufficiently to stick to
the formwork, which would then pull off bits of concrete as it rose. The height of
reinforcement should not exceed 4m above top deck. This technique has been
successfully utilized in construction of buildings, water towers, bridge piers,
chimneys. Construction of Shaft wall has been continued till the staging height has
been reached. Step1- 3: Provision of dowel bars are there at the requisite
requisite level so that
the supporting RC bracket can be constructed once the shaft construction is over. For
accessing to the top of the tank Pre-fabricated steel staircase are used which was
placed on the R.C bracket (projected from the circular shaft) by means of a crane
through the opening at the top of the central shaft.
Height of rft. above

top deck should not
exceed 4.00m C.L of Tank & shaft
Top Deck for Hydraulic Height of rft. above
Installation top deck should not
exceed 4.00m
Top Deck
Working Deck
for Concreting
2000
Pumped concrete supply
Crane
Slipform 1200
Shutter Working Deck
Slip-form
form
Shutter
Lower Deck
Lower Deck
Barrel of Green Concrete
Fig.4 – Typical arrangement

a and procedure of Slip-Forming
Forming Technique
Stage-II
Step-II-1:12
1:12 numbers of supporting columns has been constructed by the conventional
on the top of final staging height at a spacing of 30o, required for placing
cing Lifting jacks
Step-II-2:
2: Construction of A circular resting on the column has been done. The
purpose of this is to place the Lifting Jacks (each having a capacity of 70t). The deck
443
slab was provided with twenty-four

twenty numbers of embedded steel pipes of 150 mm in
diameter at equal intervals.
Yoke at suitable interval

(Depending on requirement
Jaw for opening the of jack)
yoke
Nozzle for
injecting hydraulic
oil
Flexible Joint
Rubber flap to
guide hydraulic oil
50mm provision of
Lifting the jack
position at each lift
of slip shuttering
Jack for holding
the yoke
Fig.5- Placing of Pre-fabricated

Pre Fig. 6- Typical Detail of Jack for
Structural staircase Holding and Lifting of Slip Form
s shuttering
Slip--form
Shutter
CASTING OF HATCH PORTION TO BE
DONE IN A TRADITIONAL WAY
Fig.7- STAGE II-

II Construction of Column and Decking Platform for Jack support
444
Stage-III:
Step-III-1:
1: Twelve numbers of short columns supported on ring beam has been cast.
The ring beam profile is made in such a way that it can be placed on the sloped profile
of the raft and has been cast encircling the shaft.
Step-III-2: The
he tank bowl was castcast on ground level encircling the circular shaft
supported on twelve columns resting on a ring beam constructed above the foundation
at the step III-1. Twenty
wenty-four
four embedded steel pipes of 150 mm in diameter has been
provided in the lifting beam.
beam Telescopic tie rods, two for each jack, has been provided
which are anchored at the topto of circular deck and also anchored at the bottom with
lifting beam through these pipes. The whole tank bowl was kept suspended with the
help of these tie rods.
Fig. 8- STAGE III-

III Construction of Tank bowl at Ground Level
Stage-IV
It is that stage when the whole tank bowl constructed at ground level was lifted upto
the final level. The
he tank bowl was lifted in a controlled manner by means of twelve
jacks, which were connected to a common manifold at the top circular beam. The
whole lifting operation was continued till the tank bowl positioned at the desired
level.
Stage-V
In this phase after that tank bowl has been placed at its final position
ion the construction
of knee beam which connects the tank bowl with the circular shaft has been carried
out.
Stage-VI
In this stage the lifting system has been removed followed by erection of pipes and
cables. After completion of service lines casting of top slab has been done
one to complete
the construction.
445
Fig. 9- STAGE IV- Lifting of Tank bowl and placing of Tank bowl at final position
1. Knee beam
2. Top slab
Fig. 10- STAGE V & VI- Construction of Knee Beam and top slab
446
Fig. 11- Elevation and sectional view of Completed Tank
3. Design
esign Methodology
3.1.. Geometrical Properties
1) Details of Outer Chamber 2) Details of Inner Chamber
- Capacity: 200cu.m
cu.m - Capacity: 75cu.m
- Top diameter:: 15.0 m - Top diameter: 10.0 m
- Bottom diameter: 6.2 m - Bottom diameter: 5.9 m
- Height: 5.35 m - Height: 2.765 m
Fig. 12-
12 Sectional Elevation of the tank bowl
447
3.2. Analysis
1) STAAD Pro [4] software was used for the modeling and analysis of the tank
considering the following break-up of the structure:
a) The annular shaft upto the staging height
b) Full tank with the tank bowl placed at the top
-under non-operating condition
-under operating condition
To suit the construction and erection mechanism of the whole structure the design and
detailing of the different components of the tank bowl along with central shaft, deck
slab and other supporting facilities of the tank were carried out.
. Apart from design of the tank against water load, the structure was also checked
against wind load and seismic load as prevalent at Sadat City site in Egypt. The
various factors for design against wind load and seismic load was considered as per
stipulations of the Egyptian code [5].
The salient features considered during design calculations are as follows:
a) M30 grade concrete along with high yield deformed bars of grade St 36/52 having
yield stress of 360 N/mm2 and ultimate stress of 520 N/mm2 was used in the design.
But for crack control the tensile stress un steel is limited to 100N/mm2.
b) The conical tanks were designed mainly for membrane forces due to self-weight
and hydrostatic pressure. The effect of hydro-dynamic pressure on the tank wall also
been considered for seismic load condition.
c) The walls were designed as un-cracked sections by limiting tensile stress of
concrete to 2.0 N/mm2 and reduced stress in steel, as generally adopted for water
retaining structures.
In order to crack control to limit the bar diameter of the reinforcement following
formula has been adopted [5]:
µ
Φ ≤ r[( }^2]x106 (2)
where,
f= largest bar diameter
r=coefficient to take bond properties of the steel
z=100
Ast= area of tension reinforcement
Act= area of concrete section at the tension zone
fs=tensile stress in reinforcement resulting from working loads
d) The structure was checked against Wind Load for the following intensity along the
height of the tank
Height (m) Intensity of wind pressure (kN /sq. m)

0-10 0.75
10-20 0.825
20-30 0.975
30-50 1.125
Table 1: Wind Pressure at different Height
448
e) The formula [5] adopted for calculating the Base Shear (V) due to seismic load is
as follows:
V=Z.I.K.C.S.W (1)
where,
Z= Seismic intensity factor
fa = 0.2 (Zone 2 map for Sadat City)
I= Importance factor= 1.25 for emergency services
K= Structural system and ductility co-efficient
co efficient = 1.33 for cores, shear walls or
braced frame system
C= Structure co-efficient
co = √
Where T= Fundamental Natural period
S= Soil coefficient which depends on the type of soil below foundation
= 1.15 for medium density soil
W= Total weight of structure Design of jack supporting deck slab and
jacking system.
e) Checking of lifting beams for high concentration of stresses during lifting operation
of tank bowl.
f) Checking of effect of concentrated load on circular deck slab at at +38.2m level and
supporting columns during lifting operation of tank bowl.
g) Detailing of the structure to match the requirements of construction and erection
has been done. Also the design
esign and detailing of the tank has been carried out based on
the clauses
auses as stipulated in [5].
[5
Critical detail of structural component (fig.12) followed during construction of jack
supporting platform and knee beam connecting tank bowl with the main shaft.
Fig, 12- Showing the detail of reinforcement in Knee beam and jack supporting column
column-slab
h) Special detailing of reinforcement has been done in the location of openings

provided for pipe entry and other ancillary services.
Tank wall
Fig, 13- Showing the detail of additional reinforcement

re around openings
449
Based on the design output the details of the several components of the tank are as
follows:
Sl Component Thickness Reinforcement max Allowable Tensile

No. (mm)/ provided (allowable) Spacing stress in
Width X (mm) (mm) concrete
Depth (N/mm2)
1 Top Dome 120 12T @200c/c 32.97 200c/c 0.781

(ring)
10T @200c/c 22.75
(radial)
2 Outer Ring 600 x1300 31-20T 43.68 pt,minm 1.1
Beam 10T @200c/c =0.4%
3 Outer 225 16T @200c/c 49,78 200c/c 0.99
conical (ring)
dome 12T @200c/c 17.5
(radial)
4 Inner top 150 10T @200c/c 18.2 200c/c 2.008
slab (T & B)
platform Bothways
5 Inner ring 250 x1200 16-20T 22.4 200c/c 0.38
beam 10T @200c/c
6 Inner 175 12T @125c/c 36.16 200c/c 0.71
conical (ring)
dome 12T @200c/c 22.6
(radial)
Table 2- Summary of design of different component of tank bowl
4. Time Schedule
The time required for construction was considerably reduced as against
conventional approaches to construction. Table 3 shows savings in time required for
construction of the tank as against conventional construction methods.
Methodology of construction Saving in time

Slip Form method of shaft construction 20%
& conventional method of tank bowl
construction
Slip Form method of shaft construction 50%
& heavy lifting of tank bowl constructed
at bottom
.. Table 3: Savings in time for different methodology of construction
450
5. Conclusion
Presently every project work is time bound and normally started with taking loan
from the bank with high rate of interest. In view of the above it is concluded that with
the combination of slip form and heavy lifting techniques, considerable time can be
saved for construction of overhead water tanks in order to commission the project
early. Design and detailing of the structure need to be modified to suit the
requirements of expeditious construction without hampering the aesthetics of the
structure.
Acknowledgment
Author would like to thank Dr. Abhijit Dasgupta, Director, M. N. Dastur & Company
(P) Ltd whose constant inspiration and guidance has made it possible to carry out the
design work and also to share the state of the art.
References
1. Chyuan-Hwan Jeng, S.H. Ju, Y.L. Mo and I.K. Fang, Scaffolding system for construction of
elevated water Towers, 2000 Proceedings of the 17th ISARC, Taipei, Taiwan, pages 1-6,
https://doi.org/10.22260/ISARC2000/0142
2. http://www.royam.ch/slipform/watertower
3 http://www.vslin.com/service-heavy-lifting.html
4. STAAD Pro. V8i, Structural Analysis and Design programming
5. The Egyptian Code for the Design and Construction of Concrete Structures
6. R.D. Anchor, Design of Liquid-retaining Concrete Structures, Second Edition, London,
Edward Arnold, 1992
7. W. S. Gray and G. P Manning, Water Towers, Bunkers, Silos and Other Elevated Structures.
London, Concrete Publications Limited, 1964
8. E. Reynolds and J. C. Steedman, J. C., Reinforced Concrete Designer’s Handbook, Tenth
Edition. London, E & FN Spon, 1988
9. M. Mallik, Construction of Overhead Water Tank- a cost effective approach, IJSRD, Vol. 4,
Issue 12, 2017
451
Paper ID- 104
Purview of Nonlinear Static and Dynamic Analysis

Procedures on Seismic Fragility Estimation of RC
Buildings with Masonry Infills
Trishna Choudhury1 and Hemant B Kaushik1 [0000-0001-5896-6543]

1 Department of Civil Engineering, Indian Institute of Technology Guwahati, Assam, India
c.trishna@iitg.ac.in, hemantbk@iitg.ac.in
Abstract. Seismic performance assessment of buildings can be carried out with

different levels of computational efficiency depending on the available data and
resources. Detailed nonlinear static or dynamic analyses need to be carried out
for assessing seismic capacity and performance of buildings for a given seismic
hazard. However, owing to the computational complexity involved in nonlinear
dynamic analyses, comparatively simpler nonlinear static analysis procedures us-
ing Capacity Spectrum Method are preferred for seismic performance evaluation
of buildings. The present study is an effort towards evaluating the effectiveness
of nonlinear static procedures in seismic performance assessment of RC build-
ings with masonry infills. Due importance is also given to vertically irregular RC
buildings where masonry infills walls are not provided in the ground storey since
these buildings are marked seismically highly vulnerable from consequences of
past earthquakes. Dynamic amplification factors (DAF) for the considered
frames are obtained as the ratio of the responses obtained from nonlinear static
and dynamic analyses at different seismic intensity. Finally, comparison of seis-
mic fragility is done based on the two analysis types and use of DAF is judi-
ciously appraised.
Keywords: Masonry infills, Seismic fragility, Dynamic amplification.
1 Introduction
Nonlinear static procedures (NSPs) are relatively simplified approaches for the evalu-
ation of the seismic response of existing structures. The procedure acts as a subjective
alternative to the relatively more time consuming nonlinear dynamic analysis proce-
dures. Although the detailed nonlinear time history analysis (NLTHA) is the most ac-
curate procedure to determine the nonlinear seismic demand, for most practical cases
nonlinear static analysis procedures are preferred. The NLTHA requires far more com-
putational effort and modelling data as compared to the nonlinear static analysis proce-
dures. In case, seismic performance evaluation of a large number of buildings in an area
is required in order to assess the seismic vulnerability, resorting to NLTHA becomes
quite tedious and infeasible. The application of various available NSP becomes more
relevant in such scenarios.
452
Past literature shows that a lot of work has been carried out in different parts of the
world to identify the seismic fragility of RC buildings with masonry infills. Different
group of researchers have suggested different interesting nonlinear methods for a de-
tailed seismic assessment of the buildings. Several methods, ranging from empirical to
nonlinear static analysis and nonlinear dynamic analysis, have been developed for seis-
mic assessment of building typologies. A lot of work has also been carried out to opti-
mize the efficiency and effectiveness of these structural assessment methods to suit a
variety of needs, for example, to reduce the computational power and time required for
detailed seismic assessment. In the present work, the performance of buildings is stud-
ied using nonlinear static procedures complemented with the capacity spectrum method
and the results are compared with nonlinear dynamic analyses results. For this, different
configurations of RC buildings that are representative of most mid-rise buildings in a
region are considered, and subjected to an extensive nonlinear static and dynamic anal-
yses. Further, comparison of seismic fragility obtained for the same building models is
made using the two nonlinear procedures.
2 RC Frames Modelling
2.1 Overview
For the comparative study, an internal frame of a three-bay, four-storey (3B-4S) RC
building, designed and detailed for the highest seismic zone as per the relevant Indian
Standards (BIS 2016), is considered. Three variants of RC frame: bare frame (BF), open
ground storey frame (OGS), and fully infilled frame (FI) are idealized. All the frames
have a bay width of 3 m and a ground storey height of 4.4 m; all upper stories are 3.2
m high. Though ductile detailing is provided in the RC frame members, the frame is
designed as a weak column-strong beam frame system to reflect the current design
practice adopted by designers in India as well as in many other countries (Kaushik et
al. 2009). Nonlinear static analysis complimented with capacity spectrum method
(CSM) as well as nonlinear time history analyses (NLTHA) were carried out on the
considered frames.
2.2 Analytical Idealization

For carrying out the analyses, the bare frame is modelled as per general design criteria
of considering only the weight of masonry infills on the frame members. For the OGS
frame, stiffness and strength of masonry infills are considered in the upper stories; how-
ever, the ground storey of the frame is kept open, i.e., without masonry infills. The fully
infilled (FI) frames are provided with masonry infills at all stories uniformly. Thus,
both the strength and stiffness of masonry infill walls are considered while modelling
for nonlinear analyses of FI frame. Typical structural plan, elevation, and member
cross-sectional details of the 3B-4S frame are shown in Fig. 1.
The RC frame is assumed to be fixed at the base and soil-structure interaction effects
are not considered in the study. Columns and beams of the frame are modelled using
453
two-noded frame elements with three degrees of freedom at each node. The mean com-
pressive strength of concrete cubes of 150 mm size (fck, which is also known as charac-
teristic cube strength) is considered as 25 MPa (Elastic Modulus: 25000 MPa), and the
reinforcing bars have expected yield stress of 450 MPa (Elastic Modulus: 200 GPa).
Poisson’s ratio for concrete is taken as 0.15, and modulus of elasticity Ec of RC mem-
bers of the frame is calculated using IS 456 (BIS 2000) as in Eq. (1).
Ec  5000 f ck (MPa) (1)
Fig. 1. (a) Structural elevation of the considered frame – bare, OGS, and FI, (b) Building floor
plan, and detailed sectional properties of (c) columns and (d) beams for the frames.
Infill walls are modelled as equivalent single diagonal struts using two-noded beam
elements. The diagonal struts for masonry walls are modelled such that transfer of bend-
ing moments from RC frame elements to masonry is prevented. Weight density and
Poisson’s ratio for masonry are taken as 18.0 kN/m3 and 0.20, respectively. Modulus
of elasticity of masonry, Em is taken from Kaushik et al. (2007) as in Eq. (2), where, fm´
is the compressive prism strength of masonry in MPa. The thickness of the strut is taken
as the actual thickness of the masonry walls (220 mm). An average value of strut width
equal to one-fourth of the diagonal length of infill is used in the present study. No pro-
vision of opening in the infill wall panels has been provided.
Em  550 f m' (MPa) (2)
2.3 Nonlinear Modelling

In addition to the elastic material properties required in linear analysis, nonlinear ma-
terial properties are required for nonlinear static analysis. In SAP2000 (CSI 2015), non-
linearity is modelled as lumped plasticity at desired locations on the structural members
(Fig. 2). Plastic hinges in RC members are assumed to form at a distance equal to half
454
the average plastic hinge length (lp) from their rigid ends. It can be calculated by the
following expression (Paulay and Priestley 1992):
lp = 0.08L +0.022db fy (in m) (3)
where, L is the length of the member in m (taken at the point of contra flexure from the
end), db is the diameter of longitudinal reinforcement in m, and fy is yield stress of
longitudinal reinforcement in MPa. Plastic hinges (only flexural) are assumed to form
at a distance of lp/2 from the face of beams and columns. Flexural hinge properties
involve axial force – bending moment interaction (P - M) as the failure envelope and
bending moment - rotation (M - θ) as the corresponding load – deformation relation.
Fiber hinges obtained using section designer integrated with SAP2000 that enables the
modelling and analysis of custom cross sections are used for defining the nonlinear
hinges in RC members. Masonry infill walls are modelled as equivalent diagonal struts
and their plastic hinges are assumed to develop at the center of struts along the length.
Length of the plastic hinge in struts is assumed to be equal to three-fourths of the diag-
onal length of the strut. (Kaushik et al. 2009). For concrete, Mander’s confined concrete
model (Mander et al. 1988) is used to characterize the stress-strain curve. Fig. 3a shows
a typical stress-strain curve for the RC column sections at III and IV storey of the frame.
The idealized stress-strain model proposed by Kaushik et al. (2007) is used to model
the material nonlinearity in masonry as shown in Fig. 3b.
Fig. 2. Location of the flexural hinge in RC members and axial hinge in the equivalent
strut for masonry infill walls
30 5
Confined b c
Compressive stress
concrete 4 a
Stress (MPa)
20
Unconfined 3
(N/m2)
concrete
2 (a) uncracked phase
10 (b) post-crack phase
1 (c) post peak strength
detrioration
0 0
0.000 0.005 0.010 0.015 0.020 0 0.005 0.01
Compressive strain Strain
(a) (b)
Fig. 3. (a) Mander’s stress-strain curve for concrete (Mander et al. 1988) for the column
sections at 3rd and 4th storey of the frames, (b) Idealized monotonic stress-strain model
for weak masonry infill (Kaushik et al. 2007).
455
For nonlinear dynamic analysis, the hysteretic characteristics for concrete, i.e., the
strength and stiffness degradation for every loading and unloading cycle, is defined by
the Takeda hysteresis model. The Takeda model (Takeda et al. 1970) shown in Fig. 4a
includes stiffness changes at flexural cracking and yielding and strain-hardening char-
acteristics. The Takeda model simulates dominantly the flexural behavior, and has a tri-
linear envelope curve and is designed to dissipate energy even at low cycles once the
cracking point has been exceeded. Fig. 4b shows the Pivot hysteretic model (Cavaleri
and Trapani 2014) used for the equivalent strut to model the hysteretic effects in ma-
sonry infill.
(a) (b)
Fig. 4 (a) Force-deformation relationship of Takeda’s degrading stiffness model
(Takeda et al. 1970), (b) Hysteretic Pivot law particularization for the equivalent diag-
onal strut (Cavaleri and Trapani 2014).
2.4 Input Ground Motions

In nonlinear dynamic analyses, besides consideration of an adequate structural model,
selection of a set of ground motions is also important for obtaining the seismic response.
The ground motions selected in the study differ in their characteristics, such as magni-
tude, peak ground acceleration (PGA), and strong motion duration as listed in Table 1.
Table 1. Input ground motions considered for time-history analyses.
Sl. No. Event PGA (g) Magnitude(Mw) Duration (sec)

1 1940 El Centro 0.35 7.1 24.02
2 1943 Indo-Burma 0.34 7.2 14.92
3 1978 Tabas 0.84 7.4 16.10
4 1985 Chile 0.71 7.8 33.81
5 1995 Kobe 0.82 6.9 8.56
6 1999 Kocaeli 0.25 7.6 36.83
7 1999 Duzce 0.97 7.2 12.60
Different samples of each ground motion are also created where the sample-to-sam-
ple variation comes from different band-limited white noise signals associated with the
456
frequency dependent envelope, which are different for different samples. Thus, a num-
ber of samples of the process are generated artificially. In order to evaluate the struc-
tural performance under increasing ground motion intensity, each ground motion and
their samples are again scaled for different PGA values starting from 0.1g until all the
damage states are reached.
3 Structural Response
3.1 Seismic Performance

Seismic performance of the considered frames is visualized in the form of base shear-
roof displacement curves as shown in Fig 5 using both nonlinear static and dynamic
analyses. The base shear force-displacement curve, also known as pushover curve, is
obtained by monotonically loading the building frames until failure in case of nonlinear
static analysis. In the case of NLTHA, each ground motion is scaled for different PGA
values and maximum base shear and corresponding roof displacement are obtained.
Finally, the median values of the force-displacement pair from all ground motion and
their samples are considered. It is observed that for all the three frames the pushover
curves match quite well with those obtained from NLTHA; in fact, the pushover curves
represent more or less the envelop responses for all the three frames.
400 400
BF OGS
Base Shear (kN)
Base Shear (kN)
300 300
200 200
100 100
NL Static
NL Dynamic
0 0
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
Max Roof Displacement (m) Max Roof Displacement (m)
(a) (b)
1500
FI
Base Shear (kN)
1000
500
0
0.0 0.2 0.4 0.6 0.8 1.0
Max Roof Displacement (m)
(c)
Fig. 5. Comparison of capacity curves obtained from NL static analysis with median
base shear-roof displacement curve obtained using dynamic analyses of: (a) bare, (b)
OGS, and (c) FI frame.
457
3.2 Dynamic Amplification Factor

The nonlinear time-history analysis is the most desired method to predict the force and
deformation demands in various components of the structure. However, the use of time-
history analysis is limited because the dynamic response is very sensitive to the mod-
elling and ground motion characteristics. It requires proper modelling of the cyclic
load-deformation characteristics and careful consideration of the material degradation
characteristics of all the important components. The computation time, time required
for generation of input parameters, and interpreting the voluminous output of the anal-
yses, make the use of the time-history analysis difficult for seismic performance eval-
uation. Nonlinear static analysis, on the other hand, is essentially a static analysis, in
which the static loads are applied in an incremental fashion monotonically until the
ultimate state of the structure, i.e., the structural failure defined by particular damage
state or failure of certain structural members, is attained. The non-linear material data
are provided in the form of envelope or backbone curves obtained from cyclic tests or
analysis. Thus, computationally the nonlinear static analyses are less demanding. In this
regard, the dynamic amplification factor (DAF) can be efficiently utilized which when
multiplied by static response gives the dynamic response of the structure concerned.
DAF is a dimensionless multiplying number, which describes how much the seismic
response obtained by static nonlinear analysis should be altered in order to get the re-
sponse under dynamic loads. A DAF higher than 1.0 implies dynamic response is
higher, whereas, a DAF of less than 1.0 implies static response is higher.
In the present study, in order to capture the nonlinear static deformation response,
capacity spectrum method is utilized. The CSM is carried out for different seismic in-
tensities in the form of response spectra. These response spectra are developed for the
ground motions and samples of ground motions, which were used in the nonlinear dy-
namic analyses of the frames. Finally, the dynamic amplification factor is obtained as
the ratio of the dynamic response to the static response of the building frames for a
given seismic demand at each seismic intensity measure level as shown in Fig. 6. It is
observed that there is a great variation of DAF with increasing PGA for BF and OGS
frame, however, for FI frame, the DAF is mostly constant over the range of PGA con-
sidered.
2.5
BF
2.0 OGS
FI
1.5
DAF
1.0
0.5
0.0
0.1 0.3 0.5 0.7 0.9
PGA (g)
Fig. 6. Dynamic Amplification Factors over a range of PGA for the three frames
458
4 Seismic Fragility
4.1 Overview
Seismic fragility is the reformulation of the results of the deterministic structural anal-
ysis in a probabilistic way. Seismic fragility analysis is the key component for seismic
vulnerability and risk assessment methodologies. A common nontechnical definition of
fragility is “the quality of being easily broken or damaged” (Porter 2016). The most
common form of a seismic fragility function, given the damage state thresholds and
variability associated with the damage state, is the lognormal cumulative distribution
function (CDF) as shown in Eq. (3). P[x ≥ θd | IM] represents the conditional probability
of exceedance of any seismic response parameter (here, x) with respect to θd, given the
seismic intensity measure (IM), such as PGA. Ф represents the standard normal cumu-
lative distribution function, θd is the median (threshold) value of seismic response pa-
rameter, (e.g., interstorey drift) considered for different damage states, and βT is the
normalized standard deviation of the natural logarithm of the threshold (θd).
 ln( x / θd ) 
P[ x  θd | IM ]     (3)
 βT 
A fragility curve can also be developed by fitting a statistical model to data on build-
ing damage at different values of the IM. Such statistical models are generally fitted to
the probability of exceedance for a particular IM, which is calculated as in Eq. (4),
where, NF is the number of analysis cases where the considered seismic response pa-
rameter exceeds the limit, and NT is the total number of analysis cases at that IM level.
NF (4)
Probability of exceedance at an IM level =
NT
4.2 Influence of Analysis Type on Seismic Fragility

Seismic fragility assessment can be carried out using the seismic response parameters
obtained either from the nonlinear static analyses or from the nonlinear time history
analyses of buildings. Nonlinear dynamic method of analysis is computationally very
expensive and requires much more input data than that required in case of the nonlinear
static method. Due to these reasons, nonlinear static pushover analysis methods are
sometimes preferred for assessment of lateral load behavior of buildings. It will be in-
teresting to study the differences in seismic fragilities obtained using results of nonlin-
ear static and nonlinear dynamic analysis methods. Such a study will help to develop a
method for converting fragilities obtained using static analysis methods to that corre-
sponding to dynamic analysis methods. For the RC building typologies considered in
the present study, four discrete damage states based on Barbat et al. (2008) are consid-
ered as Slight (S), Moderate (M), Extreme (E), and Complete (C) for the estimation of
seismic fragility as shown in Fig. 7 and Fig. 8. Damage states are specified in order to
physically categorize the degree of damage to members of a building. These damage
states are quantified based on the yield and ultimate displacement values of the building
frames. For fully infilled (FI) frames, failure of the ground storey infill walls is an ad-
ditional criterion considered to obtain the ultimate displacement capacity of the frame.
459
1 1
0.8 0.8
P[ds|Disp]
P[ds|Disp]
0.6 BF 0.6 BF
OGS OGS
FI 0.4 FI
0.4
0.2 0.2
Slight Slight
0 0
0 0.05 0.1 0.15 0.2 0 0.05 0.1 0.15 0.2
Roof Displacement (m) Roof Displacement (m)
1 1
0.8 0.8
P[ds|Disp]
P[ds|Disp]
0.6 0.6
0.4 0.4
0.2 0.2
Moderate Moderate
0 0
0 0.05 0.1 0.15 0.2 0 0.05 0.1 0.15 0.2
1 1
Extreme Extreme
0.8 0.8
P[ds|Disp]
P[ds|Disp]
0.6 0.6
0.4 0.4
0.2 0.2
0 0
0 0.05 0.1 0.15 0.2 0 0.05 0.1 0.15 0.2
1 1
Complete Complete
0.8 0.8
P[ds|Disp]
P[ds|Disp]
0.6 0.6
0.4 0.4
0.2 0.2
0 0
0 0.05 0.1 0.15 0.2 0 0.05 0.1 0.15 0.2
(a) (b)
Fig. 7. Seismic fragility curves obtained using (a) nonlinear static analysis, and (b) non-
linear time history analysis with respect to maximum roof displacement.
460
10
1.0 1.0
Slight Slight
0.8 BF 0.8
OGS
P[ds|PGA]
P[ds|PGA]
0.6 FI 0.6
0.4 0.4
BF
0.2 0.2 OGS
FI
0.0 0.0
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
PGA(g) PGA(g)
1.0 1.0
Moderate Moderate
0.8 0.8
P[ds|PGA]
P[ds|PGA]
0.6 0.6
0.4 0.4
0.2 0.2
0.0 0.0
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
PGA(g) PGA(g)
1.0 1.0
Extreme Extreme
0.8 0.8
P[ds|PGA]
P[ds|PGA]
0.6 0.6
0.4 0.4
0.2 0.2
0.0 0.0
0.0 0.5 1.0 0.0 0.2 0.4 0.6 0.8 1.0
PGA(g) PGA(g)
1.0 1.0
Complete Complete
0.8 0.8
P[ds|PGA]
P[ds|PGA]
0.6 0.6
0.4 0.4
0.2 0.2
0.0 0.0
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
PGA(g) PGA(g)
(a) (b)
Fig. 8. Seismic fragility curves obtained using (a) nonlinear static analysis, and (b) non-
linear time history analysis with respect to peak ground acceleration (PGA).
461
11
Fig. 7 and 8 show the comparative seismic fragility curves of the three frames with
respect to peak roof displacement and peak ground acceleration respectively. P[ds|Disp]
and P[ds|PGA] in the figures represent the conditional probability of exceedance of a
particular damage state (ds) with respect to displacement and PGA respectively. Alt-
hough, the seismic fragility is slightly overestimated by the nonlinear static analysis, a
similar pattern of seismic fragility is obtained using both nonlinear static or dynamic
analysis methods with respect to peak roof displacement. Seismic fragility of the FI
frames with respect to peak roof displacement is found to be the highest in all the dam-
age states. This is because the displacement capacity of the FI frame is very less as
compared to the BF or OGS frames.
When the seismic fragility is assessed with respect to peak ground acceleration
(PGA), nonlinear static analysis is found to slightly underestimate the seismic fragility
for slight and moderate damage states. However, it is interesting to note that the non-
linear dynamic analysis methods seem to provide realistic results as the seismic fragility
for OGS frames is found to be highest for all the damage states and for all values of
PGA. This represents the true comparative fragility of the frames when past earth-
quakes are considered.
5 Conclusion
The present study is an approach towards establishing a relation between nonlinear

static and nonlinear dynamic analysis for performance assessment and seismic fragility
analysis. Nonlinear static and dynamic analyses are carried out for three different con-
figurations of reinforced concrete (RC) frames with masonry infills considering several
real and synthesized ground motions. Based on the seismic response obtained from each
type of analysis, dynamic amplification factors are developed for each of the frames
with respect to peak ground acceleration. From the comparison of base shear-displace-
ment plot, it is clear that nonlinear static analysis could be a very good alternative for
nonlinear dynamic analysis when seismic performance assessment needs to be done.
However, care needs to be taken for the parameter with respect to which the fragility is
assessed.
References
1. Barbat, A.H., Pujades, L.G., and Lantada, N.: Seismic damage evaluation in urban areas
using the capacity spectrum method: Application to Barcelona. Soil Dynamics and
Earthquake Engineering, 28(10), pp. 851–865 (2008).
2. BIS.: IS 1893 (Part 1): 2002, Criteria for Earthquake Resistant Design of Structures: Part 1
General Provisions and Buildings. 5th Revision. Bureau of Indian Standards, New Delhi,
India (2016).
3. BIS.: IS 456, Indian standard code of practice for plain and reinforced concrete”, 4 th Revi-
sion. Bureau of Indian Standards, New Delhi, India (2000).
462
12
4. Cavaleri, L., and Di Trapani, F.: Cyclic response of masonry infilled RC frames: Experi-
mental results and simplified modeling. Soil Dynamics and Earthquake Engineering, 65, pp.
224-242 (2014).
5. CSI.: Structural Analysis Program (SAP2000) – Advanced, static and dynamic finite ele-
ment analysis of structures. Computers and Structures Inc. Berkeley, USA (2015).
6. Kaushik, H. B., Rai, D. C., and Jain, S. K.: Effectiveness of some strengthening options for
masonry-infilled RC frames with open first storey. Journal of Structural Engineering, ASCE,
135(8), pp. 925 – 937 (2009).
7. Kaushik, H. B., Rai, D. C., and Jain, S. K.: Stress-strain characteristics of clay brick masonry
under uniaxial compression. Journal of Materials in Civil Engineering, ASCE, 19(9), pp.
728-739 (2007).
8. Mander, J. B., Priestley, M. J., and Park, R.: Theoretical stress-strain model for confined
concrete. Journal of Structural Engineering, ASCE, 114(8), pp. 1804-1826 (1988).
9. Paulay, T., and Priestley, M. J. N.: Seismic Design of Concrete and Masonry Structures.
John Wiley and Sons. New York (1992).
10. Porter, K.: A beginner’s guide to fragility, vulnerability, and risk. Encyclopedia of
Earthquake Engineering, (2016).
11. Takeda, T., Sozen, M. A., and Neilsen, N. N.: Reinforced concrete response to simulated
earthquakes. Journal of the Structural Division, 96(12), pp. 2557-2573 (1970).
463
Paper ID- 107
EFFECT ON CURING TECHNIQUES ON DURABILITY PROPERTIES

OF CONVENTIONAL AND HVGGBFS CONCRETE
1
Amar R. Chougule, 2M. B. Patil and 3K. B. Prakash
1
PhD Research scholar, Visvesvaraya Technological University (VTURRC), Belagavi,
Karanataka, India, civil.amar@gmail.com, +91 9986823131.
2
Professor & HOD, Department of Civil Engineering, Govt. SKSJTI, Bangalore- 560 001,
Karanataka, India, drmbpatil@gmail.com, +91 9448144330.
3
Principal, Govt. Engineering College, Haveri, Devagiri- 581 110, Karanataka, India,
kbprakash04@rediffmail.com, +91 9448267374.
ABSTRACT
At the present time industrial by-product(wastes) is a foremost problem created for the
environment, by considering this point an experimental study is carried out to evaluate correct
curing technique on the use of high volume of slag obtained from steel manufacturing industry i.e.
known as GGBFS in place of cement. In this study, the mechanical properties such as compressive,
split tensile and flexural strengths were experimented on mixtures of 0%, 50% and 60%
replacement of cement by GGBFS in concrete and allowed to cured in different curing techniques
such as water curing, steam curing, high pressure steam curing, compound curing, lime water
curing and curing by using SAP. In addition to mechanical properties durability study in terms of
water absorption, soraptivity and low pH acidic attack tests were performed between HVGGBFS
and conventional concrete. The results of this study exposed that, adding of 50% amount of
GGBFS in concrete cured by using lime water curing technique considerably improves its
mechanical properties and also more resistant in strengths compared with conventional concrete
in acidic media.
Key words: HVGGBFS concrete, Curing techniques, Mechanical property, Water absorption,
Soroptivity, Acidic attack.
1. Introduction
After water, concrete is the highest second consumed material throughout the world. In the
modern era, it is regularly used construction material. With the improvement in various technology
and advanced research work, so many modifications were found out on the traditional concrete
mixes for the reason to obtained not only superior quality of concrete but also fulfills the
improvement in mechanical and durability properties requirements.
Now a days construction industry are in boom towards development of new projects and
replacing ones such as buildings, dams, bridges, roads, etc. Therefore, massive amount of cement
material is required to fulfill the gap between urbanization development. India is the world’s
second-largest cement producer with about 7% of global production in recent years [1]. The
production process of one tonne of cement emits about one tonne of CO2 into the atmosphere,
resulting from the de-carbonation of limestone in the kiln during the manufacture of cement and
464
the combustion of fossil fuels [2]. Yearly global cement production of 1.6 billion tonnes is
responsible for about 7% of the total CO2 emission into the atmosphere. Portland cement
production is not only one of the most energy-intensive processes but also is responsible for
enormous amount of greenhouse gas emission. The production of one tonne portland cement
utilizes nearly 4 GJ energy [3].
India is the second major cement producing country next to China and has 137 large and
365 mini cement plants. It is expected that the industry will continue to increase the annual cement
output in coming years and India’s cement production will grow at a compound annual growth
rate (CAGR) of around 12% during 2011-12 to 2013-14 to reach 303 million metric tonnes.
Cement manufacturing association (CMA) is targeting to achieve 550 MT capacities by 2020 [4].
Industrial developments in China and India are affecting the environment adversely in two ways.
First, the installation of new cement plants is increasing substantially the CO2 emissions, and
second, very large capacity thermal power stations result in availability of huge amounts of fly ash
and bottom ash that are not being recycled properly. Most of these ashes are being dumped in
lagoons, landfill sites and abandoned quarries. In addition of this developing countries like India
are facing so much problems to dump and manage industrial waste materials (by-products) such
as fly ash, GGBS, metakaolin, silaca fume, etc.
Thus, potentially valuable cementing resources are being wasted in precisely the countries
that need it most to reduce the greenhouse gas emissions, and to make economical and durable
concrete structures. Some measures through which the cement and concrete industry can contribute
towards reducing CO2 emissions are
• Use less OPC.
• Use more supplementary cementing materials.
• Use less unit water content by using more water reducers and superplasticizers.
• Incorporate recycled aggregates in concrete.
• Use lightweight concrete where possible [5-6].
As a civil engineer it our prime duty to use optimum amount of cement and replace by
using these industrial by-products which are having same chemical properties of cement in the
concrete. Hence it helps to protect the environment from the pollution.
2. Materials
In this research work locally available materials used such as 43 grade OPC cement conforming
to IS 8112-1989 specifications [7], coarse aggregate of maximum size 12mm and down, river sand
4.75mm to 150 µ size conforming to zone II of IS 383:1970 [8], BASF Master Rheobuild SP1
superplasticizer was used and GGBFS provided by Jindal steel plant, Bellary, Karnataka, India.
3. Mix Proportion
The M30 grade concrete mix proportions of conventional concrete named as MP 1 (0%
replacement of cement by GGBFS), 50% HVGGBFS concrete named as MP 2 (50% replacement
of cement by GGBFS) and 60% HVGGBFS concrete named as MP 3 (60% replacement of cement
by GGBFS) were designed based on the IS: 10262-2009 method [9]. The water to binder ratio kept
same as 0.38 during all types of concrete. The concrete mix designed is based on trial methods in
the laboratory to achieved a targeted slump of 100mm. Superplasticizer of 1.1% by weight of
cement was used to enhance the workability properties of concrete. Also the other properties of
fresh concrete such as compaction factor (CF) and VB time were checked before the casting of
each concrete. Mix proportions and the properties of fresh concrete are presented in table 1. Also
465
the variation of slump, CF and VB time w.r.t. the MP 1, MP 2 and MP 3 are shown in figure 1,
figure 2 and figure 3 respectively.
Table 1
Mix proportioning (kg/m3) and properties of fresh concrete
Properties of fresh
Mix proportioning (kg/m3)
Description of concrete
concrete VB
Cement GGBFS Water CA FA SP Slump CF
Time
MP 1 443 --- 168 1186 663 4.873 95 0.85 9
MP 2 221.5 221.5 168 1178 659 4.873 100 0.899 8
MP 3 177.2 265.8 168 1176 657 4.873 100 0.90 7
Fig. 1 Variation of slump
Fig. 2 Variation of CF
466
Fig. 3 Variation of VB time
4. Curing methods
The altered curing methods implemented in the research work are as follows-
 Water curing (WC): Specimens were cured in water tank and tested at 28 days after curing.
 Steam curing (SC): Specimens were cured for 12 hours under steam without pressure.
Steam generated in steam curing tank and tested at 1 day after curing.
 Steam curing with high pressure (SCHP): Specimens were cured for 10 hours in autoclave
chamber with a pressure of 10kg/cm2 and tested at 1 day after curing.
 Compound curing (CC): Specimens were cured by application of curing compound by
means of brush on all sides of the cube. The BASF, wax based curing compound was
selected for curing and then tested at 28 days.
 Lime water curing (LWC): Specimens were cured in lime water at room temperature and
tested at 28 days after curing. In lime saturated water, lime (3g/lit) was used to saturate the
water.
 Super absorbent polymers (SAP): The concrete were casted by adding 0.5% SAP by weight
of cement and specimen were in the concrete cubes are left in open air to be internal cured
at room temperature, until 28 days tested.
In each curing method total twenty-seven cube specimens were casted for 0%, 50% and 60%
replacement of cement by GGBFS with and without acidic attck, in order to find out the resistance
capacity of acidic attack. For acidic media, the specimens after 28 days of various curing methods,
were immersed in H2SO4 (Sulphuric acid) solution for 90 days. Sulphuric acid of pH=2.00 was
prepared and maintained throughout the 90 days of duration, then specimens were immersed to
check the acidic attack penetration. All the curing process are presented in figure 4.
467
a) Water curing tank b) Steam curing tank
c) High pressure steam curing(Autoclave chamber) d) Compound cured specimens
e) Lime water curing tank f) Swollen SAP gel into concrete

Figure 4 Various curing process
468
5. Results and discussion

5.1 Water absorption
Following table 2 gives the water absorption test results. The variation in percentage of water
absorption at different curing methods is illustrated in the form of graph as shown in figure 5.
Table 2 Water absorption test results (in %)
Curing techniques MP 1 MP 2 MP 3
WC 0.62 0.59 0.56
SC 3.38 3.37 3.22
SCHP 3.10 3.02 2.60
CC 0.63 0.63 0.62
LWC 0.605 0.578 0.548
SAP 0.61 0.582 0.56
Figure 5. Variation of water absorption

From the above table 2 and figure 5, it is clear that, the conventional concrete (MP 1) absorbed
more water than the HVGGBFS concrete (MP 2 and MP 3). Also it is seen that, for 60%
replacement of cement with GGBFS, water absorption percentage is low and which is very less
occurs in the LWC technique on the other hand the values are maximum during SC and SCHP
techniques compared with other curing techniques.
5.2 Water soraptivity

Following table 3 gives the water soraptivity test results. The variation in percentage in water
soraptivity at different curing methods is illustrated in the form of graph as shown in figure 6.
Table 3 Water soraptivity test results (in %)
Curing techniques MP 1 MP 2 MP 3
WC 4.00 3.98 3.89
SC 4.23 4.21 4.17
SCHP 4.15 4.10 3.95
CC 0.00 0.00 0.00
LWC 3.85 3.84 3.74
SAP 3.95 3.92 3.86
469
Figure 6. Variation of water soraptivity
From the above table 3 and figure 6, it is seen that, for 60% replacement of cement with GGBFS
(MP 3) water soraptivity percentage is low with reference to conventional concrete (MP 1) and
compared with curing techniques, there is no water soraptivity occurs in the CC technique as the
film of chemical compound coated on the surface of specimens.
5.3 Normal compressive strength [10]

Following table 4 gives the normal compressive strength test results for 28 days (CS @ 28 days)
and 118 days (CS @ 118 days) of curing without acidic attack.
Table 4. Normal compressive strength test results
MP 1 MP 2 MP 3 MP 1 MP 2 MP 3
Curing CS +/- I/D CS +/- I/D CS +/- I/D CS +/- I/D CS +/- I/D CS +/- I/D
methods @ 28 CS @ 28 CS @ 28 CS @ 118 CS @ 118 CS @ 118 CS
days @WC days @WC days @WC days @WC days @WC days @WC
WC 42.67 --- 48.37 --- 40.77 --- 43.27 ---- 49.17 ---- 42.67 ----
SC 35.00 -17.97 36.83 -23.86 33.63 -17.51 36.80 -14.95 38.07 -22.57 35.17 -17.58
SCHP 40.50 -5.08 42.12 -12.92 37.93 -6.96 42.00 -2.94 44.50 -9.50 39.67 -7.03
CC 32.23 -24.47 35.17 -27.29 31.70 -22.25 34.20 -20.96 36.80 -25.16 33.10 -22.43
LWC 49.40 +15.77 58.73 +21.42 46.20 +13.32 51.20 +18.33 59.80 +21.62 48.17 +12.89
SAP 45.10 +5.69 53.03 +9.63 42.03 +3.09 46.80 +8.16 54.00 +9.82 44.15 +3.47
(* +/- I/D CS @ WC = Percentage increase or decrease of compressive strength w.r.t. water
curing technique)
470
Figure 7. Variation of normal compressive strength (28 days)
Figure 8. Variation of normal compressive strength (118 days)
Table 4 represents the overall compressive strength of conventional and HVGGBFS concrete
cured using six different curing methods without subjecting to acidic attack. From the test results
it is clear that, there is an enhancement in compressive strength test results as the curing duration
increases from 28 days to 118 days period. Also it is observed that, for 50% HVGGBFS concrete
(MP 2) specimens cured via LWC method had higher compressive strength than those cured via
WC method. Maximum compressive strength extended from 28 days to 118 days as 58.73 MPa to
59.80 MPa in LWC technique when HVGGBFS concrete made with 50% replacement of cement
by GGBFS. Compared with WC method, the LWC method increases compressive strength by
18.33%, 21.62% and 12.89% at 0%, 50% and 60% replacement of cement by GGBFS in concrete
respectively. Figure 7 and 8 shows the variation of normal compressive strength at 28 days and
118 days curing period. Addition of GGBFS upto 50% mix enhances the compressive strength of
concrete in all curing methods. The increase in compressive strength of 50% HVGGBFS concrete
(MP 2) compared with that of conventional concrete (MP 1) cured via WC, SC, SCHP, CC, LWC
471
and SAP curing methods were 13.63%, 3.45%, 5.95%, 7.60%, 16.80% and 15.38% respectively
for 118 days of curing period.
5.4 Compressive strength when subjected to acidic attack

Following table 5 gives the compressive strength test results for 90 days when subjected to acidic
attack of pH = 2.0.
Table 5. Compressive strength test results when subjected to acidic attack
MP1 MP2 MP3
Curing
+/- I/D CS +/- I/D CS +/- I/D CS +/- I/D CS +/- I/D CS
methods CS CS CS
@WC @WC @ MP 1 @WC @ MP 1
WC 36.07 ---- 42.20 ---- +16.99 36.87 ---- +2.22
SC 31.23 -13.40 35.80 -15.17 +14.63 32.00 -13.20 +2.47
SCHP 33.47 -7.21 41.10 -2.61 +22.80 34.90 -5.33 +4.27
CC 28.10 -22.09 30.90 -26.78 +9.96 29.67 -19.53 +5.59
LWC 42.50 +17.84 52.40 +24.17 +23.29 43.50 +17.99 +2.35
SAP 36.80 +2.03 44.80 +6.16 +21.74 37.80 +2.53 +2.72
(* +/- I/D CS @ MP 1 = Percentage increase or decrease of compressive strength w.r.t. concrete
mix proportion 1)
Figure 9. Variation in % of increase or decrease of compressive strength

w.r.t. WC when subjected to acidic attack
472
Figure 10. Variation in % of increase or decrease of compressive strength

w.r.t. MP 1 when subjected to acidic attack
Table 5 represents the overall compressive strength of conventional and HVGGBFS concrete
cured using six different curing methods when subjected to acidic attack. From the test results it is
clear that, there is an enhancement in compressive strength test results as the cement replaced with
GGBFS in concrete. Also it is observed that, for 50% HVGGBFS concrete (MP 2) specimens
cured via LWC method had higher compressive strength than those cured via WC method.
Maximum compressive strength extended from MP1 to MP 2 concrete mixtures as 42.5 MPa to
52.4 MPa in lime water curing and from MP1 to MP3 concrete mixtures as 42.5 MPa to 43.5 MPa
in lime water curing techniques. Compared with WC method, the LWC method increases
compressive strength by 17.84%, 24.17% and 17.99% at 0%, 50% and 60% replacement of cement
by GGBFS in concrete respectively when subjected to acidic attack. Figure 9 and 10 shows the
variation of percentage of increase or decrease compressive strength with respect to WC when
subjected to acidic attack and variation in percentage of increase or decrease of compressive
strength with respect to MP 1 when subjected to acidic attack respectively. Addition of GGBFS
upto 50% mix enhances the compressive strength of concrete in all curing methods when exposed
to acidic attack. The increase in compressive strength of 50% HVGGBFS concrete (MP 2)
compared with that of conventional concrete (MP 1) cured via WC, SC, SCHP, CC, LWC and
SAP curing methods were 16.99%, 14.63%, 22.80%, 9.96%, 23.29% and 21.74% respectively and
also in 60% HVGGBFS concrete (MP 3) compared with that of conventional concrete (MP 1)
cured via WC, SC, SCHP, CC, LWC and SAP curing methods were 2.22%, 2.47%, 4.27%, 5.59%,
2.35% and 2.72% respectively.
6. Conclusions
Following conclusions may be drawn based on the experimentations conducted on the behavior of
HVGGBFS concrete cured by different curing techniques subjected to acidic attack:
473
 The workability test results clearly indicate that, as percentage replacement of cement by
GGBFS increased there is an increase in workability properties. The reason for higher
workability is due to the spherical and glassy texture of GGBFS particles.
 Water absorption and soroptivity test results indicate that, the normal concrete
(conventional concrete) absorbs more water than the HVGGBFS concrete. In addition, it
is clear that for higher GGBFS content, lower the amount of water absorption. Also from
the test results, it is indicated that, water absorption and soroptivity values are high in steam
cured specimens but very less in compound cured specimens as the filming of chemical
compound is coated on the surface of specimens.
 Many concrete structures are subjected to the fury of acids. Therefore, acidic attack
becomes an important durability test for concrete. It is observed from the test results that
there is significant improvement in compressive strength of concrete in which cement is
replaced by GGBFS when subjected to acidic attack of pH = 2 for 90 days. 50%
replacement level yields good strength. The test results also indicate that all the
compressive strengths of 50% replacements of cement by GGBFS in lime water cured
specimens yield good strength compared to normal water curing technique. At 50%
replacement when subjected to acidic attack of pH = 2 for 90 days, compressive strength
is found to increase in lime water curing by 24.17% with reference to water curing
technique.
 The test results clearly indicate that the resistance of concrete to acidic attack increases
with replacement of cement by GGBFS. 50% replacement of cement by GGBFS exhibits
better resistance to acidic attack as compared to 60% replacement.
 The test results clearly indicated that, the lime water curing technique yields good results
when HVGGBFS concrete is subjected to acidic attack. Hence, it is recommended as
effective curing technique for HVGGBFS concrete when subjected to acidic attack.
REFERENCES
1. Robbie M. Andrew, “Global CO2 Emissions from Cement Production”, Earth system
science data, 2018, Vol 10, pp 195-217.
2. Roy D., “Alkali-Activated Cements-Opportunities and Challenges”, Cement and Concrete
Research, Vol. 29(2), 1999, pp. 249-254.
3. Mehta P. K., “Reducing the Environmental Impact of Concrete”, ACI Concrete
International, Vol. 23(10), 2001, pp. 61-66.
4. Vora B. R., “Indian Cement Industry”, Inter-Connected Stock Exchange of India (ISE)
Report., 2011.
5. Malhotra V. M., “Role of Supplementary Cementing Materials and Super Plasticizers in
Reducing Greenhouse Gas Emissions”, Proceedings of International Conference on Fiber
Composites, High-Performance Concrete and Smart Materials, Indian Institute of
Technology, Chennai, India, 2004, pp. 489-499.
6. Mehta P. K., “Greening of the Concrete Industry for Sustainable Development”, ACI
Concrete International, Vol. 24(7), 2002, pp. 23-28.
474
7. _____43 Grade ordinary Portland cement – Specifications, (First revision), IS 8112: 1989,
Bureau of Indian Standards, New Delhi, May 1990.
8. _____Specifications for coarse and fine aggregate from natural sources for concrete
(Second revision), IS 383: 1970, Bureau of Indian Standards, New Delhi.
9. _____Recommended Guidelines for concrete mix design, IS 10262: 2009, Bureau of
Indian Standard, New Delhi.
10. _____Methods of tests for strength of concrete, edition 1.2, IS 516: 1959, Bureau of Indian
Standards, New Delhi, reaffirmed 1999.
475
Paper ID- 108
Comparative Study of Mixture Design Methods used for Self-Compacting Concrete

Priti patel1, L.K.Mishra2, Rakesh Kumar3
1
pritice0036@gmail.com, 2lkm@mnnit.ac.in, 3rkpat@mnnit.ac.in,
Department of Civil Engineering, MNNIT Allahabad, U. P., India.
Abstract:
Strength, durability and workability are the essential performance criteria for a mixture design.
Mixture design for a Self-Compacting Concrete (SCC) is quite challenging as concrete has to be
designed for high workability and stable composition. Several mixture design methods for
producing SCC are in use. BIS method is recently introduced.In this paper, an attempt has been
made to compare two methods of mixture design for SCC viz. EFNARC method and BIS
method The BIS method provides mix proportion for SCC using suggestive range for cement
content, nominal maximum size of aggregate and binder content. EFNARC method does not
provide direct calculation for determination of mix contents; however the mix contents are
obtained from associated mixture volume calculations. BIS Method overcomes this limitation of
EFNARC method. The SCC mix of M25 grade and class of SF3 (slump flow 760mm-850mm)
were designed using these methods. Fresh state and hardened state properties of concrete were
determined. Flowability, filling ability and passing ability of SCC were determined for fresh
state evaluation while compressive strength at 7 and 28 days was obtained using 150 mm cube.
The experimental investigation revealed that the compressive strength of SCC is nearly same
for both the methods while, the corresponding cement content, water content, for EFNARC
Method were higher when compared with BIS method. The SCC designed with BIS method
gives desirable performance in fresh state and hardened state but the performance of concrete
designed using EFNARC method exhibits reduced green state performance conforming to SF2
(660 - 750 mm) class.
Key Words: SCC, Mixture design, EFNARC method, BIS method
476
Introduction
Self compacting concrete (SCC), is a high performance concrete which has excellent property of
flowability, stability and segregation resistance.SCC results in faster construction, reduction in
use of manpower, better surface finishes and energy efficiency and energy conversation [1].It is
special type of concrete that is able to flow due to its intrinsic plasticity and fills formwork
completely, even in the presence of dense reinforcement.It compacts with its own weight and
maintain homogeneity.SCC is different from normal concrete as it does not require any external
vibration for compaction, has high fluidity and good segregation resistance.Due to these material
properties the SCC can produce highly durable and reliable concrete structures. Mixture design
for a Self-CompactingConcrete (SCC) is quite challenging as concrete has to be designed for
high workability and stable composition[2].Okamura proposed a mixture design method for
SCC in 1993 based upon the following guidelines to determine final water to powder ratio:
(1) Coarse aggregate (CA) content: 50 % of the solid volume

(2) Fine aggregate content 40 % of the mortar volume
(3) Water to powder ratio between 0.8 to 1.1 by volume, for different powder properties.
(4) Super plasticizer dosage as per requirement
The mixture design method proposed by Okamura demands quality control of paste and motar
and it is too complicated for practical implementation [3].Shi et al [4]had reviewed the different
mixture design methods of SCC, viz; Empirical Design Method (based on Okamura method),
Compressive Strength Method (based on EFNARC guidelines),Closing aggregate packing
methods (based on packing factor of aggregate), Methods based on statistical factorial model
(based on effect of different parameter), and Rheology of paste model (based on fluid behavior)
and concluded that the Empirical Design Method was easy, however, its main disadvantage was
excessive testing on available raw materials, while the Compressive Strength Method required
minimum trial mixes but required adjustments in quantity of all the ingredients to obtain an
optimal mixture proportion. The close aggregate packing method is based on the least voids
content between aggregates due to paste and aggregate interaction, however SCC produced by
this method shown reduced segregation resistance. The method based on statistical factorial
model uses statistical relationships amongst different key parameters (namely the contents of
cement and mineral admixtures, water-to powder ratio, volume of coarse aggregate, and dosage
477
of SP) influencing mix design and their statistical regression analysis for the performance of
associated concrete in terms of flowabaility stability and compressive strength for different
concrete making material .The method based on rheology of paste model can reduce the
laboratory work on material investigations[4].BIS Method is recently introduced to produced
SCC [5], BIS Method is extension of conventional concrete mixture design method considering
the air content and strength vs water to cement ratio relationship along with guidelines for
achieving other mix attributes such as guidelines selecting mineral admixture content super
plasticizer content powder content and fine aggregate content. In this method computation of
mixture composition using volumetric considerations.
Table No. 1. Comparison of various methods of mixture design for SCC

[4,6,7,8,9,10,11,12,13,14]
S.No. Method Governing factor Advantage Limitations
1. Empirical Design Content of Coarse and fine Easy to follow Excessive testing
Method aggregate, water to powder on available raw
ratio and superplasticizer materials
dosage
2. Compressive Strength Compressive strength Minimum trial Adjustments to all
Method mixes the ingredients to
obtain an optimal
mixture proportion
3. Close Aggregate Method Packing factor and least Simple and Concrete segregate
voids between aggregates require small easily
amount of
powder content
4. Statistical Factorial Content of cement and Reduction in no.
Required intensive
Model mineral admixture water to of trial batches
laboratory testing
powder ratio, volume of on available raw
coarse aggregate and dosage materials
5. Rheology of Paste Model Rheological behavior ofReduced Optimum
cement paste laboratory work evaluation involves
too many
parameters.
A comparative assessment of self compacted concrete designed using EFNARC Method and BIS
Method were used. Fresh and hardened state properties were determined on both the mixtures.
The powder is a binary blend of OPC and Metakaolin.
478
2 Experimental Plan
2.1 Self Compacting Concrete specification:
1) Grade : M25
2) Type of cement : OPC 43 grade
3) Slump flow : SF3(760mm-850mm)
4) Passing ability : >0.75
5) Viscosity : 9 -25 sec for VS2/ VF2 class
2.2 Material and material properties
Cement
Ordinary Portland cement of 43 grade was used for making concrete mix. The testing of OPC
wasdone according to IS 8112(1989)[15]. The normal consistency, initial and final setting of
cement obtained were 30%, 48 min and 240 minrespectively.The specific gravity and % retained
on 90microm sieve were found 3.12 and 0.86 respectively. Compressive strength of cement at
the age of 28 days was 48.69 N/mm 2.
Metakaolin
The specific gravity of metakaolin used in the investigation was 2.51. The specific surface area
of metakaolin was 16300 cm2/gm .The oxide composition and loss on ignition for OPC &
Metakaolin as provided by the manufacturers are illustrated in the Table 2
Table No. 2. Oxide Composition and Loss on ignition for OPC and MK
Composition (%) OPC MK

Silicon dioxide (SiO2) 21.23 72.32
Aluminium oxide(Al2O3) 4.65 16.52
Magnesium oxide (MgO) 1.35 0.18
Ferric oxide(Fe2O3) 3.42 0.76
Calcium oxide (CaO) 61.23 2.38
Sodium oxide(Na2O) 0.75 0.0
Potassium oxide(K2O) 1.25 3.78
Sulphur trioxide(SO3) 2.45 0.0
Loss of ignition LOI 3.02 2.65
479
Coarse and fine aggregates:
The specification of aggregate was done according to IS 383(1970)[16].The coarse aggregate

was used with nominal size of 16mm.The fine aggregate was used with maximum size of
4.75mm The material properties of the aggregates are given below in Table 3
Table No. 3. Material properties of the aggregates
Material Properties Coarse Aggregates Fine Aggregates

Fineness Modulus 7.3 3.05
Specific Gravity 2.68 2.56
Bulk Density 1580 kg/m3 1680 kg/m3
Water Absorption 1.2% 1.95%
Super plasticizer
Polycaboxylic-ether based chemical admixture (Master Rheobuild 817RL) was used for making
self compacting concrete with specific gravity 1.06, the relative density 1.180.02 at 25°C, pH
>6 and content of Chloride ion is less than 0.2%.
2.3 Mixture Composition for various methods
In this paper two mixture design methods viz. ENAFRC Method and BIS Method are considered
for evaluation of SCC of M25 grade concrete and class of SF3.
EFNARC guidelines for producing the SCC:
Typical ranges for mix constituents are given in the Table- 4 follows the range of binder
content(380-600 kg/m3) by weight, coarse aggregate content (750- 1000 kg/m3),fine aggregate
should be 48-55% of the total aggregate weight.The water to powder ratio range from 0.85
to1.10 by volume[17]. The following ratio obtained by EFNARC method 1:1.882:1.595 with
water/binder ratio 0.43 and binder/cementitious material ratio 0.8:0.2 on mass basis.
480
BIS guidelines for producing the SCC:
The range of mix constituents a) Ample amount of fines (< 0.125mm) preferably in the range of
400 kg/m3to 600 kg/m3 This can be obtained by having sand content more than 48 percent of
total aggregate weight and or using mineral admixtures to 25 percent to 50 percent by mass of
cementitious materials content. b) Water content between 150 to 210 kg/m3 c) Use of high range
water reducing admixture like polycarboxylate ether based high range water reducing admixture
(water reduction > 30 percent) and sometimes also using a Viscosity Modifying admixture
(VMA) [5].
The following ratio obtained by BIS method 1:2.2:1.75 [by weight] with water/binder ratio 0.43
and binder/cementitious material ratio 0.8:0.2 on mass basis.
Table No. 4. Mix proportions obtained from EFNARC and BIS Method
S.No Ingredient Composition EFNARC BIS code

. Method Method
1 Total binder content(in kg/m3) 470 442
(i) Cement content 376 354

(ii) Metakaolin 94 88
2 Aggregate Content (in kg/m3) 1635 1750
(i) Fine aggregate content 885 975

(ii) Coarse aggregate content 750 775
3 FA to \ total aggregate content (%) 54 55
4 Water content(in kg/m3) 202 190
Water to binder ratio 0.43 0.43

5 Admixture 1.2 % 1.2 %
6 Weight proportion of constituent materials 1:1.88:1.59 1:2.2:1.75

[ Total binder content : Fine Aggregate:
Coarse aggregate]
481
475
Total binder content
450
425
EFNARC Method BIS code Method
Fig.1: Comparison of binder content of EFNARC method and BIS methods
The concrete mixture designed using EFNARC method and BIS methods were evaluated for
fresh and hardened state according to EFNARC May 2005 guidelines [17] and IS516 (1959
reaffirmed ) [18].
3.1 Fresh State Evaluation
The fresh properties of concrete mixes (EFNARC & BIS Method) are given in the Table-5.It is
evident from Table- 5 that Concrete mix designed by BIS method gives enhanced fresh
properties at the same water /binder ratio 0.43.
Table No. 5. Fresh and Hardened properties of SCC
S.No. Tests EFNARC BIS CODE Codal provision

for SF3 Class
1. Slump flow 665 mm 785 mm 760mm-850mm
2. T50 sec 3.7 sec 2.6 sec >2
3. L- Box 0.82 0.89 0.8 minimum
4. V-funnel 13 sec 10 sec 9 to 25 sec
It is also observed from this Table- 5 that the concrete mix designed using EFNARC resulted in
SF2 class while the one designed using BIS Method produced SF3 class for the same
water/binder ratio. The Table- 5 indicate that better quality of concrete at fresh state is obtained
482
using BIS method as the concrete produced by BIS methods fall into the category of SF3 while
produced by EFNARC method fall into category of SF2, Slump flow is 665 mm was found
from EFNARC method while from BIS method slump flow was 785 mm.
Viscosity of SCC can be indirectly calculated byT50 and V-funnel tests. According to V-funnel
test, SCC can be categorized into two classes:1) VF1 ≤10s 2) VF2 ≥7s, ≤27s. T50 and V-funnel
tests time value does not gives the direct viscosity of SCC but it is indirectly showing the
viscosity on the form of the rate of flow. SCC with low viscosity initially have very quick flow
and after that it will be stop[6]. From the Table- 5 it can be concluded that the value of V-Funnel
obtained by both methods fall into category of VF2.
According to Passing ability SCC can be categories into two part (1) PA1 (structures with a
space of 80 mm to 100 mm) and blocking ratio ≥0.75 (2) PA2 (structures with a space of 60 mm
to 80 mm) and blocking ratio ≥0.75. L-Box,U-Box and j-ring tests were conducted on concrete
mix. It is evident from Table-5 that SCC produced by BIS method is more passable from dense
reinforcement in comparison to SCC produced by EFNARC method.
The improved fresh state properties in concrete designed with BIS method are observed
primarily due to reduced but sufficient paste content.
3.2 Hardened State Evaluation

Compressive Strength:
The hardened concrete cubes were tested for compressive using compression test machine. It is
observed from the Fig.2, that the compressive strength of concrete samples of mix designed by
BIS Method was comparable with the concrete designed by EFNARC method at age of 7 and 28
days. This increase in strength is 8.2 % more at the age of 28 days.
483
45
Compressive Strength (N/mm2
40
35
30
25
)
20
15
10
7 days 28 days
Days
Fig.2. Compressive strength of the specimens at age of 7 and 28 days
4. Conclusion
On the basis of the computations and experimental investigations conducted on the mixture
design for SCC using EFNARC method and BIS method, the following conclusion are drawn:
1) The binder content is 6% less in the BIS method in comparison to the EFNARC which
makes the BIS method is more economical than EFNARC method. Corresponding weight
proportion of constituent materials [Total binder content: Fine Aggregate: Coarse
aggregate] for BIS method and EFNARC method are 1:1.88:1.59, 1:2.2:1.75 respectively.
2) SCC obtained from BIS method is more flowable in comparison to SCC obtained from
EFNARC method at same water/ binder ratio.
3) The SCC produced from BIS method fall into category of SF3 which is applicable
tocongested reinforced structures, complicated geometry while SCC produced from
EFNARC method fall into category of SF2 which is applicable to walls, columns etc.
4) The blocking ratio is higher for SCC produced by BIS method in comparison to
EFNARC method.
5) The compressive strength of SCC produced by BIS method and EFNARC method are
comparable.
484
References
[1] H. Okamura and M. Ouchi, “Self-compacting high performance concrete,” Prog. Struct.
Eng. Mater., vol. 1, no. 4, pp. 378–383, Jul. 1998.
[2] K. Ozawa and M. Ouchi, “H. Okamura Kochi University of Technology, Kochi, Japan,”
Struct. Concr., no. 1, p. 15, 2000.
[3] H. Okamura, K. Maekawa, K. Ozawa, High Performance Concrete,Gihoudou Pub., Tokyo,
1993 (in Japanese)..
[4] C. Shi, Z. Wu, K. Lv, and L. Wu, “A review on mixture design methods for self-compacting
concrete,” Constr. Build. Mater., vol. 84, pp. 387–398, Jun. 2015.
[5] IS: 10262-2017. (Second Revision) Concrete mix proportioning- Guidelines. Bureau of
Indian Standards, New Delhi, India .
[6] Okamura H, Ozawa K. Mix design for self-compacting concrete. Concr Lib JSCE
1995;25:107–20
[7] Khaleel OR, Abdul RH. Mix design method for self compacting metakaolin concrete with
different properties of coarse aggregate. Mater Des 2014;53:691–700
[8] Ghazi FK, JadiriAl RS. New method for proportioning self-consolidating concrete based on
compressive strength requirements. ACI Mater J 2010;107(5).
[9] Dinakar P, Sethy KP, Sahoo UC. Design of self-compacting concrete with ground granulated
blast furnace slag. Mater Des 2013;43:161–9
[10] Su N, Hsu K, Chai H. A simple mix design method for self-compacting concrete.Cem
Concr Res 2001;31(12):1799–807.
[11] Sebaibi N et al. Composition of self - compacting concrete (SCC) using the compressible
packing model, the Chinese method and the European standard. Constr Build Mater
2013;43:382–8.
[12] Khayat KH, Ghezal A, Hadriche MS. Factorial design model for proportioning self-
consolidating concrete. Mater Struct 1999;32(9):679–86.
[13] Ozbay E, Oztas A, Baykasoglu A, Ozbebek H. Investigating mix proportions of high
strength self compacting concrete by using Taguchi method. Constr Build Mater
2009;23(2):694–702
485
[14] Bui VK, Akkaya Y, Shah SP. Rheological model for self-consolidating concrete. ACI
Mater J 2002;99(6).
[15] IS: 8112 (1989).Specification for 43 grade ordinary Portland cement. bureau of indian
standards, New Delhi, (India).
[16] IS 383 (1970): Specification for Coarse and Fine Aggregates from Natural Sources for
Concrete. New Delhi; bureau of indian standards.
[17] The European guidelines for self-compacting concrete; specification production and use.
EFNARC, May 2005.
[18] IS 516 (1959): specification for methods of test for strength of concrete. bureau of indian
standards, New Delhi, India .
486
Paper ID- 109
Management of sustainable development of territories
Olga Sotnikova1[0000-0003-4351-256X], Elena Zhidko1 [0000-0003-1462-4036 ],

Ekaterina Prokshits1[0000-0002-9260-4383], and Kseniia Lapshina1[0000-0002-0322-3037]
1
Voronezh State Technical University, Voronezh, Russia
hundred@vgasu.vrn.ru
lenag66@mail.ru
e.prokshits@mail.ru
k.sotnikova@vgasu.vrn.ru
Abstract. Management of development of building of territories means

activities of state institutions and/or municipality for formation and the approval
of the concept of such development, localization of important objects,
etc. Besides, a necessary condition of planning of development of urban areas is
respect for the principles of formation of the compatible with the biosphere safe
and comfortable urban environment. The urban environment is presented in the
form of open dynamic system which condition is caused by external influences,
internal interactions between clusters of system and the intra making processes.
Here the safe city environment acts as natural social and technical structure.
This concept assumes unity and interaction/interference of three main bases
(natural aspect of X, social aspect of Y and technological aspect of Z) subject to
external influences. The analysis of data of monitoring gives a qualitative and
quantitative picture of current state of the urban environment. Submission of
such information is the user-friendliest if geo-information technologies are
applied, such, for example, as GIS-mapping of the city. In this case the
electronic card – the city basis consisting of n-vector layers of separate
(independent) layers of social and economic life of the city is formed. The
number of n-vector layers is generally not limited. As a layer, for example:
housing estate; industrial building; underground construction; utilities; transport
systems; power systems; grounds of household waste; recreational recreation
areas, etc. can be taken.
Keywords: Management, Sustainable development, Geo-information systems,

Mapping, Vector layers.
1 The modern urban environment as a dynamic system
The modern urban environment is the dynamic system which open for external
managing directors of influences, that is the influences transferring system to a new
state as it is possible to tell that its state is caused by:
1. External influences (for example, climatic factors, change of requirements of
standard and legal base, social standards and qualities of life, change of education
487
level and culture of society and others, defining a possibility of cooperation with
the biosphere);
2. Internal interactions between components (for example, rendering services, social
tasks, environmental protection, etc.);
3. The intra making processes (for example, the intra production and economic
relations).
On the basis of the transformation principles of the city in Biosphere compatible

and developing the person [1-8, 9] the conceptual model of the safe urban
environment as natural-social-technical structure is offered.
As a part of this structure the components making it are allocated:
 the natural component as the part of the external environment containing resources
for the person, interacting with objects of city infrastructure and which is exposed
to negative technogenic impact;
 the social component as the part of the external environment which is also
interacting with objects of city infrastructure and expecting satisfaction of the
requirements at certain risks. (besides, the person as the representative of society,
being a part of the environment, indirectly is influenced by negative technogenic
impact);
 the technological component making power impact on natural and social
environments, and defining first of all, a possibility of formation of an adverse
ecological situation in a certain territory.
At the same time the conceptual model of ecologically safety organizational and
technical system of the city looks as follows (Fig. 1).
Fig. 1. The scheme of conceptual natural-social-technical structure of the urban environment

[9]
488
The principal novelty in creation of the safety urban environment conceptual

model in the form of open dynamic system in this case is that its state is defined by
other specific laws which are based on mutual influence of system's components
taking into account external influences.
For formalization of a system considered by the authors [9] the set-theoretic
approach was used [10], which determining parameters entered in Fig.1:
X - a set of fortune natural component, characterized as the capacity (resource) of
the biosphere, and depending on impact on it: technological objects, person as parts of
the nature, and external environment;
Z - a set of fortune technological component (production, services, pollution and
waste);
Y - a set of fortune social component, characterized as "the human potential" as a
part of the considered system, and depending on influence of two other components.
Interactions as a part of analyzed natural-social-technical system of the urban
environment according to [9] are described by sets:
XPТС, XSС – sets of impacts of a natural component on technological and social
components: (XPТС, XSС  X);
ZNС, ZSС – sets of impacts of a technological component on natural and social
environments (ZNС, ZSС  Z);
YPТС, YNС – sets of impacts of a social component on technological and natural
components: (YPТС, YNС Y).
The mathematical model natural-social-technical structure of the urban
environment as difficult dynamic system is presented in a general view by the
nonlinear equation of a look:
(1)
where - continuous or piecewise and continuous functions; - the system

coordinates reflecting the provision of some point in the phase plane or in space of
coordinates, and characterizing fortune making systems at any moment.
For the description of a condition of system and definition of components of
humanitarian balance on the basis of representation (1) various processes, situations
caused by interactions of various factors have to be formalized. Separate processes in
the itself and functional dependences describing them are known, also many of them
are normed. Nevertheless, statistically most of them are not recorded neither on the
different cities, nor on the historical periods of any one city in the ratio with the level
of development of human potential. It is one of the most important problems of the
forthcoming researches in this direction to define optimum indicators and
purposefully to make corresponding changes to mechanisms of management of the
city.
In order to develop an implementation algorithm of such task the balanced state
natural-social-technical system of the urban environment is formalized and presented
by set of three nonlinear equations:
489
(2)
where - the nonlinear functions reflecting

influence of parameters of internal interactions between components of X,Y and Z
and external impacts on dynamics of system.
Taking into account interactions in the considered system the model of the
balanced state is offered in the form of the following system of the differential
equations:
(3)
where - the pace of change a condition of a natural component;

- the pace of change a condition of a social component;
- the pace of change a condition of a technological component; ,
, - nonlinear functions; X - a set of fortune natural
making, characterized as the capacity (resource) of the biosphere, and depending on
influence of objects of technological infrastructure and the person and also the
external revolting influences; - impacts of a technological component on the
environment; Y – set of fortune of social component; – impacts of a natural
component on the social environment; – impacts of a technological component on
the social environment; Z – set of fortune of technological component; –
impacts of a social component on a motor transportation part; – impacts of a
natural component on a motor transportation part.
The system of the differential equations (3) is synthesis of the known exponential
law on types developing of V. Volterra:
, (4)
where - a condition of i-component in natural-social-technical structure,

i =1,2,3; - gain coefficient; t – time.
Coefficients of a gain of each component can be presented in the form of some
490
functions from all interacting factors. For example, we will present an absolute
increment of a natural component for a period Δt as:
, (5)
where - the coefficient reflecting change of a condition of a natural component

for the accepted unit of time (for example, one year);
- the coefficient reflecting change of a condition of a natural component for the
accepted unit of time taking into account influence of a social component;
- the coefficient reflecting change of a condition of a natural component taken
for a unit of time taking into account influence of a technological component.
Coefficients in the differential equations can be also presented in the form of some
functions or the sum of the coefficients reflecting, for example, realization of the
operating influences in relation to components natural-social-technical of the urban
environment:
(6)
The same way to form increments of other components of natural-social-technical

structure are register.
Then the system of the equations takes a form:
(7)
where - coefficient of change in the state of the natural

component; - coefficient of change in the state of the social
component; - coefficient of change in the state of the
production and technical component.
Thus, authors [9] received the system of three nonlinear differential equations
accepted to the description natural-social-technical structure of the urban environment
which has to be complemented with entry conditions at
The mathematical model in a general view gives an idea of a possibility of the

description a condition of the urban environment system and quantitative assessment
the components of humanitarian balance Biotechnosphere of the urban territory,
characteristics of the general properties models and allows to conduct numerical
researches of transformation processes of separate components and components in the
concrete region on the basis of data of monitoring [11], collecting and processing
491
corresponding statistical information and appointment on these basics of parameters

to the covered model.
2 The urbanized territories and assessment of value ZNС
Under the urbanization of the territory is understood as the relation of the area of
city lands to the total area of the region (edges, region, district). In Russia there are a
number of territories with high degree of an urbanization - Moscow (17,4%), St.
Petersburg (6,0%), Tula (14,8%), Yaroslavl (9,9%), Central-Chernozem (7,3%). A
number of industrial centers forms large agglomerations. The index of an
agglomeration is defined as the relation of number of the city settlements located in
an external zone of agglomeration to the total number of its urban population. Among
the largest agglomerations - Moscow, St. Petersburg, Novosibirsk, Kemerovo,
Chelyabinsk.
For the urban territories also calculate such parameters as: population density on
urban areas, security of citizens with green plantings and a greening of urban areas,
development of transport infrastructure, security with water and recreational
resources. In addition, for the urban territories integrated ecological assessment which
is defined by a total score of influence extent of technogenic factors and stability of
natural complexes and components (a land cover, water and air basins) is removed.
Many cities on intensity of pollution and the distribution area of toxic anomalies
elements in various environments represent technogenic geochemical provinces
already now. Pollutants of the atmosphere, waters and soils of the urbanized
territories in the form of various toxic gases, aerosols, particles of solid or liquid
substances adversely influence the environment. Being in the urban environment,
these products of production economic activity of the person form technogenic
geochemical anomalies of pollutants in various components of a landscape [4-6].
For city landscapes, under the influence of technogenic processes and an
urbanization not only considerable transformations, but also essential (including
irreversible) geochemical changes of the main physical and chemical properties of a
landscape which do not correspond to the natural analogs any more are characteristic,
and develop on the new technogenic-provoked cycles.
Certainly, degree of negative impact of the cities on natural systems in each
concrete city is very specific and depends on the size of the city, its specialization in
territorial division of labor, types of building, extent of improvement and gardening.
The urbanization involves the increased concentration of negative anthropogenic
toxicants the result of which is almost universally disturbed natural environment, air
pollution, soil, water bodies, drying up of vegetation, loss of biodiversity, increased
morbidity of the urban population.
492
3 Geo-information technologies as effective tool of complex

assessment of urban ecosystems condition ZSC
Currently, under the geo-information systems (GIS) understand hardware and

software systems that enable to collecting, processing, analysis and display of
spatially coordinated data and allow the most effective interpretation of information
and knowledge about the territories [7,8]. A cartographical method in combination
with geo-information technologies - an integral part of a modern geo-ecological
research therefore use of computer technologies in geoecology passes from the field
of theoretical and methodical developments into the field of broad practical
application.
Intensive development of GIS-technologies began with emergence of
cartographical software packages, such, for example, as «Маріnfо», which allowed
not only to create, but also on the basis of the arriving new information quickly to
update ecological cards, to form cartographic and ecological models in relation to the
solution of various problems of ecological assessment and monitoring.
Information filling of GIS comprises an electronic card file, documents with
opportunities of thematic search and a set of basic cards. Communication of
documents of an electronic card file with objects on basic cards gives to the user of
system new opportunities of cartographical information search. The GIS’s main
functions are:
 creation of electronic card files of documents;

 linking of the documents which are stored in card files with electronic maps of the
considered territory;
 carrying out thematic and cartographical search of documents;
 creation of thematic maps.
Modern computer technologies allow to process and interpret quickly big arrays of
the cartographical data accumulated in maps of different function, contents and scale.
But it’s more effective to use maps and information which is contained in them, it is
necessary to choose and put into practice such methods of the analysis, synthesis and
the forecast which combine application of computerized programs of mathematic-
cartographical modeling and traditional methods of geographical cartography.
In recent years there has been a new, higher stage of the automated geo-
information mapping providing creation and the purposeful analysis of cartographical
images on the basis of geographic information systems was outlined. The electronic
cards received by means of computers offer broad prospects for expeditious creation
of the models reflecting not only a statics, but also dynamics of the phenomena by
comparison of various objects in existential aspect that is especially important for
conducting environmental assessments. Working in the interactive mode with
electronic cards by means of manipulators, it is possible to highlight, increase certain
sites of the image for the detailed analysis and also to combine them for obtaining the
synthesized information, to scale images and by that to reflect development or
regression of the studied phenomenon. The organization of access to the separate,
493
chosen as necessary by the operator card elements, allows to analyze ratios between
components for more exact analysis of events. Existence of the editorial mode gives
an opportunity to quickly make changes and to model the existing or predicted
situation.
For information support of the solution of various ecological tasks, including
ecological safety and environmental protection, it is necessary to collect, to keep,
process and interpret various geo-ecological data. As a natural stage on the way of
further systematization of the saved-up data and to their effective use creation of
regional geo-ecological atlases has to serve. The projected satin cartographical work
has to meet to a number of the traditional and specific requirements caused by its
functional features: thematic completeness and versatility of contents; internal unity
in the relation of contents and data representation forms; geographical concreteness
and detail; optimum combination of interpretation and synthesis of data with their
help and factual reproduction; semantic orientation; present; availability, ease of
perception of information; possibility of fast obtaining reference; convenience of use;
high esthetic qualities.
The geographic information systems including parameters of health of the
population and the environment - the main tool for decision-making on formation and
improvement of policy of decrease in an adverse effect of factors of the environment
on health of the population.
Especially promising are regional GIS, which allow systemically organize regional
geo--ecological information, analyze it quickly and map it, which is important for the
purposes of regional environmental management and optimization of environmental
management.
For example, a regional GIS was created, which is a geo-information geo-
ecological Atlas of one of the regions of Russia. The Atlas includes a number of
models of the following geo-ecological maps:
1. Maps of sustainability of natural complexes-reflect information assessment of the

ability of landscapes or their components to maintain the structure and mode of
operation in the conditions of technogenesis.
2. The overall landscape-ecological (geo-ecological) maps reveal the geo-ecological
context, natural X, Y, social, technical and production Z subsystems, in the process
of economic development of landscapes.
3. Maps of landscape-ecological (geo-ecological) potential reveal the possibility of
landscape to meet the needs of society.
4. Maps of technogenic complexes and objects that have influence on the natural
territorial complexes.
5. Maps of man-made changes in landscapes-reflect the acquisition of new landscape
or the loss of previous properties under the influence of external factors or self-
development.
6. Landscape-ecological (geo-ecological) maps of regulation of economic activity -
reflect the existing and proposed restrictions in order to form cultural landscapes.
7. Maps forecasting anomalous (catastrophic) phenomena in landscapes, that is, those
that do not fit into the normal mode of landscapes functioning, accompanied by a
494
radical restructuring of their condition, the destruction of geotechnical systems, the

deterioration of welfare, health and living conditions of the population of the
region.
Other example of effective use of GIS as tool of assessment of a condition of the

habitat: GIS of control of technogenic impact on the environment of the inhabited
territory where the ecological situation is close to critical. The areas of 12 stations
located in various areas are a part of system (10 - PC platform and 2 - Unix
platform).
To work as a base layer a digital model of the topographic base scale 1:500000 as
given. The main content of the system is information on annual emissions of
pollutants into the atmosphere and water bodies, the location and content of waste
storage facilities, the state of forests and the activity of pollution sources. Regular
updating of information is carried out at the expense of monitoring data - tests and
water intake, atmospheric air, etc.
Introduction of GIS offers broad prospects for the analysis and forecasting of an
ecological situation. So, for example, combination of vector layers of model of an
arrangement of storages of waste and basins of the rivers gives complete idea of
potential danger of hit of pollutants in waterways. And by consideration of the project
of construction of any object by means of GIS the section "Conservation" allows to
demonstrate influence of an object on the environment and to pick up optimum
parameters for construction of nature protection constructions.
Other example of use of GIS at assessment of a condition of the habitat are
pollution researches heavy metals of the environment in one of the mining provinces,
most known in Russia. GIS basis is made here by the following vector layers in scale
1:10000: geological map of the area, tectonic map, hydrogeological map; map of a
hydraulic network and lakes, quarters of settlements and technogenic objects and also
relief; map of points of geochemical approbation of soils and surface water.
With the help of the Spatial Analyst module, maps of soil contamination by
elements (Zn, Si, Cr, Mn, Cd, Pb, As, Ni) were built on this basis. Further maps of
soil contamination by the specified elements in relative shares of maximum
permissible concentrations were constructed.
The result of the work was 12 vector layers of the territory assessment map for
each of the analyzed factors. For the general expert assessment of the area, these
layers were summarized and a final map of the integrated assessment of the area by
13 factors was obtained.
Thus, complex satin mapping on the basis of GIS-technologies is one of the most
effective and perspective methods of the spatial analysis allowing to estimate quickly
state of environment and to make effective administrative decisions on optimization
of environmental management and regional planning of development the urban
territories on this basis.
For assessment of the quality of the environment of the city territory created GIS
"Eco-geochemistry and Technological Hazards", which was formed with help of the
electronic geo-ecological atlas of the city. The automated City Eco-geochemistry
495
10
database and also program and cartographical tools of creation of city geo-ecological
maps forms a basis of GIS.
"Ecogeochemistry of the city" databases are the cornerstone ten years observations
of geochemical state of environment of the city air pollution and soils on priority
pollutants B various points of the city (annually - not less than 200 samplings for
assessment of air pollution and the soil).
3.1 Creation in the environment of «Mapinfo» the electronic

topographic base map of the city
Fig. 2. Type of the GIS electronic basis-map
For systematization of data on state of environment and studying of formation of

zones of environmental risk for the population of the city the Mapinfo Professional
program, version 7.8 which is most convenient for realization of a goal is used. By
means of this program the city GIS electronic basis-map was created which structure
is shown in the Fig. 2.
At the first stage of creation of the topographic base map, a raster image of the city
was registered in the Mapinfo system the raster image of the city was registered. In
the projection "NonEarth (kilometers)". A 1:20 000 scale city map was used as the
basis for the raster image.
On the basis of this raster layer 11 vector layers were created: housing estate;
industrial building; green plantings; streets; large highways; railroads; relief;
cemeteries; gardens; reservoir; borders of areas.
The layer "Housing estate" (Fig. 2) is created by means of "grounds" which were
imposed on the main residential quarters of the city.
The layer "industrial building" (Fig. 2) is created by analogy with the housing
estate by means of grounds which are imposed on contours of the industrial
enterprises of the city - the main sources of environmental pollution.
496
11
B the table of this layer are brought names of the enterprises, their postal address, a
specific contribution of emissions of pollutants B the atmosphere from city and a class
of harm of the enterprise. On the basis of this layer it is provided possible to estimate
areas of alleged local influence of industrial zones taking into account a wind rose and
also placement of fragments of the housing estate within sanitary protection zones of
the industrial enterprises that is observed B city where several thousands of people
live in sanitary protection zones of the enterprises 1 and 2 of classes of harm.
The layer "green plantings" (Fig. 2) is created by means of grounds. This layer
allows to estimate possible "retardation" of distribution of toxic substances from the
industrial enterprises and highways.
Layers "large highways" (Fig. 2) and "railroads" are created by means of lines and
polylines. The layer "large highways" displays the main transport arteries of the city.
As motor transport is the main source of lead, formaldehyde, oil products and other
pollutants, comparing this layer with a layer "housing estate", it is possible to estimate
zones of possible excess of maximum allowable concentration of these pollutants in
the environment. Railway transport is less toxic from the ecological point of view,
however, comparing a layer "railroads" with "housing estate", it is possible to reveal
residential quarters in which the noise discomfort and local chemical pollution of the
environment is observed.
The layer "relief" (Fig. 2) is created by means of smoothed polylines. This layer
allows to estimate distribution of pollutants of the environment taking into account
the relative height of this or that territory. So, for example, in hollows concentration
of pollutants of the environment is usually higher, than on heights. Also this layer
allows to predict aeration of this or that site of the urban area. Layers "gardens" and
"cemeteries" are created by means of grounds. The layer "gardens" displays the main
landings of fruit-trees in the territory of the city, and the layer of "cemetery" displays
an arrangement of the main cemeteries of the city.
The layer "reservoir" (Fig. 2) includes all reservoirs located in the territory of the
city - a city reservoir, the rivers, lakes, ponds and other reservoirs. The city reservoir,
ponds and lakes and also the large rivers were created by means of grounds, and
smaller rivers - by means of smoothed polylines.
The layer of "border of areas" (Fig. 2) created by means of lines and polylines
shows an arrangement of the borders dividing the city into 6 certain territorial areas.
The received electronic card basis formed a basis for spatial display and the
analysis of eco-geochemical and medical-statistical data. Creation of GIS was carried
out by the block (modular) principle that gives the chance to expand system due to
addition of new blocks (programs) or to work only with a certain part (module) of
GIS. The created GIS can be used as help system about geochemical state of
environment and health of the population in various parts of the city as the tool of the
mathematical analysis of data of imitating modeling and forecasting of emergence of
zones of environmental risks under the influence of eco-geochemical factors of the
environment, so and as the tool for management of urban areas.
497
12
References
1. Steve Egger: Determining a sustainable city model. Environmental Modelling & Software,
Volume 21, pp. 1235-1246 (2006).
2. Ercan Baldemir, Funda Kaya, Tezcan Kasmer Sahin: A Management Strategy within
Sustainable City Context: Cittaslow. Procedia - Social and Behavioral Sciences, Volume
99, pp. 75-84 ( 2013 ).
3. Marie-Hélène Lizée, Jean-Francois Mauffrey, Thierry Tatoni, Magali Deschamps-Cottin:
Monitoring urban environments on the basis of biological traits. Ecological Indicators,
Volume 11, pp. 353–361 (2011).
4. Bin Yang, Tong Xu, Longyu Shi: Analysis on sustainable urban development levels and
trends in China's cities. Journal of Cleaner Production , Volume 141, pp. 868-880 (2017).
5. Agata, M.L.: Municipal waste management in context of sustainable urban development.
Proced. Soc. Behav. Sci 151, pp. 244-256 (2014).
6. Guo, C.Z., Luo, L.L., Ye, M.: Empirical analysis of factors influencing the sustainable
development of resource-based cities. China Popul. Resour. Environ, Volume 24 (8), pp.
81-89 (2014).
7. Paul McCord, Francesco Tonini, Jianguo Liu: The Telecoupling GeoApp: A Web-GIS
application to systematically analyze telecouplings and sustainable development. Applied
Geography, Volume 96, pp. 16–28 (2018).
8. Luhe Wan a, Yuwei Zhang, Shaoqun Qi, Haiyan Li, Xiaohong Chen, Shuying Zang: A
study of regional sustainable development based on GIS/RS and SD model d Case of
Hadaqi industrial corridor. Journal of Cleaner Production, Volume 142, pp. 654-662
(2017).
9. V.A. Ilyichev, S.G. Emelyanov, V.I. Kolchunov, N.V. Bakaeva and others: The principles
of transformation the city in Biosphere compatible and developing people. M., ASV,
p.184 (2015).
10. W.B.Vasanta Kandasamy: Set Theoretic Approach to Algebraic Structures in
Mathematics. Florentin Smarandache Educational Publisher Inc., p. 166 (2013).
11. Kseniia Lapshina, Natalia Bakaeva, Olga Sotnikova: Assessment of economic efficiency
of acceptance of administrative decisions in the field of an urban ecology. International
Scientific Conference: Sustainable growth in small open economies, pp. 143-146. Institute
of Economic Sciences, Belgrade, Serbia (2017).
498
Paper ID- 110
EFFECT OF DIFFERENT POZZOLANIC MATERIALS ON

PARTIALLY REPLACING NATURAL SAND BY M-SAND
1KiranM. Mane, 2D. K. Kulkarni and 3K. B. Prakash

1
PhD Research scholar, S.D.M. College of Engineering and Technology, Dharwad- 580 002,
Karanataka, India, kiranmane818@gmail.com, +91 9850169818.
2
Professor, Department of Civil Engineering, S.D.M. College of Engineering and Technology, Dharwad-
580 002, Karanataka, India, dilipkkulkarni@rediffmailcom +918151832888
3
Principal, Govt. Engineering College, Haveri, Devagiri- 581 110, Karanataka, India,
kbprakash04@rediffmail.com, +91 9448267374.
Abstract
Now a day’s concrete is the extreme broadly used construction material in civil engineering
industry because of its extraordinary structural strength and stability. The overuse level of
cement and natural sand for civil industry has several undesirable social and ecological
consequences. As an answer for this, industrial wastes called as by-products (pozzolanic
materials) such as fly ash, GGBFS, silica fume and metakolin can be used to interchange
partially cement and natural sand by manufacturing sand (M-sand). This research aims to
investigate the possibility of replacing natural sand by M-sand with 20% of above pozzolanic
materials substitute in concrete. In this experimentation, natural sand was replaced by M-sand in
various percentages (0%,10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% and 100%), with
water-cement ratio of 0.45 and cement was partially substituted by 20% of pozzolanic materials.
M30 grade of concrete mix proportions were designed as per IS 10262:2009 guidelines. The
fresh concrete properties and compressive strength results were checked for the different
concrete mix proportions and compared with conventional concrete. From this research work, it
can be concluded that for replacement of 60% natural sand by M-sand and 20% cement by silica
fume yields maximum compressive strength than conventional concrete.
Keywords: Pozzolanic materials, M-sand, Fresh concrete property, Compressive strength,

Conventional concrete.
499
Introduction
Concrete is the most widely used construction material throughout the world hence concrete
technology is the backbone of the infrastructural development of every country. It has made
tremendous advancements in the western as well the eastern world. Annual worldwide
manufacture of concrete is approximately up to the 1m3 per person on earth. The main
component of concrete is associated with Portland cement. The world cement production was
about 2.6 billion tonnes in 2008. By 2020, cement requirement is estimated to be around 3.5
billion tones this will obviously cause an equal demand on the materials like sand, aggregate and
other materials required to produce huge quantity of cement concrete[1]. This will naturally
cause depletion of all the natural resources connected in making cement concrete every year. The
three major concerns associated with cement production are enviro-eco issues, sustainability
issues and intense energy needs [2]. The production of cement releases approximately an equal
amount of CO2 into atmosphere due to the calcination of limestone and combustion of fossil fuel.
In view of this, with an interest in minimizing the overall CO2 emissions associated with OPC
composites blended cements were introduced by partially replacing OPC by pozzolanic materials
such as fly ash, ground granulated blast furnace slag (GGBS), silica fume, metakaolin, rice husk
ash etc. [3]. Also due to increased levels of construction expected in the forthcoming years, it is
expected that fine aggregates suitable for use in concrete will become scarce or uneconomical
[4]. With the expected shortfall in natural sands, manufactured sands offer a viable alternative to
natural sand. Manufactured sand has to satisfy the technical requisites like workability and
strength of concrete. Since the data on this aspect of concrete using manufactured sand is scarce,
it is necessary to investigate the concrete produced with manufactured sand.[5]
Research significance
Shortfall of river sand in today’s scenario has appealed many researchers worldwide to find an
alternative. Moreover, quarrying of river sand will impact on the environmental issues, bank
slides, loss of vegetable on the river banks and it may also impact aquatic life. Considering fine
aggregate is very vital component in concrete, it necessary to develop an alternative material[5]
This article addresses this critical issue and provides information on the possibility of using
manufacture sand, blended with natural sand, to achieve the desired strength properties of
500
concrete through the use of pozzolanic materials. Such concrete can be used in structural
applications for both PCC and RC.
Materials and Methodology
OPC 43 grade cement and satisfy the requirement of IS 8112-2013 with specific gravity 3.15.
Potable water is used for concrete mixing and curing. Locally available natural sand conforming
to Zone II with specific gravity 2.61 and fineness modulus 2.24 [6,8] Locally available vertical
shaft impact (VSI) crusher sand conforming zone II with specific gravity 2.82 and fineness
modulus 2.91.[7]Coarse aggregates used in the experimental study are 10mm and 20 mm size
and having specific gravity 2.94 [8], Fly ash used in this experimentation was obtained from
JSW plant Ratnagiri Maharastra India, having 58.54% silicon dioxide (SiO2), specific gravity
2.15 [9], Silica flume is obtained from ELKEM south Asia Pvt limited Mumbai India was named
Elkem-micro silica 920 D conforming to ASTM C1240. It is available in dry densified form and
having 91.14% silicon dioxide (SiO2), Specific gravity found 2.2 [10], Ground granulated blast
furnace slag(GGBFS) is obtained from JSW plant Bellary Karnataka India having 41.61%
silicon dioxide (SiO2), specific gravity 2.85 [11]. Metakaolin is obtained from Golden micro
chemicals Mumbai, India having 54.66% silicon dioxide (SiO2), specific gravity 2.2 and
naphthalene based, free from chloride admixture used in this project was Fosroc Conplast SP430
to improve workability of concrete.
The experimental investigation was based on a reference concrete mix of grade M30 using
natural aggregates. On the basis of the material properties, the proportioning of concrete mix was
carried out in accordance to IS 456:2000 and as per the guidelines of IS 10262:2009 [12].
Preparation of specimens
The concrete was obtained by replacing natural sand in the reference mix. The natural sand were
replaced by manufactured sand in the proportion of 0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%,
80%, 90% and 100%. Along with natural sand, 20% of the cement was replaced by the different
mineral admixture by weight. To measure the workability, slump, compaction factor, flow and
vee-bee shear test were carried out for each replacement level of natural sand to manufactured
sand and cement by mineral admixtures. To determine the harden properties of concrete,
501
compression strength tests were tested in 200T compression testing machine for cubic specimens
of size 150mm x 150mm x 150mm after cured specimens for 28 day as per I.S. 516 -1959[13]
Results and discussions-
1. Test results on fresh concrete:
The workability test results carried out by using slump, compaction factor, flow and vee bee
degree with different percentage replacement of natural sand by manufactured sand and 20%
cement replaced by different pozzolanic materials are shown in tables 1 to 4 respectively and the
variation of slump, compaction factor, flow and vee bee degree with respect to various
percentage replacement of natural sand by manufacture sand are presented in figures 1, 2, 3 and
4 respectively. From the test results obtained, it is observed that as percentage replacement of
natural sand by manufactured sand is increased there is decrease in workability properties. The
reasons for reduction in workability are due the angular shape and rough surface of manufactured
sand which imparts more internal friction there by reducing the flow characteristics of concrete.
Table 1 Slump (mm) values with different pozzolanic materials.
Percentage
replacement of No replacement Cement Cement Cement Cement
natural sand by of cement replaced by replaced by replaced by replaced by
manufactured fly ash silica fume GGBFS metakaolin
sand
0 100 98 98 98 99
10 95 96 95 94 96
20 94 95 94 94 95
30 90 93 92 93 94
40 85 85 92 92 92
50 85 85 88 90 89
60 84 85 82 88 84
70 80 82 80 82 82
80 80 80 78 82 82
90 75 76 76 80 78
100 75 76 74 74 72
502
Table 2 Compaction factor values with different pozzolanic materials.
Percentage
sand
0 0.885 0.910 0.900 0.912 0.889
10 0.862 0.864 0.892 0.889 0.885
20 0.862 0.863 0.890 0.865 0.882
30 0.860 0.860 0.888 0.862 0.880
40 0.858 0.858 0.885 0.860 0.878
50 0.858 0.857 0.884 0.860 0.875
60 0.852 0.857 0.853 0.859 0.875
70 0.852 0.853 0.852 0.855 0.871
80 0.848 0.853 0.852 0.855 0.855
90 0.845 0.852 0.850 0.850 0.850
100 0.820 0.830 0.827 0.832 0.825
Table 3 Flow (%) values with different pozzolanic materials.
Percentage
sand
0 96 92 92 94 96
10 92 92 88 92 94
20 92 90 88 92 94
30 86 88 87 90 88
40 84 86 85 88 88
50 83 83 84 87 84
60 48 52 60 56 72
70 44 52 48 54 48
80 44 50 46 50 44
90 38 40 32 50 44
100 32 33 28 40 34
503
Table 4 Vee-Bee degree (second) values with different pozzolanic materials.
Percentage
sand
0 5.0 5.0 5.0 5.0 5.0
10 6.0 6.0 5.0 5.0 5.0
20 6.0 6.0 6.0 6.0 6.0
30 7.0 7.0 6.0 6.0 6.0
40 7.0 7.0 6.0 7.0 6.0
50 7.0 7.0 7.0 7.0 7.0
60 8.0 8.0 7.0 7.0 7.0
70 8.0 8.0 8.0 8.0 7.0
80 8.0 9.0 8.0 8.0 8.0
90 9.0 9.0 9.0 9.0 8.0
100 11.0 10.0 10.0 10 11.0
100
95
No replacement of cement
Slump (mm)
90
cement replaced by fly ash
85
cement replaced by silica fume
80
cement replaced by GGBFS
75
70 Cemenrt replaced by metakaolin
0% 20% 40% 60% 80% 100%
% replacement of natural sand by manufactured sand
Fig1. Variation of slump
0.92
Compaction factor
0.9
0.88 Cement replaced by fly ash
0.86 Cement replaced by silica fume
0.84
Cement replaced by GGBFS
0.82
0.8 cemet replaced by Metakaolin
0% 20% 40% 60% 80% 100%
Fig. 2 Variation of compaction factor
504
95
85
75 Cement replaced with fly ash
65
Flow (%)
Cement replaced with silica

55 fume
45 Cement replaced with GGBFS
35 Cement replaced with
25 Metakaolin
0% 20% 40% 60% 80% 100%
Fig. 3 Variation of flow
12
Vee- Bee degree (Sec.)
10
8 Cement replaced with fly ash
6 Cement replaced with silica

fume
4
Cement replaced with GGBFS
2
Cement replaced with
0 Metakaolin
0% 20% 40% 60% 80% 100%
Fig. 4 Variation of Vee-bee degree
2. Test results on compressive strength of concrete
The compressive strength test results were carried out for different replacement of natural sand
by manufactured sand and 20% cement replaced with fly ash, silica fume, GGBFS and
metakaolin in concrete are shown in table 5 and the variation of compressive strength with
different percentage replacement of natural sand by manufacturing sand is presented in fig
5.Percentage increase of compressive strength with respect to reference mix is shown in table 6.
From test results obtained it is observed that the reference concrete specimen made with 0%
manufactured sand and without any pozzolanic materials have less strength. Also compressive
strength of specimens made with 60% replacement of natural sand by manufactured sand and
20% cement replaced by silica fume had higher compressive strength than that of the reference
505
concrete. Its value is 47.9 MPa. Whereas it is 47.12MPa, 47.16MPa and 47.20Mpa in concrete
made with 60% replacement of natural sand by manufactured sand and 20% cement replaced
with fly ash, GGBFS and metakaolin respectively. Also 47.10MPa is occurs at concrete made
with 60% replacement of natural sand by manufactured sand and no replacement of cement. The
improvement for compressive strength is mainly attributed to better particle packing and reduce
surface area which increases binding capacity of manufactured sand and silica fume with cement
paste and thereby increasing the strength. Also improved bond between hydrated cement matrix,
better particle size distribution and reduce micro fines percentage in manufactured sand added
with silica fume the concrete mix become more cohesive. [14]But beyond 60% replacement of
natural sand by manufactured sand it is observed there was gradual decreasing in compressive
strength which was due to lack of bond between manufactured sand and silica fume, as more
micro fine percentage available in manufactured sand and it acts as filler materials without
contribution for strength. The table 6 shows the tests results for percentage of increase of
compressive strength for different percentage replacement of natural sand by manufacturing
sand.
Table 5 Compressive strength (MPa) results with different pozzolanic materials.
Percentage No Cement Cement Cement Cement

replacement of replacement of replaced by replaced by replaced by replaced by
natural sand by cement fly ash silica flume GGBFS metakolain
manufactured sand
0(Ref.) 38.51 39.2 42.3 40.2 40.8
10 40.14 42.3 42.8 42.4 42.6
20 41.87 42.3 43.94 43.7 43.9
30 45.92 45.94 46.12 45.98 46.0
40 45.92 45.95 46.9 45.99 46.10
50 46.52 46.55 47.2 46.8 46.9
60 47.10 47.12 47.90 47.16 47.20
70 42.88 43.8 46.9 45.20 46.80
80 41.47 43.2 46.20 44.92 46.15
90 40.81 42.6 46.20 44.80 46.00
100 39.99 42.59 46.07 44.14 45.10
506
Table 6 Percentage increase of compressive strength test results.
Percentage No Cement Cement Cement Cement

replacement of natural Replacement replaced by replaced by replaced replaced by
sand by manufactured of cement fly ash silica flume by metakolain
sand GGBFS
0(Ref.) - - - - -
10 +4.23 +7.90 +1.18 +5.47 +4.41
20 +8.72 +7.90 +3.87 +8.70 +7.59
30 +19.24 +17.19 +9.03 +14.37 +12.71
40 +19.24 +17.21 +10.87 +14.40 +12.99
50 +20.79 +18.75 +11.58 +16.41 +14.95
60 +22.30 +20.20 +13.23 +17.31 +15.68
70 +11.34 +11.73 +10.87 +12.43 +14.70
80 +7.68 +10.20 +9.21 +11.74 +13.11
90 +5.97 +8.67 +9.21 +11.44 +12.74
100 +3.84 +8.64 +8.91 +9.80 +10.53
48
Compressive strength (MPa)
47 No replacement of cement
46
45
44 Cement replaced by fly ash
43
42
41 Cement replaced by silica fume
40
39 Cement replaced by GGBFS
38
37
0% 20% 40% 60% 80% 100% Cement replaced by Metakaolin
Fig. 5 Variation of compressive strength
Conclusions
 The pozzolanic material (Silica fume, metakaolin, GGBFS and fly ash) used in the
present work have shown that for any percentage replacement of natural sand by
manufactured sand the compressive strength properties of concrete with pozzolanic
marterials are higher as compared to concrete without any pozzolanic materials.
507
 The mix of 20% silica fume with 60 % replacement of natural sand by manufactured sand
achieved a maximum compressive strength of 13.23% increase in compressive strength
when compared to reference concrete mix. Also the mix of 20% silica fume with 100%
replacement of natural sand by manufactured sand shows a greater compressive strength
of 8.91% increase in compressive strength when compared with reference mix.
 The mix of 20% fly ash with 60 % replacement of natural sand by manufactured sand
 The mix of 20% GGBFS with 60 % replacement of natural sand by manufactured sand
 The mix of 20% metakaolin with 60 % replacement of natural sand by manufactured sand
 By using manufactured sand, blended with natural sand, it is possible to achieve the
desired strength and durability properties through the use of pozzolanic materials. Such
concrete can be used in structural applications for RC construction.
REFERENCES
1) Parvati V.K., K.B. Prakash “Feasibility of fly ash as a replacement of fine aggregate in
concrete and its behavior under sustained elevated temperature.” International Journal Of
science & Engineering Research May 2013, vol.4.issue 5, pp 87-90.
2) Naredra Kumar B. “Development of high strength self compacting concrete using quartz
sand as an alternative of natural river sand.” Indian concrete Journal April
2017,vol.91,Issue 4, pp. 43-50
508
3) Verma K.and Pajgade P.S. “Effect of partial replacement of natural sand with crushed
sand along with supplementary cementing materials(fly ash and GGBS)” International
Journal of Research in Engineering and Technology, January 2015,Vol.4,Issue 1,pp.288-
92
4) Nataraja M.C., Manu A.S.and Girih,G.“Utilization of different types of manufactured sand
as fine aggregate in cement mortar”. Indian concrete journal January 2014,vol.88,Issue
1,pp 19-25
5) Nagendra R. From ICI-KBC Chairman’s desk Proceeding of One day seminar on
“Alternatives to river sand a sustainable approach.” 13th December 2013.
6) Lokeswaran M.R., Natarajan C. “Study on the properties of cement concrete using
manufactured sand Advances in structural engineering” December 2014, pp 1803-1809.
7) Shivakumar G. “Manufactured sand –a solution and an alternative to river sand in
concrete.” Proceeding of One day seminar on Alternatives to river sand a sustainable
approach P.P. 42 ,Friday 13th December 2013 Karnataka state pollution control board
parisara Bhavana,Tejasvi Hall, 3rd floorNo.49, Church street, Bangalore.
8) Indian standard code of practice for coarse and fine aggregate from natural sources for
concrete IS.383: 1970 Bureau of Indian Standards , New Delhi.
9) Manjunath H. and. Patagundi B.R. , Performance of concrete by partially replacing fine
aggregate with granulated furnace slag and cement with fly ash Indian Concrete Journal,
June 2017,Vol,91,Issue 6,pp.41-47.
10) Shanmugapriya T., Uma R.N. Optimization of partial replacement of M-sand by natural
sand in high performance concrete with silica fume International journal of engineering
science & Emerging Technologies June 2012 Vol.2, Issue 2,pp73-80
11) Samanthula R. Polimreddy M.R. Performance study on Ggbs concrete with Robosand
International journal of science &Technology research , February 2015,V ol.4,Issue 2,
pp 1-5
12) Indian standard code of practice for Guidelines for concrete mix design
IS 10262: 2009. Bureau of Indian Standards , New Delhi.
13) Indian standard code of practice for Strength of testes for strength of concrete I.S.516:
1959 Bureau of Indian Standards , New Delhi.
509
14) Shehdeh Ghannam, Husam Najm, Rosa Vasconez “Expermental study of concrete made
with granite and iron powder as partial replacement of sand” Sustainable Materials and
Technologes 9 2016 pp 1-9.
510
Paper ID- 111
Finite Element Analysis of Unconfined Brick Masonry with Experimental

Verification
Anjani Kumar Shukla*1 (0000-0003-1449-2539), Saurav2 (0000-0002-9015-0285), P.R. Maiti3

(0000-0002-0327-1354)
1,3
Department of Civil Engineering Indian Institute of Technology (BHU), Varanasi, India
2
Department of Civil Engineering, JAYPEE University, Solan, H.P.,India
*akshukla.rs.civ16@iitbhu.ac.in
Abstract
The survival of low rise Unconfined masonry structure is very rare in earthquake of even low or moderate
intensity. To make Unconfined masonry structures earthquake resistant, a software analysis with experimental
verification have been done. This study starts with developing a 3D finite element model of unconfined brick
masonry. A finite element model for brick masonry subjected to Vertical loading is presented. The four brick
is used to model the unconfined masonry in ANSYS as well as in the Laboratory. In the Software the model
of stacks was tested under different loads and they were verified by applied the same type of load in the
laboratory. The cracking patterns, week zones and the failure regions of Software and Experimental stacks
were compared and presented. The overall result of stress was also presented by the Von-Mises (distortion
energy method) stress contour. The ANSYS modeling of plain Brick masonry shows that masonry structure
fails at the joint.
Keywords: Finite Element Analysis, Masonry Structure, Earthquake resistant structure, ANSYS
Modelling, Unconfined Brick Masonry
1. Introduction
The major portion of the masonry construction are unconfined around the world. At the time of
Earthquake these structures have showed very poor performance. Based on past studies, the most
common modes of failure of unconfined masonry buildings are:
1) Failures at the joints between vertical tie columns and horizontal bond beams.
2) Horizontal cracks at the joints.
3) Shear cracks into the tie-columns of unconfined Masonry
4) Cracks in window piers and walls.
Present study covers Finite element modelling of Brick masonry, Stress analysis of Brick Masonry
using ANSYS and Analysis of different stresses in Brick Masonry wall for different boundary
conditions.
511
2. Modelling of Bricks units using ANSYS
Modelling of four bricks of size 190mm*90mm*10mm is presented in the ANSYS Software
based on the Finite Element Analysis.
2.1 Choice of Elements
The ANSYS software contains more than 100 different element types in its element library. Each
element has a unique number and a prefix that identifies the element category, such as BEAM3,
PLANE42, SOLID45, etc. ANSYS classifies the elements into 21 different groups, out of which our
main concern is of structural group. SOLID45 is used for the 3-D modelling of solid structures. The
element is defined by eight nodes having three degrees of freedom at each node: translations in the
nodal x, y, and z directions. The element has plasticity, creep, swelling, stress stiffening, large
deflection, and large strain capabilities. The geometry, node locations, and the coordinate system for
this element are shown in Figure 1: "SOLID45 Geometry". The element is defined by eight nodes and
the orthotropic material properties. Orthotropic material directions correspond to the element coordinate
directions. Pressures may be input as surface loads on the element faces as shown by the circled numbers
on "SOLID45 Geometry".
Figure 1. Solid45 Element
SOLID45 and BRICK 8NODE 45 have a different numbers of nodes per element this is major
dissimilarity among both, by which both employ different interpolation functions for the different the
degrees of freedom. The interpolation functions for the BRICK 8NODE 45 element are linear.
2.2 Material properties
For the brick element material properties which are assigned are modulus of elasticity (EX) and
Poisson’s ratio (PRXY). Values of EX and PRXY are take according to Ali and page 198612 and are
tabulated below in Table 1 and Table 2.
512
Table 1. Material properties of bricks
Properties Mean
(a)Bricks
Modulus of elasticity 14700 Mpa
Poisson’s ratio .16
Tensile strength 1.20 Mpa
Table 2. Material properties of mortar
Properties Mean
(a) Mortar
Modulus of elasticity 7400 Mpa
Poisson’s ratio .21
Tensile strength 0.78 Mpa
2.3 Model Preparation on ANSYS

2.3.1 Model detail of specimen M1 - To model the masonry, rectangular blocks are used for bricks
and also for mortar. Four numbers of bricks of standard size i.e 190mm×90mm×90mm are used and
mortar of thickness 10mm is placed in between them as shown in the Figure 2.
Dimension of bricks: Dimension of mortar:

Length =190mm Length= 190mm
Width = 90mm Width = 90mm
Height =10mm
Height = 90mm
513
Figure 2. Model of specimen M1
2.3.2 Model detail of specimen M2: A brick wall panel is modelled with rectangular blocks of size
190mm×90mm×90mm and mortar as shown in the fig4. In wall is modelled as a stretching bond as in
Figure 3, in which all bricks are laid as stretchers. Dimension of the wall is 1.99m×0.79m×.09m.
Figure 3. Model of specimen M2
2.3.3 Meshing
To mesh this brick masonry volume 4 to 6 sided mapped meshing is used. The mapped meshing concept
is valid only in case of two- and three- dimensional problems (no line elements). The solid model
entities (area and volumes) meshed with these option use quadrilateral area elements or hexahedral
(brick) volume elements. The reason why mapped meshing is desirable is that it generates regular thus
computationally well-behaving, meshes. Only “regular” areas or volumes can be mapped meshed. The
mapped meshed volume is shown in the figure 4 and figure 5 below
514
Figure 4. Meshed model for specimen M1 Figure 5. Meshed model for specimen M2
2.4 Model Preparation in Laboratory
2.4.1 Material Specification:
Bricks: Locally available burnt clay bricks from a kiln are used in this experiment. Average Size
of the bricks was 22cm×10.5cm×6.5cm. Water absorption test was performed to determine the
water absorption of bricks
Cement: Portland Pozzolana cement PPC (Brand Name - Ultratech Cement), standard consistency
test was performed and it was found out to 36%.
Sand: Sand used in the preparation of mortar was natural sand.
Mortars: Cement Mortar; Proportion: 1cement: 4 sand
Casting and Curing: For Experimental verification the 4 brick specimen were casted and cured
for 28 days in tank. Refer Figure 6 for casting and Figure 7 for curing.
Figure 6. view of the bricks with mortar Figure 7. Curing in laboratory
515
3. Result and discussion
Post processing includes defining boundary condition and application of loads. For specimen M1, the
one end of the brick masonry is fixed. The load is applied on the other end of the blocks. Pressure
loads of 70 kN, 80kN; 90kN is applied on the top face of the brick masonry. In addition to the vertical
load, horizontal load is also applied to the specimen M1 different stress contours are drawn and graph
is been plot between different parameters. For specimen M2, one end is fixed to make it confined and
vertical load is applied.
3.1 Stress computation
3.1.1 Stress computation for specimen M1 with vertical loading
Figure 8. X component stress Figure 9. Y component stress
From Figure 8 it can be seen that X component of stress is maximum at the middle points. It increases
parabolically from ends towards the center. For middle bricks this stress is more as compared to top and
bottom bricks and figure 9 Y component of stress has more or less same values except at the lower
points where it is more. And the joints of all bricks shows the higher stress with red marks. It means
these locations are at high risk at the time of Y Loading.
516
3.1.1 Various Stress computation for specimen with Experimental Verification
3.1.1.1 1st Principal Stress: In Figure 10 it can observed that Variation of 1st principal stress is
similar to x component stress but its values are lesser than that of x component of stress and the similar
pattern can be verified by laboratory experiment in Figure 11.
Figure 10. 1st principal stresses Figure 11. Experimental Verification
3.1.1.2 XY shear stress: XY shear stress is more critical at the bottom corner of the masonry.
Masonry Prism has the tendency to crack at the bottom first in compressive testing machine which can
be verified in Figure12. The variation of XY shear stress along Y axis is shown in Figure13 Peak values
are obtained at the place of brick mortar interface. The variation of XY shear forces along Y axis can is
plotted in Figure 14.
Figure 12. XY shear stress Figure 13. Experimental Verification of XY Shear
517
Figure 14. Variation of XY shear stress along Y axis
3.1.1.3 YZ shear stress: When the force was applied on the YZ Direction the stake feel the force at
its face on bottom (Figure 15). Variation of stress with distance along Y axis can be observe in Figure
16.
Figure 15. YZ shear stress Figure 16. Variation of YZ shear stress along Y axis
518
3.1.1.4 XZ shear stress: When the force was applied on the XZ Direction the masonry prism feel
the force at its joint corner of every brick (Figure 17). Variation of stress with distance along Y axis
can be observe in Figure 18.
Figure 17. XZ shear stress Figure 18. Variation of XZ shear stress Y axis
3.1.1.5 Von Misses stress: Von Misses stress on the specimen can be obtained by ANSYS
Software and the critical points can be followed (Figure 19) and its Experimental verification can be
seen in figure 20. Variation of stress with distance along Y axis can be observe in Figure 21.
Figure 19. Von misses stress Figure 20 Experimental Verification of Von Misses Stress
519
Figure 21. Variation of Von Misses stress along Y axis
It can be observed that shear stress increases with distance along Y axis. At a distance of 90mm stress
is approximately 1400 Mpa but from 90mm to 100mm i.e at the level of mortar it decreases abruptly to
-1563Mpa. Maximum shear stress is also observed at 28mm which is equal to 1693Mpa.
3.2 Stress computation for specimen M1 with vertical loading and horizontal loading
In this model in addition to vertical load of 70kN, horizontal load of 30kN is also applied. At first the
horizontal load is applied at the upper most brick. From Y component of stress it is observed that
maximum stress occurs at the lowest level. Variation of Von misses stress can be observed from Figure
22 and Figure 23.
Figure 22. Y component of stress Figure 23. Experimental Verification
520
Similarly, The stress in XY direction can be observe in Figure 24 and verified which can be found in figure 25.
And its Von Misses stress Verified the experimental output of stress in laboratory (Figure 26).
Figure 24. XY shear stress Figure 25. Experimental Verification
Figure 26. Von misses stress
Again in this model horizontal load is applied at the middle two bricks. It is observed from Figure27
that Y component of stress at the bottom most point has decreased. There is increased in Von misses
stress when wind load is applied in the middle two bricks as shown in the Figure28.
521
Figure 27. Y component stress Figure 28. Von misses stress
3.4 Stress computation for specimen M2 with vertical loading and horizontal loading
For specimen M2, simultaneously vertical and horizontal loading is applied to the wall and its stress
variations are observed. From Figure27 it is observed that due to application of horizontal loads wall
has the tendency to deflect away in the direction of load. But other end of the wall has not defected due
to confinement. XY shear stress also decreases from top to bottom.
Figure 29. XY shear stress
4 Conclusion
When we will analyze all above stress contours we can see that at every boundary layer of mortar
ANSYS shows the maximum stress and in Danger zone. It means if we want to increase the strength of
masonry structure we have to strengthening our brick mortar joint with adequate ductility so that at the
time of earthquake it can shows some deflection before failure. So that we can save the human life as
well.
522
References:
1. Rots, J.G., "Computer Simulation of Masonry structure: Contiuum and Discontinuum

Models ", pp. 93-103, Proceedings of the International Symposium on Computer Methods in
Structural Masonry Swansea, UK, Apnl 1999.
2. Stockl, S., Bienvirth, H., and Kupfer, H., "The influence of Test Method on the Resirlts
of Compression Tests on Mortar ",Proceedings,10" IBMAC, University of Calgary, pp.1397-
1406, 2000.
3. Alcocer, S., Arias, J.G., and Flores, L.E. (2004). Some Developments on Performance-Based
Seismic Design of Masonry Structures. International Workshop on Performance-Based Seismic
Design, Bled, Slovenia.
4. EERI (2006). The Tecomán, México Earthquake January 21, 2003, An EERI and SMIS Learning
from Earthquakes Reconnaissance Report, Technical Editors S.M. Alcocer and R.E. Klingner,
Earthquake Engineering Research Institute, Oakland, California, March 2006.
5. Schultz, A. E. (1994). Performance of Masonry Structures During Extreme Lateral Loading
Events, Masonry in the Americas, ACI Publication SP-147, American Concrete Institute,
Detroit, pp.21-55.
6. Tomazevic, M. and Klemenc, I. (1997). Seismic Behaviour of Confined Masonry
Walls.Earthquake Engineering and Structural Dynamics, Vol. 26, pp.1059-1071.
7. Yoshimura, K. et al. (2004). Experimental Study For Developing Higher Seismic Performance
of Brick Masonry Walls, Proceedings of the 13th World Conference on Earthquake Engineering,
Vancouver, Canada, Paper No. 1597.
8. Alcocer, S.M. (2006). Personal Communication.
9. Tomazevic, M. (1999). “Earthquake-Resistant Design of Masonry Buildings”. Imperial College
Press, London, U.K.
10. Blondet, M. (2005). Construction and Maintenance of Masonry Houses – For Masons and
Craftsmen. Pontificia Universidad Catolica del Peru, Lima, Peru (http://www.world-
housing.net/)
11. Mohammed S. Mohammed (2006) Finite Element Analysis Masonry Walls of Unreinforced
12. Ali,S. and Page,A.,W. “Finite Element Model for Masonry Subjected to Concentrated Loads”,
J., Struct. Division ASCE, Vol.114, No.8, 1998, pp.1761-1784
13. Hilsdorf, H.K.(1969) “An investigation into the failure mechanism of brick masonry loaded in
axial compression.” pp 34-41
14. Khoo,C.L. and Hendry,A.W.(1973) “ A failure criterion for brick work in axial compression”
proc.,3rd int. Brick masonry conference .
15. Atkinson, R.H., Noland,J.L., and Abrams, D.P.(1982) “A deformation theory for stack bonded
masonry prism in compression” proc., 7th int. Brick masonry conference., Melbourne university,
Australia,565-576
16. Suter, G. T., and Naguib, E. M. F., "Effect of Brick Stiffness Orrhorrapy on the Lateral
Stress in Stuck-Bonded Brick Masonry Prisms ",Proceedings, Fourth North American Masonry
Conference, paper 18, 1997.
17. Page,A.W.,and Brooks, D.S.,"Load Bearing Masonry -Review ",Proceedings of the 7th
International Brick Masonry Conference, pp. 81-99: 1995.
18. Hendry,A.W., "TestingMethods in Masonry Engineering", Proceedings, Fourth North
American Masonry Conference, paper 49,1997
19. Drysdale,R. G., & Wong, H.E., "interpretationof the Compressive Strcngth of Masonry
Prisms ",Proceedings of the 7thInternational Brick Masonry Conference, 1995.
20. Brooks, J.J.,and Abu Baker,B. H.,"The Modulus of Elasticity of Masonry, The British Masonry
Society Journal. Vol.12.No2. November 2000
21. Mario Paz, “Structural Dynamics - Theory and Computations”, 6th Edition, Pearson Education,
2005.
523
22. Pankaj Agarwal & Manish Shrikhande, “Earthquake Resistant Design of Structures”, 5th
Edition, Prentice Hall of India, New Delhi, 2009. 140 G V P College of Engineering
(Autonomous) 2013
23. Jai Krishna A.R, Chandrasekharan A.R, Brijesh Chandra, “Elements of Earthquake
Engineering”, 2nd Edition, South Asian Publishers, New Delhi, 2001.
24. Chopra A.K., “Dynamics of Structures”, 5th Edition, Pearson Education, Indian Branch, Delhi,
2007.
25. Clough & Penzien, “Dynamics of Structures”, 4th Edition, McGraw Hill, International Edition,
2008.
524
Paper ID- 112
Analysis of Moment and Torsion in skew plates using ABAQUS
Anjani Kumar Shukla*1(0000-0003-1449-2539), Vishal Koruthu Philip2 (0000-0002-5006-7823) P.R.

Maiti (0000-0002-0327-1354),3
Department of Civil Engineering, Indian Institute of Technology (IIT-BHU), Varanasi, India

*1
akshukla.rs.civ16@iitbhu.ac.in
Abstract
The increased utilization of skew plates in thin walled structural components of aircrafts, submarines,
automobiles and other high-performance application areas have necessitated a strong need to understand
their characteristics. This paper presents the bending analysis on isotropic skewed plates using the finite
element software ABAQUS Simulia. Modelling was done in ABAQUS Simulia using SC8R element as it
gave better results for shell structures. Analysis is done on skew plates of various skew angles and aspect
ratios with all ends clamps subjected to uniform pressure and the results are generated. The effect of skew
angle, aspect ratio and thickness of the plate on various parameters like bending and torsional moments,
support shear and deflection are separately presented which may serve as benchmark for further research.
From the results, it is observed that the maximum central deflection decreases with increase in skew angle.
Negative reaction was found to develop at the acute angle corners and it increased as the skew angle
increased. Bending moments were found to decrease and torsion was found to increase with increase in
skew angle.
Key words: Skew plates, finite element, uniform pressure, Plate Analysis using Abaqus
1. Introduction
Plates and shells are extensively used as structural parts in civil, aerospace, automotive and marine
engineering structures. Specific application of isotropic skew plates includes aircraft wings and aircraft tail-
fins. The edges of these plates are often so mounted that their boundaries can be assumed to be equivalent
to various classical flexural boundary conditions. Due to the rapid development of aviation and aerospace
programs, the analysis of plates of different skew angles and with different loading have become necessary.
Experience showed that such structures may fail in many cases not on account of high stresses, surpassing
the strength of material but owing to insufficient stability of slender members. Due to its significance in
525
structural mechanics, large number of references in the published literature deal with vibration and static
stability behavior of plates subjected to uniform in-plane stresses.
A general theory about the bending of skew plates and shells is given in the book of Timoshenko [1]. Das
et al. analyzed the large deflection static behavior of isotropic skew plates under uniform pressure loading
using appropriate domain mapping technique [2]. Girish et al. studied the buckling behavior of skew plates
using finite element method using Hyper-Mesh as pre and post processor and MSC/NASTRAN as solver
[3]. Srinivasa et al. studied the buckling behavior of Laminated Composite Skew Plates and obtained time
history for the nonlinear free vibration of composite skew using Newmark’s time integration technique [4].
Naghsh and Azhari employed the element-free Galerkin (EFG) method to analyze the large amplitude free
vibration of point supported laminated composite skew plates [5]. Singha and Daripa studied the large
amplitude free flexural vibration behavior of symmetrically laminated composite skew plates using the
finite element method [6]. Dey and Singha investigated the dynamic stability characteristics of simply
supported laminated composite skew plates subjected to a periodic in-plane load using the finite element
approach with a four-noded shear flexible high precision plate-bending element [7]. Farag and Ashour
developed a fast converging semi analytical method for assessing the vibration effect on thin orthotropic
skew (or parallelogram/oblique) plates [8]. Chun and Lim derived analytical solutions for the problem of
skewed thick plates under transverse load on a Winkler foundation [9].
Although much research has been done on the behavior of skew plates very little insight have been obtained
about the variation of parameters like moments, shear and torsion with skewness which may be useful for
the understanding the behavior of skew slabs and bridges. The objective of this study is to investigate the
bending behavior of skew plates using finite element method. The behavior with respect to three parameters
(skew angle, aspect ratio and thickness) is analyzed in detail.
2. Finite element modelling

In the present study, ABAQUS Simulia software is used for the analysis. Skew plates of different skew
angles are modelled using continuum shell type element SC8R Figure 1. The analysis of the plates was
done with all ends clamped subjected to uniform pressure of 20kPa for different aspect ratios and thickness.
The material properties of the plate modelled were that of mild steel. (Young’s Modulus E=2x10 11N/m2,
Poisson’s ratio µ=0.3.)
526
Figure. 1: Finite element model of plate in ABAQUS
For a constant plate thickness (6mm), the skew angle Ф was varied from 0o to 75o for different aspect
ratios and variation of different parameters like maximum deflection, moments and shear force at the
supports is analyzed. To understand the effect of thickness, analysis was done for a square skew plate by
varying the thickness from 1mm to 10mm for different skew angles (figure 2).
Figure. 2: Applied load on the plate model
The results obtained from the analysis in ABAQUS are tabulated below (Table 1-4). The results are also
graphically plotted to understand the variation in different parameters.
It can be seen in all the tables below that with the increment of aspect ratio (i.e. a/b length ratio) with
increase in skew angle, both Maximum Negative and Maximum positive reaction force is increasing.
527
3.1 Variation of parameters with skew angles for different aspect ratios
Table 1: Variation of parameters for aspect ratio a/b=0.5
skew Max Max negative Max positive

angle deflection Mx My Mxy Reaction Force Reaction Force
(degree) (mm) (Nm) (Nm) (Nm) (N) (N)
0 10.84 513.7 235.2 46.1 -381.5 172.8
15 10.27 492.7 235.3 79.05 -970.8 510.7
30 8.68 432.9 234.7 94.32 -1765 1295
45 6.46 344.6 229.3 102.3 -2562 2113
60 5.26 294.6 222.3 102.6 -2876 2442
75 3.044 194.8 196.2 110.5 -3169 2774
Table 2: Variation of parameters for aspect ratio a/b=1

0 69.56 962.9 962.9 73.23 -1075 281.5
15 66.24 933.1 945.9 162 -3018 1438
30 56.94 846.3 893.7 220.6 -6036 4220
45 43.46 710.6 804.4 234.3 -9592 7658
60 36 628.7 745.5 222.9 -11240 9285
75 21.59 449.2 601.5 242.3 -13460 11510
Table 3: Variation of parameters for aspect ratio a/b=1.5

0 132.1 1000 1630 59.2 -1418 327.7
15 125.6 974.6 1606 167.9 -4098 2568
30 107 880.6 1496 222.8 -7624 6217
528
45 81.18 734.4 1317 226.4 -13330 10600

60 65.79 650.4 1199 232.1 -12780 11800
75 38.52 468.2 932.4 248.3 -14500 13800
Table 4: Variation of parameters for aspect ratio a/b=2

0 173.2 935.5 2053 44.62 -1638 756.3
15 164.1 904.4 2003 99.41 -4437 2666
30 138.7 815.1 1853 139.7 -8205 6434
45 103 680.1 1613 156.2 -12010 10370
60 83.86 601.2 1463 154.2 -13540 12000
75 48.37 435.1 1125 165.2 -15140 13840
3.2 Variation of maximum deflection : When the chart of deflection Vs Skew angle for maximum
deflection is analyzed it is found that with increment in skew angle, deflection is reducing for every aspect
ratio (Figure 3).
200 a/b=2
180
a/b=1.5
160
140 a/b=1.0
Deflection (mm)
120
a/b=0.5
100
*
80
Applicable for all the
Charts
60
40
20
0
0 15 30 45 60 75
Skew angle (degree)
Figure. 3: Variation of maximum deflection with for different aspect ratios
529
3.3 Variation of longitudinal moment Mx and My:When Moment Vs Skew angle chart were analyzed
it can be observed that the aspect ratio of a/b= 2.0 shows the lower moment than the aspect ratio 1.0 and
1.5 respectively (Figure 4 and 5).
1200
1000
800
Moment (Nm)
600
400
200
0
0 15 30 45 60 75
Skew angle (degree)
Figure. 4: Variation of longitudinal moment Mx for different aspect ratios
2500
2000
Moment (Nm)
1500
1000
500
0
0 15 30 45 60 75
Skew angle (degree)
Figure. 5: Variation of longitudinal moment My for different aspect ratios
530
3.4 Variation of torsional moment Mxy, Maximum positive and negative reaction force : When the
torsion Vs skew angle chart were analyzed it has been observed that ab = 1.0 and a/b= 1.5 is alternatively
defeated each other at 0,15,30,45,60,75 degree of skew angles, while a/b=2 is leading in torsion upto
300Nm to a/b=0.5 at every interval of skew angle (Figure 6). While torsion increases to 2000 to 14000, the
a/b= 1.5 and 2.0 defeated each other at a regular interval while a/b=1.0 leads a/b=0.5 throughout. (Figure
7). And it inverse when torsion is in Negative Zone (Figure 8).
300
250
200
Torsion (Nm)
150
100
50
0
0 15 30 45 60 75
Skew angle (degree)
Figure. 6: Variation of torsional moment Mxy for different aspect ratios
16000
14000
12000
Torsion (Nm)
10000
8000
6000
4000
2000
0
0 15 30 45 60 75
Skew angle (degree)
Figure. 7: Variation of maximum positive reaction force for different aspect ratio
531
0
0 15 30 45 60 75
-2000
-4000
Torsion (Nm) -6000
-8000
-10000
-12000
-14000
-16000
Skew angle (degree)
Figure. 8: Variation of maximum negative reaction force for different aspect ratios
3.5 Variation of parameters with thickness

If the thickness of the plate were increased, the moment in all direction as well as the all force were
decreased. In the all the Tables (Table 5-10) and in the chart it can be observed-
Table 5: Variation of different parameters with thickness for skew angle Ф=00
Max Max negative Max positive

Thickness deflection Mx My Mxy Reaction Force Reaction Force
(mm) (mm) (Nm) (Nm) (Nm) (N) (N)
1 292.6 233.89 513.1 10.03 0 226.7
2 86.6 233.9 513.2 11.16 0 189.1
4 20.84 234 513.7 15.51 0 172.8
6 5.217 234.42 514.3 17.72 0 170.2
8 3.36 234.93 514.9 18.74 0 169.3
10 1.12 234.55 515.6 19.21 0 165.8
532

(mm) (mm) (Nm) (Nm) (Nm) (N) (N)
1 250.6 213.79 493.34 21.03 -256.02 436.97
2 76.6 212.9 492.78 22.49 -230.5 418.81
4 15.84 213 493.59 24.34 -200.34 400.28
6 4.217 213.95 494.3 29.27 -187.05 370.82
8 2.96 213.49 493.29 30.47 -157.8 349.36
10 1.03 213.59 493.6 32.12 -146.68 321.59

(mm) (mm) (Nm) (Nm) (Nm) (N) (N)
1 241.6 200.64 458.51 40.15 -416.02 659.54
2 68.6 203.9 456.28 44.61 -409.5 635.1
4 11.84 201 458.47 46.15 -381.5 614.98
6 3.62 201.52 457.93 47.62 -352.2 601.76
8 2.56 202.3 458.19 50.74 -325.8 582.15
10 0.98 200.85 458.06 53.81 -302.8 565.48
533

(mm) (mm) (Nm) (Nm) (Nm) (N) (N)
1 228.6 189.9 427.81 70.86 -569.28 816.7
2 55.6 187.17 426.2 74.28 -548.95 784.41
4 10.24 190.74 426.7 76.98 -532.15 752.98
6 3.12 189.92 427.76 79.2 -500.65 723.62
8 2.16 188.3 428.9 81.4 -487.98 709.13
10 0.91 190.75 426.46 85.56 -479.41 695.28

(mm) (mm) (Nm) (Nm) (Nm) (N) (N)
1 212.6 173.16 395.15 94.32 -800.02 1021.57
2 48.6 175.79 396.2 96.16 -776.4 1000.91
4 9.44 173.79 395.47 100.51 -746.39 972.38
6 2.82 174.52 395.3 107.72 -721.53 945.9
8 1.56 173.93 397.69 110.74 -688.95 926.31
10 0.85 173.5 396.36 119.21 -676.19 897.46
10
534

(mm) (mm) (Nm) (Nm) (Nm) (N) (N)
1 182.6 155.84 359.71 120.6 -1040.54 1325.97
2 38.6 155.9 358.42 127.81 -1016.97 1267.1
4 8.74 154.46 359.37 135.51 -991.25 1239.28
6 2.12 153.2 359.3 137.72 -965.82 1197.42
8 1.24 154.83 357.09 145.49 -942.38 1153.79
10 0.69 154.95 359.66 150.42 -908.68 1105.08
In the figure 9 the deflection Vs Thickness for different skew angle i.e. 0,15,30,45,60 and 75 is presented.
350
300
Deflection (mm)
250
200
150
100
50
0
1 2 4 6 8 10
Thickness (mm)
∅=0 ∅=15 ∅=30 ∅=45 ∅=60 ∅=75
Figure. 9: Variation of maximum deflection with thickness
11
535
In the figure 10 and 11 the Moment Mx, and My Vs Thickness for different skew angle i.e. 0,15,30,45,60
and 75 is presented respectively.
250
200
Moment (Nm)
150
100
50
0
1 2 4 6 8 10
Thickness (mm)
∅=0 ∅=15 ∅=30 ∅=45 ∅=60 ∅=75
Figure. 10: Variation of moment Mx with thickness
600
500
Moment (Nm)
400
300
200
100
0
1 2 4 6 8 10
Thickness (mm)
∅=0 ∅=15 ∅=30 ∅=45 ∅=60 ∅=75
Figure. 11: Variation of moment My with thickness
12
536
In the figure 12, 13 and 14 the torsional Moment Mxy, Positive and negative reaction Vs Thickness for
different skew angle i.e. 0,15,30,45,60 and 75 is presented respectively.
160
140
120
Moment (Nm)
100
80
60
40
20
0
1 2 4 6 8 10
Thickness (mm)
∅=0 ∅=15 ∅=30 ∅=45 ∅=60 ∅=75
Figure. 12: Variation of torsional moment Mxy with thickness
0
1 2 4 6 8 10
-200
-400
Reaction (N)
-600
-800
-1000
-1200
Thickness (mm)
∅=0 ∅=15 ∅=30 ∅=45 ∅=60 ∅=75
Figure. 13: Variation of negative support reaction with thickness
13
537
1400
1200
1000
Reaction (N) 800
600
400
200
0
1 2 4 6 8 10
Thickness (mm)
∅=0 ∅=15 ∅=30 ∅=45 ∅=60 ∅=75
Figure. 14: Variation of positive support reaction with thickness
In the Figure 15 (a, b, c, d, e and f) we can observe the principal stress contours at different skew
angle. Which shows the point of Maximum and minimum stress locations on the plate. Which
can be used at the time of designing of plate at particular skew angle and load.
(a) 00 (b) 150 (c) 300
14
538
(d) 450 (e) 600
(f) 750
Figure. 15: Principal stress contours for different skew angles
4. Conclusion
In the present work, an effort has been made to analyze the behavior of skew plates with material properties
E=2x1011 N/m2, poisons ratio µ=0.3 under uniform pressure of 20kPa. From the results, it is observed that
the maximum deflection at the center of skew plates decreases with increase in skew angle for any fixed
value of aspect ratio and this behavior is consistent for all aspect ratios. The maximum deflection is found
to increase with increase in aspect ratio of the plate. The moments Mx and My at the mid span of the plate
decreases with increase in skew angle for any fixed value of aspect ratio. The torsional moment Mxy
increases with increase in skew angle for a fixed value of aspect ratio and this behavior is consistent for all
aspect ratios. Negative reaction force is found to develop at the obtuse corners of the plate. Both positive
and negative reaction forces are found to increase with increase in skew angle for all values of aspect ratios.
Variation in different parameters with thickness of the plate is also analyzed in the present work. Deflection
at the center was found to decrease rapidly with increase in thickness of the plate. With increase in thickness
of the plate torsional moment Mxy was found to increase marginally but the longitudinal moments Mx and
My were found to remain fairly constant. The reaction forces at the supports was found to decrease with
increase in thickness of the plate.
15
539
References
1. Timoshenko S. and Woinoswksy-kreiger S., “Theory of plates and shells”, Mc Graw hills
Publication, 1959.
2. Das D., Sahoo P. and Saha K., “Large deflection analysis of skew plates under uniformly
distributed load for mixed boundary conditions”, International Journal of Engineering, Science
and Technology Vol. 2 No. 4(2010), pp. 100-112.
3. Girish G.B., Srinivasa C.V., and Suresh Y. J., “Buckling of skew plates subjected to linearly
varying in-plane edge load using finite element method”, Proc. of Int. Conf. on Advances in
Robotic, Mechanical Engineering and Design 2011, DOI: 02.ARMED.2011.01.18.
4. Srinivasa C.V., Suresh Y.J. and Kumar W.P.P., “Buckling studies on laminated composite skew
plates, International Journal of Computer App.lications (0975 – 8887) Volume 37– No.1 (2012),
pp. 35-47.
5. Naghsh A. and Azhari M., “Non-linear free vibration analysis of point supp.orted laminated
composite skew plates”, International Journal of Non-Linear Mechanics 76 (2015), pp. 64–76.
6. Singha M.K. and Daripa R., “Nonlinear vibration of symmetrically laminated composite skew
plates by finite element method”, International Journal of Non-Linear Mechanics 42 (2007), pp.
1144–1152.
7. Dey P. and Singha M.K., “Dynamic stability analysis of composite skew plates subjected to
periodic in-plane load”, Thin-Walled Structures 44 (2006), pp. 937–942.
8. Farag A. M. and Ashour A. S., “Free vibration of orthotropic skew plates”, Journal of Vibration
and Acoustics Vol. 122 (2000), pp. 313-317.
9. Chun P.J. and Lim Y. M., “Analytical Behavior Prediction for Skewed Thick Plates on Elastic
Foundation”, Mathematical Problems in Engineering (2011), Article ID 509724, 19 pages.
10. Lu P. and Shao C., “Simplified analysis of a skew-plate bridge based on grillage analogy model”,
The IES Journal Part A: Civil & Structural Engineering, 5:4 (2012), pp. 253-262, DOI:
10.1080/19373260.2012.695251.
16
540
Paper ID- 113
Experimental Study of the Use of Alccofine as a Corrosion Inhibiter in Steel

Fiber Reinforced Concrete
Saurav1(0000-0002-9015-0285) Anjani Kumar Shukla*2(0000-0003-1449-2539) and Nimisha

Sharma (0000-0002-4843-2721)3
1
Jaypee University of Information Technology, Waknaghat, Solan, India
*2
Indian Institute of Technology (BHU), Varanasi, India
3
APG Shimla University, Shimla, India
2*
Abstract. Rusting is a process in which unstable metals react with atmospheric gases resulting in formation of metal
oxides, sulphides, carbonates and sulphates etc. In civil engineering industry corrosion of steel plays a major
problem due to use of reinforced cement concrete (RCC) in large scale. As a result prevention of corrosion requires
considerable effort and extensive study. In this paper experimental study is carried out to find the effect of alccofine
(ultra fine slag) used as corrosion inhibiters in steel fibre reinforced concrete (SFRC). The entire experiments were
divided into three different categories. The first set of samples corresponds to conventional RCC beams designated
as “RC” of size 100x100x500 mm. Second set was prepared by addition of steel fibre (1% by volume as per
previous literatures) in conventional concrete designated as “SF” in this study and last one was made by partial
replacement of cement with alccofine (12% by volume as per previous literatures) in SFRC beam which is
designated as “ASF” in this study. In these entire samples three different diameters 12mm, 16mm and 20mm of steel
bars were used. Accelerated corrosion method was used to introduce corrosion into these samples and the amount of
corrosion was measured using impressed current technique. Comparative study is done between the actual mass loss
(Mac) of steel bars and theoretical mass loss (Mth). Depth of penetration of corrosions is also computed for different
concrete samples for RC, SF and ASF. From this study it is found that use of alccofine not only increases the
mechanical strength of concrete but also helps in decreasing the corrosion rate when compared to conventional
concrete. It is found that ASF concrete samples shows approximately 2% less corrosion when compared with SF
concrete samples.
Keywords: Impressed Current Technique, Depth of Corrosion, Degree of Induced Corrosion, Steel Fiber
1 Introduction
Corrosion of steel occurs because of electrochemical reaction, when two dissimilar metals are electrically connected
in the presence of moisture and oxygen. This same process takes place in steel alone because of differences in the
electro-chemical potential on the surface, which forms anodic and cathodic regions, connected by the electrolyte in
the form of the salt solution in the hydrated cement. Carbonation is the result of the interaction of carbon dioxide gas
in the atmosphere with the alkaline hydroxides in the concrete. If the concrete is permeable, then carbonation
spreads to the concrete in contact with the steel or soluble chlorides can penetrate up to the reinforcement and if
water and oxygen are present, then corrosion of reinforcement takes place. The pH of concrete pore solution is
normally above 12. In such an alkaline environment, an oxide film is formed on steel surface, which protects steel
from corroding. It is referred to as steel passivation. Both carbonation and chloride ingress because this oxide film to
breakdown [1-2] steel is then depassivated and corrosion process is initiated. Normally concrete provides a good
resistance to corrosion due to its another important characteristic i.e. the high alkalinity of the pore solution, which is
541
comprised of mainly sodium and potassium hydroxides, with a pH ranging from 12.6 to 13.8. At this pH level, a
protective (or passive) film is spontaneously formed during the early stages of cement hydration. This passive film
may grow to a thickness of the order of 10 -3 to 10-1 µm and contains hydrated iron oxides [3]. A corrosion model
proposed by Tuutti [4] can be divided in two distinct time phases: initiation phase and propagation phase (Figure 1).
The initiation phase corresponds to the progressive ingress of the aggressive agents like carbon dioxide CO 2 and
chloride ions Cl- through the concrete cover. The quality of the concrete cover is also involved in the physical
protection of steel from environment, because concrete transport properties control the ingress kinetics of aggressive
agents. During this phase, no corrosion occurs and it usually takes many years for aggressive agents to reach steel
surface and depassivate steel [5]. The laboratory acceleration of corrosion is primarily consists on the acceleration of
the initiation phase so that the depassivation of steel is quickly achieved.
Figure 1: Corrosion model [36]
Two methods frequently used in laboratory for the determination of corrosion are Artificial climate technique and
Impressed current technique. Many researchers have used these techniques to induce the corrosion in reinforced
concrete, and studied the concrete behavior after corrosion, e.g. Ormellese[6]. In this paper impressed current
technique is used to induce corrosion in Concrete samples and the degree of induced corrosion is determined
theoretically using Faraday’s law.
2 Literature review
In order to study the durability of marine concrete structure Ahn [7] used impressed current technique to induce the
corrosion. Galvanostatic technique was also used to accelerate corrosion in reinforcement by impressing anodic
direct current. The beams samples were connected in series with the constant current flowing through all the beams.
Working electrode and counter electrode (stainless steel bars) were mounted in parallel near the concrete surfaces
where bending moment was maximum. Current levels applied during the wet cycle were changed to observe the
change of behavior of the beams at different current levels. Ahmad [8] performed accelerated corrosion test for
inducing reinforcement corrosion through impressed current technique. The positive terminal of the DC power
source was connected to the steel bars (anode) and the negative terminal was connected to the counter electrode
(cathode). The current was impressed from counter electrode to the rebars through concrete with the help of the
electrolyte normally sodium chloride solution. C. Alonsol et al [9] worked on quantifying the relation between the
amount of corrosion and cover cracking. Various parameters were considered during the study including cover to
diameter ratio, proportions of cement, water cement ratio, cast position of the bar, transverse reinforcement and
corrosion rate. The initiation of cracking took place in two steps, generation and propagation. Two different types of
specimens were casted in their study. Specimen 1 was concrete beam of size 15 × 15 × 38 cm with different
diameters of bars with different dimensions of cover. The reinforcements were placed at the center and corner.
Specimen 2 was concrete cubes of 30×30×30cm with and without the transverse reinforcement and the diameter of
reinforcement was 16 mm and that of rebar was 6 or 8 mm. Study on corrosion of reinforcement in concrete and
effect of inhibitor on service life of RCC was studied by V. Kumar et al [10]. The authors did extensive research to
develop models that predicts the time for corrosion initiation which reveals that though Calcium palmitate and its
combination with calcium nitrite decrease the concrete strength but inhibition to the corrosion of the rebar increases
542
the service life of the Reinforced concrete by 8 to 10 times. N. V. Chanh [11] worked on the effect of steel fiber
reinforced concrete in crack resistance and propagation. Due to ability to arrest cracks, fiber composites possess
increased extensibility and tensile strength, both at first crack and at ultimate, particular under flexural loading; and
the fibers are able to hold the matrix together even after extensive cracking. The paper discussed the mechanical
properties, technologies, and applications of SFRC. Fibers do little to upgrade the static compressive strength of
concrete, with increments in strength extending from basic nil to maybe 25%. Indeed, even in individuals that
contain ordinary reinforcement notwithstanding the steel fibers, the fibers have little impact on compressive
strength. In any case, the fibers do generously expand the post-splitting pliability or energy assimilation of the
material, Fibers adjusted toward the ductile pressure may achieve extensive increments in direct tensile strength, as
high as 133% for 5% of smooth, straight steel fibers. In any case, for haphazardly disseminated fibers, the expansion
in strength is considerably small, running from as meager as no increase in a few cases to 60% in others. Analysis on
compressive strength of concrete with partial replacement of cement with alccofine was done by P.R.K
Chakravarthy and R.R. Raj [12]. In this paper presence of Alccofine in the ordinary cement in ideal measurements
can be relied upon to enhance the compressive strength and give protection against chloride ingress. The primary
objective of the work was to find the compressive strength of cement with partial replacement of cement with
Alccofine. Replacement of cement in M25 mix with alccofine in varying percentage was done and compressive
strength was determined at 7 and 28 days.
3 Experimental Investigation
3.1 Material
Ordinary Portland cement of 43 grade from single source is used in this research work. Table 1 shows the different
properties of cement conforming to IS: 8112-1989 [13]. Coarse aggregate with maximum size of 20mm having
specific gravity 2.75 and fineness modulus 6.7 was used in this investigation. Locally available river sand (fine
aggregates) having specific gravity 2.67 and fineness modulus 2.5 was also used. Alccofine 1203 is used as
cementitious material in this work. Amount of alccofine to be used as partial replacement of cement was determined
from previous literature survey which comes out to be 12% by weight of cement [14-15]. Apart from all these basic
raw materials steel fibers were having diameter 0.8mm were cut to a fixed length of 0.35mm. All other properties of
steel fibers are tabulated in Table 2.
Table 1. Physical properties of cement.
Property Result
Specific gravity 3.17
31%
Normal consistency (P):
Setting times
40mins
Initial setting times
466mins
Final setting times
Fineness of cement 5% retained
(By 90 micron sieve)
Soundness of cement 2mm
Compressive Strength
17.1MPa
7 Days
42.1MPa
28 Days
Rectangular moulds of 100mm x 100mm x 500mm (Figure 3.8) were used for casting beams samples and in
accordance with Indian standards. A total of 36 samples were casted out of which 18 were made to corrode and rest
for comparison purposes. Reinforcement of varying diameters (12mm, 16mm and 20mm) of grade Fe500 was used.
The experiment was classified into two sets by varying clear cover of 35mm and 45mm that were further classified
543
into three sets varying on the mix. The first set was a simple reinforced concrete (RC), next reinforced concrete by
addition of steel fiber (SF) varying 1% by volume of concrete mix and the last one by addition of ultra-fine slag by
replacement of cement (ASF) by 20% along with 1% steel fiber addition (Figure 3.9).
Table 2: Properties of steel fiber used
Length 50mm
Diameter 1mm
Appearance Clear and Bright
Tensile strength 800-2500 MPa

Undulated along
Shape
its length
Size 0.8mmx0.35mm
Aspect ratio 43.75 Figure 2:Steel fibers

Rectangular moulds of 100mm x 100mm x 500mm (Figure 3.8) were used for casting of testing samples of beams.
All the specimens were prepared in accordance with Indian Standard Specifications. A total of 36 samples were
casted out of which 18 were made to corrode and rest for comparison purposes. Reinforcement of varying diameter
(12mm, 16mm and 20mm) of grade Fe500 was used. The experiment was classified into two sets by varying clear
cover of 35mm and 45mm that were further classified into three sets varying on the mix. The first set was a simple
reinforced concrete (RC), next reinforced concrete by addition of steel fiber (SF) varying 1% by volume of concrete
mix and the last one by addition of ultra-fine slag by replacement of cement (ASF) by 20% along with 1% steel fiber
addition(Figure 3.9). Different concrete specimen with their designations are tabulated in Table 3
Table 3: Specimen classification
Specimen Diameter Cover

name (mm) (mm)
RC1 12 35
RC2 16 35
RC3 20 35
RC4 12 45
RC5 16 45
RC6 20 45
SF1 12 35
SF2 16 35
SF3 20 35
SF4 12 45
SF5 16 45
SF6 20 45
ASF1 12 35
ASF2 16 35
ASF3 20 35
544
ASF4 12 45
ASF5 16 45
ASF6 20 45
Figure 3(a) shows how a wooden plank of size (100x100) mm was used to form support on the reinforcement
protruding side. Holes in the wooden end allowed a single length of rebar to be suspended at a central location in the
concrete beam. The coarse aggregates and fine aggregates were weighed first with an accuracy of 0.5 grams. OPC
from single source was used in the whole process. Dry fine aggregates were introduced first in the mixer and mixed
thoroughly. After that coarse aggregates were added to it. Then water was added carefully so that no water was loss
during mixing. For addition of steel fiber in the respected samples the steel fiber was distributed randomly during
mixing. Ultra fine slag was added after partial replacement of cement by 12%. All the specimens were left in the
steel mould for the first 24 hours at ambient condition (Figure 3(a)). After that, they were de-moulded with care
upon requirement of aging so that no edges were broken and were placed in the curing tank at the room temperature
for curing (Figure 3(b)).
Figure 3(a) : Casting of samples Figure 3(b): Curing of samples.
3.2 Impressed current Method

The impressed current technique, likewise called the galvanostatic technique, comprises of applying a steady current
from a DC source to the steel inserted in concrete to initiate critical corrosion in a brief timeframe. After applying
the current for a given length, the degree of initiated corrosion can be resolved hypothetically utilizing Faraday's
law, or the level of real measure of steel lost in corrosion can be computed with the assistance of a gravimetric test
led on the removed bars subsequent to subjecting them to quickened corrosion. Utilizing the genuine measure of
steel lost in corrosion, an equal corrosion current density can be resolved. Set-ups utilized for prompting
reinforcement corrosion through impressed current comprise of a DC control source, a counter anode, and an
electrolyte. The positive terminal of the DC control source is associated with the steel bars (anode) and the negative
terminal is associated with the counter anode (cathode) (Figure 4(a)). A current of 0.05 Ampere is inspired from
counter anode to the rebars through concrete with the assistance of the electrolyte (ordinarily sodium chloride
solution 3.5%).The power supplies had a current limit of 500 mill amperes in augmentations of 1 mA. The power
supplies permit utilization of a steady current activity with programmed hybrid, which changes the non-constant
parameters i.e. voltage or power, consequently to make the current constantly steady. These power supplies have a
545
current precision of ±1% at 500 mA full scale. The current intensity was selected in order to achieve the desired
theoretical degree of corrosion of the steel within a certain time frame. Faraday’s law was used to determine the
theoretical mass loss. After curing, the specimen is immersed in 3.5% NaCl solution for 24 hours before testing for
accelerated corrosion. The concrete specimens are immersed in a tank (Figure 4(b)) such that water is below the top
of concrete and not touching the reinforcement. After 24 hours specimens is introduced to accelerated corrosion
process in the same tank by applying constant voltage (10V) until the development of first visible crack on the
surface of specimen. A constant current density of 0.005 Ampere/sq. cm is induced with a built-in Multimeter to
monitor the current. The water solution in the tank was changed after every 5 days. The direction of the current
adjusted so that the reinforcing steel became an anode and the reference electrode as cathode, placed in the tank in
sodium chloride solution that acted as electrolyte
Figure 4(a):DC voltage source Figure 4(b): Impressed current set up
4.1 Calculation of induced corrosion

The mass of rust produced per unit surface area of the bar due to applied current over a given time can be
determined theoretically using the following expression based on Faraday’s law [16]:
𝑊𝐼𝑎𝑝𝑝𝑇
Mth= ……………………. Eq. 1
𝐹
Where, Mth= theoretical mass of rust per unit surface area of the bar (g/cm2), W = equivalent weight of steel, Iapp=
applied current density (Amp/cm2), T = duration of induced corrosion (sec) and F = Faraday’s constant (96487
Amp-sec). The actual mass of rust per unit surface area determined by gravimetric test on rebars extracted from the
concrete by breaking the specimens after the accelerated corrosion test is completed:
𝑊𝑖−𝑊𝑓
Mac= ……………………. Eq. 2
𝜋𝐷𝑙
Where, Mac= actual mass of rust per unit surface area of the bar (g/cm2); Wi= initial weight of the bar before
corrosion (g); Wf= weight after corrosion (g) for a given duration of induced corrosion (T); D = diameter of the rebar
(cm); and L = length of the rebar sample (cm). The degree of induced corrosion is also expressed in terms of the
percentage weight loss (ρ) calculated as [21]:
𝑊𝑖−𝑊𝑓
ρ= ×100……………………… Eq. 3
𝑊𝑖
546
The mass loss calculated by Faraday’s law is directly proportional to the current applied. After the visibility of first
surface crack on specimens, the steel reinforcing bars were retrieved, cleaned of rust, using stiff metal brush and
then weighed to determine the actual mass loss of the steel bars. After removal of the concrete, it was clear that a
concentration of corrosion was present in correspondence to both the crack location and the ends of the concrete
specimens. The reinforcement used was measured before and after corrosion and measured by weighing machine.
Later by using empirical formulas induced corrosion is calculated which varied maximum to 7.6% in terms of
percentage of weight loss (Table 4).
Table 4: Calculation of Degree of induced corrosion
Specimen Dia. Length Initial Final Time (s) Current Loss Degree of
(cm) ofbar weight of weight of (amp/cm2) (gm) induced
(cm) bar(gm) bar (gm) corrosion
Wi Wf (Wi-Wf) (%)
D L T Iapp ρ
RC1 1.2 48 420 389 86400 0.005 31 7.381
RC2 1.6 48 680 658 103680 0.005 22 3.235
RC3 2.0 48 1100 1089 138960 0.005 11 1.000
RC4 1.2 48 420 394 95040 0.005 26 6.190
RC5 1.6 48 680 661 120960 0.005 19 2.794
RC6 2.0 48 1100 1093 129600 0.005 7 0.636
SF1 1.2 48 420 409 972000 0.005 11 2.619
SF2 1.6 48 680 669 1042560 0.005 11 1.618
SF3 2.0 48 1100 1087 1274400 0.005 13 1.182
SF4 1.2 48 420 411 975600 0.005 9 2.143
SF5 1.6 48 680 671 1044000 0.005 9 1.324
SF6 2.0 48 1100 1095 1296000 0.005 5 0.455
ASF1 1.2 48 420 416 993600 0.005 4 0.952
ASF2 1.6 48 680 678 1123200 0.005 2 0.294
ASF3 2.0 48 1100 1098 1296000 0.005 2 0.182
ASF4 1.2 48 420 418 1123200 0.005 2 0.476
ASF5 1.6 48 680 679.5 1728000 0.005 0.5 0.074
ASF6 2.0 48 1100 1099.6 2073600 0.005 0.4 0.036
This actual mass loss (Mac) of steel bars when compared with the theoretical mass loss (Mth) calculated by Faraday’s
law is tabulated in Table 5 and the corresponding graphical representation is shown in Figure 5. It is evident that the
actual value resulted are a bit higher than the theoretical values, which certainly does fulfill the theoretical criteria.
This also resulted because some time is required for depassivation of the steel bar.
547
3.00
Mass Loss (gms) 2.50
2.00
1.50
1.00
0.50
0.00
RC1 RC2 RC3 RC4 RC5 RC6 SF1 SF2 SF3 SF4 SF5 SF6 ASF1 ASF2 ASF3 ASF4 ASF5 ASF6
Concrete Samples
Mac Mth
Figure 5: Comparison of Actual mass loss and Theoretical mass loss
Table 5: Comparison of actual mass loss (Mac) of steel bars with the theoretical massloss (Mth).
Specimen Diameter Mac Mth
(mm) (gm/cm sq.) (gm/cm sq.)
RC1 12 0.171 1.250
RC2 16 0.091 1.500
RC3 20 0.036 2.011
RC4 12 0.144 1.375
RC5 16 0.079 1.750
RC6 20 0.023 1.875
SF1 12 0.061 1.406
SF2 16 0.046 1.508
SF3 20 0.043 1.844
SF4 12 0.050 1.412
SF5 16 0.037 1.511
SF6 20 0.017 1.875
ASF1 12 0.022 1.438
ASF2 16 0.008 1.625
ASF3 20 0.007 1.875
ASF4 12 0.011 1.625
ASF5 16 0.0021 2.500
ASF6 20 0.0013 3.000
It is observed that for various cover depths the mass loss also varied along with variation in the bar diameter. The
maximum mass loss was in the bar with minimum diameter of 12mm and decreasing linearly with decreasing
diameter (Figure 6). In steel fiber reinforced concrete, there is very little mass loss in the bar with the highest
diameter with maximum mass loss in 12mm diameter bar (Figure 7). In terms of cover, there was a slight more mass
loss in 35mm cover samples than the 45mm ones. Time required for the development of first visible crack on the
surface of concrete sample increased slightly as the cover is increased. After the addition of ultrafine slag and steel
548
fiber in concrete mix, the mass loss decreased in the bar with higher diameter as almost there was little mass loss
(Figure 8). The samples with more cover depth showed less corrosion.
0.18
0.16
Mass loss (gm/cm2)
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
Cover(35mm) Cover(45mm)
Dia. (12mm) Dia. (16mm) Dia. (20mm)
Figure 6: Variation of Actual mass (Mac) for different cover depths in RCC beams
0.07
0.06
Mass loss (gm/cm2)
0.05
0.04
0.03
0.02
0.01
0
Figure 7: Variation of Actual mass (Mac) for different cover depths in Steel fiber reinforced
concrete beams.
549
0.025
0.02
Mass loss (gm/cm2)
0.015
0.01
0.005
0
Figure 8: Variation of Actual mass (Mac) for different cover depths in Steel fiber reinforced
concrete along with ultra-fine slag beams
By means of the accelerated corrosion method, the steel bar can be treated to attain a certain level of corrosion. The
degree of corrosion or corrosion level is measured as the loss in weight of the steel bar relative to the weight of unit
length before corroding. To calculate the corrosion depth, the following equation (Eq. 4) is used [17]:
d = s × db/ 4……………………..Eq. 4
Where, d = the corrosion depth (mm); s = Loss of bar/Original weight of bar (grams); db = diameter of steel bar
(mm). The depth of penetration calculated is shown in Table 6. The corresponding pictures in Figure 9 shows the
corrosion depths in. This implies that the trend followed by corrosion penetration has decreased by addition of steel
fiber and ultra fine additive. The bar with larger diameter seems to create less corrosion depth by addition of fiber
and admixtures therefore; a large bar needs a longer duration for producing a similar corrosion level.
Table 6: Depth of Penetration of corrosion in the concrete samples

Loss of
Diameter Initial Final Depth of
bar/Original
Specimen of bar weight of weight of Loss(gm) penetration
weight of
(mm) db bar (Wi) bar (Wf) (mm)
bar (s)
RC1 12 420 390 31 0.074 0.221
RC2 16 680 655 22 0.032 0.129
RC3 20 1100 1086 11 0.010 0.050
RC4 12 420 362 26 0.062 0.186
RC5 16 680 633 19 0.028 0.112
RC6 20 1100 1089 7 0.006 0.032
SF1 12 420 409 11 0.026 0.079
SF2 16 680 669 11 0.016 0.065
SF3 20 1100 1087 13 0.012 0.059
SF4 12 420 411 9 0.021 0.064
SF5 16 680 671 9 0.013 0.053
SF6 20 1100 1095 5 0.005 0.023
ASF1 12 420 416 4 0.010 0.029
550
ASF2 16 680 678 2 0.003 0.012

ASF3 20 1100 1098 2 0.002 0.009
ASF4 12 420 418 2 0.005 0.014
ASF5 16 680 679.5 0.5 0.001 0.003
ASF6 20 1100 1099.6 0.4 0.000 0.002
Figure 9: Corrosion depth of the concrete samples
5 Conclusion
 According to the accelerated corrosion method adopted in this research, large steel bars need longer
duration, compared for producing a similar corrosion level, but tend to exhibits high corrosion depth.
 The variation in clear cover on the sample produced a little effect on corrosion, as it took almost similar
time to induce corrosion in the samples.
 The reinforced concrete samples showed less time to corrode than the ones with extra additives.
 Mix of ultra-fine slag and steel fiber samples of concrete showed approximately 2 percent less corrosion
than the steel fiber reinforced concrete and approximately7.6 percent more than normal reinforced concrete
sample
References
1. Tamer, A., Maaddawy, E., and Khaled, A.S. (2003), “Effectiveness of Impressed Current Technique to Simulate
Corrosion of Steel Reinforcement in Concrete.” Journal of Materials in Civil Engineering, 15(1), 41- 47.
2. Shetty, M.S. (2005). “Concrete Technology-Theory and Practice.” S. Chand and Co, New Delhi.
3. Ramachandran, S., Campbell, S., and Ward, M.B. (2001). “The Interactions and Properties of Corrosion
Inhibitors with Byproduct Layers.” NACE International (57).26-31
4. Tuutti, K. (1982). “Corrosion of steel in concrete.” Swedish Cement and Concrete Research Institute, Stockholm
5. Yingshu, Y., Yongsheng, J., and Shah, S.P. (2007). “Comparison of Two Accelerated Corrosion Techniques for
Concrete Structures.” ACI Structural Journal, (104), 344–347.
6. Ormellese, M., Berra, M., Bolzoni, F., and Pastore, T. (2006).“Corrosion inhibitors for chlorides induced
corrosion in reinforced concrete structures.” Cement and Concrete Research, 36 (3), 536–547.
551
7. Ahn, W. (2001). “Galvanostatic testing for the durability of marine concrete under fatigue loading.” Cement and
Concrete Research, (31), 343–349.
8. Ahmad, S. (2009). “Techniques for inducing accelerated corrosion of steel in concrete.” The Arabian Journal for
Science and Engineering, 95–104.
9. Alonsol, C., Andydel, C., Rodyigtrez, J., and Diex, J. M. (1998). “Factors controlling cracking of concrete
affected by reinforcement corrosion.” Material and structures, 31(1), 435-441.
10. Kumar, V., Singh, R., and Quraishi, M. A. (2013). “A Study on Corrosion of Reinforcement in Concrete and
Effect of Inhibitor on Service Life of RCC.” Journal of Materials and Environmental Science, 4 (5), 726-731.
11. Chanh, N., V. “Steel Fiber Reinforced Concrete.”Research note, Ho Chi Minh City University of Technology,
(108-116).
12. Chakravarthy, P.R. K., and Raj, R.R. (2017). “Analysis on Compressive Strength of Concrete with partial
replacement of Cement with Alccofine.” ARPN Journal of Engineering and Applied Science, 12(8), 2392-2395.
13. IS:8112-1989, “Ordinary Portland Cement, 43 grade — specification”, Bureau of Indian Standards, New Delhi.
14. Saurav, Gupta, A.K. (2018). “Experimental investigation to find the optimum dose of steel fibers in concrete
incorporating ultrafine slag”, Journal of Engineering Science and Technology, 13 (1), 187-195.
15. Saurav, Gupta, A.K. (2014). “Experimental study of strength relationship of concrete cube and concrete cylinder
using ultrafine slag Alccofine”, International Journal of Scientific & Engineering Research 5 (5), 102
16. Ahmad, S. (2009). “Techniques for inducing accelerated corrosion of steel in concrete.” The Arabian Journal for
Science and Engineering, 95–104.
17. Balouch, S.U., Forth, J.P., and Granju, J.L. (2010). “Surface corrosion of steel fibre reinforced concrete.”Cement
and Concrete Research, 40 (1), 410–414.
552
Paper ID- 114
Utilization of Lime in Cement-Based Materials: Review of Fresh,

Mechanical and Durability Properties
Aniket S. Patil1* (https://orcid.org/0000-0002-9743-1274) and Dr. B. Kondraivendhan2 (https://orcid.org/0000-0002-1589-919X)

1
Research Scholar, Applied Mechanics Department, S. V. National Institute of Technology, Surat 395007, India,
E-mail: er.patilaniket@gmail.com, Tel: +91-7768893322
2
Assistant Professor, Applied Mechanics Department, S. V. National Institute of Technology, Surat 395007,
India
E-mail: kondraivendhan78@yahoo.co.in
Abstract
Concrete has versatility, and the relatively low cost is the most popular man-made construction material.
However, this popularity comes at a price: concrete production has an enormous impact on the environment, a fact
which at the past has been significantly overlooked. First of all the production of concrete requires vast amounts of
natural resources each year. Then, the production of each tonne of Portland cement releases almost one tonne of
carbon dioxide into the atmosphere. The reduction of cement content in concrete is one of the persistent global
sustainability concerns of the 21st century. As such, finding an alternative material to cement has become imperative.
A lot of researches have been carried out on the use of industrial wastes as substitute material for different
ingredients of concrete such as coarse aggregate, fine aggregate and cement. Results of various researches indicated
that substitution of an alternative material in concrete not only improves properties of concrete but also lead to
sustainable concrete development which in turn strengthens economy as well as ecology.
Lime is the one amongst all versatile chemicals in the world. Some proofs exhibit the usage of hydrated lime,
and quicklime in construction was extensive as early as 1000 B.C., amongst plenty of civilizations, inclusive of the
Greeks, Egyptians, and Romans. The applications of lime have increased steadily over a period; this may be due to
its ease of procurement, and reasonable cost. This paper contends with the impact of lime on fresh, hardened, and
durability properties of cementitious materials.
Keywords
Concrete, Cement, CO2 Emission, Sustainability, Lime.
1 Introduction
Lime is one of the most versatile chemicals in the world with several chemicals, industrial and environmental
utilities. It is an ivory alkaline matter comprising of CaO (calcium oxide), which is acquired by limestone heating;
also it is employed for the manufacture of Ca(OH)2 (hydrated lime). Because of its diverse geological sources and
the different environment encircling its making, there are numerous distinctions regarding physical and chemical
553
specifics. Some proofs exhibit the usage of hydrated lime, and quicklime in construction was extensive as early as
1000 B.C., amongst plenty civilizations, inclusive of the Greeks, Egyptians, and Romans [1]. Lime has been
practiced as a cementing material worldwide since days of yore. Lime is inexpensively and adequately obtainable in
the environment [2]. The lime cycle is the process through which calcium carbonate (limestone) is heated to convert
it into quicklime, then hydration to transformed into slaked lime, and by carbonation naturally regresses to calcium
carbonate; which is shown in Figure 1. The circumstances exhibit throughout the lime cycle have a substantial
effect on the end product thereby the intricate and diverse physical character of lime products are formed [3].
Figure 1 Lime cycle

Lime is extensively employed in products and uses, each fitted to specific market necessities. Lime and its
derivatives are utilized in steel refining as a fluxing agent and in construction as a binder. The pivotal role of lime
has been found in certain domains of environmental contamination prevention and mitigation; few amongst these
involve treacherous waste treatment, remediation of air emissions and purification of drinking and wastewater [1].
The chief construction-relevant utility of lime is earthen dams, building foundations, and soil stabilization for roads.
Low-quality soils combined with Lime to make a useable base and sub-base. Hydrated lime has been recognized to
be a superlative anti-stripping adjunct for tar pavements. Also, it assists to oppose the corrugation and fracture surge
at miserable temperatures, abate age hardening, and ameliorate the moisture resistance and durability. Latest
research reveals that lime has crucial benefits comparing with other additives for hot mix bitumen roads, comprising
economy and improved durability. Mayhap the ancient and most common usage of lime has been in plaster and
mortar, due to its excellent workability and plasticity [4].
This paper covers the physical, chemical properties of lime, and its types. It contends with the influence of lime
on fresh, hardened, and durability properties of cementitious materials.
2 Types of Lime
The term “lime” is usually wrongly practiced to explain products of limestone. That is a common reason for
confusion. Lime incorporates slaked lime and quicklime, and this term ‘lime’ is equivalent to the term ‘lime
products’ [1].
554
Quicklime is CaO (calcium oxide) made by the decarbonization of CaCO3 (calcium carbonate). Slaked lime is
manufactured through reacting or ‘slaking’ quicklime along with water and comprises Ca(OH) 2 (calcium hydroxide)
[5]. The attributes of different classes of lime are epitomized in Table 1.
Table 1 Lime Nomenclatures [1]
Name Formula Magnesium Hydrated Kiln Temperature Synonyms
Quicklime CaO No No 900-1200° C Calcium oxide, lime, burnt lime
Dolomatic lime, calcium magnesium
Dolime CaO MgO Yes No 900-1200° C
oxide
Sintered
CaO MgO Yes No Up to 1800° C Dead-burned dolime
dolime
Calcium hydroxide, calcium hydrate,
Slaked lime Ca(OH)2 Yes Yes Not Applicable
caustic lime, hydrated lime
Hydrated
CaMg(OH)4 Yes Yes Not Applicable Calcium magnesium tetrahydroxide
dolime
There are varied features of lime products. These elements are persuaded by:
 The existence of magnesium in the crude material, and thereby the product;
 The kiln temperature;
 A hydrating stage.
Various forms of lime are utilized for a wide range of usage. A distinct diversity is made from hydrated,
dolomitic, and calcium limes. Calcium limes are undoubtedly the gigantic class and are provided in hydrated
composes, ground, and lump. Dolomitic limes are too specialized outcomes and are generally delivered in lesser
amounts in a ground, lump, dead-burned and hydrated forms. Hydraulic lime is partly hydrated and comprises
cementitious composites, and is used widely in construction applications. The vastly manufactured product of lime
is the quicklime [1].
3 Properties of Lime
3.1 Chemical Properties
Limestone is a general term for rocks that contain 80% or more of calcium or magnesium carbonates, inclusive
of chalk, marble, marl, and oolite. Further categorization is by composition as argillaceous (clayey), high calcium,
conglomerate, silicious, dolomite, magnesian, and other limestones. Another origins of lime contain seashells, coral,
ankerite, and calcite [3]. The chemical composition of different limes by several authors is given in Table 2.
3.2 Physical Properties
The color of limestone usually mirrors the degrees and characteristics of the impurities present. White deposits
are mostly of superior purity. Iron sulfide and/or carbonaceous material often led to different tints of grey and dark
hues. Red, cream and yellow hues are denotative of manganese and iron. Impurities in marble commonly yield
diversity in colors and patterns. Limestone frequently has an earthy or stale odor, that is reasoned by its
carbonaceous matter content [5]. The physical properties of limestone are presented in Table 3.
555
4 Effect of Lime on Fresh Properties
Fresh properties of Self Consolidated Concretes (SCCs) shown in Table 4 were studied by Ghafoori et al.
During the investigation it was noticed that, to achieve uniform consistency of SCC (by replacing 5, 10, 15, 20, 25,
and 30% of all cementitious materials content with limestone powder), as the amount of limestone powder increases,
the necessary quantity of high range water reducing admixtures (HRWRA) increases. The inspected behavior can be
attributed to the decrease in water quantity to maintain consistent water to cementitious materials ratio, 635 ± 25 mm
slump flow was kept constant for the SCC mixtures by using an appropriate amount of
Table 2 Chemical Composition of Various Limes
Chemical
CaCO3 MgCO3 Fe2O3 Al2O3 SiO2 SO3 MnO Na2O3 K 2O TiO2 P2O5 LOI Ref.
Properties
(Limestone
97.63 0.96 0.13 0.32 0.71 0.13 - - - - - - [6]
Powder)
Slaked Lime 74.2 0.7 0.06 0.08 0.10 0.01 - - - - - - [7]
Natural
Hydraulic 96.00 1.0 0-1.5 0-1.5 1.5 - - - - - - - [8]
Lime
Hydrated
54.3 7.7 1.2 5.4 12.6 - - - - - - - [8]
Lime
Natural
Hydraulic 62.0 1.0 1.22 1.84 5.7 1.0 0.02 0.08 0.49 0.14 0.03 26.0 [9]
Lime
Limestone 53.47 1.02 0.17 0.40 0.55 - - 0.01 0.03 43.13 [10]
Limestone 52.58 0.05 0.38 0.93 2.61 0.12 - - - - - - [11]

Limestone A
54.4 0.62 0.09 0.22 - - - - - - - - [12]
3.11µm
Limestone B
54.5 0.65 0.13 0.24 - - - - - - - - [12]
7.53µm
Limestone C
53.1 0.88 0.15 0.28 - - - - - - - - [12]
18.62µm
Natural
Hydraulic 62 1 3.5 5 21 - - 0.3 0.7 - - - [13]
Lime 5
Slaked Lime 62.33 0.36 0.18 0.31 1.01 0.52 - - 0.03 - 0.01 35.13 [14]
Natural
Hydraulic 62.0 1.0 1.22 1.84 5.70 1.0 0.02 0.08 0.49 0.14 0.03 26.0 [15]
Lime 3.5
Table 3 Physical Properties of Lime [5]

Property Value
Specific Gravity 2.3 to 3.4
Bulk Density 1400 to 1450 kg/m3
Porosity 0.1 to 30%
Hardness 2 to 4 Mohs scale
Apparent density 1.5 to 2.9 g/cm3
Water absorption 0.4 to 20%
556
HRWRA. The J-ring of all SCC was within the limit of 50 mm. Alternate to cementitious materials with limestone
powder ensued in an increase in T50 flow time of the studied mixtures. T50 increased by 9.4, 26.7, 12.9, 7.8, 12.8,
and 15.3%, when limestone powder content increased from 0 to 5, 5 to 10, 10 to 15, 15 to 20, 20 to 25, and 25 to
30%, respectively [6].
Table 4 Fresh properties of SCC mixtures [6]
HRWRA, Slump flow,
Mixture ID w/cm* w/p# T50, s J-ring, mm
kg/m3 mm
Reference 0.45 0.45 0.87 622 1.06 25.0
L5 0.45 0.43 1.03 635 1.16 12.5
L10 0.45 0.40 1.22 622 1.47 26.3
L15 0.45 0.38 1.44 660 1.66 37.5
L20 0.45 0.36 1.55 660 1.79 25.0
L25 0.45 0.34 1.58 648 2.02 37.5
L30 0.45 0.31 1.87 648 2.33 31.3
*: Water-cementitious materials ratio (cement + fly ash).
#: Water-powder ratio (cement + fly ash + limestone powder).
Tsivilis et al. examined the influence of limestone on the workability of concrete. The specimens containing
limestone, even though having greater fineness, show acceptable workability. The slump and flow of the concretes
are shown in Table 5 [10]. El-Alfi et al. reported enhancement of water in achieving consistency of sulfate resistant
cement paste; by adding silica fume and limestone; as the silica fume increases. They observed that the setting time
of cement pastes moderately reduces with limestone content. However, the setting times enhanced as the silica fume
increments [11]. Moon et al. determined the setting times of concrete with 10, 20 and 30% cement replacement by
limestone powder. The initial and final setting times of Limestone Powder concrete is faster than that of control
concrete, regardless of replacement ratios [12].
Table 5 Slump and flow properties of Concrete [10]
Sample Limestone (%) Slump (mm) Flow (mm)
LC1 0 130 460

LC2 10 120 440
LC3 15 120 420
LC4 20 110 420
LC5a 35 110 400
Faria et al. produced lime mortar with 0.05 to 1 wt% of graphene oxide (GO) and observed a significant
decrease in flow. The addition of Graphene Oxide (0.05, 01, 0.5 1% in dispersed and powder form) in Natural
Hydraulic Lime (NHL) mortars does not show a noteworthy effect on the mortar consistency [15]. Billong et al.
observed the similar results by adding slaked lime to metakaolin (20 and 30%) [16]. Silva et al. evaluated the effect
on blended lime-hydraulic lime mortar’s properties containing hydraulic lime. The lower quantity of water was
required to attain satisfactory workability with escalating hydraulic lime content, as shown in Figure 2, because of
the less specific surface of particles of hydraulic lime as compared to aerial lime one [17].
Ismael and Ghanim examined the effect of various percentage of metakaolin, and hydrated lime is as shown in
Table 6 on fresh properties of blended cement. The w/b ratio of normal consistency decreased in mixes with no
hydrated lime, when compared to mixes with 10% hydrated lime. Metakaolin exhibited good soundness properties,
557
despite the addition of unsound material (hydrated lime); however, the mixes still show zero expansions using the
Le Chatelier method, and all results were less than 0.8% according to the autoclave method, as shown in Table 7.
The addition of metakaolin replacement without hydrated lime in mixes B4, B5 and B6 proved that the time of
setting was increased by using metakaolin without hydrated lime when compared to the reference mix BO. The
other mixes B1, B2, and B3 showed different results from the previous mixes, as indicated in Figure 3, owing to the
combined effects of both materials (metakaolin and hydrated lime); these materials played a major role in decreasing
the setting time. They observed the enhancement in workability of mortar by adding hydrated lime and metakaolin,
with the increment of water [18].
Figure 2 Water/binder (W/B) ratio used in the preparation of the mortars [17]
Table 6 Materials percentages of mixes with metakaolin [18]

Mix symbol Portland cement % Gypsum % Lime % Metakaolin %
B0 97 3 0 0
B1 83 3 10 4
B2 79 3 10 8
B3 75 3 10 12
B4 93 3 0 4
B5 89 3 0 8
B6 85 3 0 12
Nunes et al. examined the consequence of nano silica (nS) on the fresh density of lime (L)-pozzolana (P) paste.
The density of LnS was little more than of L. The density of LPnS is notably superior than LP, regardless the
porosity is only faintly low [19]. Granneman et al. evaluated the consequence of crystallization modifiers
additivated lime mortar. The insignificant impact on the flow of the fresh mortar was seen with the modifiers [20].
Barrera et al. investigated the potential utilization of waste graphite powder in hydraulic lime pastes. Because of
the low density of graphite as contrasted with the hydraulic lime and graphite particle’s jagged structure and the
coarse surface; the pastes flow diameter was remarkably decreased with the graphite addition [21]. When was used,
558
Barbhuiya et al. observed decrease in the workability by adding hydrated lime to modify the fly ash contained
specimens [22].
Table 7 Properties of the blend [18]
Setting Time (Min) Soundness Water
Normal
percentage
Sample Consistency
Initial Final mm % according to
(W/C)
flow test %
B0 0.25 237 277 1 0.0612 100
B1 0.287 149 242 0 0.0469 105.4
B2 0.296 165 239 0 0.0671 107.5
B3 0.313 133 248 0 0.0604 110.8
B4 0.271 243 351 0 0.0675 104
B5 0.272 217 336 0 0.0673 104.3
B6 0.273 225 315 0 0.0669 104.5
Bentz et al. observed the little delay in the pastes setting times mixed with silica powder (SA) in comparison of
those obtained by limestone powder (LS), despite the surface area of the SA is 20 % more than LS. This can be
attributed to some carboaluminates formation in the LS pastes and a probably intrinsically superior capability of the
LS grains than the SA concerning the nucleation and increment of cement hydration products, mainly C-S-H. The
Figure 3 Relation between setting time and percentage of metakaolin and hydrated lime [18]
limestone powder shows moderately higher efficiency at accelerating hydration as compared to the silica; it
contributes more advantages in rheological characteristics by diminishing the yield stress and lessening the
consistency factor, whenever employed to substitute cement. Given rising in cement hydration, these fillers
substitute up to 40 % of cement have least affect initial setting times [23].
5 Effect of Lime on Mechanical Properties
Ghafoori et al. explored the impact of limestone powder on mechanical characteristics of SCC. The limestone
powder-containing SCCs produced similar or slightly greater compressive strengths as compared to reference
concrete. These slight improvements, which ranged between 2 and 8%, can be partially assigned to the improved
packing density of the studied SCC mixtures because of the filler effect of limestone powder, which is shown in
559
Figure 4. The 28-day absorption of the studied SCCs having 5, 10, 15, 20, 25, and 30% limestone powder was 7.2,
9.6, 37.9, 43.2, 63.1, and 60.8% lower than that of reference mixture, respectively. The same trend was observed for
90 days. The overall trend decreased in primary capillary absorption coefficient with increased content of limestone
powder replacing a portion of the cementitious materials [6]. Koteng and Chen reported that the lime-pozzolana
paste has good physical properties containing 30% Fly Ash, 10% Silica Fume, and 60% natural hydraulic lime
(NHL) for concrete production [8].
Grilo et al. examined mechanical characteristics of hydraulic lime mortar in varied curing situations. At 28
days the humid curing results in higher compressive strength, as expected in consideration of greater relative
humidity supports a higher degree of hydration. The dynamic modulus of elasticity, compressive and flexural
strengths of mortars aged 28 and 180 days are shown in Table 8. As seen in Table 8 with the increment in
metakaolin, mortars mechanical demeanour is even superior in humid curing, showing the certitude that greater
relative humidity supports hydration as well as pozzolanic reactions. This incorporation also resulted in a clear
improvement regarding decreasing capillary water absorption devoid of fading the drying capability to a deficient
degree. Water absorption curves of all mortars in marine (a), humid (b) and standard (c) curing situation are
represented in Figure 5 [9]. Tsivilis et al. reported that the cement concretes with limestone betoken identical
sorptivity as shown in Table 9 with the pure cement concrete [10].
Mechanical characteristics of limestone and silica fume incorporated sulfate resistant cement pastes with 90
days curing were studied by El-Alfi et al. The increment in compressive strength with curing period for each
specimen was observed. This is due to the increase in the number of hydrated products. The pure cement paste has
reduced compressive strength at 28-90 days.
Figure 4 Compressive strength of SCC mixtures [6]

The 5 % addition of limestone, as well as silica fume, exhibits good mechanical strength at all curing ages [11].
Moon et al. determined the compressive strength of concrete with the substitution of cement by 10, 20 and 30%
limestone powder, and noted that the compressive strengths of concrete specimens containing limestone powder
560
decreased with increasing replacement ratio irrespective of fineness [12]. Morsy et al. studied mechanical properties
of binder mixed with metakaolin and slaked lime in the proportion of 4:1. Anhydrous Gypsum was incorporated as
an activator with 0, 5, 10, 15 and 20% by weight of the binder. The initial increase of 0 to 5% anhydrous gypsum
content shows a fast rise in the flexural and compressive strengths but the reduced strengths for the subsequent
increase in gypsum percentage. The utmost bulk density and the least water absorption was observed at 5% dosage
of anhydrous gypsum [14].
Faria et al. investigated the incorporation of GO in cement composites. The similar flexural strength and the
slightly increased compressive strength for mortars with 0.05 % and 0.1 % of dispersed GO was observed in
comparison with the control mortar. The flexural and compressive strength was somewhat reduced for further
supplements of dispersed GO. The compressive strength marginally increased, on the other hand, the flexural
strength marginally reduced with the addition of powder GO the as shown in Table 10. NHL-GO mortars have
lowered the capillary absorption of water especially for lesser doses of the dispersed GO presented in Figure 6.
Furthermore, with escalating doses of GO, hardened bulk density faintly decreased. Only 0.05% dispersed GO
mortar asserts the computed drying parameters; remaining specimens have drying rates moderately over and a
drying index faintly under the reference one, which is positive [15].
Silva et al. evaluated the influence on the blended lime-hydraulic lime mortar’s properties containing hydraulic
lime. Almost no change was observed in flexural strength but a reasonable increment in compressive strength as the
hydraulic binder content increases [17]. Ismael and Ghanim studied the influence of hydrated lime and metakaolin
on hardened properties of blended cement mortar. The mixes containing 10% lime and metakaolin show good
results for compressive strength is presented in Table 11 as compared to the reference mix [18]. Nunes et al. noted
the effect of Nanosilica particles (nS) as a pozzolanic admixture in lime (L) and to synergistically improve the
pozzolanic reactivity of lime-pozzolana (LP) by incorporating nS to L and LP pastes. Table 12 represents the
mechanical properties of the lime paste specimens. A substantial increase was observed in the compressive strength
of LnS paste, however, no change in the flexural strength. On the other hand, the enhanced flexural strength of LPnS
paste was observed, and the compressive strength of LPnS was improved to a lesser extent. The dynamic modulus of
elasticity was substantially lowered with the addition of nS to L, and to a small magnitude to LP. On the other hand,
LnS shows higher static modulus of elasticity than L. The improved initial mechanical strength for both LnS and
LPnS pastes was observed [19].
Granneman et al. investigated the physical properties of lime mortar added with crystallization modifiers
(sodium ferrocyanide and borax). No Considerable changes have been measured in water absorption, or drying
behavior between the crystallization modifiers added mortars and the control one. The mortars contained with borax
or ferrocyanide has a somewhat high strength than the control mortar. Such dissimilarities may be attributed to the
little changes in water content, and not inevitably of the modifiers mien [20].
561
Table 8 Dynamic modulus of elasticity, flexural and compressive strength (average values) of mortars aged 28 and 180
days [9]
Modulus of Elasticity (MPa) Flexural strength (MPa) Compressive strength (MPa)
Mortar
28 days 180 days 28 days 180 days 28 days 180 days
NHL_M 4142 6746 0.50 1.14 1.19 2.54
NHL_H 5181 7243 0.87 1.27 1.51 2.50
NHL_S 4094 4694 0.52 0.58 1.01 1.14
NHL_10MK_M 5457 4788 0.88 0.88 3.16 4.02
NHL_10MK_H 9185 8285 0.75 1.38 3.75 3.62
NHL_10MK_S 4951 4559 0.84 0.64 4.07 3.76
NHL_20MK_M 8904 5875 1.14 0.88 6.54 5.10
NHL_20MK_H 12786 8828 1.39 1.33 7.10 5.09
NHL_20MK_S 7746 6959 1.11 1.13 6.93 4.53
M- natural marine environment curing at the experimental station
H- laboratory controlled the humid curing
S- laboratory controlled standard curing
10MK- 10% metakaolin
20MK- 20% metakaolin
Table 9 Sorptivity of the tested samples, Charge passed through the concrete specimens (rapid chloride permeability
test), Carbonation depth of the cement mortars [10]
Sorptivity Permeability Carbonation depth (mm)
Sample Charge (C)
(mm/min0.5) class 9 months 12 months
LC1 0.237 6100 High 3 5
LC2 0.238 5800 High 0 0
LC3 0.226 6000 High 0 0
LC4 0.220 6400 High 0 0
LC5 0.224 6600 High 0 0
Table 10 Flexural and compressive strength, and bulk density of mortars [15]
Flexural strength Compressive strength Bulk density
Mortar
(MPa) (MPa) (kg/m3)
Without GO R 0.25 0.56 1809
Dispersed GO 0.05D 0.24 0.59 1791
0.1D 0.25 0.63 1781
0.5D 0.19 0.55 1784
1D 0.22 0.51 1784
GO in powder 0.05P 0.22 0.59 1810
0.5P 0.21 0.57 1791
10
562
Figure 5 Water absorption curves of all mortars in marine (a), humid (b) and standard (c) curing condition [9]
Table 11 Compressive strength of the mixes [18]
Table 12 Main mechanical properties of the lime paste specimens [19]

Flexural strength Compressive Dynamic modulus Static modulus of
Paste
(MPa) strength (MPa) of elasticity (GPa) elasticity (GPa)
L 1.33 2.64 6.69 3.41
LnS 1.33 3.06 3.08 4.60
LP 2.11 4.20 6.48 Not determined
LPnS 2.45 4.41 5.26 Not determined
11
563
Figure 6 Capillary water absorption curves of mortars [15]

Barrera et al. examined the mechanical properties of hydraulic lime pastes additivated with graphite waste
powder as shown in Table 13. At 28 and 90 days cured specimens with graphite content up to 25% rose up to 7.4%
and 10%, in the bulk density respectively. In contrast, for above 25% graphite content, the bulk density curtailed by
10 to 17%, in consequence of the greater water requirement to attain the uniform workability. The bulk density
could be enhanced by adding the plasticizer. With the graphite content up to 25% Sorptivity reduces. In contrast, for
the graphite content above 25%, the increment in the sorptivity of the blend was observed. The flexural strength was
more with the graphite content up to 25% after the age of 28 days. But, this fashion was contradictory for 90 and
360 days of cured samples. At inconstant water to binder ratio and the graphite content was above 25%, the flexural
strength dropped with the rise in graphite content. The compressive strength gained up to 86% up to 25% graphite
content with a 0.60 w/b ratio. But after 28 days of curing a clear contraction of 47% in the compressive strength was
noticed, for further increase in graphite content. Contradictory to the flexural strength, an insignificant rise in the
compressive strength was observed at 90 and 360 days of curing period without plasticizer and with graphite
addition to specimens. Dynamic modulus of elasticity as shown in Figure 7 was nearly consistent for the various
graphite doses after 360 days curing period [21].
Barbhuiya et al. investigated the properties of silica fume, or hydrated lime added to fly ash concrete. The
merger of silica fume and hydrated lime enhanced the compressive strength of fly ash concrete at an early age as
shown in Figure 8 and 9. It also improved the sorptivity of the samples as seen in Figure 10 [22]. Pavia and Aly
examined the impact of secondary cementitious materials on the mechanical characteristics of hydrated lime
mortars. The replacement of lime by supplementary cementitious materials increases the strength with increase in
substitution level [24]. The properties of slaked lime mortar contained with sodium carbonate and metakaolin were
examined by Pascual et al. Improvement in mechanical strength, Shore C hardness, and bulk density was observed
for 10 to 30% increase in lime on an adjusted sample of slaked lime, metakaolin, and sodium carbonate, compared
to control mortar [25].
12
564
Table 13 Mechanical properties of the different lime pastes [21]

Flexural strength (MPa) Compressive strength (MPa)
Sample
Ch_0%g 0.030 0.022 0.887 1.699
Ch_5%g 0.020 0.023 0.915 1.820
Ch_10%g 0.015 0.028 1.123 2.200
Ch_15%g 0.023 0.031 1.350 3.604
Ch_20%g 0.017 0.020 0.648 1.283
Ch_25%g 0.017 0.016 1.307 2.739
Chp_0%g 0.025 0.031 1.686 3.934
Chp_5%g 0.021 0.029 2.660 3.744
Chp_10%g 0.027 0.063 1.610 4.017
Chp_15%g 0.028 0.056 4.064 7.256
Chp_20%g 0.028 0.047 4.105 8.350
Chp_25%g 0.028 0.077 4.600 10.317
Ch: hydraulic lime pastes
Chp: hydraulic lime pastes with a plasticizer
g: graphite powder
Figure 7 Dynamic modulus of elasticity vs. graphite content by percentage (%) [21]
Figure 8 Compressive strength development for Series A (30% replacement) [22]
13
565
Figure 9 Compressive strength development for Series B (50% replacement) [22]
Figure 10 Sorptivity of concrete [22]

Richardson et al. performed a limited amount of testing to examine the effect of lime on High-Volume Fly Ash
Mixtures. The 5% lime improved strength somewhat for the 50% and 70% fly ash mixtures at later ages; at other
ages, there was no improvement or even small decreases. The 10% lime mixture strength generally was slightly
lower than the 5% lime. At the 4% gypsum level, two levels of lime were explored: 5 and 10%. The effect of
gypsum-lime on 1-day strengths of 50 and 70% fly ash contents mixtures was mixed: for the 50% fly ash, gypsum-
lime reduced strengths at both the 5 and 10% levels of lime. However, for the 70% fly ash, the gypsum-lime
increased strengths [26]. Silva et al. prepared mortars with hydrated lime powder, Portland cement, and natural
hydraulic lime. The natural hydraulic lime mortars bestowed more progressive capillarity water absorption. Almost
no change was observed in the strength up to 25% hydraulic binder contents. An increment in the strength for
hydraulic binder from 25% to 75%, a lot more prominent in lime-cement mortars [27].Pavoine et al. studied the
effect of limestone filler (LF) on cement concrete. After 1 day, the control cement concrete exhibited the maximum
compressive strength. LF concretes achieved same compressive strength after 7 days compared with the control
[28].
14
566
Kanellopoulos et al. evaluated the effectual recycling of construction waste materials, such as recycled lime
powder (RLP). At cement substitution with 5% and 8% RLP, the reduction of strength fails to surpass 18%. Up to
20% cement substitution shows a notable contraction of compressive strength. These results suggest that RLP has
the ability to replace the cement at low substitution levels in concretes approved for structural uses. Even though at
greater replacement levels the strength is decreased remarkably, RLP can be consumed as cement alternate in non-
structural concrete members. Replacing 5% cement by RLP shows 14% increment in the sorptivity Table 14.
Further substitution of cement by RLP (8%) shows an extra 13% increment in the sorptivity values [29]. El-Shimy
et al. investigated lime-silica fume pastes employing initial CaO/SiO2 molar ratios of 0.80, 1.0, 1.30, 1.70 and 2.0.
The strength of each paste enhances with the time of hydration for all the mixes investigated. As for hydration
proceeds, increased hydrated calcium silicates are created and lodged in the open pore system of the hardened paste
leading to a continual enhancement in compressive strength of the hardened paste. With the exception of the low-
lime mix made with an initial CaO/SiO2 molar ratio of 0.80, the strength values decrease with increasing initial
CaO/SiO2 molar ratio of the lime-silica fume mixture [30].
Table 14 Sorptivity and RCP values for control and RLP concrete mixtures [29]
28 days
Sample
Sorptivity (mm/min0.5) RCP (Coulombs)
Control 0.087 5181
RLP 5% 0.101 5294
RLP 8% 0.119 5400
6 Effect of Lime on Durability Properties
Ghafoori et al. examined the effects on attributes of SCC comprising limestone powder as a partial substitution
for cementitious materials. The depth of water penetration was reduced considerably through the use of limestone
powder as shown in Figure 11. Rapid chloride penetration of SCC mixtures was improved by the partial substitution
of cementitious materials with limestone powder. As can be seen in Figure 12, the passed charge reduced by 13.7,
4.5, 8.0, 14.6, 3.9, and 5.2%, when limestone powder replacement level increased from 0 to 5, 5 to 10, 10 to 15, 15
to 20, 20 to 25, and 25 to 30%, respectively. For these increases in limestone powder level, the improvements were
17.6, 5.4, 10.6, 10.2, 6.5, and 12.2% for 28 days duration; and 9.8, 3.6, 5.4, 19.0, 1.3, and –1.8% for 90 days
duration. The usage of limestone powder up to 25% led to marginal improvements of 2 to 4% in the rapid chloride
migration at 28 days. It was a larger decrease (approximately 13%) with 30% of limestone powder replacing
cementitious materials. More improvements were observed in rapid chloride migration with the inclusion of
limestone powder for 90 days duration. The results are shown in Figure 13. The 28-day chloride diffusion
coefficients of the studied SCCs increase by the inclusion of limestone powder. This observation was attributed to
the reduction in chloride binding capacity through consumption of C3A by limestone powder. The chloride diffusion
of all studied SCCs was almost similar for 90 days duration [6].
15
567
Figure 11 Water penetration depth of selected SCC mixtures [6]
Figure 12 RCPT results of selected SCC mixtures [6]

The properties of cement concrete and mortar containing limestone are studied by Tsivilis et al. The limestone
contained cement concrete show lesser defiance to freezing and thawing as equated with the control cement
concrete. This certitude possibly ascribed to the higher W/C ratio. The rapid chloride permeability test results are
presented in Table 9. The total electrical charge gone through the experimented samples containing up to 20%
limestone was near 6000 Charge. The concrete containing 35% lime exhibits the high permeability to chloride. All
specimens are graded as high chloride ions permeability, and this is primarily due to the higher water to cement
ratio. The cement specimens containing limestone show a decrement in the corrosion potential, in comparison with
the control samples. It means that the limestone cement contributes to anti-corrosive protection. It can be seen from
Figure 14 the anti-corrosion effect is higher as the limestone content increases. Relating to the carbonation depth,
lime contained cement specimens show no carbonation at 9 to 12 months an exposure time whereas the control
cement specimens had a 3-5 mm carbonation depth as can be seen in Table 9. [10].
16
568
Figure 13 RCMT results of selected SCC mixtures [6]
Figure 14 Corrosion potential vs. exposure time and limestone content [10]
The chloride ions diffusion in cement made with and devoid of limestone powder was investigated by Moon et
al. The diffusion coefficient of concretes with 10% replacement of cement by limestone powder was significantly
smaller than that of concretes with 20 and 30% replacement irrespective of fineness. The concrete specimens
containing limestone powder indicate a clear increase in total passed charge and penetration depth of chloride ions,
compared to control concrete specimen [12]. Silva et al. observed a tendency only to a small reduction in the
permeability to water vapor of the blended mortars as the hydraulic binder content increases [17]. Barbhuiya et al.
investigated the effect of either hydrated lime or silica fume incorporation in fly ash concrete. The air permeability
of silica fume and hydrated lime contained fly ash concrete diminished at 30% fly ash content whereas it stayed
constant at the content of 50% fly ash in comparison with the pure fly ash concretes as shown in Figure 15 [22].
17
569
Figure 15 Air permeability of concrete [22]

Silva et al. noted an increment in cement was accountable for a considerable diminution of the water vapor
permeability of the blended lime-cement mortars; similar results were observed by the blended lime-natural
hydraulic lime mortars as natural hydraulic lime increases [27]. An experimental program was carried by Pavoine et
al. to check the effect of limestone filler on concrete. The use of LF does not yield a significant impact on the
chloride ion penetration concerning that of the use of OPC only. The accelerated corrosion tests on concretes
containing reinforcement exhibit that tested concrete replaced by Limestone filler were rapidly damaged [28].
Kanellopoulos et al. explored that alternating 5% of the cement by RLP show a 2% rise in the rapid chloride
permeability value. Further substitution of cement by RLP (8%) exhibits a total increase of 4% in the rapid chloride
permeability as shown in Table 14 [29].
7 Conclusion
The paper reviews literature related to the effect of addition or replacement of lime in the cement-based
materials. There are various forms of lime which can be effectively used in cement-based materials. Researchers
have used other materials or admixtures such as Metakaolin, Silica Fume, Nano silica, Fly Ash, High Range Water
Reducing Agents, Super plasticizers along with lime. Based on the literature review, the following conclusion has
been drawn:
 Use of Lime accelerates the early-age hydration process.
 The consistency of lime-based cementitious materials can be enhanced by the addition of HRWRA or silica
fume.
 As hydraulic lime content increases, necessary water content decreases to attain satisfactory workability.
Also, it was found that there was a rise in the workability with an introduction of lime and metakaolin, with
the inflation of water.
 The initial and final setting times of lime powder concrete is swifter than setting times of control concrete.
The observed fresh density of lime concrete is notably higher when nano silica was annexed in lime-
pozzolana.
18
570
 Addition of lime increases the compressive strength. Furthermore, it was detected that there was less gain
or rather no change in the flexural strength. Diminution or immutable in Modulus of elasticity was marked.
 The comprehensive fashion was diminishing in water absorption as the limestone powder content replacing
the cementitious materials increases. An adjunct of lime resulted in considerable reduction in the sorptivity.
 The recycled lime powder can be consumed as cement alternate in non-structural concrete members.
 The limestone contained cement concrete show lesser defiance to freezing and thawing in comparison with
the control cement concrete.
 The anti-corrosive performance is improved as the limestone increments. Relating to the carbonation depth,
lime contained materials, carbonation was not observed at 9 to 12 months exposure time.
 Addition of lime resulted in significant reduction in the water penetrability and air permeability.
 It was a larger reduction in rapid chloride migration with limestone powder replacing cementitious
materials.
 Addition of lime results in increased chloride diffusion coefficient and rapid chloride penetration.
It was observed that more work was carried out in Lime mortar as compared to Lime added concrete or pastes.
There are a lot of scopes to study about lime incorporation in pozzolana blended concrete to improve the mechanical
and durability performance of concrete.
8 References
1. Dowling, A., Dwyer, J.O., Adley, C.C.: Lime in the limelight. J. Clean. Prod. 92, 13–22 (2015).
doi:10.1016/j.jclepro.2014.12.047
2. Acharya, P.K., Patro, S.K.: Use of ferrochrome ash ( FCA ) and lime dust in concrete preparation. J. Clean.
Prod. 131, 237–246 (2016). doi:10.1016/j.jclepro.2016.05.042
3. www.wikipedia.org, https://en.wikipedia.org/wiki/Lime_(material)
4. www.lime.org, https://www.lime.org/lime-basics/uses-of-lime/construction/
5. Oates, J.A.H.: Lime and Limestone. Wiley-VCH (1998)
6. Ghafoori, N., Spitek, R., Najimi, M.: Transport Properties of Limestone-Containing Self- Consolidating
Concrete. ACI Mater. J. 114, 527–536 (2017). doi:10.14359/51689673
7. Sébaïbi, Y., Dheilly, R.M., Beaudoin, B., Quéneudec, M.: The effect of various slaked limes on the
microstructure of a lime – cement – sand mortar. Cem. Concr. Res. 36, 971–978 (2006).
doi:10.1016/j.cemconres.2005.12.021
8. Koteng, D.O., Chen, C.: Strength development of lime – pozzolana pastes with silica fume and fly ash.
Constr. Build. Mater. 84, 294–300 (2015). doi:10.1016/j.conbuildmat.2015.03.052
9. Grilo, J., Faria, P., Veiga, R., Silva, A.S., Silva, V., Velosa, A.: New natural hydraulic lime mortars –
Physical and microstructural properties in different curing conditions. Constr. Build. Mater. 54, 378–384
(2014). doi:10.1016/j.conbuildmat.2013.12.078
10. Tsivilis, S., Batis, G., Chaniotakis, E., Grigoriadis, G., Theodossis, D.: Properties and behavior of limestone
cement concrete and mortar. Cem. Concr. Res. 30, 1679–1683 (2000)
19
571
11. E. A. El-Alfi, A. M. RADWAN, S.A.E.-A.: EFFECT OF LIMESTONE FILLERS AND SILICA FUME
POZZOLANA ON THE CHARACTERISTICS OF SULFATE RESISTANT CEMENT PASTES. Ceram. −
Silikáty. 29–33 (2004)
12. Han-Young Moon, Ho-Seop Jung, and J.-P.K.: Diffusion of Chloride ions in Limestone Powder
Concrete.pdf. J. Korea Concr. Inst. 16, 859–865 (2004)
13. Nardi, C. De, Cecchi, A., Ferrara, L., Benedetti, A., Cristofori, D.: Effect of age and level of damage on the
autogenous healing of lime mortars. Compos. Part B. 124, 144–157 (2017).
doi:10.1016/j.compositesb.2017.05.041
14. Morsy, M.S., Almusallam, T.H., Abbas, H.: Mechanical Properties , Phase Composition and Microstructure
of Activated Metakaolin-slaked Lime Binder. KSCE J. Civ. Eng. 21, 863–871 (2017). doi:10.1007/s12205-
016-0667-2
15. Faria, P., Duarte, P., Barbosa, D., Ferreira, I.: New composite of natural hydraulic lime mortar with
graphene oxide. Constr. Build. Mater. 156, 1150–1157 (2017). doi:10.1016/j.conbuildmat.2017.09.072
16. Billong, N., Melo, U.C., Njopwouo, D., Louvet, F., Bonnet, J.P.: Effect of mixture constituents on properties
of slaked lime – metakaolin – sand mortars containing sodium hydroxide. Cem. Concr. Compos. 31, 658–
662 (2009). doi:10.1016/j.cemconcomp.2009.06.001
17. Silva, B.A., Pinto, A.P.F., Gomes, A.: Influence of natural hydraulic lime content on the properties of aerial
lime-based mortars. Constr. Build. Mater. 72, 208–218 (2014). doi:10.1016/j.conbuildmat.2014.09.010
18. Ismael, N.S.: Properties of blended cement using metakaolin and hydrated lime. Adv. Cem. Res. ICE. 27,
321–328 (2015)
19. Nunes, C., Slí, Z., Stefanidou, M.: Microstructure of lime and lime-pozzolana pastes with nanosilica. Cem.
Concr. Res. 83, 152–163 (2016). doi:10.1016/j.cemconres.2016.02.004
20. Granneman, S.J.C., Lubelli, B., Van Hees, R.P.J.: Characterization of lime mortar additivated with
crystallization modifiers. Int. J. Archit. Herit. 00, 1–10 (2018). doi:10.1080/15583058.2017.1422570
21. Barbero-barrera, M.M., Medina, N.F., Guardia-martín, C.: Influence of the addition of waste graphite
powder on the physical and microstructural performance of hydraulic lime pastes. Constr. Build. Mater. 149,
599–611 (2017). doi:10.1016/j.conbuildmat.2017.05.156
22. Barbhuiya, S.A., Gbagbo, J.K., Russell, M.I., Basheer, P.A.M.: Properties of fly ash concrete modified with
hydrated lime and silica fume. Constr. Build. Mater. 23, 3233–3239 (2009).
23. Dale P. Bentz, Chiara F. Ferraris, Scott Z. Jones, Didier Lootens, and F.Z.: Limestone and Silica Powder
Replacements for Cement: Early-Age Performance. Cem. Concr. Compos. 78, 43–56 (2017)
24. Pavía, S., Aly, M.: Influence of aggregate and supplementary cementitious materials on the properties of
hydrated lime ( CL90s ) mortars. Mater. CONSTRUCCIÓN. 66, 1–10 (2016)
25. Arcones-Pascual, G., Hernández-Olivares, F., Sepulcre-Aguilar, A.: Comparative properties of a lime mortar
with different metakaolin and natron additions. Constr. Build. Mater. 114, 747–754 (2016).
20
572
26. Richardson, D.N., Beckemeier, K.W., Volz, J.S.: Effects of Powder Additive on High-Volume Fly Ash
Mixtures. ACI Mater. J. 535–546 (2015). doi:10.14359/51687395
27. Silva, B.A., Pinto, A.P.F., Gomes, A.: Natural hydraulic lime versus cement for blended lime mortars for
restoration works. Constr. Build. Mater. 94, 346–360 (2015). doi:10.1016/j.conbuildmat.2015.06.058
28. Pavoine, A., Harbec, D., Chaussadent, T., Tagnit-hamou, A., Divet, L.: Impact of Alternative Cementitious
Material on Mechanical and Transfer Properties of Concrete. ACI Mater. J. (2014). doi:10.14359/51686828
29. Kanellopoulos, A., Nicolaides, D., Petrou, M.F.: Mechanical and durability properties of concretes
containing recycled lime powder and recycled aggregates. Constr. Build. Mater. 53, 253–259 (2014).
30. E. El-Shimy, S.A. Abo-El-Enein, H. El-Didamony and Osman, T.A.: Physio-Chemical and Thermal
Characteristics of Lime-Silica Fume Pastes. J. Therm. Anal. Calorim. 60, 549–556 (2000)
21
573
Paper ID- 116
Finite Element Analysis of A Beam With Corrugated Web of A Jib Crane
Saurav (0000-0002-9015-0285) Anjani Kumar Shukla* (0000-0003-1449-2539) and Damini Talwar
(0000-0002-7495-9359)
Jaypee University of Information Technology, Waknaghat, Solan, India
*
Indian Institute of Technology (BHU), Varanasi, India
*
Abstract
In this study a Jib Crane Boom is analysed on FEM software ANSYS. Conventional plane web cantilever steel I
beam S10@25.4lb/feet having a span of 2.54m has been analysed to study the buckling behaviour of the beam
under the action of self-weight and vertical load at free end. Models are created on AutoCAD and analysed on
ANSYS. Buckling load is calculated by using a load multiplier. Further a new design approach of using a web
with trapezoidal corrugations has been proposed to study the lateral torsional buckling behavior of the beam
subjected to self-weight and vertical load at free end. A total of 32 models with different corrugation angles, web
thicknesses and web widths are created and analysed to see the effect of variations in corrugations on the buckling
capacity of the beam. From the analysis results, it is also observed that not only web thickness, but also the
corrugation plate length, corrugation angle and the corrugation width influence the buckling capacity of the beam.
Further the corrugated beam is compared to flat beam on the basis of weight and buckling load and it is observed
that corrugated beam is economical and has higher buckling capacity than the flat beam.
Keywords: Jib cranes, lateral torsional buckling, finite element analysis.
1. Introduction
Today’s industry demands versatile, efficient and cost effective equipment while at the same time providing
more flexibility along with significant savings through increased productivity. A jib crane is a type of crane
making use of a mounted arm to lift and move the material. The arm is mounted either perpendicular to or at an
acute angle upwards from a pillar or wall, may rotate along its central axis via a limited arc or possibly a full circle
[1]. A crane is a mechanical lifting device equipped with a winder, wire ropes and sheaves that can be used both
to lift and lower materials and to move them horizontally. Cranes are commonly employed in the transport industry
for the loading and unloading of freight; in the construction industry for the movement of materials; and in the
manufacturing industry for the assembling of heavy equipment’s. It serves a larger area of floor space within its
own travelling restrictions than any other permanent type hoisting arrangement [2]. Jib crane have three degrees
of freedom. They are vertical, radial, and rotary. However, they cannot reach into corners [3]. Lifting capacity of
such cranes may vary from 0.5 ton to 200 ton and outreach from a few meters to 50 meters [4]. Considerable
research studies have been carried out about structural and equivalent stresses in order to provide safety under
static loading and dynamic behavior of cranes. Finite element analysis is a powerful technique originally
developed for numerical solution of complex problems in structural mechanics, and it remains as a method of
choice for complex systems. The basic principal of this numerical method is dividing all the large structure into
small elements having simple shapes. The unknown variables of an element are the displacement values for each
nodal point [5].
574
1.2 Classification of Jib Crane
Cranes may be classified on the basis of the load carrying capacity, the height to which the load is lifted and the
frequency of lifting the loads. For example, the Crane Manufactures Association of America (CMAA) has
classified the cranes as follows [6].
1. Class A1 (standby service)
2. Class A2 (infrequent use)
3. Class B (light service)
4. Class C (moderate service)
5. Class D (heavy duty)
6. Class E (serve duty cycle service)
7. Class F (steel mill)
This research is focus on basically the three common type of jib cranes-
a) Wall-mounted jib
b) Floor-mounted jib
c) Articulating jib
All three jib types can be broken down into specific installation categories and customized to meet your needs.
a) Wall-Mounted Jib Crane: Provide 270° rotation for a circular coverage area. With capacities up to 5 tons,
these systems are not light weight. However, they are not nearly as heavy-duty as a freestanding system. Wall-
mounted jibs require no floor space.
b) Floor Mounted Jib Crane: There are many types of jib cranes under the category of floor-mounted and each
serves its own unique purpose. For instance, a freestanding (also known as a stand-alone) jib crane is foundation-
mounted, which means that it can be installed almost anywhere inside or outside. Freestanding systems offer
higher capacities, longer spans, and 360° and 270° rotations to cover a large circular area within your facility.
These systems are tough and heavy-duty compared to other floor-mounted types, but they are also more expensive
and require a special foundation for proper mounting. Floor-mounted jib cranes are an excellent option whether
they are freestanding or mast mounted depending on the needs of your application.
c) Articulating Jib Crane: These versatile material handling solutions are able to lift and move loads around
corners and columns, reach into machinery, and service virtually any point between the pivot anchor and the far
reach of the boom. They offer multiple installation options, including floor-mounting, wall mounting, ceiling-
mounting, and even bridge-mounting. With multiple installation types and excellent lift flexibility and range of
motion, these systems are a great option for nearly any type of application .
1.3 Stability Analysis
Stability analysis of a jib crane is required to avoid the accidents that can be caused in case of failure of beam of
a jib crane. From studies and investigations, it is observed that the jib crane requires breakdown maintenance most
of the times in a year due to the reasons given below [7]:
 Lateral shift of cantilever I-type beam with respect to axis of mounting and because of that movement of
trolley is restricted.
 Bending at free end stuck the trolley at tip position.
575
 Bending caused damage to the beam.
1.3.1 Lateral Torsional Buckling

• When an applied load causes both lateral displacement and twisting of a member lateral torsion buckling has
occurred.
• Lateral torsional buckling may occur in an unrestrained or laterally unsupported beam.
• A beam is considered to be unrestrained when its compression flange is free to displace laterally and rotate
[8]
Lateral-buckling is a major design aspect of flexural members of thin walled I-girders. When a slender I-girder is
subjected to flexure about its strong axis with insufficient lateral bracing out of plane bending may occur as applied
load approaches its critical value. At this critical value, lateral buckling occurs [9]. However, the structural action
of a beam is predominantly bending, with other effects such as shear, bearing and buckling also being present
[10].
Lateral Torsional Buckling is effective on laterally unrestrained beams which are loaded so as to be under bending
about their major axis. Calculating the smallest load causing LTB of the beam, which is known as critical LTB
load, is a hard problem to solve. Especially for the sections which are under greater warping moments during
torsion such as I sections, developing a closed form solution is not practical. Many studies were conducted in
search of this solution [11]. The use of corrugated webs is potential method to achieve adequate out-of plane
stiffness and shear bulking resistance without using stiffeners. Therefore, further lateral restraints have to be
provided to control lateral-buckling using corrugated web.
1.3.2 Purpose of Corrugations:
To allow the use of thin plates without stiffeners in buildings, bridges construction, crane design, the corrugated
webs are introduced. Reduction in beam weight and cost can be achieved because the usage of larger thickness
and stiffeners gets eliminated [12]. Providing corrugations in the web considerably reduces the cost of beam
fabrication and improves the fatigue life [13]. It also improves the aesthetics of structures. There is less literature
available on application of corrugated web. The results of available studies indicate that the strength of such
girders can be higher as compared to girders with stiffened or un-stiffened web [14]. In present research a web
with trapezoidal corrugations is used. In this research the finite element models of plane web as well as corrugated
webs are developed and analysis is performed on ANSYS.
2. Literature Survey
C. C. Dandavatimath and H.D. Sarode [15] They investigated the bending of a cantilever I section beam of a
jib crane subjected to self weight and load at free end. New designs of web shapes were proposed and analyzed
using FEM software ANSYS to see the effect of geometrical parameters on bending and load carrying capacity
of the beam. Further results of proposed designs were compared with the conventional beam results. It is observed
that with change in web designs load carrying capacity and resistance to bending increases. M. Dhanoosha and
V. Gowtham Reddy [16] In this study, a jib crane was designed analytically and was analyzed on FEM
software.Using analytical design dimensions, models are made on CREO and analyzed on ANSYS. Furthermore,
the stress regions in a floor mounted jib crane were investigated by using 2 different steel i.e. Structural steel and
ASME A36 steel. Results were then compared and it is concluded that ASME A36 steel gives less deformation
and stress values than Structural steel. Fatimah De’nan et al. [17] In this paper a three-dimensional finite element
model using LUSAS 14.3 had been made to study the outcome of the steel beam with triangular web shape for
576
the shear buckling behavior. Beams of different thickness were compared to a normal flat beam. Eigen value
buckling analysis was used in calculating the buckling load of the beam with flat web and that with a triangular
web shape. Results show that the various web thicknesses gave a significant impact on the shear buckling. Also,
the corrugation thickness of the web is also efficient in increasing the shear buckling capacity of the beam.
C. I. Gerdemeli, S. Kurt, B. Tasdemir [18] In this study a Jib Crane is designed analytically as well as using
FEM. Analytical design results and those that were obtained by finite element method have been compared. The
consistency of the finite element method for Jib crane design had been looked for. Notably, it has been observed
that, FEM is the most practical and efficient method which can be used for Jib crane design. According to the
comparison results, it is seen that, the error margins were between the acceptable limits. S.M. Rajmane and A.
Jadhav [19] In this research work, finite element analysis of a column mounted jib crane was done. Software
ANSYS was used for modelling & analysis. During the work, effects of variation of web thickness, web height
on deflection of beam & Von Mises stresses were studied. It is found that with an increase in web thickness,
deflection and Von Mises stress decrease. With an increase in web height, initially the deflection reduces but later
it increases. Von Mises stress also reduces initially, but later increases and becomes constant Fatimah Denan,
M.H. Osman and S. Saad [20] In this study experimental and numerical study on lateral torsional buckling
behaviour of steel section with trapezoid web were performed. In the experimental work, I sections with dimension
200 x 80 mm and 5 m length were loaded vertically, and were laterally unrestrained to allow lateral torsional
buckling to occur. In the analytical study, Eigen-value buckling analysis in the finite element method was
performed to determine the critical buckling load of the steel section.. S.S. Kiranalli and N. U. Patil [21] In this
study analysis of a jib crane was done using FEM software ANSYS. Initially the crane was modelled by a simple
2D element as a beam and column, and results were compared with the analytical solutions to check suitability of
the mesh and element for 3D model analysis.. R. K. Amreeta and V. Singh [22] In this study stress analysis of
single girder wall mounted jib crane having load carrying capacity of 1000 kg and a 2.5 m span length and 180
degrees swing range was done. Amit S. Chaudhary, Subim N. Khan [23] This paper deals with the bending
behavior, shear capacity and lateral torsional buckling of I section cantilever beam of a jib crane tested to an
eccentric point load and self-weight of the beam using Finite element analysis. Different cross sections and
different web shaped cantilever beams were proposed in the study. A corrugated trapezoidal web cantilever beam
was compared to a flat web cantilever beam. Kavita R. Kapadni and S.G. Ganiger [24] This paper examines
the shear stresses, deflection, lateral torsional buckling in a regular I section simply supported gantry crane beam
subjected to a concentrated load at the centre of the beam and the self-weight (Figure 1).
Figure 1: Tapered beam with corrugated web [7]
577
N. S. Khetre, et al. [25] In this paper an I section boom was designed for a rotary jib crane. Using Static stress
analysis and displacement analysis, yield strength and displacement was calculated analytically for I section.
Further the models were made on Solid works and analyzed on COSMOS for stress and displacement and
compared with the analytical results. Both the results agreed well with each other. Jiho Moon et al. [26] In this
study result of the theoretical and finite element analyses of the lateral–torsional buckling of I-girders with
corrugated webs under uniform bending were presented. R. Divahar and P. S. Joanna [27] In this study
experimental study was performed on load carrying capacity of cold-formed steel section with trapezoid web. Six
cold-formed steel beams with plain and trapezoidal webs were tested under two-point loading for its pure flexural
behaviour. J. Lee et al. [28] In this study a one-dimensional finite element model was developed to study the
lateral buckling of a composite I-section. The model developed is capable of predicting lateral buckling loads and
moments for various configurations. The effects of different loading, difference in location of applied load and
the fibre angle of web on buckling loads and moments of composite were studied. From the results it was
concluded that orthotropic closed-form solution is not suitable for knowing about lateral buckling loads because
of the presence of coupling stiffness for a beam under pure bending with off-axis fibre orientation. Gupta
Pratyoosh et al [29] In this study a finite element program was developed to analyze the lateral torsional buckling
of continuous, tapered I beams. This program was used to analyze beams with linear as well as quadratic taper for
different support conditions (simply supported and fixed). The program predicted the buckling loads for web depth
taper, flange width tapers as well as flange thickness taper for continuous I beam where it has been observed that
the maximum buckling load is achieved for flange thickness or flange width taper, and that the load increases
gradually when the web depth is tapered. A study on the effect of increase in number of spans on buckling loads
of non-prismatic beams was also done where results show that the there is an increase in the buckling load from
a single to double span beam, but beyond 2 spans, the change in the bucking capacity is very little.
3. Designing of Crane
The objective of conducting finite element analysis is to predict the lateral torsional buckling behavior of I -section
beam of a jib crane subjected to self-weight and lifted load at the free end. A new design approach of using
trapezoidal web corrugations in place of flat web of beam of the jib crane is proposed to predict the buckling
behaviour of the beam and to check the economy and compare it with the flat web I beam. For the analysis purpose
32 3-D models of I - section beam with corrugated web are generated in AutoCAD and analysis is carried out in
ANSYS Workbench 15.0. Further effect of web corrugations parameters and on buckling capacity, stress, has
been studied-
3.1 Dimension of proposed Crane
Gorbel Crane FS-350 (Insert Mounted) is selected with the following crane parameters [29]:
 Capacity: 500 kg
 Rotation: 360⁰
 Span of Boom: 2.54 m
 Mast diameter 300 mm
Boom Dimensions: From ASTM A6, I beam S10@25.4lb/feet is selected with following section parameters [26]
shown in Table 1.
578
Table 1: Beam Properties

Designation Depth Width of Thickness Flange
(mm) flange of web thickness
(mm) (mm) (mm)
S10×25.4 254 12.5 8 118
3.2 Load Calculations:

Loads acting on the crane beam are calculated as follows [3]:
1 Trolley weight: the weight of the trolley and equipment attached to the trolley.
2 Dead load: the self-weight of the beam and trolley.
3 Hoist load: The working load and the weight of the lifting devices.
4 Vertical Inertia Forces: Dead Load Factor + Hoist Load Factor. According to CMAA [25], Dead load factor
equals to 1.2 and Hoist load factor equals to 0.15.
5 Inertia Forces from drive: The inertia forces occur during acceleration or deceleration or crane motions and
depend on the driving and braking torques. Inertia forces from drives equals to 2.5% of the vertical load.
Load Calculation is shown in Table 3.3. For calculation of load, trolley position is assumed to be at the free end
of the beam.
Table 2: Load Calculations
1. Lifted Load = 500 x 9.81= 4905N
Due to Impact load factor = 0.25x4905 = 1226.25N
Due to Dead load factor = 0.5 x 4905 = 2452.5 N
Total =8583.75 N
=8.58 KN
2. Trolley Load 149.68 x 9.81 =1.46 KN
Due to Dead load 1.2 x 1.46 = 1.752 KN
3. Total Vertical Load= Lifted load + Trolley 8.58 + 1.46 + 1.752

load
= 11.792 KN
4. Inertia forces from drive = 2.5% of Vertical 2.5 x 11.792/100

Load
= 0.2948 KN
5. Total Test Load =11.792 + 0.248
=12.04 KN or 12040 N
579
3.3 I - Section Beam Analysis

Conventional beam of the crane will be first analyzed. For that the modeling is done on AutoCAD and analysis
performed on ANSYS, procedure of which is explained later in this section. Further beams with trapezoidal
corrugations in the web have to be developed in order to study and analyze the buckling behaviour of the beams.
Figure 2 shows the top view of the proposed beam. For this purpose a total of 32 models with the following
parameters are prepared and analyzed.
3.3.1 Parameter selection:
For trapezoidal web following parameters are selected:
i. Web thickness (tcw) – 6, 8 mm
ii. Corrugation angle (θ) – 30⁰, 45⁰, 60⁰, 75⁰
iii. Infill corrugation plate length (b) – 150, 250 mm
iv. Corrugation web width (h) – 25, 35 mm
Figure 2: Top View of proposed web

3.3.2 Modeling
AutoCAD is used to create the models. First a conventional I beam with a plane web is modeled on AutoCAD as
shown in Figure 3(a). Further 32 models (Figure 3(b)) with trapezoidal corrugated web are prepared with
following parameters shown in Table 3.
Table 3: Parameters of corrugated web section of various proposed beams

Model no. Web Thickness Corrugation Angles Web Width (h) Corrugation
(tcw) (ᶿ) mm plate length (b)
Mm mm
1-8 6,8 30⁰,45⁰,60⁰,75⁰ 25 150
9-16 6,8 30⁰,45⁰,60⁰,75⁰ 35
17-24 6,8 30⁰,45⁰,60⁰,75⁰ 25 250
25-32 6,8 30⁰,45⁰,60⁰,75⁰ 35
580
Figure 3(a): Conventional I beam with a plane web Figure 3(b): I beam with trapezoidal corrugations in
web
3.4 Analysis on ANSYS: For analysis ANSYS software is used. The procedure of the analysis is explained
below. The same procedure is followed for flat as well as corrugated beam:
Step 1: Engineering Data
First step of the analysis involves feeding the engineering data into the software ANSYS as shown in Figure 4.
Step 2: Creating the Geometry
For creating geometry models are imported from AutoCAD on ANSYS as shown in Figure 5.
Figure 4: Engineering data fed into ANSYS Figure 5: Geometry imported in ANSYS
Step 3: Generate Mesh

Mesh tool provides a number of options to mesh the object. The mesh can be 2D or 3D depending upon the
requirement of the user. Given areas or volumes can be selected and meshed by proper selection. To determine
the size of the element a convergence study was conducted. Table 4 shows the data obtained after conducting the
convergence study. From Figure 6 it can be seen that convergence is obtained at an element size of 10 mm.
Therefore element size of 10mm is used in the analysis. Figure 7 shows the mesh details in ANSYS.
581
Table 4: Data obtained from convergence study

Mesh Size (mm) Number of elements Maximum
Displacement(mm)
70 5598 7.5194
60 5793 7.5176
50 7406 7.4189
40 7763 7.5174
30 13509 7.5077
20 17015 7.4632
10 30536 7.6598
5 121776 7.6599
Displacement vs Number of
elements
7.7
7.65
7.6
Total 7.55
Deformati
on 7.5
7.45
7.4
0 50000 100000 150000
Number of elements
Figure 6: Convergence study graph Figure 7:Mesh Details
Step 4: Boundary Conditions and Load Application.

Once the system is fully designed, the last task is to burden the system with constraints, such as physical loadings
and boundary conditions. For support conditions left end of the beam is fixed and every movement is restricted in
X, Y, Z directions as shown in Figure 8.For purpose of load application, total lifted load of 12040 N is distributed
in an area of 118 × 100 mm from the free end applied in transverse direction to avoid any stress concentration.
Figure 9 shows the application of load in ANSYS software.
582
Figure 8: Fixed Support at the left end Figure 9: Load Application
STEP 5: Obtain Solution

After load application, the models are then analyzed for buckling load, shear stress, Von Mises Stress and
principal stress.
4. Analysis of ANSYS Software Results
4.1 Finite Element Analysis (FEA) Results
4.1.1 FEA Results of a Flat Beam:
After Finite element analysis the results of buckling load, Maximum Shear Stress and Equivalent Stress have been
tabulated in Table 5. Buckling load is calculated only for first mode by multiplying the load multiplier with the
applied load.
Table 5: FEA results of a flat beam

Buckling Load Maximum Equivalent Stress Maximum Principal Stress
(KN) Shear Stress (MPa) (MPa)
(MPa)
81.0 56.864 106.82 160.27
4.1.2 FEA Results of Corrugated Beams:
After Finite element analysis the results of buckling load, Maximum Shear Stress and Equivalent Stress have been
tabulated in Table.6-9 to see the effects of corrugation web dimensions on structural capacities.
Table 6: FEA results of corrugated beam for Buckling load
Corrugation Web Web Buckling load (KN)

Plate Width Thickness
Length 30⁰ 45⁰ 60⁰ 75⁰
25 6 71.32 84.73 86.48 87.84
583
8 88.25 90.24 92.00 93.82
150 35 6 85.44 87.71 94.48 94.76
8 91.01 93.34 95.95 99.39
25 6 70.22 83.22 83.00 83.83
8 86.57 88.81 89.80 89.85
250 35 6 84.15 84.00 87.49 87.56
8 89.28 90.05 93.45 93.731
Table 7: FEA results of corrugated beam for Maximum Shear Stress
Corrugation Web Web Maximum Shear Stress (MPa)

Length 30⁰ 45⁰ 60⁰ 75⁰
6 64.36 61.32 60.27 60.00
25 8 58.01 57.30 55.59 50.79
150 6 61.31 60.93 55.34 53.00
35 8 56.32 55.79 54.54 49.14
6 64.64 62.31 60.99 60.64
250 25 8 58.07 57.33 56.143 55.82
6 62.18 60.82 60.48 60.00
35 8 57.28 56.64 55.71 54.11
Table 8: FEA results of corrugated beam for Equivalent Stress
Corrugation Web Web Equivalent Stress (MPa)

Length 30⁰ 45⁰ 60⁰ 75⁰
25 6 118.5 113.39 111.10 110.93
8 109.24 107.81 108.25 108.00
150 35 6 108.42 107.57 106.61 105.99
8 107.96 106.81 106.18 105.00
25 6 121.73 115.78 114.66 114.01
8 109.99 109.34 109.27 108.72
250 35 6 116.96 115.32 114.00 113.28
584
8 110.32 109.08 108.35 107.23
Table 9: FEA results of corrugated beam for Maximum Principal Stress
Corrugation Web Web Maximum Principal Stress (MPa)

Length 30⁰ 45⁰ 60⁰ 75⁰
6 157.73 157.00 156.77 156.22`
25 8 156.55 156.43 155.07 155.01
150 6 156.91 156.12 155.00 155.96
35 8 156.42 156.00 154.20 154.82
6 158.06 157.93 157.91 156.99
25 8 157.43 156.63 156.11 156.00
250 6 157.96 157.45 156.65 156.28
35 8 157.34 157.26 156.13 156.10
Further the results have been plotted in the form of graphs (Figure 10-21) to see the effect of web corrugations on
structure capacities.
102
58
100
56
98
54
96
Buckling Shear 52
94 Stress
Load (KN)
(MPa) 50
92
90 48
88 46
86 44
30⁰ 45⁰ 60⁰ 75⁰ 30⁰ 45⁰ 60⁰ 75⁰
Corrugation Angles Corrugation Angles
8 mm thick web 8 mm thick web
Figure 10: Effect of Corrugation Angles on Buckling Figure 11: Effect of Corrugation Angles on Shear
Load Stress
585
108.5 96
108 94
107.5 92
107 90
106.5
88
Equivalent 106 Buckling 86
Stress 105.5 Load (KN)
(MPa) 84
105
82
104.5
80
104
103.5 78
30⁰ 45⁰ 60⁰ 75⁰ 30⁰ 45⁰ 60⁰ 75⁰
Corrugation angles
Corrugation Angles
150mm(Corrugation plate
8 mm thick web length)
250 mm Corrugation plate
length
Figure 12: Effect of Corrugation angles on Equivalent Figure 13: Effect of Corrugation plate length on
Stress Buckling Load
65 124
64 122
120
63
118
62
116
Shear 61 114
Stress Equivalent
60 112
(MPa) Stress
59 (Mpa) 110
108
58
106
57 104
30⁰ 45⁰ 60⁰ 75⁰ 30⁰ 45⁰ 60⁰ 75⁰
Corrugation angles Corrugation angles
150mm(Corrugation plate length) 150mm(Corrugation plate length)
250 mm Corrugation plate length 250 mm Corrugation plate length
Figure 14 :Effect of Corrugation plate length on Shear Figure 15:Effect of Corrugation plate length on
Stress Equivalent Stress
586
100 65
90 64
80
63
70
60 62
Shear
Buckling
50 Stress 61
Load (KN)
40 (MPa)
60
30
59
20
10 58
0 57
30⁰ 45⁰ 60⁰ 75⁰ 30⁰ 45⁰ 60⁰ 75⁰
25mm(Corrugation plate 25mm(Corrugation plate

width) width)
35 mm (Corrugation plate 35 mm (Corrugation plate

width) width)
Figure 16: Effect of Corrugation plate width on Figure 17: Effect of Corrugation plate width on
Buckling Load Shear Stress
110 105
109 100
108
95
Equivalent107 Buckling
Stress 106 Load (KN)90
(Mpa) 105
85
104
80
103
102 75
30⁰ 45⁰ 60⁰ 75⁰ 30⁰ 45⁰ 60⁰ 75⁰
25mm(Corrugation plate width) 6mm web thickness

35 mm (Corrugation plate width) 8mm web thickess
Figure 18 :Effect of Corrugation plate width on Figure 19: Effect of Corrugation plate thickness on
Equivalent Stress Buckling Load
587
70 109
60 108
50
107
Shear 40 Equivalent
Stress Stress 106
(MPa) 30 (Mpa)
105
20
10 104
0 103
30⁰ 45⁰ 60⁰ 75⁰ 30⁰ 45⁰ 60⁰ 75⁰
6mm web thickness 6mm web thickness

8mm web thickess 8mm web thickess
Figure 20: Effect of Corrugation plate thickness on Figure 21:Effect of Corrugation plate thickness on
Shear Stress Equivalent Stress
4.2 Comparison between Flat Beam and corrugated beam:

To choose the best model we have to first compare the flat beam with the 32 corrugated beams on the basis of
weight and buckling load. The model which is to be chosen depends on two conditions.
Condition 1: The weight of the proposed beam should be less than that of the flat beam i.e. 95.348 kg.
Condition 2: The buckling load of the proposed beam should be greater than that of the flat beam i.e. 81 KN.
Table 10: Weight comparison of the proposed corrugated beams.
Corrugation Web Web Weight (kg)
Plate Width Thickness 30⁰ 45⁰ 60⁰ 75⁰
Length
6 87.15 87.78 88.56 89.527
25 8 96.60 97.43 98.46 99.76
150 6 74.93 88.24 89.32 90.56
35 8 96.96 98.05 99.48 101.17
6 86.79 87.225 87.61 88.07
250 25 8 96.11 96.69 97.21 97.21
6 87.02 87.51 88.17 88.82
35 8 96.42 97.07 97.96 98.82
Table 10 shows the weight of the proposed corrugated beams extracted from the ANSYS software. From Table 5
and 6, based on above 2 conditions the corrugated beam is compared to the flat beam and the best model is chosen.
Table 11 shows the results obtained after the comparison of the flat beam and corrugated beam.
588
Table 11: Comparative Results between the flat beam and corrugated beam
Model Weight Buckling Maximum Equivalent Maximum
(Kg) Load(KN) Shear Stress (MPa) Principal Stress
Stress(MPa) (MPa)
Flat Beam 95.48 81.0 56.864 106.82 160.27
Corrugated 89.32 94.48 55.34 106.61 155.00
Beam
After comparing the flat beam to the corrugated beam it can be seen model number 13 having a corrugation plate
length of 150mm, web width of 35mm, web thickness of 6mm and corrugation angle 60⁰ offers higher buckling
load and is more light and economical than the flat beam. Result files of both these models i.e. the flat beam and
best optimized corrugated beam is shown in appendix.
5. CONCLUSION
In this study a new design approach of using a web with trapezoidal corrugations for a beam of a jib crane is
proposed, to study the lateral torsional buckling behavior of the beam subjected to self-weight and lifted load at
free end. After conducting the finite element analysis of the flat beam of the jib crane and comparing with the
proposed corrugated web beams, following points are concluded:
1. Steel I beam with trapezoidal corrugated web offers higher resistance to lateral torsional buckling as
compared to the conventional steel I beam having a flat web.
2. Steel I beam with trapezoidal corrugated web produces less shear, equivalent and principal stresses than the
conventional steel I beam with a flat web.
3. Model number 13 i.e. beam having a corrugation plate length of 150 mm, web width of 35 mm, web thickness
of 6 mm and corrugation angle 60⁰ offers higher buckling load and is more light and economical than the flat
beam.
4. Trapezoidal web thickness, corrugation angle, length of corrugated plate, width of corrugated web influences
the resistance to lateral torsional buckling and shear stress in following ways:
a. Higher trapezoidal web thickness results in higher resistance to lateral torsional buckling.
b. Increasing the length of corrugated plate reduces the resistance to lateral torsional buckling.
c. Increasing the width of corrugated web increases the resistance to lateral torsional buckling.
d. Corrugation angle of 75⁰ results in higher resistance to lateral torsional buckling
REFERENCES
[1] Fatima denan, Keong, C.K., and Hashim, N.S. (2017). “The effects of the depth of web on the bending
behaviour of triangular web profile steel beam section”, American Institute of Physics, AIP Conference
Proceedings 1892,020022, 1-8.
[2] Gerdemeli, I., Kurt, .S. and Tasdemir, B. (2012). “Design and Analysis with Finite Element Method of
Jib Crane”, Mechanical Engineering Istanbul Technical University – Turkey, 565-568.
[3] Limaye, A. A. and Alandkar, P.M. (2013). “Strength of Welded Plate Girder with Corrugated Web
Plate”, International Journal of Engineering Research and Applications, 3(5), 1925-1930.
589
[4] Bollimpelli, K.S. and Kumar, V.R. (2015). “Design and Analysis of Column Mounted JIB Crane”,
International Journal of Research in Aeronautical and Mechanical Engineering, 3(1), 32-52.
[5] Arvind Kumar, V.C. (2015). “Design and Analysis of Beam for Deformation of Floor Mounted Jib
Crane”, M.Tech. Thesis, Gujarat Technical University.
[6] Brettle, M. (2005). “Lateral Torsional Buckling and slenderness’, Technical Paper, New Steel
Construction, 30-34.
[7] Divahar, R. and Joanna, P.S. (2014). “Lateral Buckling Of Cold Formed Steel Beam with Trapezoidal
Corrugated Web”, International Journal Of Civil Engineering And Technology, 5(3), 217-225.
[8] Ozbasaran, H. (2014). “A parametric study on lateral torsional buckling of European IPN and IPE
cantilevers”, International Journal of Civil, Architectural, Structural and Construction Engineering,
8(7), 783-788.
[9] Fatimah De’nan, Mustar, M., Hassan, A.B. and Omar, N. (2013). “Effect of Triangular Web Profile on
the Shear Behaviour of Steel I-Beam, Iranica Journal of Energy & Environment, 4 (3), 219-222.
[10] Suratkar, A., Shukla, V. and Zakiuddin, K.S. (2013). “Design Optimization of Overhead EOT Crane
Box Girder Using Finite Element Analysis”, International Journal of Engineering Research &
Technology, 2(7), 720-724.
[11] Guide lines from user manual of ANSYS 14.0
[12] Gorbel User Manual www.gorbel.com/authorizedusers.asp
[13] ASTM A6: Standard Specification for General Requirements for Rolled Structural Steel Bars, Plates,
Shapes and Sheet Piling.
[14] Crane Manufacturers Association of America Specifications-74
[15] Dandavatimath, C.C. and Sarode, H.D. (2017). “Finite element analysis and Optimization of Jib Crane
Boom”, International journal of innovative research in Science, Engineering and Technology, 6(7),
14287-14294.
[16] Dhanusha, M. and Reddy, V.G. (2016). “Detail Design and Analysis of A Free Standing I Beam Jib
Crane’’, International Research Journal of Engineering and Technology, 3(12), 193-203.
[17] Fatimah De’nan and Hashim, N.S. (2011). “The effect of web corrugation angle on bending performance
of triangular web profile steel beam section”, International Journal of Environmental Protection, 1(5),
53-56.
[18] Gerdemeli, I., Kurt, S. and Alkan, H.O. (2010). “Main Girder Beam Design and Finite Element Analysis
of 2x160 Ton Gantry Crane”, 14th International Research/Expert Conference Trends in the
Development of Machinery and Associated Technology, Mediterranean Cruise, 565-568.
[19] Rajmane, S.M. and Jadhav, A. (2015). “Finite element analysis of jib crane’’, International Journal of
Innovative Research in Technology, 2(6), 404-407.
[20] Osman, M.H. and Saad, S. (2010). “The Study of Lateral Torsional Buckling Behaviour of Beam With
Trapezoid Web Steel Section By Experimental And Finite Element Analysis”, International Journal of
Recent Research and Applied Studies, 2 (3), 232-240.
[21] Kiranalli, S.S. and Patil, N.U. (2015). “Jib Crane Analysis using FEM”, International Journal for
Scientific Research and Development, 3(4), 185-189.
590
[22] Amreeta, R.K. and Singh, V. (2015). “Design and stress analysis of single girder jib crane’’,
International Journal of Engineering Research and Technology, 4(9), 932-936.
[23] Chaudhary, A.S. and Khan, S.N. (2015).”Design and analysis of varying cross sectional cantilever beam
with trapezoidal web for Jib crane”, International Engineering Research Journal, Special Issue 2, 4475-
4480.
[24] Kavita, Kapadni, K.R. and Ganiger, S.G. (2015). “Review paper on design and structural analysis of
simply supported gantry crane beam for eccentric loading”, International Research Journal of
Engineering and Technology, 2(8), 1622-1626.
[25] Khetre, N.S., Bankar, P.S. and Meshram, A.M. (2014). “Design and Static Analysis of I section Boom
for Rotary Jib Crane”, International Journal of Engineering Research and Technology, 3(8), 1071-1074.
[26] Moon, J., Yi, J.W., Choi, B.H., Lee, H.E. (2009). “Lateral Torsional Buckling of I girder with corrugated
webs under uniform bending”, Thin Walled structures, Elsevier, 47(1), 21-30.
[27] Divahar, R. and Joanna, P.S. (2014). “Lateral Buckling Of Cold Formed Steel Beam with Trapezoidal
Corrugated Web”, International Journal Of Civil Engineering And Technology, 5(3), 217-225.
[28] Lee, J., Kim, S.E. and Hong, K. (2002). “Lateral buckling of I section composite beams”, Engineering
Structures 24, Science Direct, Elsevier, 955–964.
[29] Gupta Pratyoosh., Wang, S.T. and Blandford, G.E. (1995). “Lateral Torsional buckling of Non Prismatic
I Beams”, Journal of Structural Engineering, 122(7), 748-755.
591
Paper ID- 118
Behaviour of Liquid Storage Tank Under Multi-

Directional Excitation.
Sourabh Vern1,
Mahendra Kumar Shrimali2, Shiv Dayal Bharti3, and Tushar Kanti Datta3
1 Research Scholar, National Centre for Disaster Mitigation and Management, MNIT Jaipur,
sourabh.vern@gmail.com
2 Professor, National Centre for Disaster Mitigation and Management, MNIT Jaipur,
shrimalimk@ gmail.com
3 Professor, National Centre for Disaster Mitigation and Management, MNIT Jaipur,
sdbharti@gmail.com
4 Emeritus Professor, Department of Civil Engineering, IIT Delhi,
tushar_k_datta@ yahoo.com
Abstract. By revoking the effect of vertical component in the analysis different

response quantities of LST may differ from the true response of structure during
a seismic event. This paper aims to study the effect of vertical component of the
earthquake on various response quantities namely tank wall displacement, sur-
charge at the free surface of the water, overturning moment and base shear. The
study is compiled with the help of Nonlinear time history analysis in an explicit
finite element module on ABAQUS platform. For illustration, a 10m long rec-
tangular liquid storage tank has been modelled by using a solid element with fluid
media modified by the arbitrary Lagrangian and Eulerian (ALE). Investigation
for response is done first with bidirectional and then with the vertical component
is considered. Some notable conclusions of the study include vertical component
alone, increases the responses to an order of about 30%.
1 Introduction
An important civil structure must be able to keep functioning or should be partially able
to work even after an earthquake hazard because the complete collapse in any of life-
supporting structure can be very damaging for recovery of a nation from the disaster.
One of such important structure is Liquid Storage Tanks: there omnipresent application
and implementation rivets extra attention than that of normal load-bearing structures.
There are several historical evidence that shows that failures of LST are proven deadly,
and their repercussions can last from a few months to more than a decade. Some of such
examples in which LST are witnessed to collapsed are 1933 Long Beach Earthquake,
California, USA,1960 Chile Earthquake, Chile, 1964 the great Alaska earthquake,
Alaska, USA,1978, Miyagi Prefecture offshore earthquake, Japan and 2003, Tokachi
offshore earthquake, Japan. As most of these tanks are used to store liquefied gases and
592
fluids which have a lower flash point a simple spark developed by contact between tank
roof and wall can set a full surface fire in the tank.
Authors in the past have taken different approaches to tackle analysis of this problem.
[1] represented more defined and exact analyses in which pressure can be subdivided
into two major parts, Convective and Impulsive. Impulsive pressure is exerted by the
inertial response of the fluid which is because of the inertial response of the tank walls.
Whereas the convective response is produced due to the oscillation of the liquid. [2] the
author presented a simplified procedure which has been prescribed by API standard
650. Additional information for measuring the sloshing height, the impact forces due
to the sloshing on the columns of the roof is given along with the change in the magni-
tude of the hoop tension due to the earthquake ground motion. the shift in the design
approach is observed from the Housner's rigid tanks in which pseudo-acceleration's
spectral value is taken instead of maximum ground acceleration.
[3] Taken up the fluid-structure interaction inside a liquid storage tank by finite element
method with emphasis on dynamic and buckling analyses. Lagrangian-Eulerian kine-
matical description for modeling fluid subdomains in fluid-structure interaction prob-
lems, the finite rotation effects in the numerical integration of constitutive rate equa-
tions arising in large deformation analysis and the implicit-explicit finite element tech-
niques for transient analysis. Disarray is reported while extracting response for high
Reynolds number flows and complex shell buckling. [4] explained the design criteria
for the simplified cylindrical ground support liquid storage tank. While examining the
design criteria, impulsive and convective part of the liquid in flexible steel or concrete
tanks fixed to rigid foundations is considered. [5] extracted the sloshing natural periods
and their modal pressure distributions by the influence of nonlinear wave theory for
two-dimensional behavior rectangular tanks. [6] studied the paralleled SPH codes are
programmed to study the liquid sloshing in both two-dimensional and three-dimen-
sional tanks of single and multi-degrees of freedom. [7] a dynamic response analysis of
vertically excited liquid storage tanks including both liquid-tank and liquid-soil inter-
action. The ground excited tank-liquid-soil system is transformed by the generalized-
coordinate approach’ I and then it is analyzed by the complex frequency response
method.
[8] Evaluate dynamic response of an elastic circular cylindrical tank having a rigid base
under a vertical excitation taking into consideration the interaction with the foundation
soil, a more representative solution for the problem in the frequency domain is obtained
where the soil is appropriately modeled by frequency-dependent parameters. The ef-
fects of vertical excitations have shown that significant hoop stresses can be developed
in the walls of liquid storage tanks, and for intense motion, these cannot be ignored in
comparison with the hydrostatic hoop stresses used in the design of such tanks. [9]
Presented a technique for evaluating the dynamic response of an upright circular cylin-
drical liquid storage tank to a vertical component of ground shaking, considering the
flexibility of the supporting medium. It is shown that soil-structure interaction reduces
the hydrodynamic effects and that the consequences of such interaction may be approx-
imated with good accuracy by a change in the natural frequency of the tank-liquid sys-
tem and by an increase in damping. [10] Investigated the method for analyzing the
earthquake response of elastic, cylindrical liquid storage tanks under vertical excita-
tions. tank response under the simultaneous action of both vertical and lateral excita-
tions is calculated to evaluate the relative importance of the vertical component of
593
ground acceleration on the overall seismic behavior of liquid storage tanks. It should
be noted that vertical excitation is important in the seismic analysis of reinforced con-
crete tanks since these structures are more susceptible to the increase in hoop stresses.
[11] conducted the study to determine the response of concrete rectangular liquid stor-
age tanks subjected to vertical ground acceleration is investigated. The maximum re-
sponse due to vertical acceleration can be as high as 45% of that due to the horizontal
component. It is concluded that the effect of the vertical component of ground motion
should be considered in the analysis of rectangular tanks for liquid containing struc-
tures. This is especially of significance for the near-field zones which require further
investigation. [12] developed the finite-element method is used to investigate the seis-
mic behavior of rectangular liquid tanks in two- dimensional space. The method can
consider both impulsive and convective responses of the liquid-tank system. Also, ap-
plying the vertical excitation will lead to an increase in the convective response of the
system. However, it does not affect the impulsive behavior significantly. This increase
is more noticeable in tall tank mode. [13] reported a baseline of the percentage increase
when switching to the vertical response period and an understanding of how each of
the different code’s provisions can impact the design forces. As other codes move to
adopt the design vertical response spectrum the provisions used to predict the demand
should also be reviewed for accuracy.
For controlling the responses of the liquid storage tank, there are several techniques
that are being employed in the tanks. Common approach witnessed by several authors
is to construct various types of obstruction inside the tank or along the wall to prevent
or completely stop the sloshing of the inertial mass of the fluid due to earthquake ground
motion. [14] surveyed the Miles' method for determining the damping produced by ring
baffles in cylindrical tanks was conducted. O'Neill's modification of Miles' equation
which eliminates free surface wave height from this equation is determined. [15] The
potential of baffles in increasing the hydrodynamic damping of sloshing in circular-
cylindrical storage tanks is investigated, the ability of baffles in reducing the sloshing
effects in storage tanks that are especially broader than fuel containers were under ques-
tion. [16] A numerical model has been developed to study three-dimensional (3D) liq-
uid sloshing in a tank with baffles. The numerical model solves the spatially averaged
Navier–Stokes equations. [17] systematic numerical simulations are carried out to in-
vestigate the sloshing dynamics of liquid in a storage tank, subjected to seismic excita-
tion. To suppress the free surface fluctuations and the associated slosh force, two types
of baffles viz., ring and vertical baffle are examined. Spectral analysis of free surface
displacement and temporal variation of pressure demonstrate dominant contribution
from the fundamental sloshing mode.
Another method which seems to have high reliability and a longer lifespan is the isola-
tion of the structure. The isolation of superstructure can be developed by use of a
passive device commonly referred to as base-isolator at the base. There are several base
isolation techniques, but due to their large area of applications only two stands out. First
being the lead rubber bearing (LRB) and the second is the frictional pendulum system
(FPS) which is a sliding type of isolator. [18] A parametric study is conducted to study
the effects of important system parameters on the effectiveness of seismic isolation of
the liquid storage tanks. It has been found that the bi-directional interaction of frictional
forces has noticeable effects and if these effects are ignored, then the sliding base dis-
placements will be underestimated which can be crucial from the design point of view.
594
[19] The seismic response of liquid storage tanks isolated with variable friction pendu-
lum system is investigated under six recorded near-fault ground motions. [20] To im-
prove the safety of concrete RLSS under earthquake action, an energy dissipation
method of sliding isolation and limiting devices for concrete RLSS is proposed, the
dynamic responses of sliding isolation concrete RLSS under bidirectional earthquake
are studied.
The present study aims to investigate the effect of the tri-directional earthquake motion
on a liquid storage tank and compare the response quantity of interest with that of the
bi-directional earthquake. Thus, by concluding the change in the response quantities by
the vertical component of the earthquake. The effect of vertical component is studied
for two strong ground motion and one little weaker ground motion, to conclude the
vertical component will result in an increase of response quantities in both cases.
2 Methodology
The analysis of the liquid storage tank under the earthquake motion involves different
physical phenomena. In the sloshing motion, the major part of the interaction is the
motion of the container. Apart from the interaction between the two different states
namely, liquid and solid, there is the interaction of the fluid media with itself which
causes the ripple effect which in effect increases the sloshing height.
For a nonlinear solution, a method for analysis required for extracting response quanti-
ties must be able to handle the non-linearity effect along with fluid-structure interaction
(FSI). Thus, a robust finite element application method is adopted for long duration
nonlinear time history analysis.
The non-linearity is solved in ABAQUS by the help of explicit central-difference inte-
gration rule. The simplicity in solving the problem doesn’t provide the necessary com-
putational efficiency which is involved with the explicit dynamic procedure. The input
of the diagonal mass matrix at initial of the time step increment are resulted from the
equation. (1)
N 𝐽 𝐽
𝑢̈ (ⅈ) = (𝑀𝑁𝐽 )−1 (𝑃(ⅈ) − 𝐼(ⅈ) ) (1)
Here MNJ is the mass matrix, PJ is the load vector, and IJ is the force vector. As given
in the equation (2) the inverse of the lumped mass matrix is first solved, the number of
iterations requires for the solution is equal to the number of degrees of freedom which
is obtained by the multiplication between the inverse of the mass matrix and the inertial
force.
1 1 𝛥𝑡 (ⅈ+1) + 𝛥𝑡 (ⅈ) (ⅈ) (2)
𝑢̇ (ⅈ+𝑧) = 𝑢̇ (ⅈ−𝑧) + 𝑢̈
2
1
(3)
𝑢̇ (ⅈ+1) = 𝑢̇ (ⅈ) + 𝛥𝑡 (ⅈ+1) 𝑢̇ (ⅈ+2)
The equation of motion for the solution is integrated using the central difference inte-
gration rule as stated in equation (2), where 𝑢̇ and 𝑢̈ are the velocity and acceleration
595
values respectively. The superscript (i) indicates the incremental series number and (i-
1/2) and (i+1/2) refers as mia d incremental series number.
2.1 Modelling of Liquid Storage Tank
The finite element modeling for the liquid storage tank is defined. The analysis needs
to be compatible with the fluid part of the storage tank which also accounts for the fluid-
structure interaction. The tank is modeled with a solid element which is an eight noded
linear brick element with reduced integration and hourglass control (C3D8R). The sec-
ond part of the assembly is the fluid that is modeled with the solid element and the
combined hourglass control, which smoothens the analysis as fluid is imparted to a
higher level of distortion. The fluid element is refined with the powerful feature, i.e.
Arbitrary Lagrangian and Eulerian (ALE). The tank taken in the analysis is flexible so
that the inertial effect of the fluid media due to the ground motion can be measured. To
extract result of higher accuracy, a mesh convergence is taken up, and convergence is
reported at a mesh size of 0.05m for the fluid media and 0.1m for the tank elements.
To stimulate an Earthquake ground motion in a liquid storage tank earthquake time
history in the form of acceleration time history is given and applied at the base of the
tank. A single dynamic explicit step is defined for the earthquake motion, the value of
period is taken differently for respective ground motion. While considering the effect
of the union of two different media the need of an interaction property is necessary. A
surface to surface contact is defined between the fluid and inner tank surface. The
advantage of defining the surface to surface contact is that ,it takes into the
consideration of the sharp edges which develops an issue of geometry cumbersome that
prolongs the computational time.
For the comparative study, the ground motion is applied in the base concerning the
X, Y and Z axis. The explicit procedure involves the use of many small-time incre-
ments. It uses the central difference operator which is conditionally stable. The value
of stable time increment is given by equation 4. The presence of high-frequency oscil-
lations increases the stable time increment thus, to mitigate the effect of these oscilla-
tions of minuscule order of damping is added into the operator.
2 (4)
𝛥𝑡 ≤ (√1 + 𝜉 2 − 𝜉)
𝜔𝑚𝑎𝑥
During the analysis progress, a global estimation program in the explicit module finds
out the maximum frequency which is the maximum of the whole system. In the nonlin-
ear problems, the frequency of the system is subjected to changes which in turn changes
the stability limit.
The modeling of the fluid media is taken care by the aid of Mie-Gruneisen equation
of state in Abaqus explicit module. The equation of state provides a media characteristic
of hydrodynamics with respect of volumetric strength. Its ability to calculate the pres-
sure as a function of the mass density and the specific energy makes it a robust tool in
dealing with the fluid media.
The Mie-Gruneisen Equation of State represented in first-order form of polynomial
equation (5).
596
𝑝 − 𝑝𝐻 = 𝛤𝜌 (𝐸𝑚 − 𝐸𝐻 ) (5)
Here, 𝑝𝐻 and 𝐸𝐻 are Hugoniot pressure and specific energy as a function of density,
whereas 𝛤 is known as Gruneisen ratio and is equals to the value given in equation (6)
𝜌0
𝛤 = 𝛤𝑜 (6)
𝜌
The above equation is then simplified into the equation (7)
𝛤𝑜 𝜂
𝑝 = 𝑝𝐻 (1 − ) + 𝛤𝑜 𝜌0 𝐸𝑚 (7)
2
To fit the Hugoniot equation in a linear form, a common Hugoniot equation then
becomes as equation (8)
𝜌0 𝑐20 𝜂 (8)
𝑝𝐻 =
(1 − 𝑠𝜂) 2
In the equation (9), 𝑐0 and s represents the single order relationship between the linear
shock velocity, Us and particle velocity, Up in the form of
𝑈𝑠 = 𝑐0 + 𝑠𝑈𝑝 (9)
By considering the above equation of state Us -Up Hugoniot form is presented in the
equation (10) given below.
𝜌0 𝑐20 𝜂 𝛤𝑜 𝜂 (10)
𝑝𝐻 = (1 − ) + 𝛤𝑜 𝜌0 𝐸𝑚
(1 − 𝑠𝜂) 2
2
The ALE is an adaptive meshing tool which gives the advantage to maintain the integ-
rity of the original mesh. A result of the unaffected topology of the fluid media is
observed which is allowed due to the independent motion of the mesh nodes, therefore
even the during the high scale deformation, the analysis doesn’t lose its stability and a
steady flow of analysis is maintained. The fluid mass is given the adaptive mesh prop-
erties in the present problems. The movement of the mesh covers the same domain of
material in the direction which is normal to the motion of the material boundary. Adap-
tive meshing proved to be pragmatic in this analysis as the fluid media is expected to
be deformed severely which leads to termination of the simulations. Not only the ALE
posses the property to solve the dramatic change in the material mesh but also it pro-
vides results of faster and higher accuracy as compared to the pure Lagrangian ap-
proach. Due to the plethora amount of the element distortion and convolution of the
mesh, which induces a loss in computational precision and augments the size of the
stable time step.
597
3 Numerical Study
For the present study, a flexible rectangular tank is taken with ground support. To study
the vertical effect of earthquake component in the liquid storage tank, different earth-
quakes are taken into the consideration with different peak ground acceleration
(PGA)values. The tank is taken as a steel tank with a thickness of 0.2m and with a
cross-sectional size of 10m by 10m. The steel structure is provided with 3% damping.
The height of the fluid stored in the tank is of 7.5m. The sundry properties of the tank
and fluid media are given in the Table.1
Table 1. Material properties for the tank and Fluid media.
STEEL FLUID
Modulus of Elasticity, Density,
Es = 200GPa ρw=983.204Kg/m3
Density, ρs=7900Kg/m3 Equation of state:
c0= 1450, s =0, γ0=0
Poisson’s ratio, υ=0.3 Dynamic Viscosity
= 0.001 N-sec/m2
Table 2. Earthquake record data taken for the analysis.
Name of Earth- Recording Time In- PGA in x- PGA in y- PGA in z-

quakes Station terval direction direction direction
(sec)
Bhuj Earth- IIT-R 35 0.69g 0.64g 0.6g
quake (2001)
Tabas Earth- Iran 40 0.41g 0.35g 0.32g
quake (1978)
Taiwan Earth- Hualien 35 0.195g 0.2g 0.17g
quake (1986)
To develop a lucid conclusive study for the effects of vertical components of

earthquakes, various response quantities are extracted. The response of the liquid stor-
age tank is represented with respect to the period of the earthquakes. The paramount
study of the interaction between the fluid and tank structure surface with first only the
bi-directional horizontal components and then after with tri-directional interactions
along with vertical components of various ground motions are taken. The impact of
material flexibility in tank walls in the analysis of the present study is included while
extracting the responses. The responses which are taken for the comparative study are
the top board displacement of the tank wall, sloshing height of the fluid media, base
shear, overturning moment of the liquid storage tank, shear stress and hoop stress in the
tank wall. The development of the time histories of the various response quantities are
first extracted and are then are put are in a juxtaposition. The sloshing height of the
fluid is extracted at the corner of the tanks. The top board displacement of the tank wall
598
is computed at the right side of the wall in the y-direction. Various stresses which are
shear stress and hoop stresses are calculated for the tank wall. To get a profound detail
of the tank overall stability and behaviour under the given interactions both the shear
force and overturning moment of the tank at the base are provided.
1(a) 1(b)
1(c) 1(d)
1(e) 1(f)
Fig. 1. Various Response Quantities for the Bhuj Earthquake, (a) Sloshing Height (b) Top Board
Displacement (c) Overturning Moment (d) Base Shear (e) Shear Stress and (f) Hoop Stress
It can be observed from the fig 1(a) for the bi-directional interaction alone that motion
of the fluid mass tends to increase first as the time history of the earthquake is proceed.
The motion of the sloshing height in the tri-directional scenario can be observed some-
what same, but as the analysis is progressed towards the ends, it can be seen that mag-
nitude of the sloshing height is more than that of the bi-directional interaction.
599
The top board displacement is discovered for the Bhuj earthquake in figure 1(b) to fold
the relative pattern in both the bi-directional and tri-direction interaction.it can be seen
from the graph that the peak values here is by the tri-directional interaction of the
ground motion. In the case of the tank structural stability, the time history of the over-
turning moment and base shear shows the peak values at the relatively the nascent stage
of the time history. A lucid bifurcation can be marked between the peak values of the
tri-directional and bi-directional earthquake interaction. The evidence in the literature
review supports the response quantities for the present study are in synchronization, as
the vertical component of the ground motion has weakened the stable phase of the liquid
storage tank. By examining the stresses time history of the tank wall, it can be clearly
justified that due the third component of the ground motion a hike in the sloshing height
is observed which in turn impact the tank walls and that results in augmentation in the
shear stress and hoop stress as described by the fig 1(e) and 1(f).
For the case of the Tabas Earthquake, one of the highlighted properties is that the PGA
level of the ground motion is considerably lower than that of the Bhuj ground motion.
The response quantities seem to be dormant by a change in PGA level and display the
same trend as that of the higher PGA ground motion. The various response quantities
show an unbalance in the peak values. The absolute peak values of the Tri-directional
interaction seem to be the dominant one. Although the tri-directional interaction leads
the peak values at almost every increment, in case of the top board displacement of the
tank wall the difference is small. The effect of the light vertical contribution in the tri-
directional response can be seen for the overturning and base shear time history plot.
The peak values the stability variables, i.e., overturning moment, and base shear are
achieved between the initial and mid-interval of their time history as shown in fig.2 (c)
and 2(d). The peak values of these response quantities are observed at the somewhat
same location as that of the sloshing height.it can be evidently because of the unbal-
anced inertial force that is generated by the sloshing mass. Thus, to mitigate this unbal-
anced mass an energy ameliorating obstruction need are demanded. While determining
the shear and hoop stresses in the tank wall, the response displayed peak values at a
prolonged time. The values are found near the latter stage of the time history. The peak
values are of order twice of that are observed at the initial passage of the time. A sig-
nificant change in time history values is seen before the half interval of the hoop stress
time history as given fig. 2(e) and 2(f). The response graph reaches a maximum low
value. The order of the difference between both interactions is seen of the highest order
here.
The results obtained by the Taiwan ground motion holds the trends, as observed in the
above couple of ground motion. In Figure 3(a) the time history response of the sloshing
in the fluid media is plotted. The maximum value in the sloshing height is marked by
the tridirectional interaction. The PGA level taken in this ground motion analysis is the
lowest. The value of the sloshing height in the initial stage of the program has developed
a crescent look in the tri-directional part. After that on somewhat every peak developed
in the response are shadowed by the tri-directional response. The value uncovered in
the top board displacement of the tank wall seems to be the quite small it may be be-
cause of the elastic strength of the steel. The top board displacement is achieving its
zenith value at the nearly same position when the sloshing in the tank is at its highest
coordinate. The values in the stability parameters follow the quite a similar path as
shown in the Fig.3 (c) and 3 (d). The increase in these values suggests that to get a
600
realistic value while describing the behavior of the liquid storage tank under earthquake
ground motion, we cannot ignore the vertical component. The nature of the stresses in
the tank due to the unbalanced mass of the fluid media is a complex phenomenon driv-
ing from the combined effects of the fluid-structure interaction and the inertial behavior
of the fluid media. Here also the maximum values are achieved at the relatively the
same time interval as in the various time responses reported for this ground motion.
(a) (b)
(c) (d)
(e) (f)
Fig. 2. Various Response Quantities for the Tabas Earthquake, (a) Sloshing Height (b) Top Board
Displacement (c) Overturning Moment (d) Base Shear (e) Shear Stress and (f) Hoop Stress
601
(a) (b)
(c) (d)
(e) (f)
Fig. 3. Various Response Quantities for the Taiwan Earthquake, (a) Sloshing Height (b) Top
Board Displacement (c) Overturning Moment (d) Base Shear (e) Shear Stress and (f) Hoop Stress
From table 3, the maximum absolute values of the various output extracted are shown.
For the present study is concluded with the help of the three different earthquakes hav-
ing different peak ground acceleration magnitude. To comprehend the change in the
output values for the various interaction for the ground motion under consideration a
graphical representation is done as shown in figure 4. It can be seen that a considerable
amount of the imbalance is present between the values of bi-directional and tri-direc-
tional interaction for the various values responses.
602
Table 3. Peak values of the various quantities taken under study represented along with their
respective ground motion.
Name of Type of Slosh- Top Over- Base Shear Hoop
Earthquakes Interac- ing Board turning Shear Stress Stress
tion Height Displace- Moment (N) (Pa) (Pa)
(m) ment (m) (N-m)
Bhuj Earth- Bi-Direc- 0.853 0.014 1.61e07 6.09e06 1.00e05 2.080e05
quake (2001) tion
Tri-Di- 0.980 0.0159 1.81e07 6.35e06 2.25 e05 3.37e05
rection
Tabas Earth- Bi-Direc- 0.36 0.011 1.62e07 4.40e06 8.084e4 2.66e05
quake (1978) tion
Tri-Di- 0.41 0.010 1.75e07 4.71e06 1.680e5 3.26e05
rection
Taiwan Bi-Direc- 0.083 0.0028 4.92e06 1.05e06 2.55e04 6.24e04
Earthquake tion
(1986)
Tri-Di- 0.090 0.0030 5.26e06 1.75e06 3.65e4 7.78e4
rection
Wave Height(m) Top Borad Displacement (m)

Overturning Moment(N-m) Base Shear (N)
Shear Stress (Pa) Hoop Stress (Pa)
PERCENTAGE CHANGE
140
120
100
80
60
40
20
0
BHUJ EQ TABAS EQ TAIWAN EQ
Fig. 4. Percentage change recorded between the tri-directional and bi-directional interactions for
the various extracted response outputs.
603
Conclusions
The behavior of the liquid storage tank resting on the ground is taken up for the analysis
in a finite element modeling software ABAQUS. A rectangular tank having both sides
of equal length is taken as a model with fluid capacity at 75 percent. The nonlinear time
history analysis is done for the conclusive remarks for the impact of the vertical accel-
eration in the contribution for the various structural failures during the ground motions.
Three types of ground motion are taken up, with different PGA values. The following
conclusions can be drawn from the numerical study:
1. The pattern of the tri-directional ground motion follows bi-directional interaction,
but in every case, the peak values of tri-directional interaction seem to surpass that of
the bi-directional interaction.
2. The small-scale change can be seen in the response case of top board displacement.
3. The sloshing response seems to hold the most changed response in graphical form
for the tri-directional interaction than that of the bidirectional interaction.
4. The values of the peak are generally developing when the height of the sloshing fluid
is at maximum. Thus, it gives the physical sense of various stresses that are resulted
due to the complex FSI and increase in the inertial force of fluid media during seismic
activity.
5. The dramatic increase can be seen for the shear stress due to the tri-directional inter-
action case. The value shows to increase with higher PGA levels and lowers with lesser
PGA levels. The hoop stress holds a relatively stable change in the values a maximum
percentage change is reported for the Bhuj earthquake.
References
1. Housner GW. Dynamic pressures on accelerated fluid containers. Bulletin of the

Seismological Society of America [Internet]. 1957;47(1):15–35. Available from:
http://www.bssaonline.org/cgi/content/abstract/47/1/15
2. Wozniak RS, Mitchell W. Basis of Seismic Design Provisions for Welded Steel Oil
Storage Tanks. In: Advances in Storage Tank Design, American Petroleum Institute,
Washington, USA. 1978.
3. Liu WK. Finite element procedures for fluid-structure interactions and application to
liquid storage tanks. Nuclear Engineering and Design. 1981;65(2):221–38.
4. Malhotra PK, Wenk T, Wieland M. Simple Procedure for Seismic Analysis of Liq-
uid-Storage Tanks. Structural Engineering International [Internet]. 2000;10(3):197–
201.Available from: https://www.tandfonline.com/doi/full/10.2749/101686600780
481509
5. Virella JC, Prato CA, Godoy LA. Linear and nonlinear 2D finite element analysis of
sloshing modes and pressures in rectangular tanks subject to horizontal harmonic
motions. Journal of Sound and Vibration. 2008;312(3):442–60.
6. Cao XY, Ming FR, Zhang AM. Sloshing in a rectangular tank based on SPH simu-
lation. Applied Ocean Research [Internet]. 2014; 47:241–54. Available from:
http://dx.doi.org/10.1016/j.apor.2014.06.006
7. Fischer FD, Seeber R. Dynamic response of vertically excited liquid storage tanks
604
considering liquid-soil interaction. 1988;16(September 1986):329–42.

8. Haroun MA, Abdei-hafiz EA. A simplified seismic analysis of rigid base liquid stor-
age tanks under vertical excitation with soil- structure interaction. 1986;5(4):217–
25.
9. Veletsos BAS, Asce M, Tang Y. Dynamics of vertically excited liquid storage tanks.
1986;112(6):1228–46.
10. Haroun MA, Tayel MA. Response of tanks to vertical seismic excitations.
1985;13(August 1984):583–95.
11. Kianoush MR, Chen JZ. Effect of vertical acceleration on response of concrete rec-
tangular liquid storage tanks. Engineering Structures. 2006;28(5):704–15.
12. Ghaemmaghami AR, Kianoush MR. Effect of Wall Flexibility on Dynamic Re-
sponse of Concrete Rectangular Liquid Storage Tanks under Horizontal and Vertical
Ground Motions. Journal of Structural Engineering [Internet]. 2010;136(4):441–51.
Available from: http://ascelibrary.org/doi/10.1061/%28ASCE%29ST. 1943-
541X000 0123
13. Morris J, Almanzar L, Chu R. Seismic Analysis of Ground-Supported Tanks Using
the Vertical Response Spectrum. Geotechnical and Structural Engineering Congress
2016. 2016;1404–13.
14. G. P. Stricklin, Baird JA. A survey of Ring baffle damping in Cylindrical Tanks.
1966;38.
15. Maleki A, Ziyaeifar M. Sloshing damping in cylindrical liquid storage tanks with
baffles. Journal of Sound and Vibration. 2008;311(1–2):372–85.
16. Liu D, Lin P. Three-dimensional liquid sloshing in a tank with baffles. Ocean En-
gineering. 2009;36(2):202–12.
17. Sanapala VS, Velusamy K, Patnaik BSV. CFD simulations on the dynamics of liq-
uid sloshing and its control in a storage tank for spent fuel applications. Annals of
Nuclear Energy [Internet]. 2016; 94:494–509. Available from:
http://dx.doi.org/10.1016/j.anucene.2016.04.018
18. Shrimali MK, Jangid RS. Seismic response of liquid storage tanks isolated by slid-
ing bearings. Engineering Structures. 2002;24(7):909–21.
19. Panchal VR, Jangid RS. Variable friction pendulum system for seismic isolation of
liquid storage tanks. Nuclear Engineering and Design. 2008;238(6):1304–15.
20. Jing W, Cheng X, Shi W. Dynamic Responses of Sliding Isolation Concrete Rec-
tangular Liquid Storage Structure with Limiting Devices Under Bidirectional Earth-
quake Actions. Arabian Journal for Science and Engineering. 2018;43(4):1911–24.
605
Paper ID- 120
Effect of Nitric Acid on Metakaolin Concrete
Anjali Singh1, Rakesh Kumar2, P.K.Mehta3, Raghvendra Agrahari4
Department of Civil Engineering, MNNIT Allahabad India
ABSTRACT
Concrete is an artificial construction material which is used in different types of constructions

worldwide. Cement is one of the binding materials used in the concrete. Huge amount of energy
is required during the production of cement. It is reported that in the production of one ton of
cement approximately one ton of carbon dioxide is emitted in the environment. So, it is desirable
to reduced green house gas emission from the cement industry. It is now established that some
pozzolanic materials may be used to partially replace cement without compromising the strength
and durability of concrete. Metakaolin (MK) is one such pozzolanic material which is obtained
from the clay mineral kaolinite. The addition of Metakaolin increases the mechanical properties
of hardened concrete.
This paper presents the results of an experimental programme to investigate and compare the
durability of Normal concrete (NC) and Metakaolin concrete (MKC). M25 grade of concrete
using normal constituents was designed and used as NC. The Ordinary Portland Cement (OPC)
was partially replaced with Metakaolin, upto15% by mass, at an interval of 2.5%. The addition
of Metakaolin increases the water demand and thus the workability reduces as the replacement
level increases. The optimum replacement level was 7.5% with reference to the compressive
strength. The cubes of size 100 mm were cured in tap water as well as in 5% Nitric acid solution
till 90 days. The compressive strength of concrete cubes was determined after 7, 28, 56, and 90
days curing.
Key words: Referral / Normal Concrete (NC), Metakaolin concrete (MKC), Compressive
strength, Nitric acid.
606
1. INTRODUCTION
Concrete is the second most consumed per capita material on the earth after water. It is an
artificial construction material, used for different construction activities worldwide. It is a
mixture of binding/cementitious material, fine aggregates, coarse aggregates and water in
required proportions. Mineral or chemical admixtures are sometimes used in concrete to enhance
the properties of fresh concrete, mixing process and hardened concrete properties. Cement is the
primary ingredient in concrete, which is produced artificially. It is the need of time to reduce the
use of OPC in construction industry due to its higher cost and higher global carbon dioxide
emissions. There is a need of time to search for such an alternative materials which may be used
in place of cement. These may be used as partially replace the cement without compromising the
strength and durability of the concrete. Researchers have indentified some pozzolanic materials
which may be used to partially replace cement and contribute to the development of strength and
durability of concrete. Fly ash (FA), ground granulated blast furnace slag (GGBS), rice husk ash
(RHA), metakaolin (MK), silica fume (SF) are some of the pozzolanic materials which can be
used for partial replacement of cement in concrete making.
Gruber et al [1] reported that the replacement of 8% and 12% of OPC with highly reactive MK
lowered the chloride ion diffusion coefficient by 50% and 60%, respectively, in comparison to
NC. Ding and Li [2] reported that MK offered better workability and compressive strength in
comparison to the NC. The incorporation of MK in concrete result in increase in strength,
especially during the early age of curing [3]. Khatib and Clay [4] reported that water absorption
by capillary action was reduced when up to 20% cement was replaced with MK in concrete.
Parande, et al. [5] reported that improvement in mechanical properties of concrete was obtained
when 15% of OPC was replaced by MK. Siddique and Kadri [6] studied the effect of foundry
cement and MK on initial surface absorption, sorptivity, water absorption and compressive
strength at the ages of 35, 56, and 84 days. It is reported that the initial surface absorption
decreased, when MK content was increased from 5 to 15%. The sorptivity was observed to
decrease till 10% OPC was replaced by MK. However, increase in sorptivity was observed at
15% replacement level. It is also reported that all mixes showed low water absorption
characteristics and compressive strength and an inverse relation with sorptivity was observed.
Dinakar, et al. [7] reported that the optimum replacement level of OPC by MK was 10% with
607
respect to compressive strength in high strength concrete. A slight reduction in the densities of
MK mixed concretes was observed due to lower specific gravity of Metakaolin compared to
cement alone. Pavlik [8] studied the effect of acetic acid and nitric acids on the corrosion of
hardened cement paste. It is reported that the corroded layer was soft, porous and had visible
cracks. The corner of the corroded layer formed was white in nitric acid solutions with
concentrations ranging between 0.025 and 0.5mol/litre. Olusola and Joshua [9] reported a
significant reduction in compressive strength with increased acid concentration, immersion
period and laterite content in concrete.
In this paper, the relative performance of NC and MKC, when exposed to the acidic environment
(5% HNO3) is compared. The MKC was prepared using supplementary cementitious material
MK. NC of grade M25 was prepared for reference.
2. EXPERIMENTAL INVESTIGATION
Materials and Their properties
The 43 grade OPC, Brand Jaypee was used for this experimental investigation. The physical
properties of OPC was determined as per IS 8112-1989 [10].The normal consistency and specific
gravity of OPC were found to be 31% and 3.15, respectively. The initial and final setting times
were 92 and 240 minutes, respectively. The compressive strength of cement was 52.50 N/mm2.
The chemical composition of OPC is presented in Table-1. The coarse aggregates (CA), of sizes
10 mm and 20 mm conforming to IS: 383-1987 [11], were used. The specific gravity of 10 and
20 mm aggregates were 2.68 and 2.54, respectively. The water absorption of 10 mm and 20 mm
aggregates were 1.2% and 1.36%, respectively. The fineness modulii of 20 and 10 mm
aggregates were 7.27 and 6.25, respectively. The fine aggregate (FA) (river sand), conforming to
Zone II, IS: 383-1987 [11], was use in this experimental work. Its fineness modulus, specific
gravity and bulk density values were found to be 2.49, 2.48 and 1680 kg/m3 respectively. The
MK (M/S Karanwal Infra, Faridabad) of grey colour specific gravity 2.45 and having specific
surface area 16300 cm2/gm was used. The chemical composition of MK is included in Table 1.
The nitric acid used for creating the acidic environment was of brand RENKEM and was
procured from civil lines, Allahabad, its specific gravity and normality were 1.41 and 72%,
respectively.
608
Table 1: Chemical Composition of Cement and Metakaolin
Chemical Composition OPC Metakaolin
Silicon Di oxide 20.40% 62.62%

Aluminum Oxide 6.20% 28.63%
Iron oxide 3.09% 1.07%
Calcium oxide 63.90% 0.006%
Magnesium oxide 1.50% 0.15%
Sodium oxide 0.50% 1.57%
Potassium oxide 0.49% 3.46%
Source: Manufacturer’s literature
3. EXPERMENTAL METHODOLOGY
Mix proportioning
The concrete mix design (M25) was carried out as per the procedure given in IS: 10262 (2009)
[12] and mix proportion was OPC: FA: CA:: 1:1.31:2.46 with water content of 186 liter/m3. The
water/ cement ratio was 0.40. The OPC was replaced in part with MK at different levels (2.5, 5,
7.5, 10, 12.5, and 15% on equal mass basis) to get the optimum replacement level with respect to
compressive strength. The cubes (150 mm) of MKC and NC were cast to determine the
compressive strength after at 7 and 28 days. Thereafter cubes of 100 mm size were cast for
further study. These were submerged in tap water as well as in 5% HNO3 solution for curing
periods of 7, 28, 56 and 90 days.
4. Results and Discussions
4.1. Workability of MKC:

The effect of MK inclusion on the workability of concrete is shown in Fig.1. The
workability of concrete mixes is found to decrease with the increase in MK content was
found to decrease by 39 mm when the replacement level was changed from 0% to 15%.
This decrease in slump may be due to more fineness of MK.
609
100
90
80
70
slump (mm)
60
50
40
30
20
10
0
2.5 5 7.5 10 12.5 15
Metakaolin (%)
Fig. 1: Variation in Workability of Concrete at Different replacement levels
4.2 Optimum dose of MK inclusion

The variation of compressive strength at 7 and 28 days, with different percentages of MK in
concrete are presented in Fig. 2. In this study, the replacement level of OPC with Metakaolin was
varied from 0 to 15%, at an interval of 2.5%. It is observed from Fig.2 that compressive strength
increases with MK content up to 15%, and after that it decreases. The optimum replacement
level of OPC by MK is 15% with respect to compressive strength. The maximum gain in
compressive strength at 7 and 28 days is 20 and 11%, respectively. This gain in strength is
primarily due to the formation of additional C-S-H gel in MKC.
610
45
7 Day Compressive Strength
40
Cmpressive Strength (MPa) 28 Day Compressive Strength
35
30
25
20
15
0 2.5 5 7.5 10 12.5 15
OPC replaced by MK (%)
Fig. 2: Variation of 7 and 28 Day Compressive Strength at Different replacement levels.
4.3. Compressive Strength of Concrete cured in tap water and nitric acid
solution:
The compressive strength of different type of concrete mixes viz; referral/ normal concrete
(NC), metakaolin concrete (MKC) was determined at 7, 28, 56 and 90 days and plotted in Fig.3.
It is observed that the concrete specimens cured in 5% nitric acid solution had lower
compressive strengths than those cured in tap water. It was also observed that the inclusion of
MK enhanced the compressive strength in comparison to NC at all the ages. The loss in
compressive strength of NC after 7, 28, 56, and 90 days of nitric acid exposure is 15.61, 13.0,
and 11.0, and 9%, respectively, while for MKC these changes are 9.23, 7.82 and 11.00, and
10.51%, respectively.
611
45
40
Compressive Strength (MPa)
35 RC(in water)
30
RC(in nitric acid)
25
MKC(in water)
20
MKC(in nitric acid)
15
7 Day 28 Day 56 Day 90 Day
Age of Concrete
Fig 3: Compressive Strength of Different type of Concrete in Water and 5% Nitric acid
6. Conclusions
The following conclusions are drawn from the present investigation.
1. Workability of concrete decreases with increase in replacement level of OPC.

2. The compressive strength of water cured MKC is higher than the NC.
3. The optimum replacement level of OPC by MK is 15%.
4. The optimum replacement level, the gain in compressive strength at 7 and 28 days is
about 20 and 11.00% respectively.
5. The compressive strength of all the mixes is reduced in acidic environment at all the
exposure periods.
References:
1) K.A Gruber, Terry Ramlochan, Andrea Boddy, R.D Hooton, M.D.A Thomas “
Increasing Concrete Durability With High-Reactivity Metakaolin” Cement and Concrete
Composites, Vol.23, No.6, 2001, pp.479-484
2) Jian-Tong Ding, “Zongjin Li., Effects of metakaolin and silica fume on properties of
concrete”, Materials Journal: 99 (4): 393-98.
612
3) J.M. Khatib, J.J. Hibbert , “Selected engineering properties of concrete incorporating slag
and metakaolin”, Construction and Building Materials ,2005: 460–72.
4) Jamal M. Khatiba, Roger M. Clayb, “Absorption characteristics of metakaoline
concrete”, Cement and Concrete Research, 2004: (34):19-29.
5) Anand Kuber Parande, B. Ramesh Babu, M. Aswin Karthik, Deepak Kumaar K.K.,
Palaniswamy, “Study on strength and corrosion performance for steel embedded in
metakaolin blended concrete/mortar, Construction and Building Materials”, 2008:22 (3):
127–34.
6) Rafat Siddique, EI-Hadj Kadri “Effect of metakaolin and foundry sand on the near
surface characterstics of concrete” Construction and Building Materials, Volume 25,
Issue 8, August 2011, Pages 3257-3266
7) P. Dinakar, Pradosh K. Sahoo, And G. Sriram (2013), “Effect of Metakaolin Content on
The Properties of High Strength concrete” International Journal of Concrete Structures
and Materials,Vol.7, No.3, pp.215–223, September 2013.
8) V. Pavlik “Corrosion of Harden Cement Paste by Acetic and Nitric Acids” Cement and
Concrete Research, Volume26, Issue 3, March 1996, Pages 475-490
9) Kalapo O. Olusola, Opeyemi Joshua, Effect of nitric acid concentration on the
compressive strength of laterized concrete, Civil and Environmental Research,
2012:2(10):48-57.
10) IS: 456-2000 “Code of Practice for Plain and Reinforced Concrete”, BIS, New Delhi.
11) IS: 8112-1989 “Specification for 43 Grade ordinary Portland cement”, BIS, New Delhis.
12) IS: 383-1970 “Specification for Coarse and Fine aggregates”, BIS, New Delhi.
13) IS: 10262-2009 “Guidelines for Concrete Mix Design Proportioning”, BIS, New Delhi.
14) IS: 8112-1989 “Specification for 43 Grade ordinary Portland cement”, BIS, New Delhi.
613
Paper ID- 121
Prospective Construction Products Using Sugarcane Bagasse Ash- A

Review
Lisa Mary Thomas1, J.Karthikeyan1

1National Institute of Technology Tiruchirappalli, Trichy - 620015, India.
lisathomas90@gmail.com
Abstract. The increase in demand of construction products for various construction activities has increased in recent
times due to urbanisation and population rise. Eventually this would lead to the exhaustion of raw materials used for
the production of construction materials. A possible way to tackle this problem is to utilise the agricultural wastes
generated in construction materials after suitable research. Agricultural by-products are abundantly available,
especially in India and recent research shows the possible use of these agricultural by-products as supplementary
cementetious materials, replacement for aggregates and soil. Some of the widely experimented agricultural by-
products are rice husk ash, coconut shells, groundnut shells, corn cob ash, bamboo leaf ash, sugarcane bagasse ash
etc. This paper presents a collective report on various construction products developed using Sugarcane bagasse ash
(SCBA). Few construction products developed using SCBA are bricks, tiles, compressed boards, paver blocks,
geopolymer concrete, mortar etc. India is one the largest producers of sugarcane, also produces high amounts of
sugarcane bagasse ash which is a by-product that arises out of burning sugarcane bagasse in the cogeneration boiler
for the production of sugar. This utilization of sugarcane bagasse ash provides a solution against dumping these in
landfills which otherwise causes health hazards, air pollution, wastage of land. Moreover, utilizing SCBA leads to
environmental benefits such as reduced pollution, waste utilization, saving land used for dumping of the waste,
reduction in greenhouse gas emissions, economic benefits such as cost saving and sustainable development. The
development of sustainable and green construction products is the need of the hour. Though SBCA ash has the
potential to be used for product manufacturing, lot of research is required finally move the material for commercial
production.
Keywords: SCBA, construction products, sustainability
1 Introduction
Large quantities of natural occurring raw materials are used for the production of
various construction products like bricks, cement, aggregates, steel, aluminium, wood,
and partitioning materials[1]. The raw materials are mined from various locations,
processed in industries to meet the required criteria for the construction product and
transported to meet the demands of construction industry. Extracting raw materials
from nature has posed problems such as exhaustion of quarries or mines, high rates of
CO2 emissions, deforestation and climate change. These have paved way for research
to utilise alternative construction materials which are eco-friendly, sustainable,
economical and serves its function as a construction product in terms durability, good
structural and mechanical properties. Over 600MT of agricultural wastes are produced
in India [2]to name a few are sugarcane bagasse ash, groundnut shell, oyster shell,
sawdust, giant reed ash, rice husk ash, cork and tobacco, which poses disposal problems
leading to environmental issues [3]. Efficient Utilization of these agricultural wastes
can solve these issues to a certain extent [4]. In the present paper the existing
knowledge on use of SCBA in developing construction products has been summarised.
Also, the design process and physcio-mechanical properties of the reviewed
construction products are explained further.
2. Sugarcane Bagasse Ash

SCBA is the second highest solid waste generated in India , next to coal combustion
residues [3, 5]. Bagasse is the fibrous matter obtained after extraction of juice from
614
sugarcane and sugarcane bagasse ash (SCBA) is the by product from the cogeneration
and combustion process at certain temperatures of sugar cane bagasse [6, 7] as shown
in Fig. 1. The quantity of SCBA is expected to increase due to rapid implementation of
new cogeneration plants and expansion of cogeneration capacity of existing plants in
sugar industries[8]. The microstructural analysis of SCBA reveals that it is mainly
composed reactive silica with a chemical composition of major oxides comprising
mainly of silicon dioxide (SiO2), aluminium oxide (Al2O3), iron oxide (Fe2O3), calcium
oxide (CaO), and the potassium oxide (K2O) and others found in traces, such as sodium
oxide (Na2O), sulphur trioxide (SO3), and Magnesium oxide (MgO) [3, 7, 17–26, 9, 27,
10–16] as shown in Table 1.
Fig. 1. Raw Bagasse ash (Picture Courtesy: Deepika et al. [10])
Table 1. Oxide Composition of SCBA (Table Courtesy: Deepika et al. [10])
Oxide Raw SCBA

Composition (%)
SiO2 72.95
Al2O3 1.68
Fe2O3 1.89
CaO 7.77
MgO 1.98
K2O 9.28
Na2O 0.02
These composition makes it suitable for its use in significant number of construction
materials such as supplementary cementetious material , sugarcane bagasse ash sand
(SCBA sand) [17, 28, 29], ceramic floor tile [14, 30], Stabilized/Sintered Earth Blocks
and Tiles, compressed earth block masonry[15], Particle boards, cement stabilized soil
blocks[31, 32], alkali-activated concrete, paver blocks, unburnt bricks[10, 33], burnt
clay bricks [11, 34, 35], compacted soil blocks[16] as listed in Table 2.
3. Development of Construction Products from Agro-waste

2.1 Sugarcane Bagasse Ash Sand
The availability of natural sand is on a decline in the past few decades due to over
ecological and environmental limitations [29]. Only few experimental studies have
been done using SCBA sand as a partial replacement for natural sand. Almir et al.[36]
studied concrete specimen with SCBA as fine aggregate. The physicochemical
615
characterization of SCBA sample gave similar results as that of natural sand , moreover
the compressive strengths of concrete specimens were found to increase with
replacement levels of 20 % and 30% [36]. Fernando et al.[17] investigated the
replacement of natural sand with SCBA sand in mortar, with replacement levels up to
50%. Compressive strength of mortar specimens were equivalent to the reference
sample, with better pore refinement and the carbonation depths of mortar samples with
30% replacement levels were found to be similar to reference samples whereas the
chloride penetration resistance increased with higher levels of replacement[17].
Table 2. List of raw materials used for construction products
Type of product/ material Materials used Authors

Sugarcane Bagasse Ash Portland-composite cement with blast Juliana et al. [28]
Sand furnace slag, Quartz sand, Basalt stone, SCBA
sand, construction waste (CW)
high early strength Portland cement sulphate Almir et al.[36]
resisting, quartz sand, SBA
SCBA sand, Natural Sand, Portland- Fernando et al.[17]

composite Cement
Ordinary Portland cement, Natural Sand, SBA Prashant et al.[29]
Supplementary SCBA, Ordinary Portland cement Ganesan et al. [20],
Cementetious Material Bahurudeen et al. [7],
Suvimol et al[18],
Nuntachai et al.[22],
Noor-ul Amin[37],
Rajasekar et al.[38],
Singh et al.[24],
Bahurudeen et al.[26],
Aukkadet et al.[27]
Ceramic Floor Tile Floor tile with quartz replaced with 2.5 wt.% Myrian et al.[30]
of SCBA
SCBA, Ceramic raw material Souza[14]
Compressed Earth Blocks Limestone Portland cement, SCBA, Sandy Sofia et al. [15]
earth (Soil)
Compacted soil blocks Soil, Hydrated lime, Pozzolanic Portland Rafael[16]
cement, SCBA
Soil, Cement, and SCBA Jijo et al. [32]
Paver Blocks Sieved bagasse ash ground to cement fineness Deepika et al.[10]
and blended with cement to make blended
cements
Unburnt Bricks Rice husk ash (RHA) ,SCBA, Clay Syed et al.[11, 12]
Common clay ,SCBA Faria et al.[39]
Unburnt Bricks Raw SCBA, River sand, Crusher Deepika et al.[10]
sand, OPC 53 grade cement
SCBA, Quarry dust, Lime Mangesh et al. [33]
Juliana et al. [28] evaluated the use of SCBA sand by partially replacing in fine
aggregate and use of construction waste (CW) by partially replacing coarse aggregate.
The concrete specimens prepared were subjected to physical, mechanical and durability
tests. Test results revealed a decrease in compressive strength in comparison with
reference mortar but this reduction decreased with increase in age of specimens[28].
Moreover, the increase in replacement levels increased the mixing water requirement
in order to maintain the consistency[17, 28]. Carbonation depths of concrete samples
with 30% SCBA sand and 30% CW reference concrete showed similar depths[28].
Prashant et al.[29] conducted compressive strength, tensile and sorptivity on concrete
616
specimens with natural sand replaced with SCBA sand from 0 to 40%. The compressive
strength was found to increase up to 10% replacement levels and thereafter decreased
whereas tensile strength decreased with the addition of SCBA sand. Also, they [29]
reported that sorptivity increased with the addition of SCBA sand as it contains fibrous
unburnt matter which absorbs water.
2.2 Supplementary Cementetious Material
Ganesan et al. [20] investigated the use of SCBA as supplementary cementetious
material. For the experimental procedure OPC was replaced with different amount of
SCBA (5, 10, 15, 20, 25 and 30 wt.%) and specimens cast using this blended cement
were tested for physical and mechanical properties. Compressive strength of SCBA
blended cement mortars and concrete were found to decrease with more than 20% ,
similar results were obtained in tensile strength test replacement levels hence they
suggested 20% as the optimum replacement level. Bahurudeen et al. [7] assessed the
pozzolanic performance of SCBA (raw and processed) from which they found that
processed SCBA exhibited more pozzolanic activity due to removal of coarse carbon
containing fibrous particles. The workability and strength was found to increase with
the addition of SCBA in concrete. They concluded that SCBA could be used as a SCM
after proper processing of the raw ash. Similar study was conducted by Suvimol et
al[18] and they suggested that pozzolanic activity was higher for finer processed SCBA
and up to 40% of cement could be replaced with SCBA. Nuntachai et al.[22] explored
the use of SCBA as a pozzolanic material in concrete. For the experimental analysis
ASTM Type I ordinary Portland cement was replaced with SCBA at 10, 20, and 30%
and from the results it was suggested that 20% is the optimum level of replacement and
higher levels of replacement resulted in concrete with low water permeability and a low
compressive strength. Similarly 20% replacement levels of Portland cement with
SCBA gave better mechanical properties and durability characteristics was reported by
Noor-ul Amin[37]. Also the heat evolution from SCBA replaced concrete was lower
than reference concrete and this increased with higher levels of replacements[22, 26].
Rajasekar et al.[38] studied the durability characteristics of Ultra high strength concrete
produced using treated or processed SCBA replacing cement. The replacement dosage
of 15% was concluded as the optimum level of replacement. Singh et al.[24]
investigated hydration of bagasse ash blended portland cement and reported that 10%
replacement levels gave better compressive strength results at ages of hydration.
Studied the effect of SCBA in mortar. It was reported that compressive strength
decreased with the addition of SCBA but later age strength was found to increase and
the durability tests gave the following results Ultrasonic Pulse Velocity (UPV)
decreased, Electrical Resistivity (ER) increased after 14days and the level of
permeability decreased. Bahurudeen et al.[26] tested the performance of SCBA
blended cement in concrete and presented results that 25% replacement of cement could
be achieved with good durability characteristics. Aukkadet et al.[27] studied the
strength and durability of SCBA concretes and found that 20 % is the optimum level of
replacement of cement with SCBA but 50% replacement was found not to affect the
modulus of elasticity. They [27] also suggested that replacing cement with SCBA was
needed for improving the durability of concrete than improving the strength.
Researched the combination of ultrafinely-ground bagasse ash for partially replacing
Portland cement in a high-performance concrete. It was reported that though there was
617
no quantifiable change in mechanical behaviour, the rheology and resistance to

penetration of chloride ions was improved.
2.3 Ceramic Floor Tile

Myrian et al.[30] prepared floor tile containing up to 2.5 wt.% of sugarcane bagasse ash
waste as a partial substitute of quartz and the quality of the floor tiles were assessed
based on its test results for linear shrinkage, apparent density, water absorption, and
flexural strength. Flexural strength values increased slightly with the addition of SCBA.
Though linear shrinkage, apparent density, water absorption values were found to
increase , it was well within the required range as per ISO 13006 standard for ceramic
tile product[30]. In SCBA incorporated ceramic materials Souza[14] observed a
decrease in flexural strength, which might be due to the presence of quartz particles in
ash. It was also found that these specimens showed an increase in linear shrinkage,
apparent density and reduction in porosity, water absorption with increase in sintering
temperatures above 1000°C [14]. The results indicated that the SCBA may be used as
an additive to produce ceramic materials.
2.4 Earth Blocks

Sofia et al. [15] investigated the mechanical properties of compressed earth block
masonry prepared using sugarcane bagasse ash and cement for its application in non-
structural masonry elements. Better structural performance was observed and addition
of SCBA was found not to affect the mechanical properties of compressed earth block
masonry[15]. Rafael[16] suggested that a combination of lime and SCBA could be used
for stabilisation of compacted soil blocks. Experimental analysis shows improved
flexural and compressive strength water saturated compacted blocks using lime, cement
and SCBA due to the formation of chemical stable compounds, mainly CSH and CAH,
resulting from the reaction between the added materials [16].Jijo et al. [32]conducted
study on ordinary Portland cement (OPC) stabilized soil blocks using sugarcane
bagasse ash (SBA). These blocks gave higher compressive strength, showed no
efflorescence and gave slightly higher water absorption[32].
2.5 Paver Blocks

Deepika et al.[10] casted paver blocks of 6mm thickness by replacing cement with
SCBA. These were designed in accordance with IS15658[40] for light to medium
traffic. While checking for durability performance, SCBA blended paver blocks
performed better whereas its compressive strength was found to be lower but this
reduction was well above the minimum requirement specified in IS 15658.
2.6 Unburnt/ Burnt Bricks

Syed et al.[11, 12] assessed the effect of addition of rice husk ash and sugarcane bagasse
ash in fired clay bricks. Though the manufactured clay bricks gave lower compressive
strengths and lower flexural strength, both the values were above the minimum
requirement and the microstructural analysis showed porous structure which resulted
in higher water absorption rates and lesser unit weight of the specimen. For lower levels
of replacement better sulphate resistance was exhibited by the specimen. Faria et al.[39]
carried out research to use SCBA in fired/burnt clay bricks. Properties such as linear
shrinkage, water absorption, apparent density, and tensile strength were studied. Test
results of the specimens showed that its linear shrinkage and mechanical strength
618
decreased, water absorption increased. The microstructure analysis of specimens

showed more porous structure[11, 12, 39]. Authors concluded that SCBA could be used
as a filler material in clay bricks but high amounts of replacement has to be avoided[11,
12, 39]. Deepika et al.[10] examined the compressive strength and water absorption of
unburnt clay bricks manufactured using SBCA, river sand and crusher sand. Higher
compressive strength was observed for unburnt bricks using SCBA & crusher sand than
compared with unburnt bricks using SCBA & river sand brick specimens with similar
proportion. In SCBA-based unburnt bricks, efflorescence was not observed but Water
absorption in SCBA unburnt bricks was higher. However, the water absorption values
of SCBA-based unburnt bricks with crusher sand were within the permissible limit as
stated in the standard. Mangesh et al. [33] prepared unburnt bricks using sugarcane
bagasse ash, quarry dust, and lime with various compositions and the specimens were
tested for physic-mechanical, functional, durability and environmental properties. No
efflorescence was observed and water absorption rates were high due to porous nature
of SCBA but was within acceptable levels. Observations during the tests showed that
bricks prepared using the following compositions and proportions SBA (50% by
weight), quarry dust (30% by weight), and lime (20% by weight) exhibited very high
compressive strength compared to commercially available clay bricks. The specimens
were also found to be durable, resistant to weathering from the following durability
tests for chloride, sulphate and carbonation, moreover bricks leachates were found to
be within permissible limits.
4. Conclusion
Based on the review of the various studies on manufacturing of construction products
from sugarcane bagasse ash, the following conclusions can be drawn:
• Sugarcane bagasse ash is abundantly available and the chemical composition
suggests that it is pozzolanic and could be a significant and potential material used
for the production of a variety of construction materials
• Though studies show that it is possible to use SBCA as supplementary
cementetious material or as a replacement for clay/soil, much research is required
in these area to commercially manufacture these products
• Utilisation of unmanaged waste such as sugarcane bagasse ash, will provide viable
solution to environmental concerns, wastage of land, health issues, sustainable
development, green construction and energy savings in terms of mining of raw
material to the manufacture of finished product.
References
1. Raut, S.P., Ralegaonkar, R. V., Mandavgane, S.A.: Development of sustainable construction material using
industrial and agricultural solid waste: A review of waste-create bricks. Constr. Build. Mater. 25, 4037–4042
2. TN, G.: Building materials in India: 50 years, a commemorative volume. New Delhi (Government of India). Build.
Mater. Technol. Promot. Counc. (998)
3. Prusty, J.K., Patro, S.K., Basarkar, S.S.: Concrete using agro-waste as fine aggregate for sustainable built
environment – A review. Int. J. Sustain. Built Environ. 5, 312–333 (2016). doi:10.1016/j.ijsbe.2016.06.003
4. Wang, B., Dong, F., Chen, M., Zhu, J., Tan, J., Fu, X., Wang, Y., Chen, S.: Advances in Recycling and Utilization
of Agricultural Wastes in China: Based on Environmental Risk, Crucial Pathways, Influencing Factors, Policy
Mechanism. Procedia Environ. Sci. 31, 12–17 (2016). doi:10.1016/j.proenv.2016.02.002
5. Madurwar, M. V., Ralegaonkar, R. V., Mandavgane, S.A.: Application of agro-waste for sustainable construction
materials: A review. Constr. Build. Mater. 38, 872–878 (2013). doi:10.1016/j.conbuildmat.2012.09.011
6. Aprianti, E., Shafigh, P., Bahri, S., Farahani, J.N.: Supplementary cementitious materials origin from agricultural
619
wastes - A review. Constr. Build. Mater. 74, 176–187 (2015). doi:10.1016/j.conbuildmat.2014.10.010

7. Bahurudeen, A., Wani, K., Basit, M.A., Santhanam, M.: Assesment of Pozzolanic Performance of Sugarcane
Bagasse Ash. J. Mater. Civ. Eng. 28, 04015095 (2015). doi:10.1061/(ASCE)MT.1943-5533.0001361
8. Bahurudeen, A., Marckson, A. V., Kishore, A., Santhanam, M.: Development of sugarcane bagasse ash based
Portland pozzolana cement and evaluation of compatibility with superplasticizers. Constr. Build. Mater. 68, 465–
475 (2014). doi:10.1016/j.conbuildmat.2014.07.013
9. Katare, V.D., Madurwar, M. V.: Experimental characterization of sugarcane biomass ash – A review. Constr. Build.
Mater. 152, 1–15 (2017). doi:10.1016/j.conbuildmat.2017.06.142
10. Deepika, S., Anand, G., Bahurudeen, A., Santhanam, M.: Construction Products with Sugarcane Bagasse Ash
Binder. J. Mater. Civ. Eng. 29, 04017189 (2017). doi:10.1061/(ASCE)MT.1943-5533.0001999
11. Kazmi, S.M.S., Abbas, S., Munir, M.J., Khitab, A.: Exploratory study on the effect of waste rice husk and
sugarcane bagasse ashes in burnt clay bricks. J. Build. Eng. 7, 372–378 (2016). doi:10.1016/j.jobe.2016.08.001
12. Kazmi, S.M.S., Abbas, S., Saleem, M.A., Munir, M.J., Khitab, A.: Manufacturing of sustainable clay bricks:
Utilization of waste sugarcane bagasse and rice husk ashes. Constr. Build. Mater. 120, 29–41 (2016).
13. Chusilp, N., Jaturapitakkul, C., Kiattikomol, K.: Experimental characterization of binary and ternary blended-
cement concretes containing ultrafine residual rice husk and sugar cane bagasse ashes. Constr. Build. Mater. 29, 41–
49 (2017). doi:10.1016/j.conbuildmat.2005.09.011
14. Souza, A.E., Teixeira, S.R., Santos, G.T.A., Costa, F.B., Longo, E.: Reuse of sugarcane bagasse ash (SCBA) to
produce ceramic materials. J. Environ. Manage. 92, 2774–2780 (2011). doi:10.1016/j.jenvman.2011.06.020
15. Lima, S.A., Varum, H., Sales, A., Neto, V.F.: Analysis of the mechanical properties of compressed earth block
masonry using the sugarcane bagasse ash. Constr. Build. Mater. 35, 829–837 (2012).
16. Alavéz-Ramírez, R., Montes-García, P., Martínez-Reyes, J., Altamirano-Juárez, D.C., Gochi-Ponce, Y.: The use of
sugarcane bagasse ash and lime to improve the durability and mechanical properties of compacted soil blocks.
17. Almeida, F.C.R., Sales, A., Moretti, J.P., Mendes, P.C.D.: Sugarcane bagasse ash sand (SBAS): Brazilian
agroindustrial by-product for use in mortar. Constr. Build. Mater. 82, 31–38 (2015).
18. Suvimol, S., Daungruedee, C.: Bagasse Ash : Effect of Pozzolanic Activity and Application in Cement Use Aspect.
3rd ACF Int. Conf. 165–173 (2008)
19. Cordeiro, G.C., Kurtis, K.E.: Effect of mechanical processing on sugar cane bagasse ash pozzolanicity. Cem. Concr.
Res. 97, 41–49 (2017). doi:10.1016/j.cemconres.2017.03.008
20. Ganesan, K., Rajagopal, K., Thangavel, K.: Evaluation of bagasse ash as supplementary cementitious material.
Cem. Concr. Compos. 29, 515–524 (2007). doi:10.1016/j.cemconcomp.2007.03.001
21. Cordeiro, G.C., Tavares, L.M., Toledo Filho, R.D.: Improved pozzolanic activity of sugar cane bagasse ash by
selective grinding and classification. Cem. Concr. Res. 89, 269–275 (2016). doi:10.1016/j.cemconres.2016.08.020
22. Chusilp, N., Jaturapitakkul, C., Kiattikomol, K.: Utilization of bagasse ash as a pozzolanic material in concrete.
23. Jagadesh, P., Ramachandramurthy, A., Murugesan, R.: Evaluation of mechanical properties of Sugar Cane Bagasse
Ash concrete. Constr. Build. Mater. 176, 608–617 (2018). doi:10.1016/j.conbuildmat.2018.05.037
24. Singh, N.B., Singh, V.D., Rai, S.: Hydration of bagasse ash-blended portland cement. Cem. Concr. Res. 30, 1485–
1488 (2000). doi:10.1016/S0008-8846(00)00324-0
25. Arenas-Piedrahita, J.C., Montes-García, P., Mendoza-Rangel, J.M., López Calvo, H.Z., Valdez-Tamez, P.L.,
Martínez-Reyes, J.: Mechanical and durability properties of mortars prepared with untreated sugarcane bagasse ash
and untreated fly ash. Constr. Build. Mater. 105, 69–81 (2016). doi:10.1016/j.conbuildmat.2015.12.047
26. Bahurudeen, A., Kanraj, D., Gokul Dev, V., Santhanam, M.: Performance evaluation of sugarcane bagasse ash
blended cement in concrete. Cem. Concr. Compos. 59, 77–88 (2015). doi:10.1016/j.cemconcomp.2015.03.004
27. Rerkpiboon, A., Tangchirapat, W., Jaturapitakkul, C.: Strength, chloride resistance, and expansion of concretes
containing ground bagasse ash. Constr. Build. Mater. 101, 983–989 (2015). doi:10.1016/j.conbuildmat.2015.10.140
28. Moretti, J.P., Sales, A., Almeida, F.C.R., Rezende, M.A.M., Gromboni, P.P.: Joint use of construction waste (CW)
and sugarcane bagasse ash sand (SBAS) in concrete. Constr. Build. Mater. 113, 317–323 (2016).
29. Modani, P.O., Vyawahare, M.R.: Utilization of bagasse ash as a partial replacement of fine aggregate in concrete.
Procedia Eng. 51, 25–29 (2013). doi:10.1016/j.proeng.2013.01.007
30. Schettino, M.A.S., Holanda, J.N.F.: Characterization of Sugarcane Bagasse ash Waste for Its Use in Ceramic Floor
Tile. Procedia Mater. Sci. 8, 190–196 (2015). doi:10.1016/j.mspro.2015.04.063
31. James, J., Pandian, P.K.: A Short Review on the Valorisation of Sugarcane Bagasse Ash in the Manufacture of
Stabilized/Sintered Earth Blocks and Tiles. Adv. Mater. Sci. Eng. 2017, (2017). doi:10.1155/2017/1706893
32. James, J., Pandian, P.K., Deepika, K., Manikanda Venkatesh, J., Manikandan, V., Manikumaran, P.: Cement
stabilized soil blocks admixed with sugarcane bagasse ash. J. Eng. (United States). 2016, (2016).
doi:10.1155/2016/7940239
33. Madurwar, M. V, Mandavgane, S.A., Ralegaonkar, R. V: Development and Feasibility Analysis of Bagasse Ash
Bricks. J. Archit. Eng. 141, 1–9 (2015). doi:10.1061/(ASCE)EY.1943-7897.0000200.
34. Muñoz Velasco, P., Morales Ortíz, M.P., Mendívil Giró, M.A., Muñoz Velasco, L.: Fired clay bricks manufactured
by adding wastes as sustainable construction material - A review. Constr. Build. Mater. 63, 97–107 (2014).
620
35. Muñoz V., P., Morales O., M.P., Letelier G., V., Mendívil G., M.A.: Fired clay bricks made by adding wastes:
Assessment of the impact on physical, mechanical and thermal properties. Constr. Build. Mater. 125, 241–252
36. Sales, A., Lima, S.A.: Use of Brazilian sugarcane bagasse ash in concrete as sand replacement. Waste Manag. 30,
1114–1122 (2010). doi:10.1016/j.wasman.2010.01.026
37. Amin, N.: Use of Bagasse Ash in Concrete and Its Impact on the Strength and Chloride Resistivity. J. Mater. Civ.
Eng. 23, 717–720 (2011). doi:10.1061/(ASCE)MT.1943-5533.0000227
38. Rajasekar, A., Arunachalam, K., Kottaisamy, M., Saraswathy, V.: Durability characteristics of Ultra High Strength
Concrete with treated sugarcane bagasse ash. Constr. Build. Mater. 171, 350–356 (2018).
39. Faria, K.C.P., Gurgel, R.F., Holanda, J.N.F.: Recycling of sugarcane bagasse ash waste in the production of clay
bricks. J. Environ. Manage. 101, 7–12 (2012). doi:10.1016/j.jenvman.2012.01.032
40. BIS:15658: Indian Standard for Precast Cocrete Blocks For Paving. Bur. Indian Stand. Dehli. (2006)
621
Paper ID- 124
Osdag: A Structural Steel Design Software for

IS 800:2007
Siddhartha Ghosh1 , Danish Ansari1 , Ajmal Babu MS1 , Nuli Dharma Teja1 ,
Reshma Konjari1 , Swathi M1 , and Subhrajit Dutta2
1
Indian Institute of Technology Bombay, Mumbai 400076, India
2
National Institute of Technology, Silchar 788010, India
sghosh@civil.iitb.ac.in
https://osdag.fossee.in/
Abstract. This paper discusses the research and development under-

taken to develop a free and open source software called Osdag 3 (Open ®
Steel Design and Graphics) for the design and detailing of steel struc-
tures, at the Indian Institute of Technology Bombay. Osdag is a model-
based platform-independent software having several modules of struc-
tural steel design following the specifications of IS 800:2007 and other rel-
evant standards. It is being developed on interpreted, high-level, general-
purpose object-oriented programming (OOP) language Python and other
Python-based tools. An interactive graphical user interface (GUI) is pro-
vided with proper validations of user input and maintenance of log files.
The log messages provides the user with valuable information on the fail-
ure of trial design and suggests required modification(s). The software
also features the inclusion of graphics for real-time 3D visualization of
the structural component details. The CAD files developed from Osdag
can be exported to conventional CAD manipulation software(s). Osdag
generates a design report, showing details of all the inputs, design con-
siderations, fabrication drawings, and – most importantly – details of
each design and detailing checks according to the Indian Standards with
reference to the appropriate clause number.
The current version of Osdag contains connection design modules for
shear connections (fin plate, end plate, cleat angle, and seated angle) and
beam-to-beam moment connections namely cover plate and extended end
plate connections.
Keywords: Osdag · steel · design · free and open-source · Python · IS

800 · CAD.
1 Introduction: Why Osdag

1.1 Steel in construction
Over the past one and half centuries, steel has emerged as probably the most
trusted building material. Modern architectural marvels make significant use
3
Osdag ® and its logos are trademarks of the IIT Bombay foundation.
622
of steel as the primary building material. Of these, many bridges, buildings,

stadia, etc. are built with what is known as ‘structural steel’ in the domains of
the construction industry, structural engineering, and building architecture. The
Forth Bridge near Edinburgh, Scotland [1], is one of the earliest examples of such
technical marvels, which has been serving rail traffic from 1890 till date. From
the pre-World War II, skyscrapers (such as the Chrysler Building [2] and the
Empire State Building in New York, USA) to modern technologically advanced
masterpieces (such as the“Gherkin” in London, UK [3]), structural steel has been
valued for the necessary qualities it provides: strength to weight ratio, durability,
sustainability, and ductility. Steel also allows a designer the flexibility to use it in
versatile ways. Construction with structural steel is significantly faster – and with
better quality control – than most common building materials. Steel buildings
can be modular and can efficiently adapt to changes in the use pattern during
a buildings life cycle. At the end of the cycle, with about 90% or more of steel
being fully recyclable [4], steel appears to be the best choice for the sustainable
design of structures.
Even with all these advantages, we hardly see steel structures in India, except
for industrial sheds, railway bridges, and transmission towers. India is currently
the third largest country in terms of production of crude steel (India produced
101.4 metric tonne steel in the year 2017 [5]), but on the other hand, it is far
behind in terms of domestic steel consumption. The per capita consumption of
steel in India is as low as 65.2 kg against the world average of 214.5 kg [5]. Also,
the steel to cement ratio of India is 0.29 [6], which is very less as compared
to other developed countries. In the fast-growing construction sector of India,
reinforced concrete structures outnumber steel structures by a very large margin.
The reasons for this disparity are diverse; here is a quick look at the technological
and (technical) education-related aspects.
1.2 Indian Standard for steel design
The Indian standard IS 800 provides general guidelines and the standard speci-
fications for the design and detailing of structures using structural steel. It was
first published in 1956 [7] with the goal of achieving “economy in the use of
structural steel”. This was last revised in 2007, shifting the design philosophy
from the Working Stress Method (WSM) [8] to the Limit State Method (LSM)
[9] and bringing our design specifications closer to those of the modern steel
design standards.
After the release of the 2007 revision, the Institute for Steel Development
and Growth (INSDAG) [10] along with academics and experienced profession-
als, conducted training programmes for college teachers and design engineers
across the country. Although this has resulted in some confidence building in
the academia and in the industry, till date most college teachers and design
professionals are not comfortable in steel design. Also, experienced professionals
resist the transition from a design code they are already very familiar with to a
new one. Besides, there are complaints against too many specifications to learn
623
Osdag: A Structural Steel Design Software for IS 800:2007 3
and implement and too few helpful resources to make use of.
1.3 Motivation for Osdag

The industry and academia’s adaption to revised British and American steel
design codes were eased by making design software/tools available for the new
standards. Availability of such design tools has been too few and too late for IS
800:2007 [9]. Commercial softwares are costly and beyond the affordable range of
many design professionals and young practicing students. This results in piracy
of the licensed software both in the industry and the academia, without the user
many a time realising that he/she is committing a criminal offence. Those who
stay away from the piracy route rely on in-house (handed down) spreadsheet
like solutions or hand calculations, that are not validated and/or are very te-
dious. The issue of non-validated solutions is most evident in the design of steel
connections. Many of the common structural design software does not include
modules for steel connection design, for which one typically relies on special-
ized connection design and detailing software, such as RAM Connection [11].
These are costly software. For example, a single license of a single connection
(beam-to-column shear connection using a fin plate) design module of the Praxis
software [12] costs Euro 150.00 (approximately INR 12,110.00, as per the current
exchange rates) exclusive of taxes.
Considering this state-of-practice, Osdag project was excited in 2014 for de-
veloping a software dedicated to structural steel design and detailing as per the
Indian standard(s), which does not make a big dent in the users pocket and
can be one simple and effective solution to bring forth the necessary confidence
among steel designers, college teachers and students of civil engineering.
2 About Osdag
Osdag is a cross-platform, free and open-source software for the design (and
detailing) of steel structures, primarily following the Indian Standard IS 800:2007
[9]. It also adheres to the other relevant standards of structural steel design in
India. It allows the user to design steel connections, members and systems in
the form of design modules using an interactive graphical user interface. The
design modules in Osdag are categorized as per design logic for the ease of the
user. The interactive GUI provides a 3D visualization of the designed component
and creates images for construction/fabrication drawings. The design is typically
optimized following industry best practices.
The beta version of Osdag is released under the terms and conditions of the
GNU lesser general public license (LGPL) Version 3 [13].
2.1 Osdag GUI

Osdag has an interactive and easy to use graphical user interface (GUI) (Fig. 1).
The GUI is designed to be user-friendly and almost all the operations have an
624
assigned shortcut key. The design modules in Osdag which are categorized and
listed systematically in the opening window of the GUI can be selected by the
user by simply clicking on radio buttons. Each radio button is associated with a
design module and has a graphical representation of the module to ensure that
the user has selected the correct or desired module. Once the user selects the
module, Osdag prompts to select or create a ‘workspace directory’ where all the
design files would be saved for further reference. The design window GUI (Fig. 1)
is mainly divided into four parts: Input Dock, Output Dock, CAD window and
Message window (along with a menu bar and a quick access bar).
Fig. 1. Design Window of Osdag GUI
The Input Dock on the left side of the design window GUI collects all the
input data from the user, validates, stores and sends it to the design and detailing
algorithm. The Input Dock window is dynamic and whenever a data is entered
as an input it is validated by a specially designed algorithm at the backend
of the software. The algorithm designed for input data validation is based on
the guidelines provided by the relevant standards and industry best practices.
‘Error’ or ‘Warning’ windows pop up when the user input does not satisfy these
validations. The feature of input data validation in Osdag enables the user to
create a more realistic design and also assists in self-learning. The input data
can also be saved into a file (with the ‘.osi’ extension) for recreating the design,
proof checking or sharing with clients/students.
The Output Dock on the right side of the design window GUI gets populated
with design results upon the successful completion of a design. Osdag generates
these results from a uniquely written ‘design and detailing algorithm’ which
625
performs each check as per the codal provisions. All the necessary parameters
associated with the design are displayed in the form of a checklist into the Output
Dock.
A menu bar is provided at the top of the design window GUI which has
options for File, Edit, View, Graphics and Help. The ‘File’ menu consists of
various options for saving and managing the files (including 3D CAD model, 2D
drawings and design report) associated with the design. The ‘Edit’ menu has an
option for setting ‘Design Preferences’ which can be used by an advanced user
to specify special preferences for the design. The ‘View’ menu has an option for
setting the font size, type and style along with options to dock and undock the
input, Output Docks. The ‘Graphics’ menu has various options associated with
the 3D CAD model. The ‘Help’ menu is loaded with video tutorials to assist
a novice user on how to use the software and design examples to learn sample
problems on structural steel design. The user can also ask/report queries related
to the software or design to the Osdag team using the ‘Ask Us a Question’ option
from the Help menu.
A ‘quick access bar’ is provided at the top of the 3D CAD window. This
bar has an option for docking and undocking the input and Output Docks,
viewing and saving the 2D drawings and viewing the sub-components of the
design individually.
2.2 CAD Model
An interactive ‘CAD window’ present at the center of the design window GUI
displays a 3D CAD model (Fig.2), which provides a clear visualization of the
designed component. You can pan, zoom and rotate the model or view each of
the sub-components separately. The sub-components in the CAD model, like,
the connecting member, connected member, connector, bolt, and weld are color-
coded in order to distinguish them individually. The created CAD model can be
saved into various formats such as IGS, STL, STEP, and BREP. This CAD model
can be later imported to other CAD manipulation software such as AutoCAD
[14] or FreeCAD [15].
2.3 Message Log
A text window for message(s) known as the ‘Message Log’ (Fig. 3) displays the
status of the design; but more importantly, it also suggests necessary changes if
a trial design is found unsafe. This message window at the bottom of the design
window GUI, displays color-coded error/warning/info messages and also suggests
possible rectifications, all of which can be saved for a design session. Warning,
error, and information are displayed in orange, red and green color respectively
for the understandability of the user. These messages also refer to the relevant
clause or a design guideline which helps in the self-learning capability of the user
to a great extent.
626
2.4 Structural Drawings
The conventional practices of design require a detailer or a draftsperson to create

fabrication drawings. This adds to the cost and duration of the project. Also the
human involvement is likely to introduce new errors. Osdag readily creates 2D
views (Fig. 4) of the designed component in the format of typical construction
drawings. These drawings follow the standard detailing practices adopted by
expert consultants, designers, erectors, and fabricators. The views can be saved
in vector (SVG) and raster (PNG) formats. The 2D images can also be used by
design engineers to submit reports or by teachers as teaching aids.
2.5 Design Preferences
Detailed Design Preferences (user choices or specific design based requirements)

can be set using the ‘design preference’ toolbox (Fig. 5). For a novice user,
default values can be used in the design process in lieu of these preferences. The
experienced user can set preferences for bolt, weld, detailing and other design
aspects respectively.
2.6 Design Report
Osdag creates a professional and very comprehensive ‘Design Report’ (Fig. 6

and Fig. 7) which can be viewed and saved in the pdf format. The design report
compiles all the necessary information like the company/designer profile, detailed
list of checks, 2D drawings, and CAD model. The salient feature of the design
Fig. 2. 3D CAD model created automatically in Osdag
627
Fig. 3. Osdag Message Log
Fig. 4. A structural drawing created automatically in Osdag
report is the inclusion of all the design and detailing checks as per IS 800:2007 [9]
628
Fig. 5. Osdag design preferences toolbox
and other relevant standards into the design report. In each check, the ‘provided’
value is compared against its ‘required’ value and a remark is provided against
the check, where, PASS indicates a safe design and FAIL indicates an unsafe
design. Unlike other commercial software(s), the design report created in Osdag
gives information on the reference of the check, which is a very useful tool not
just for a student/learner, but also for an expert proof-checker/design reviewer.
3 Osdag Architecture
The block diagram (Fig. 8) depicts the architecture of the code(s) and its
functioning in Osdag. All the design modules in Osdag follow this architecture
in order to create a successful design along with a 3D CAD model, 2D drawings
and Design Report. The design data is collected in the form of input from the user
through an interactive ‘Input Dock’ from the design window GUI. Other inputs,
like the section properties, are collected from the Osdag database present at the
backend of the software. The database contains properties of the standard Indian
steel sections as per IS 808:1989 [16] (rolled sections) and globally used parallel
flange sections [10]. The data is collected and stored in the ‘Main File’ of Osdag
in an appropriate form only after its validation by uniquely written functions and
conditional statements. If the input data is found to be ‘inappropriate’ in terms
of design or programming sense then Osdag pops up with errors and suggestions
to correct the user input. The Main File is the core of the architecture as it
629
Fig. 6. Design Report
Fig. 7. Design Report
630
10
DISPLAY
Steel Section Database Input Dock Design Preferences
Output Dock
2D Drawings
3D CAD Model
Main File
Design Report
Calculation File Log Messages
2D Drawing Generator 3D CAD Model Generator Design Report Generator
Fig. 8. The architecture of Osdag
integrates all the supporting files into it, which enables the functioning of the
module. The input data is then called into the ‘Calculation File’ where all the
design and detailing checks are made. These checks are written in an algorithmic
way and include appropriate ‘log message(s)’. When a check does not satisfy a
certain design or detailing criterion, the log message associated to this check
is displayed in the message window of the GUI. All the calculated parameters
in the Calculation File are stored in the form of a dictionary as an ‘output
dictionary’. This dictionary is called into the ‘2D Drawing Generator’ file, ‘3D
CAD Model Generator’ file and the ‘Design Report Generator’ file from the
Calculation File. The 2D Drawing Generator file creates 2D views (front, side,
and top) of the designed component, the 3D CAD Model Generator file creates
the 3D CAD model and the Design Report Generator file creates the design
report by echoing the inputs and compiling all the checks, views and the CAD
model. These outputs are sent back to the Main File where it connects the
output to the ‘Display’ unit. The display unit populates the output results into
the Output Dock, sends 2D views and 3D CAD model to its display window,
loads design report and displays respectively. In case of an ‘Unsafe’ design, only
the Output Dock and a partial design report are shown.
4 Development Process
The development process in Osdag begins with identifying the most common and
widely used design module. The Osdag team, along with its industrial and aca-
demic collaborators, decides the modules to be incorporated into Osdag through
a review process. Based on a thorough review of existing Indian design stan-
dards, design guidelines, detailing guidelines, and standard textbook examples,
631
the ‘Design and Detailing Checklist’ (DDCL) and the ‘Design and Detailing
Query’ (DDQ) are prepared. These documents include a list of checks to be
carried out for a particular module in an algorithmic way and the queries as-
sociated with the design. Using a web-based interface, feedback on DDCL and
DDQ are collected from the ‘Expert Reviewers’. The feedbacks received are
analyzed/discussed by the team and then incorporated into the design and de-
tailing algorithm. During the review process, the team works on writing the
Python code for the Main File, Calculation File, 3D CAD generator file, 2D
Drawing Generator file, report generator file, etc. The code management and
continuous integration (CI) is done using version control systems like Git and
GitHub (https://github.com/osdag-admin/Osdag). Once the module is inte-
grated into Osdag, it goes through the necessary checks and tests before being
compiled into the installer. The Osdag team also works on creating self learn-
ing and e-learning materials such as design examples and video tutorials. The
installers and the learning materials are made available to its users free of cost
through the Osdag website (https://osdag.fossee.in/). The team also takes
feedback and resolves user queries through the FOSSEE [17] discussion forum
(https://forums.fossee.in/accounts/login/).
5 Resources
Adhering to todays learning/education techniques, Osdag provides a variety of

self-learning tools and contents through its website: http://osdag.fossee.
in. Besides Osdag installers (Fig. 9) for different operating systems (https:
//osdag.fossee.in/resources/downloads), these web pages come with di-
verse user tools, video tutorials (on different aspects of installing and operating
the software, https://osdag.fossee.in/resources/videos), design examples
(https://osdag.fossee.in/resources/sample-design), discussion forum https:
//osdag.fossee.in/forum, developer tools etc. The video tutorials are also
available on Osdags YouTube channel http://tinyurl.com/OsdagVideos. While
these videos and design examples are excellent self-learning materials, these can
also be used in conducting training programmes on Osdag efficiently.
6 Summary
In this paper, the authors have described the various salient features of a free
and open source software called Osdag. Osdag is used for design and detailing of
steel structures as per the Indian Standards (IS 800:2007). Currently Osdag is
the only Free and Open Source Software (FOSS) in India for the design of steel
structures as per the latest provisions of the Indian Standards. The ‘structure’ of
the software and its development process have also been briefly discussed. The
various learning/resource materials and tools available for students, teachers and
design professionals have been highlighted, as well.
632
12
Fig. 9. Osdag installers available online on the Osdag Resources page
Acknowledgement
Funding provided by the Ministry of Human Resources and Development (MHRD),

Government of India through the (FOSSEE) [17] project under the National Mis-
sion on Education through Information and Communication Technology (NME-
ICT) [18] is used for the Osdag project. The authors acknowledge the contribu-
tion of Expert Reviewers of Osdag from academia and industry. The authors also
would like to acknowledge the FOSSEE team from IIT Bombay for its continued
support. Most importantly the authors would like to acknowledge the contribu-
tion of Deepa Chaudhary, Jayant Patil, Hashmi Suhel, Jeffy Jaffar, Aravind P,
Sasir Pentyala, Siddhesh Chavan and the advices from Prof. Kannan Moudgalya
and Prof. Prabhu Ramachandran.
References
1. The Forth Bridges, https://www.theforthbridges.org Last accessed 14 Dec 2018
2. Massey, K.: History of Walter P. Chrysler and the Chrysler Building. Allpar. https:
//www.allpar.com/history/chrysler-building.html. Last accessed 14 Dec 2018
3. Gherkin London, http://www.30stmaryaxe.com. Last accessed 16 Dec 2018
4. World Auto Steel, https://www.worldautosteel.org. Last accessed 7 Nov 2018
5. World Steel Association report, Steel Statistical Yearbook 2018, worldsteel.org.
Last accessed 21 Dec 2018
6. CCI Annual Report, Cement corporation of India ltd., New Delhi (2017), https:
//www.cciltd.in/
7. IS 800: General Construction in Steel – Code of Practice. Bureau of Indian Stan-
dards, New Delhi. (1956)
633

10. Institute for Steel Development And Growth - INSDAG, http://www.
steel-insdag.org. Last accessed 23 Nov 2018
11. RAM Connection Standalone V8i 7.0, Bentley Systems, Inc.(2010, https://
www.bentley.com/en/products/product-line/structural-analysis-software/
ram-connection. Last accessed 13 Dec 2017)
12. EC - PRAXIS 3J, http://www.gsi-eng.eu. Last accessed 4 Oct 2018
13. GNU Lesser General Public License version 3, https://opensource.org/
licenses/lgpl-3.0.html. Last accessed 7 Jan 2018
14. AutoCAD: A computer-aided design and drafting software, Autodesk, Inc. https:
//www.autodesk.in. Last accessed 18 Dec 2018
15. FreeCAD- An open-source parametric 3D CAD modeler, http://www.freecadweb.
org. Last accessed 18 Dec 2018
16. IS 808: Dimensions for Hot Rolled Steel Beam, Column, Channel and Angle Sec-
tions. Bureau of Indian Standards, New Delhi. (1989)
17. Free and Open Source Softwares for Education, https://fossee.in. Last accessed
19 Dec 2018
18. National Mission on Education through Information and Communication Technol-
ogy, https://www.worldautosteel.org. Last accessed 19 Dec 2018
19. Ghosh, S.: FOSSEE News. vol. 1, pp. 1–4. FOSSEE, IIT Bombay, Mumbai (2018).
634
Paper ID- 128
Experimental and Analytical Investigations on Two-way

Flexural Capacity of Biaxial Voided Slab
R Sagadevan1[0000-0002-0086-6651] and Rao B N2

1Ph. D. Scholar, Structural Engineering Laboratory, Department of Civil Engineering, Indian
Institute of Technology Madras, Chennai 600036, India.
sagadevan.ceg@gmail.com
2 Professor, Structural Engineering Laboratory, Department of Civil Engineering, Indian Insti-
tute of Technology Madras, Chennai 600036, India.

bnrao@iitm.ac.in
Abstract. Rise in urbanisation and subsequent demand for infrastructure accel-

erates the use of concrete as a construction material. Conventional design prac-
tices need large member sizes to fulfil architectural and structural requirements.
Hence, the size of structural members such as slab, beam, column, etc., signifi-
cantly affects the quantity of concrete used. This calls for the need to optimize
use of concrete to reduce the self-weight of structure. Present study explores the
possibility of biaxial voided slabs as a substitute to conventional reinforced con-
crete slabs. A typical biaxial voided slab eliminates concrete from middle of floor
slab by introducing voids and it leads to significant reduction in self-weight,
about 30-50%. However, these voids reduce sectional area of slab concrete,
which leads to reduction in flexural stiffness and shear resistance. The behaviour
of biaxial voided slab subjected to two-way flexure is discussed. Experiments are
carried out to determine two-way flexural behaviour of two-way slab with sphere
shape voids by adopting sixteen-point loads. The obtained experimental results
of strength and serviceability criteria are compared for slabs with and without
voids. It is found that flexural capacity of two-way slab with biaxial voids is same
as solid slab, with minor reduction in its flexural stiffness. Furthermore, analyti-
cal study is carried out based on yield line analysis to predict the flexural capacity
of slab and compared with experimental results. It is evidenced that sixteen-point
load is equivalent to 89 % of uniformly distributed load and flexural capacity of
void slabs can be predicted by using yield line analysis as used in conventional
solid slab.
Keywords: Biaxial Voided Slab, Flexural Capacity, Flexural Stiffness, Sixteen-

Point Load, Yield Line Analysis.
1 Introduction
Reinforced concrete (RC) structures being constructed across the world to overcome
the increase in the demand of urbanisation and infrastructure development. It rises the
usage of concrete as a construction material. The demand in architectural and structural
635
requirements such as aesthetic, number of stories, span, etc., increases the size of struc-
tural members which further increases the requirement of concrete. In past, many re-
searchers, structural and design engineers came up with many types of slab systems to
overcome the structural and non-structural requirements. One-way voided slab system
is one among them which reduces considerable quantity of concrete without significant
change in its structural performances. However, the demand of two-way slab system
were not fulfilled. In early 2000s, the two-way voided slab system were introduced with
plastic void formers having various shapes such as spherical, donut, cuboid, etc. It leads
to reduction in self-weight up to 50 % [1, 2, 11, 3–10].
The biaxial voided slab reduces self-weight by 44 % with the same flexural capacity
of the solid slab [1]. The one-way flexural capacity of the biaxial voided slab was stud-
ied by various researchers [1, 4, 11–15]. These investigations showed that the capacity
of RC voided slab is the same as RC solid slab with slightly lower stiffness. Similarly,
the two-way flexural behaviour of voided RC slab system was investigated [6, 16].
These studies showed that, the flexural stiffness of voided slabs was 25 % lower that of
an RC solid slab, however, the flexural capacity remains the same.
In the past, various types of loading configuration were adopted to study the two-
way flexural capacity of RC slabs, such as single point load at the centre of slab [17,
18], five-point load in small-scale specimens [6], twelve-point load in full-scale speci-
mens [16]. These tests resulted in premature localised punching shear failure and the
load was concentrated at the central region of the slab. A methodology to design and
analyse the biaxial voided slab needs to be studied for practical application. Suitable
loading configuration which overcome the premature failure and load concentration
needs to be verified.
In this study, the flexural behaviour of biaxial RC voided slabs was studied through
experiments and compared with analytical expressions. Experiments on three full-scale
specimens were conducted, (i) solid slab (SP 1), (ii) slab with 90 mm diameter sphere
voids (SP 2), and (iii) slab with 180 mm diameter sphere voids (SP 3). The uniformly
distributed load (UDL) was simulated by adopting a sixteen-point load. The obtained
experimental results are compared with theoretical calculations which are based on the
yield line analysis (YLA) in conjunction with Indian code IS 456: 2000 [19] provisions.
In addition to that, the flexural behaviour of RC solid and biaxial voided slab having
identical dimensions and reinforcement details was compared.
2 Experimental Study
The two-way flexural behaviour of slab systems was expressed in terms of ultimate
load carrying capacity, flexural stiffness and overall load versus deflection plot. The
void formers manufactured by recycled polypropylene were used in the slab specimens
to create voids. These sphere shape void formers were fabricated with 3 mm wall thick-
ness and it ensures any unintended change in shape while handling, placing the concrete
and using vibrator for compaction. The details of test specimens, materials’ properties,
load configurations and test methodology are summarised below.
636
2.1 Details of Specimens
Fig. 1. Details of test specimen SP 1 (one-quarter of slab)
Three full-scale slab specimens were tested. The plan dimensions of test specimens
were 3300 mm × 3300 mm. Minimum reinforcement ratio was provided to ensure flex-
ure failure dominating shear failure as voided slabs are vulnerable to shear. In longitu-
dinal and transverse directions reinforcements were arranged in the form of mesh. The
637
size of reinforcement bars was restricted to 6 mm. The specifications of specimens are
summarised in Table 1. The details of slab specimens are shown in Fig. 1 – 3.
638
The M 25 grade concrete confirming to IS 456 [19] was used to cast the test specimens
with the mix proportion of 1 : 2.25 : 4.5 (cement : fine aggregate : course aggregate)
and water-cement ratio of 0.55. Concrete cubes were cast and cured under similar ex-
posure conditions as that of slab specimens. The compression test on cubes was carried
at the time of the flexure test on the companion slab specimen. The observed mean
639
strength is summarised in Table 1. The reinforcements of grade Fe 500D confirming to

IS 1786 [20] was used. The properties of reinforcements were obtained by conducting
tensile tests which are summarised in Table 2.
Table 1. Details of specimens
Specimen Depth Reinforcement in both directions Self-weight Mean cube

(mm) Top Bottom Spacing (kN) strength
(mm2/m) (mm2/m) (mm) (N/mm2)
SP 1 150 343 343 80 39.6 31.2
SP 2 150 343 343 80 36.0 31.0
SP 3 250 257 257 105 54.8 29.4
Table 2. Mechanical properties of reinforcement
Diameter Strength (N/mm2) Strain

(mm) Nominal Yield Ultimate Yield Ultimate
6 500 560 650 0.0033 0.0591
2.2 Experimental Test Set-up and Instrumentation

Test Setup. The flexural behaviour of the slab was studied through sixteen-point load
test. Fig. 4 shows the actual test set-up. The localised pre-mature punching shear failure
was avoided by applying the point load as patch load through steel plate of size 170
mm × 170 mm × 12 mm as shown in Fig 5. A pair of hydraulic jack of capacity 1500
kN, each was used to apply the load. The load was transferred through hot rolled steel
sections to the slab specimens. The steel sections were allowed to rotate freely with the
specimen’s deformation by proving the rollers at its ends. The line-type reaction hinge
of length 2800 mm was provided at its all four sides at 150 mm from specimen edges
to simulate simple support condition. The experimental errors from the support condi-
tion such as stress concentration and generation of fixed end moment was minimised
by the support discontinuity at corner [16].
Instrumentation. Instruments were provided to measure the applied load, deflection,

and strain in reinforcements. The applied load was measured by a pair of load cells of
capacity 1000 kN each. Deflection was measured at mid-span using linear variable dif-
ferential transformer (LVDT) with a measuring range of ± 100 mm. The strain was
obtained using strain gauges of 10 mm gauge length in longitudinal and transverse di-
rections of the bottom reinforcements located at the centre of slab specimens. A data
acquisition system, having the facility to record load, deflection, and strain at same
instance was used.
640
Fig. 4. Two-way flexural test set-up (sixteen-point load)
Test Procedure. Load controlled monotonic test was performed with a pair of hydrau-
lic jack. The jacks were synchronised and operated using a single master control system
to ensure equal load distribution across each jack. The loading was paused at every 50
kN interval to perform the visual inspection, mark cracks, and take photographs. The
test was terminated immediately after observing major drop in applied load; it ensures
the safety of measuring and loading devices, and avoids any permanent damage to load
transfer members.
2.3 Test Results and Discussions
Load Deflection Behaviour. All the three slab specimens showed typical flexural be-
haviour. Initially, all specimens remained elastic until cracking, after that inelastic ac-
tions started, i.e. yielding of bottom reinforcements at bottom of slab and crushing of
concrete at top of slab. The load versus mid-span deflection plot is shown in Fig. 6. It
can be observed that, the behaviour of solid slab (SP 1) and voided slab (SP 2) speci-
mens show identical load-deflection behaviour after cracking. It implies that the pres-
ence of void is not affecting load carrying capacity of the slab but the initial stiffness is
influenced by the voids. The initial stiffness of SP 1 is 37 % more in comparison with
the SP 2, however, the secant stiffness corresponding to yield load of voided and solid
slab specimens were comparable.
641
Fig. 5. Position of sixteen-point (patch) load (one-quarter of slab)
Load Carrying Capacity. The load carrying capacity of solid and voided slabs was
observed to be equal. Ultimate load and corresponding mid-span deflection are sum-
marised in Table 3.
Table 3. Results based on experimental study

Specimen Sixteen-point load Equ. UDL (kN)
Pu (kN) δu (mm) LL, Wu1 DL+LL, Wu2
SP 1 997.5 73.64 1121 1160
SP 2 1008.7 74.08 1133 1169
SP 3 1184.9 95.88 1331 1386
Note: Pu is observed ultimate load; δu is mid-span deflection corresponding to Pu.
642
1200
1000
800
16-point load (kN)
600
SP 1
400 SP 2
SP 3
200
0
0 10 20 30 40 50 60 70 80 90 100
Mid-span deflection (mm)
Fig. 6. Load versus mid-span deflection plot of tested specimens
3 Analytical Study
3.1 Moment Capacity of Slab Section
0.67fck
x̅=0.42xu
xu C=0.54fckxu
d' N.A
Tt=Ast,t fst,t
de
d
Tb=Ast,b fst,b
Fig. 7. Stress and force in slab at ultimate
The stress distribution across the depth of slab is shown in Fig. 7 which corresponds to
linear strain variation. Based on the forces in concrete and reinforcements, the equation
643
10
to estimate the ultimate moment capacity of a slab section per unit width is given in Eq.
(1). The contribution from the top and bottom reinforcements were considered for the
estimate of capacity.
m = Tb (de - x̅) + Tt (d′ - x̅) (1)
The moment capacity of tested slab specimens are estimated and given in Table 4. It
can be observed that, the moment capacity in orthogonal direction remains same. This
is primarily because of the orientation and position of the reinforcements.
3.2 Load Carrying Capacity of Slab

The theoretical load carrying capacity of slab specimens were estimated based on YLA.
The capacity of the slab was calculated based on the suitable crack pattern at ultimate
stage. In this study, the square shape of the slabs, under two-way flexural action were
tested Therefore, the yield line was assumed to form X-shape along the diagonals; it
results in dividing the slabs into four equal triangular parts as shown in Fig. 8. The
expression for collapse load was derived based on the equilibrium of slab parts.
The ultimate load carrying capacity (Pu) of the slab specimen subjected to sixteen-
point load is calculated and given by Eq. (2).
Pu = 64ml / 3le (2)
Similarly, the ultimate load carrying capacity (Wn) of the slab specimen subjected to
UDL is calculated and given by Eq. (3).
Wn = 24ml / le (3)
By comparing Eq. (2) and (3), it is observed that the ratio of the ultimate load carry-
ing capacity of the slab with sixteen-point load (Pu) and UDL (Wn) is 89 %. The esti-
mated theoretical ultimate load carrying capacity of slab specimens by YLA is summa-
rised in Table 4.
Table 4. Theoretical capacity of slab specimens

Specimen Moment capacity (kNm/m) Capacity by YLA (kN)
mx my UDL (Wn)
SP 1 30.32 30.32 800
SP 2 30.30 30.30 800
SP 3 38.97 38.97 1029
644
11
Fig. 8. Assumed yield line pattern with deformation contour
4 Summary and Conclusions

Table 5. Comparison of experimental and theoretical capacity
Specimen Capacity - UDL (kN) Ratio
Experimental Analytical Wu2 / Wn
(Wu2) (Wn)
SP 1 1160 800 1.45
SP 2 1169 800 1.46
SP 3 1386 1029 1.35
From Table 5, it can be observed that the ratio of experimentally obtained ultimate load
capacity of slab specimens and theoretically estimated ultimate load of a solid slab is
in the range of 1.36 – 1.46. Experimentally obtained ultimate load capacity is the same
645
12
for solid and voided slab. The same was observed in theoretical estimation as well.
Thus, the ultimate capacity of voided slab can be estimated by yield line theory similar
to a conventional solid slab.
Based on the experimental and analytical investigations of the biaxial voided slab
under two-way flexure, the following conclusions are drawn.
1. The solid and voided slab specimens showed typical two-way flexure behaviour with
major cracks in the form of X, similar to assumed yield line pattern.
2. The experimentally observed ultimate load carrying capacity of solid and voided
slab specimens were remain the same. The ratio of experimentally obtained ultimate
load capacity of slab specimens and theoretically estimated ultimate load of a solid
slab is in the range of 1.36 – 1.46.
3. The analytical investigation by yield line analysis showed that the ratio of the ulti-
mate load carrying capacity of the slab with sixteen-point load and uniformly dis-
tributed load is 89 %.
4. The analytical estimation of the ultimate load carrying capacity by yield line analysis
in congestion with IS 456 of solid and voided slab are equal. Thus, the ultimate ca-
pacity of voided slab can be estimated by yield line theory similar to a conventional
solid slab.
5. The initial stiffness of solid slab specimen is 37 % more in comparison with the
voided slab specimen, however, the secant stiffness corresponding to yield load of
voided and solid slab specimens were comparable.
6. The ratio of experimentally observed and estimated ultimate capacity of slab speci-
mens was observed to be in the range of 1.36 - 1.46. The influence of the orientation
and ultimate tensile strength of reinforcements on the enhancement of capacity of
slab need to be investigated, as researchers reported that these factors are influencing
the ultimate load carrying capacity of solid slabs [21, 22].
7. The strength enhancement due to tensile membrane action at large deflection in solid
slab was studied by various researchers [21, 23–25]. Similar kind of enhancement in
the voided slab need to be verified.
Acknowledgement
This work was supported by Department of Science & Technology, India

(SR/S3/MERC/0040/2012). The authors wish to acknowledge the assistance and facil-
ities offered by Technical Staff, Structural Engineering Laboratory, IIT Madras.
Notation
The following symbols are used in this paper:

Ast,b = area of bottom reinforcement;
Ast,t = area of top reinforcement;
C = force in concrete per unit width of slab;
d = overall depth of slab;
646
13
de = effective depth to bottom reinforcement;

d′ = effective depth to top reinforcement;
fck = concrete cube compressive strength;
fst,b = stress in bottom reinforcement;
fst,t = stress in top reinforcement;
l = overall length of slab;
le = effective length of slab;
m = ultimate moment capacity per unit width of slab;
mx = ultimate moment capacity per unit width of slab in x-direction;
my = ultimate moment capacity per unit width of slab in y-direction;
Pu = ultimate load (sixteen-point load) or observed ultimate load in experiment;
Tb = force in bottom reinforcement;
Tt = force in top reinforcement;
xu = neutral axis depth at ultimate;
x̅ = depth of resultant compressive force in concrete;
Wn = theoretical ultimate load (UDL) of solid slab based on YLA;
Wu1 = equivalent UDL of ultimate load without considering self-weight of slab;
Wu2 = equivalent UDL of ultimate load inclusive of self-weight of slab;
δu = mid-span deflection of slab corresponds to ultimate load.
References
1. BubbleDeck Technology: BubbleDeck Voided Flat Slab Solutions - Technical

Manual & Documents. , UK (2008)
2. Chung, J.H., Park, J.H., Choi, H.K., Lee, S.C., Choi, C.S.: An Analytical Study
on the Impact of Hollow Shapes in Bi-axial Hollow Slabs. In: FraMCoS-7. pp.
1729–1736. Korea Concrete Institute (2010)
3. Churakov, A.G.: Biaxial Hollow Slab with Innovative Types of Voids. Constr.
Unique Build. Struct. 6, 70–88 (2014)
4. Cobiax Technologies: Cobiax Engineering Manual. , Switzerland (2010)
5. Daliform Group: U-Boot Beton® System Study: Lightened Concrete Slab by
using U-Boot Beton®. , Italy (2014)
6. Ibrahim, A.M., Ali, N.K., Salman, W.D.: Flexural Capacities of Reinforced
Concrete Two-Way Bubbledeck Slabs of Plastic Spherical Voids. Diyala J.
Eng. Sci. 06, 9–20 (2013)
7. Kim, B.H., Chung, J.H., Choi, H.K., Lee, S.C., Choi, C.S.: Flexural Capacities
of One-Way Hollow Slab with Donut Type Hollow Sphere. Key Eng. Mater.
452–453, 773–776 (2011). doi:10.4028/www.scientific.net/KEM.452-453.773
8. Kim, S.H.: Flexural Behavior of Void RC and PC Slab with Polystyrene
Forms. Key Eng. Mater. 452–453, 61–64 (2011).
doi:10.4028/www.scientific.net/KEM.452-453.61
9. Sagadevan, R., Nageswara Rao, B.: Numerical Study on Flexural Capacity of
Biaxial Hollow Slab. In: Rao, A.R.M. and Ramanjaneyulu, K. (eds.) Recent
Advances in Structural Engineering, Volume 1. pp. 97–105. Springer
647
14
Singapore, Singapore (2019)

10. Taskin, K., Peker, K.: Design Factors and the Economical Application of
Spherical Type Voids in RC Slabs. In: International Scientific Conference
People, Buildings and Environment 2014, An International Scientific
Conference. pp. 448–458. , Kroměříž, Czech Republic (2014)
11. Valivonis, J., Jonaitis, B., Zavalis, R., Skuturna, T., Šneideris, A.: Flexural
Capacity and Stiffness of Monolithic Biaxial Hollow Slabs. J. Civ. Eng. Manag.
20, 693–701 (2014). doi:10.3846/13923730.2014.917122
12. Chung, J.H., Bae, B. Il, Choi, H.K., Jung, H.S., Choi, C.S.: Evaluation of
Punching Shear Strength of Voided Slabs Considering the Effect of the Ratio
b0/d. Eng. Struct. 164, 70–81 (2018). doi:10.1016/j.engstruct.2018.02.085
13. Daliform Group: U-Boot Beton® Disposable Formworks for Lightened
Structures in Reinforced Concrete Cast on Site. , Italy (2012)
14. Sagadevan, R., Nageswara Rao, B.: Analytical Studies on Flexural Capacity of
Biaxial Hollow Slab. In: Proceedings of the International Conference on
Composite Materials and Structures (ICCMS). , Hyderabad, India (2017)
15. Ibrahim, A.M., Ali, N.K., Salman, W.D.: Finite Element Analysis of
Reinforced Concrete Slabs with Spherical Voids. Diyala J. Eng. Sci. 6, 15–37
(2013)
16. Chung, J.H., Jung, H.S., Bae, B. il, Choi, C.S., Choi, H.K.: Two-Way Flexural
Behavior of Donut-Type Voided Slabs. Int. J. Concr. Struct. Mater. 12, (2018).
doi:10.1186/s40069-018-0247-6
17. Fall, D., Shu, J., Rempling, R., Lundgren, K., Zandi, K.: Two-way Slabs:
Experimental Investigation of Load Redistributions in Steel Fibre Reinforced
Concrete. Eng. Struct. 80, 61–74 (2014). doi:10.1016/j.engstruct.2014.08.033
18. Matešan, D., Radnić, J., Grgić, N., Čamber, V.: Strength Capacity of Square
Reinforced Concrete Slabs. Mater. Sci. Eng. 43, 399–404 (2012).
doi:10.1002/mawe.201200972
19. IS 456: Plain and Reinforced Concrete - Code of Practice. , Bureau of Indian
Standards, New Delhi (2000)
20. IS 1786: High Strength Deformed Steel Bars and Wires for Concrete
Reinforcement - Specification. , Bureau of Indian Standards, New Delhi (2008)
21. Wood, R.H.: Plastic and Elastic Design of Slabs and Plates. Thames and
Hudson, Londan (1961)
22. Taylor, R., Maher, D.R.H., Hayes, B.: Effect of the Arrangement of
Reinforcement on the Behaviour of Reinforced Concrete Slabs. Mag. Concr.
Res. 18, 85–94 (1966). doi:10.1680/macr.1967.19.58.58
23. Sawczuk, A., Winnicki, L.: Plastic Behavior of Simply Supported Reinforced
Concrete Plates at Moderately Large Deflections. Int. J. Solids Struct. 1, 97–
111 (1965). doi:10.1016/0020-7683(65)90019-3
24. Taylor, R.: A Note on A Possible Basis for A New Method of Ultimate Load
Design of Reinforced Concrete Slabs. Mag. Concr. Res. 17, 183–186 (1965).
doi:10.1680/macr.1965.17.53.183
25. Kemp, K.O.: Yield of A Square Reinforced Concrete Slab. Struct. Eng. 45,
235–240 (1967)
648
Paper ID- 130
Review on Limestone Calcined Clay Cements (LC3)

S. M. Gunjala,b* and B. Kondraivendhanc
a
Research Scholar, Applied Mechanics Department, S.V.National Institute of Technology, Surat 395007, India.
b
Department of Civil Engineering, Sanjivani College of Engineering, Savitribai Phule Pune University, Kopargaon-
423603, Maharashtra, India.
c
Applied Mechanics Department, S.V. National Institute of Technology, Surat 395007, Gujarat, India.
Abstract. Limestone calcined clay cement (LC3) is low budget, eco-friendly creative ternary blended cement. It is
made by using the combination of low grade calcined clay, limestone and gypsum. It can be replaced by 50% of
clinker which is beneficial for reduction of carbon dioxide (CO 2) emission at the time of production of cement. In
the existing literature the physical and chemical characteristics, sources of raw clay, methods used for calcination,
temperature effect on raw clays, mechanical properties of LC 3 i.e. compressive strength is checked with respect to
the portland pozzolana cement (PPC) and ordinary portland cement (OPC). Durability properties by different
researchers were studied such as rapid chloride penetration test, permeability and sorptivity. Production cost and
field application of this blended cement were checked. An Industrial trial was carried out in India, Cuba and
Switzerland which shows significantly good performance of all the above mentioned properties. In the present study,
a review of literature on limestone calcined clay cement carried out for the purpose of future research.
Keywords: Calcined clays; limestone; blended cement; compressive strength; durability.
1 Introduction
Day by day demand of concrete is increasing due to industrialization, globalization and infrastructure development.
Developing countries require a large quantity of concrete. Concrete is the homogenous mixture of cement, coarse
aggregate, fine aggregate, and water. For the production of 1 ton of cement, it emits nearly 1 ton of carbon dioxide
(CO2) into atmosphere [1]. As compared to the world, India is second highest cement producer, in which near about
280 million tons of cement is produced in India [2]. Annually from cement production nearly about 1.35 billion tons
of greenhouse gas releases, which is represented in the 2013 World Business Council for Sustainable Development
Energy Agency (WBCSDEA) [3]. Clinkers contain nearly about 20% of the supplementary cementitious material
(SCM). These are fine limestone, granulated blast-furnace slag (GBFS) and coal fly ash (FA). The GBFS and FA
sources are limited globally to only about 15–25% of cement consumption. Metakaolin is also used as a partial
substitution for cement in concrete [4]. A combination of metakaolin to limestone as 2:1 in portland cement (PC)
and the mixture was prepared, in addition of 15, 30, 45 and 60% of quartz in PC by Antoni et al. [5] blended cement
and calcined clay are used up to 30% obtained from low-grade kaolinite clay and limestone (addition of limestone in
cement ranges of 0-10%) cement are produced and the filler effect and pozzolanic effect on compressive strength,
hydration phases and pore size distribution were studied [6]. A recently developed alternative low-CO2 SCM system
uses optimized combinations of calcined clays with ground lime.
*
Corresponding author, Email:gunjalsachin20@gmail.com
649
2 Utilization of Cement in India and Other Industrial Nations

According to International Energy Agency (IEA), the future cement consumption for other industrial nations will
increase day by day up to 2050 [7]. Growth in cement production is higher in developing countries where the
population increases. Cement production in the whole world 10 years ago which is nearly equal to today’s
production in China. In India demand of cement has increase but consumption per capita is only 1/6th that in China
[3] as shown in Figure 1.
Fig. 1. Utilizations for future cement by region [3] .The lower section shows the evolution of cement production
distribution among different regions
The utilization of other building materials such as glass, asphalt, iron, ceramic, wood etc. is very less as compared to
cement and other cementitious materials as shown in Figure 2.
Fig. 2. Estimated utilization of common materials 2001-2005. Data from [7] except from ceramics asphalt and
cementitious, all other materials are shared with other industries
650
3 Limestone Calcined Clay Cement (LC3)

LC3 is an innovative blended portland cement contains clinker, calcined clay, low grade limestone and gypsum [8].
In which clinker, calcined clay, limestone and gypsum are 50%, 30%, 15% and 5% respectively are as shown in
Figure 3. The main invention in LC3 is to mix the easily found low-quality (percentage of kaolin content is less)
kaolinite clay with the substitution of 15% limestone without deduction in mechanical properties [9]. Scrivener et
al.[10] Studied the reaction of metakaolin and limestone in which alumina present reacts with limestone, to make a
low permeable material giving same compressive strength when higher replacement of clinker.
5%
15%
Clinker
Calcined Clay
50% Limestone
Gypsum
30%
Fig. 3. Percentage of LC3 content
4 Temperature Effect on Raw Clays

One of the SCM i.e. raw clays is abundantly available worldwide. Clays have sufficient proportion of kaolinite and
becomes highly pozzolanic if calcined between 700 oC to 850oC. A very reactive material is added in cement known
as metakaolin produced by calcining high purity kaolinitic clays [11]. Raw kaolinite clays produced an excellent
pozzolanic activity when clay is calcined at a temperature of 500oC to 800oC due to the creation of the active
amorphous phase called metakaolin [12]. Similarly, the calcination of good kaolinitic clays at temperature of 550 oC
to 900oC develops an amorphous silica compound aluminosilicate which is a highly reactive pozzolana. Tironi et al.
[13] investigated temperature required for calcination of clay A2, A3, A4, and A5 at 700 oC to complete conversion
of kaolin to metakaolin and A1 clay required the temperature of 800 oC for the calcination. Where A1 and A2
contain high kaolinite poor kaolinite i.e. 16% before calcination of clay.
The methods used for calcination of calcined clay are either by rotary and vertical shaft kiln for the small production
of cement. An oil fired shuttle kiln was used for the calcination process[14].The two clays A1 and A2 were used for
calcination using laboratory furnace fixed bed technique, the complete kaolinite conversion of clay was checked by
X-ray diffraction (XRD) [12]. Depending on sources of clay the chemical composition may vary. The calcined clay
which is used in making LC3 the chemical composition, calcination process and origin of clay used by different
651
authors are as listed in Table 1. The tests carried out on calcined clay such as pozzolanic activity, mineralogical
composition, and microstructural properties are as shown in Table 2.
Table 1. Chemical composition of Calcined clay, Calcination Process and Origin of Clay.
Sr Chemical SiO2 Al2O3 Fe2O3 CaO MgO Na2O K2O TiO2 LOI
No Composition (%)
1 Tironi et al.,2012
A1(94% Kaolinite 45.9 37.0 0.77 0.08 0.12 0.06 0.40 0.99 13.30
clay)
A2(76% Kaolinite 51.4 31.3 0.92 0.40 0.19 0.36 0.38 1.42 12.15
clay)
A3(65% Kaolinite 59.4 27.1 0.76 0.15 0.12 0.14 0.41 0.26 9.65
clay)
A4(48% Kaolinite 65.7 21.1 0.85 0.26 0.22 0.07 0.68 0.43 7.77
clay)
A5(16% Kaolinite 74.8 14.8 1.10 0.30 0.26 0.92 3.71 0.18 3.44
clay)
Calcination Process Laboratory furnace fixed bed techniques

Origin of Clay Argentina (La Rioja,Chubut and Santa Cruz)
2 Antoni,Rossen,Martirena, & Scrivener, 2012
Metakaolin 50.62 46.91 0.38 0.02 0.09 0.28 0.18 1.29 0.00
Calcination Process Flash
Origin of Clay Burgess
3 Tironi et al., 2014
A1(94% Kaolinite 45.9 37.0 0.77 0.08 0.12 0.06 0.40 0.99 13.30
clay)
A2(76% Kaolinite 51.4 31.3 0.92 0.40 0.19 0.36 0.38 1.42 12.15
clay)
Calcination Process Laboratory furnace fixed bed techniques
Origin of Clay Argentina (La Rioja,Chubut and Santa Cruz)
4 Bishnoi et al., 2014
70 to 80% kaolinite 43.30 36.35 2.26 0.46 - - - - 13.94
clay
20 to 30% kaolinite 55.78 17.46 8.89 4.84 - - - - 9.49
clay
Calcination Process Oil fired shuttle kiln
Origin of Clay West Bengal (eastern India)
5 (Vizcaíno-Andrés et al., 2015)
Sample 1 54.7 27.8 12.1 1.7 0.9 0.3 1.5 0.8 10.4
Sample 2 54.2 28.2 12.3 1.7 0.9 0.3 1.6 0.8 10.3
Sample 3 55.0 26.0 13.4 1.8 1.0 0.3 1.6 0.8 9.8
Calcination Process Wet process Rotary Kiln
Origin of Clay Pontezuela,Cuba
6 Anders L et al., 2015
Calcined Clay 48.4 29.5 16.5 0.4 1.0 0.0 0.7 0.6 1.5
Calcination Process Laboratory Oven
Origin of Clay Pontezuela, Cuba
7 Hollanders, Adriaens, Skibsted, Cizer, & Elsen, 2016
Kaolinite clay 1 45.82 38.79 0.56 0.01 0.05 0.03 0.20 0.42 14.12
Kaolinite clay 2 46.24 40.03 0.27 0.03 0.04 0.03 0.02 1.59 11.76
Kaolinite clay 3 44.77 38.45 1.12 0.02 0.06 0.05 0.01 2.23 13.29
Kaolinite clay 4 44.85 38.62 0.17 0.04 0.04 - - 0.01 16.27
652
Calcination Process Fixed bed electrical furnace

Origin of Clay Capim and Halloysite Industrial supplier
8 Emmanuel et al., 2016
Kaolinite clay 1 63.44 24.47 0.68 0.1 0.11 0.22 0.12 1.44 9.21
Kaolinite clay 2 54.67 27.69 4.93 0.06 0.13 0.25 0.12 1.68 10.28
Kaolinite clay 3 44.59 37.59 0.817 0.034 0.09 0.04 0.15 3.01 13.46
Kaolinite clay 4 43.95 36.92 1.29 0.1 0.13 0.03 0.23 3.44 13.66
Kaolinite clay 5 41.1 37.13 1.39 0.1 0.14 0.09 0.47 4.67 14.37
Calcination Process Rotary kiln
Origin of Clay Kutch Gujarat,India
9 Tironi, Scian, & Irassar, 2015
Kaolinite clay 65.7 21.1 0.85 0.26 0.22 0.07 0.68 0.43 7.77
Calcination Process Laboratory Oven
Origin of Clay Santa Cruz, Argentina.
10 Dhandapani & Santhanam, 2017
Calcined Clay 58.43 24.95 5.08 0.09 0.19 0.05 0.21 1.41 9.58
Origin of Clay Bhuj Gujarat,India
11 Dhandapani, Sakthivel, Santhanam, Gettu, & Pillai, 2018
Calcined Clay 58.43 24.95 5.08 0.09 0.19 0.05 0.21 1.41 9.58
Origin of Clay Bhuj Gujarat,India
Table 2. Different test carried out on calcined clay

Sr. Authors Pozzolanic Mineralogical Microstructural Properties of LC3
No activity composition Properties
1 Tironi et Frattini Test X-ray diffraction Mercury instrusion Compressive strength of
al., (2012) Electrical Fourier transformed porosimetry(MIP) Mortar.
Conductivity infra-red Blaine Fineness
spectrometry(FTIR)
2 Tironi et Frattini Test X-ray diffraction Mercury instrusion Compressive strength of
al.,(2014) Scanning Electron porosimetry(MIP) Mortar
Microscopy Scanning electron Blaine Fineness
microscopy(SCM)
3 Scrivener, Pozzolanic ---- Mercury Intrusion Compressive strength
(2014) reactivity Porosity.(MIP) Chloride Penetration
4 Bishnoi et Lime X-ray diffraction Standard consistency
al.,(2014) reactivity test Fourier transformed Initial setting time
(IS1727-1967) infra-red ---- Compressive strength of
spectrometry (FTIR) Mortar
Blaine Fineness
5 Vizcaíno- Pozzolanic X-ray fluorescence Mercury Intrusion Compressive strength of
Andrés et reactivity X-ray diffraction Porosity.(MIP) mortars.
al., (2015)
6 Anders L et X-ray fluorescence Mercury Intrusion Compressive strength of
al., (2015) X-ray diffraction Porosity.(MIP) mortars.
Isothermal Calorimetry
7 Hollanders Chapelle Test X-ray fluorescence Blaine Fineness
et al., (French norm X-ray diffraction
(2016) NF-P 18-513) Fourier transformed ----
infra-red
spectrometry (FTIR)
653
8 Emmanuel Lime X-ray fluorescence Mercury instrusion Standard consistency

et al., reactivity test X-ray diffraction porosimetry(MIP) Initial setting time
(2016) (IS1727-1967) Scanning electron Final setting time
microscopy(SCM) Compressive strength of
Isothermal Calorimetry Mortar .Blaine Fineness
Specific gravity
9 Tironi, Frattini Test X-ray fluorescence Mercury instrusion Compressive strength of
Scian, & X-ray diffraction porosimetry(MIP) Mortar
Irassar, Scanning electron
(2015) microscopy(SCM)
10 Dhandapan Lime X-ray fluorescence Mercury intrusion Compressive strength of
i& reactivity test X-ray diffraction porosimetry (MIP) and Mortar
Santhanam, electrical impedance
(2017) spectroscopy(EIS),
11 Avet & Pozzolanic X-ray fluorescence Mercury intrusion Compressive strength of
Scrivener, reactivity X-ray diffraction porosimetry (MIP) and Mortar
(2018) Scanning electron
microscopy (SCM).
12 Dhandapan Lime X-ray fluorescence Oxygen permeability, Compressive strength of
i et al., reactivity test X-ray diffraction rapid chloride Mortar and concrete,
(2018) penetration, chloride Standard consistency,
migration, resistivity Initial setting time,
development and water Final setting time,
sorptivity Blaine Fineness,
Specific gravity.
5 Comparison of LC3 with Ordinary Portland Cement

5.1 Production Process
Limestone and raw material are taken from quarry then broken into small pieces and fed into the rotary kiln to burn
up to 1450oC to create clinker which is after crushed and blended with gypsum to form OPC. For fly ash based
portland pozzolana cement, is fly ash added in the preparation of cement. By using same technology LC3 is
produced.The substitution of calcined clay and low quality of limestone blending together with clinker to form
LC3.The typical production process of ordinary portland cement (OPC), fly ash based portland pozzolana cement
(PPC) and limestone calcined clay cement (LC3) are as shown in Figure 4.
Process Map of OPC
Removal and
Transportation Preparation Preparation Preparation Packing and
of raw of Raw Meal of Clinker of Cement Dispatch
Limestone from
Quarry
Process Map of PPC
Removal and Preparation of Preparation of Preparation of Packing and

Transportation of Raw Meal Clinker Cement Dispatch
raw Limestone
from Quarry
Addition of
Fly ash
654
Process Map of LC3

Removal and
Transportation Preparation Preparation Preparation Packing and
Preparation
of raw of Raw of Clinker of Clay of Cement Dispatch
Limestone from Meal
Quarry
Addition of Calcined
Clay and Low quality
of Limestone
Fig. 4. Process Map for a Typical Cement Plant in India, OPC, PPC and LC 3
Following are the advantages of LC3
1. Emissions of CO2 in LC3 are considered to be 20-30% less than ordinary portland cement
because of decreased in clinker content leads to less emission of carbon dioxide in the
atmosphere. Crushing of limestone takes less energy than burning it. Calcination of clay takes
place at 800°C. Total 55% of the energy required for clinkerisation at 1450°C.
2. LC3 can be manufactured with presently available equipment, leading to only marginally
additional investments for calcination of clay.
3. Low cost of production, less clinker content, reduced fuel consumption for calcination
contributes to smaller production costs.
5.2 Clay Minerals Occurrence in India and World Soil Map
In India, kaolinite clays commonly known as china clays according to Indian Bureau of Mines in 22 states are
available [15]. India has estimated clay resources of about 2803 Million tons as per Indian Bureau of Mines report
(2014-2015 Mineral Year Book), .as is given in Table 3.
Table 3. Clay Stock in India Category Wise (Indian Bureau of Mines report (2014-2015 Mineral Year Book)
Type of Clay Existence Stock Probable stock in Possible stock in Total stock in Million
in Million Tons Million Tons Million Tons Tons
China Clay 124 53 2527 2704
Ball clay 12 4 67 83
Shale 15 0.5 0.5 16
Total 151 57.5 2594.5 2803
The reserve constitutes only about 7% of resources out of 208.5 Million tons in terms of proved and possible reserve
estimation system as per UNFC classification. China clay available in the state of Kerala is 25% similarly in West
Bengal, Rajasthan and Orissa States is 16% and Karnataka state is 10%.In case of Ball clay resource from Andhra
Pradesh Contributes 60% & 38% from Rajasthan state. World Soil Map is as shown in Figure 5
655
Fig. 5. World Soil Map. Kaolinite clays are yellow and pale green region likely occur on top surface [9]
Kaolinite clay most likely occurs near the top surface in the yellow and pale green region, Alfisols (pale
green) and Ultisols (yellow) as on the map. Clay is the weathering product from all type of rocks which are
abundantly available near the earth surface. Particularly kaolinite clay is available in plenty in India and South East
Asia.
6 Compressive Strength of LC3 as Compared to Other Cement

LC3-50 contains 50% of clinker, 30% of calcined clay, 15% of limestone and 5% of gypsum which give same
mechanical properties as compared to ordinary portland cement. The compressive strength of mortar (cement to
standard sand 1:3) with water to the cement ratio of 0.50 give good compressive strength than portland cement [8,
13]. As the percentage of kaolin in calcined clay increases the growth of compressive strength varies accordingly at
7 days and 28 days than the OPC. The early stage of compressive strength i.e. 1 day is smaller because of pozzolanic
reaction is slow at an initial stage and same behavior through a graph of compressive strength as shown in Figure 6
and 7. In India [14] first trial production of LC3-50 was carried out in the state of West Bengal (Eastern India) due to
the availability of clay. Clay 1 and 2 contain 70 to 80% and 20 to 30% of kaolinite content respectively. The
limestone A and B contain 36.29% and 42.84% of CaO was used in this study. The four blends of LC 3, A, B, C, and
D (blend A contain clay 1 and limestone A, blend B contain clay 1 and limestone B, blend C contain clay 2 and
limestone A, blend D contain clay 2 and limestone B.) were prepared and compared with OPC and PPC with a slight
change in water-cement ratio. The blend A and B have highest compressive strength than OPC and PPC while blend
C and D has more compressive strength than PPC at 3, 7 and 28 days. The compressive strength of concrete at the
water-cement ratio of 0.45, the LC3 A has the highest strength than all other blends, such as OPC and PPC at 3, 7
and 28 days [14].An Industrial trial was carried out in Cuba to form an LC 3 in which clinker is replaced by 50% of
SCM. Such as calcined clay and limestone. Andres et al. [16] studied the compressive strength of new production
cement is more than OPC at 3, 7 and 28 days. Second trial production of LC3 was carried out in India at Gujarat.
656
The clay which contain 58% kaolin content was used for the production of LC 3.The colour of this clay is reddish
because it contains high iron. Compressive strength of mortar containing LC3, water to cement ratio (w/c) taken as
0.35, 0.40, 0.45 and 0.50 .as compared to OPC .The Initial strength of LC 3 at 3,7,28 days are higher than that of
OPC, while the strength gain later age 56 days is less than OPC [17].
Fig. 6. Compresssive strength of OPC and LC3 Cement ranging from lower clay content to higher clay content.The
calcined clay originate from India,Brazil,Thailand and Cuba [9]
Compressive strength of LC3 at 28 days as compared to ordinary portland cement and 30% fly ash used in OPC i.e.
FA30 was higher result[18]. Mechanical properties and durability properties were studied [19] on OPC, FA30 and
LC3 with the addition of clinker 50%, calcined clay 31%, limestone 15% and gypsum 4% respectively. The
compressive strength of mortar at 28 days of OPC, FA30, and LC 3 is 61.0, 46.0 and 43.7Mpa. The compressive
strength of Concrete OPC- M30, OPC-M50, FA30-M30, FA30-M50, LC3-M30 and LC3-M50 was also checked.
Tironi et al. [13] studied compressive strength on mortars taking cubes (Size 25x25x25 mm) made with standard
sand (1:3) and w/c ratio of 0.50. The 5 clay A1, A2, A3, A4 and A5 were used having kaolinitic content 94,76,65,48
and 16% respectively. The Clay A2 strength is higher than OPC at 7, 28, and 90 days. Similarly, clay A1
compressive strength nearly equal to OPC and other clay has less compressive strength than OPC. Tironi et al. [12]
for blended cement two clay A1 and A2 replaced by 15% and 30% by mass of cement. Clay A2 has highest
compressive strength at 30% replacement than other clay at 2,7 and 28 days because of clay A2 has a high
pozzolanic activity which was carried out by Frattini test[20]. Antoni et al. [5] taken has metakaolin to limestone
657
ratio 2:1 and prepared four mixes using the combination of portland cement, limestone and metakaolin B15, B30,
B45 and B60 such that the replacement of portland cement 15, 30, 45 and 60% and adding metakaolin to limestone
ratio 2:1. The compressive strength of mortars compared to with reference portland cement The B15 has highest
compressive strength than other mixes and reference concrete at 1, 7, 28 and 90 days. The calcium carbonate reacts
with alumina present in metakaolin making important amounts of hemicarboaluminate and less amount of
monocarboaluminate at early as 1 day. Similarly, limestone and metakaolin absorb calcium hydroxide in the blends
which could be completely missing with the high level of substitution at past ages.
Fig. 7. Relation between calcined kaolinite content and compressive strength [9]
7 Durability of LC3
Durability is an important property to measure performance life of a structure. The detailed study on durability of
LC3 is going in Switzerland, Cuba and India [9]. The phase compositions of LC3 are very similar to Portland and
blended cement. Calcium silicate hydrate (C-S-H) is principal products after reaction, also forms aluminate phases
monohemicarbonate and ettringite in limestone cement. LC3 has a small porosity as compare to other blended
cement as shown in Figure 8 [9] by using mercury instrusion porosimetry.
658
Fig. 8. Mercury Instrusion Porosity results for LC3blend and reference Portland cement [9]
Similarly, LC3 kept in 0.5M NaCl solution shows good resistance to penetration of chloride iron after two years of
ponding as shown in figure.9 [9]
Dhandapani and Santhanam studied pore structure development by using mercury instrusion porosimetry (MIP) as
compared to Portland cement, FA30 and LC3.In that LC3 binder has significantly improved pore structure. Also,
rapid chloride permeability test (as per ASTM C1202) for LC3 binder was done and observed that reduced
permeability as compared to OPC and FA30.Similarly, LC3 system has critical pore size evaluation and conductivity
greater improvement was in 7 days.X-ray diffraction prove that at early age calcium hydroxides consume due to the
sensitivity of calcined clay monocarboaluminate and hemicarboaluminate phases were seen at early age beneficial to
microstructure development which reduces the porosity [18].
Fig. 9. Chloride profiles after 2 years ponding in 0.5 NaCl solutions [9]
659
8 Field Application of LC3

Using LC3 first trial production carried out in India for testing of concrete in the laboratory as well as in the field.
The various application is such as microcracking roofing tiles, RCC door and window frame, hollow concrete
blocks, low duty paving blocks and construction of two storey houses[14] are as shown in Figures 10 and 11.
Industrial trials were carried out in Cuba for the manufacture of hollow concrete blocks, many cubic meters of
25MPa concrete used under standard condition[16].
Fig. 10. Micro concrete roofing tiles produced using different blends of LC 3 [14]
Fig. 11. Construction of two story house using LC3 [14]
660
9 Production cost of LC3

Last few years various studies have been going on LC3 [7, 12]. It is observed that LC3 is more economical than fly
ash based portland pozzolana cement in India [21]. Following point is to be consided for economical production of
LC3:
1. Using combination of limestone and calcined clay 50% of reduction in clinker is possible and lower energy
is required for burning.
2. Crushing limestone takes lower energy than burning it
3. At 800oC calcination takes place and nearly 55% of energy required for clinkerisation.
4. Existing equipment can be used.
In India high kaolin content clay known as china clay available at least in 22 states [22]. Economical production of
LC3 with respect of OPC and PPC depends upon the cost, availability and quantity of fly ash used.
10 Concluding Remarks
With low capital investment, alternative cement production is possible in developing country which faces an
increase in population, infrastructure development and greenhouse gas emissions. This article tries to present a
review of sources of raw clay, calcination methods, compressive strength, durability properties, chemical
composition, field application and physical performance of limestone calcined clay cement.
The most of the work in LC3 is going on to study durability, transport properties, corrosion due to carbonation and
chloride ingress, alkali-silica reaction, sulphate attack and leaching also, mechanical properties, flexure strength,
compressive strength, tensile strength and creep and shrinkage However there is no work on high performance
concrete (HSC) and ultra-high performance concrete (UHPC), reported in the literature. Therefore, there is a lot of
scope for studying these properties of HSC and UHPC in near future.
Acknowledgement:
Authors are thankful to Sanjivani College of Engineering Kopargaon-423603,Maharashtra,India,for providing
technical facilities to complete the present research study.
REFERENCES
[1] Gartner, E.: Industrially interesting approaches to ‘low-CO2’ cements. Cem. Concr. Res. 34,
1489–1498 (2004). doi:10.1016/j.cemconres.2004.01.021
[2] Planning Commission. Low Carbon Strategies for Inclusive Growth. Government of India, 2011.
http://planningcommission.gov.in/reports/genrep/index.phprepts=report_carbon.html
[3] Initiative, WBCSD Cement Sustainability Getting the Numbers Right, Project Emissions 2014,
Report, 2016
[4] Sabir, B., Wild, S., Bai, J.: Metakaolin and calcined clays as pozzolans for concrete: A review.
Cem. Concr. Compos. 23, 441–454(2001). doi: 10.1016/S0958-9465(00)00092-5
[5] Antoni, M., Rossen, J., Martirena, J., Scrivener, K.: Cement substitution by a combination of
metakaolin and limestone, Cem. Concr. Res. 42, 1579–1589 (2012).
doi:10.1016/j.cemconres.2012.09.006
[6] Tironi, A., Scian, A.N., Irassar, E.F.: Blended Cements with Limestone Filler and Kaolinitic
661
Calcined Clay: Filler and Pozzolanic Effects. J. Mater. Civ. Eng. 29, 195–201(2017).
doi:10.1007/978-94-017-9939-3_24
[7] IEA, Energy Technology Perspectives 2015 Mobilizing Innovation to Accelerate Climate Action,
Energy Technol., 2015. http://www.iea.org/etp/etp2015
[8] Scrivener, K., Martirena, F., Bishnoi, S., Maity, S.: Calcined clay limestone cements (LC3), Cem.
Concr. Res. (2017). doi:10.1016/j.cemconres.2017.08.017
[9] Scrivener, K.L.: Options for the future of cement. Indian Concr. J. 88, 11–21(2014).
http://www.lc3.ch/wp-content/uploads/2014/09/0851_ICJ_Article.pdf
[10] Limestone Calcined Clay Cements. https://www.lc3.ch
[11] Fernandez, R., Martirena, F., Scrivener, K.L.: The origin of the pozzolanic activity of calcined
clay minerals: A comparison between kaolinite, illite and montmorillonite. Cem. Concr. Res. 41,
113–122(2011). doi:10.1016/j.cemconres.2010.09.013
[12] Tironi, A., Castellano, C.C., Bonavetti, V.L., Trezza, M.A., Scian, A.N., Irassar, E.F.: Kaolinitic
calcined clays - Portland cement system: Hydration and properties. Constr. Build. Mater. 64, 215–
221(2014). doi:10.1016/j.conbuildmat.2014.04.065
[13] Tironi, A., Trezza, M.A., Scian, A.N., Irassar, E.F.: Kaolinitic calcined clay: Factors affecting its
performance as pozzolance. Constr. Build. Mater. 28, 276–281(2012)
[14] Bishnoi, S., Maity, S., Mallik, A., Joseph, S., Krishnan, S.: Pilot scale manufacture of limestone
calcined clay cement : The Indian experience, Indian Concr. J. 88, 22–28(2014)
[15] Indian Minerals Handbook, Ministry of Mines, Government of India, 2011, 2011
[16] Vizcaíno-Andrés, L.M., Sánchez-Berriel, S., Damas-Carrera, S., Pérez-Hernández, A., Scrivener,
K.L., Martirena-Hernández J.F.: Industrial trial to produce a low clinker, low carbon cement,
Mater. Construcción. 65, e045 (2015). doi:10.3989/mc.2015.00614
[17] Arun, S.B., Emmanuel, C., Haldar, P., Maity,S.: Second pilot production of limestone calcined
clay cement in India: The experience. Indian Concr. J. 90, 57–64(2016)
[18] Dhandapani, Y., Santhanam, M.: Assessment of pore structure evolution in the limestone calcined
clay cementitious system and its implications for performance. Cem. Concr. Compos. 84, 36–47
(2017). doi:10.1016/j.cemconcomp.2017.08.012
[19] Dhandapani, Y., Sakthivel, T., Santhanam, M., Gettu, R., Pillai, R.G.: Mechanical properties and
durability performance of concretes with Limestone Calcined Clay Cement (LC3). Cem. Concr.
Res. 107, 136–151 (2018). doi:10.1016/j.cemconres.2018.02.005
[20] BS EN 196-5:1995, British Standard Methods of testing cement Part 5: Pozzolanicity test for
pozzolanic cements
[21] Maity, S., Joseph S., Bishnoi, S.: An economic analysis of the production of limestone calcined
clay cement in India, Indian Concr. J. 90, 22–27(2016)
[22] Indian Bureau of Mines. Indian Minerals Yearbook 2012, Part-III: Mineral Reviews, Government
of India. 2011
662
Paper ID- 131
Performance issues and durability criteria of concrete

structures under marine exposed condition: A global trend of
research
Rahul Shah
Assistant Professor
School of Engineering & Technology
Navrachana University
Vadodara, Gujarat, India
rahulshahedu@gmail.com
J.D.Rathod
Associate Professor
Faculty of Technology & Engineering
The M.S.University of Baroda
Vadodara, India, Gujarat
Abstract : The infrastructure development has crossed the boundaries of land across the world since
many decades. The shoreline and off shore structures are mainly constructed using reinforced
cement concrete. Under the marine environment, performance of concrete has been very
different especially the durability aspects when compared to inland structures. Many factors
including wave height, exposure conditions and zones, chloride transport mechanism,
temperature and deterioration mechanism etc also play a crucial role in performance of
concrete. Around the world different codal provisions and criteria govern the design of the
structures. Researchers across the world are trying to establish the relationship of deterioration
rate and service life prediction by incorporating all the factors and aspects in separate
conditions.
This paper is an attempt to address factors and issues related to concrete performance under
the marine environment considering parameters and criteria for improvement of durability
performance using current trend of global research.
Key words: Marine environment, durability of concrete structures, chloride ingress,
carbonation, supplementary cementitious materials
663
Introduction: The construction sector is also one of the prime sector which is developed parallel to
industrialization across the world. The construction on land is not a only place where the humankind’s
visualization and creation has seen the reality in terms of infra structures. The requirement of all kind of
structures supporting naval navigation, industries along coastline has emerged has large construction on
sea shore or in the sea itself.
From last 50 to 60 years concrete, because of its great properties and economy is the first choice of
general constructions. Most of the sea-line structures may be commercial or industrial structures are
made up of RCC as a prime constituent. Apart from all beneficial aspects, the concrete shows
deterioration when it is subjected to continuous contact of marine atmosphere. Several researchers
around the globe are determining the root causes of deterioration and methods to prevent the damage
to the structures due to marine atmosphere. The global research suggests several factors responsible for
deterioration of marine concrete. The factors atmosphere, exposure, transport and other material
factors as well.
Factors accountable to marine effects on concrete:
Exposure conditions: The complex interplay of physical and mechanical mechanism of chloride ingress
and sulphate play a crucial role for deterioration of the concrete. The long term durability of concrete
structure primarily depends on the exposure condition of the concrete under the atmospheric
conditions. The major exposure conditions are atmospheric zone, tidal zone and submerged zone.
In case of atmospheric zone the major cause of deterioration is chloride induced corrosion to steel
reinforcement.[6] The airborne chloride generally attacks concrete and chloride induced corrosion
starts, although carbonation induced corrosion may take place based on some favorable conditions like
adequate relative humidity and temperature, but its amount is significantly smaller compared to
chloride induced corrosion. In case of concrete exposed to splash or tidal zone, which is generally
considered as poor and most detrimental condition considering all exposure conditions. Both
mechanical and chemical processes like diffusion, erosion and sorption etc. take place and may lead to
major deterioration like salt crystallization, erosion etc [7]. For exposure conditions of fully submerged
condition, concrete deteriorate due to more of chemical processes like sulphate attack under the effect
of permeation.
664
Transport Mechanism: The mechanism of transferring salt crystals from sea to concrete structure is
generally performed by absorption, capillary and permeation action. Several factors like splash zone,
height of structure, distance, water table are responsible to in the process. To predict the mathematical
model of transfer rate is complex and more often the empirical formulas are generated considering the
localized condition and experimental approach.
Chemical mechanism: The chemical composition of the sea water generally include several ions like Na+,
Mg+2, Cl-, So3- and other dissolved salts [2]. The major predominant ion is chloride among all ions. Due to
contact of such ions with cementitious compounds, some of original compounds like calcium hydroxide
turns to gypsum, CSH to MSH and CAH converts to ettringite. These all secondary products affects the
chemical and physical composition of the concrete and lead to accelerate physical process of
permeation and chemical process of de-pessivation of steel and sub sequent corrosion [4]. The
understand the phenomenan the diffusion of chloride can be modeled by non steady state condition
using Fick’s second law or some empirical approach suggested by global researchers. Some experimental
investigation has suggested use of sulphate resistant cement but it may not much useful to counter the
chemical mechanisms.
Temperature effects: Temperature has impact-full effects on marine concrete subjected to exposed
conditions. Lower temperature accelerate freeze thaw cycles and increase the deterioration of concrete
[1]. The increased wetting drying cycle or temperature change cycle increases the frost attack and lead
to crack progression due to increased volume of freezed water. This action generally accelerates the
chances of chloride ingress.
Role of material and mix design : The selection of material selection starting from aggregates to cement
play a vital role in performance of concrete. Several researchers have investigated role of
supplementary cementitious materials like metakoline and silica fume in concrete and received positive
results in preventing the chlore ingress in the concrete [7]. More refined structures of pores of the
microstructures help in restriction of formation of etringite and other byproduct of secondary reaction
of cementtitous materials [1]. The mix design approach also has very pivotal role in meeting the
durability requirements. Instead of providing design requirements, performance based approach should
be adopted considering the local material and atmospheric conditions [11].
Codal guidelines: Despite of high importance of requirement most of the design standards do not
provide in depth guidelines for area of concern. IS:456-2000 only limits up to classification of marine
665
environmental conditions and provision of concrete cover. Where as, ACI 318 M 08 and BS 8000- 2006
only additionally provide design life cycle apart from cover requirements and environmental
classification. AS 3600-2009 provides more information definition of coastal zones based of distance
from sea shore and definition of tidal and spray zones apart from general criteria [6]. Considering the
very serious issues of diffusion, rate of carbonation, temperature effects more details standards are
highly awaited.
Current and expected future research trends: Considering the current scenario of literature
availability, codal guidelines, experimental research and need of the hour of the requirement to
understand the processes of deterioration following listing is proposed.
Table 1 : Impact of parameters
Sr No Parameters Challenge Impact in deterioration

1 Exposure conditions 1. Rate of chloride ingress Very High
in various zones
2. Complex interplay
between chemical and
physical mechnism
2. Transport Mechanism 1. Determination of rate Very High
of transfer
2. Empirical approach
against mathematical
model
3. Chemical and Physical Mechanism 1. Issues of Ci ions Very High
ettringite formation
process , steel corrosion
2. Salt crystals in splash
zone
4 Materials and perspective 1. Role of SCM in concrete Moderate
approach 2. Role of aggregates
5 Codal provisions 1. Mix design of marine Moderate
concrete
666
2. Chloride ingress and

carbonation prediction
6 Temperature 1. Freeze thaw cycles Low
2. Salt crystels
Considering the above factors and current issues, the requirement from global research and challenges
can be highlighted.
Conclusion: The following conclusions can be drawn based on present study related to issues and
challenges
1. Service life prediction of concrete structures subjected to marine conditions is major concern
for engineers and scientists. Though several methods are developed but they are localized and
based on empirical approach. More emphasize must be given to issues like chloride diffusion
and carbonation rate in the concrete.
2. There is a huge scope for perspective mix design, water binder ratio, checking the usability of
SCM and chloride threshold values.
3. Prediction of transport mechanism by mathematical model, relationship between physical and
mechanical mechanism is still to be discovered thoroughly.
References:
1. Neville A.M., 2013. Properties of Concrete, 5th edition, Pearson

2. Galan ,Perron L., Glasser F.P.,2015,Impact of chloride rich environment on cement past
mineralogy .Cement and concrete research 68,147-183
3. Lea, F.M., 1970. Chemistry of cement and concrete. John Wiley& Sons,New York
4. Angst,U.,Elsener,B.,Larsan ,C.K.,Vennesland ,2009.Critical chloride content in reinforced
concrete – A review. Cement and concrete research 39 (12),1122-1138
5. De Weerdt,K.,Justnes H.,Geiker,M.R.,2014.Changes in phase essemblenge of concrete exposed
to sea water. Cement and concrete composites 47 ,53-63
6. Santhanam, M.,Otieno,M.,2016. Deterioration of concrete in marine environment, Marine
concrete structures (5), 137-149
667
7. Thomas, M., 2016, Durability of concrete for marine constructions: Materials and properties.
Marine concrete structures (6), 137-149
8. Moffat, E.G., Thomas,2018. Performance of 25 years old silica fume and fly ash light weight
concrete in harsh marine environment, Cement and concrete research. In press
9. Chalee,W., Ausapanit, P., Jaturapitakkul, C., 2009. Utilization of fly ash concrete in marine
environment for long term design analysis, Materials and design 31,1214-1249
10. Medeiros,M.H.F., Gobbi,A.,Reus,G.,C.,Helene,P.,2014.Reinforced concrete in marine
environment: Effect of wetting and drying cycles, height and positioning in relation to sea shore.
Construction and building materials 44,452-457
11. Nanukuttan,2015.Performance of concrete exposed to marine environment: Predictive
modeling use of laboratory/insitu methods. Construction and building materials, In press
668
Paper ID- 135
Sustainable Mortar Production by Utilizing Post Consumed PET Waste

and Copper Slag
Satish M. Waysal1 (orcid.org/0000-0001-5737-9494), Dr. Yogesh D. Patil2, Dr. Bharatkumar Z. Dholakiya3, Akshay M
Bhoi4
1,4
Research Scholar. Applied Mechanics Department, Sardar Vallabhbhai National Institute of Technology,
Surat, Gujarat, India, e-mail: satish.waysal@gmail.com
2
Assistant Professor in Applied Mechanics Department, Sardar Vallabhbhai National Institute of Technology,
Surat, Gujarat, India
3
Assistant Professor in Applied Chemistry Department, Sardar Vallabhbhai National Institute of Technology,
Surat, Gujarat, India
Abstract
Plastic, one of the most significant innovations of the 20th century, is a ubiquitous material. PET is a kind of
polyesters made of the ethylene glycol and terephthalic acid’s composition and its chemical name is Polyethylene
Terephthalate or ‘‘PET’’. PET is one of the most widely used plastics in the packaging industry because of high
stability, high-pressure tolerance, non-reactivity with substances and great quality of gas trapping which can
preserve the gas in the gaseous drinks. Incineration PET bottles may produce poisonous gasses. Another problem
arises from the fact that these materials slowly decompose and they need hundreds of years to return to the cycle
of nature. So it seems that recycling is the best way because of environmental compatibility and economic benefits.
Recycling of polyethylene-terephthalate (PET) to produce materials like concrete or mortar is one of the best
solutions for disposal of such waste plastics.
Cement is the most extensively used construction material. There is an increasing demand for the Cement in the
construction industry due to rapidly increasing civilization. With the increasing demand for construction materials,
the natural resources are depleting worldwide. On the other front, PET waste and copper slag, i.e., industrial waste
generated in the copper industry are increasing substantially and polluting the environment to a great extent. If we
could use PET waste and copper slag in a construction, it will be a great relief for the environment. In this work,
waste PET bottles were used to produce a PET resin by glycolysis in Polyethylene glycol (Molecular weight 100).
These resins were prepared with a ratio of 30:70 PET to glycol and were added in proportion 5%, 10%, and 15%
by mass of cement in a mortar. Fine aggregate is replaced with 20%, 40% 60% and 80% by copper slag. The
effect of this replacement on standard consistency, setting time, soundness of cement together with slump flow, a
compressive strength, tensile strength, and abrasion resistance of mortar is studied. It is observed that the addition
of PET resin and copper slag suggest that PET waste and copper slag can be potentially utilized in the production
of mortar.
Keywords: - PET Resin, Copper Slag, Cement mortar, Sustainability
1 Introduction
Plastics utilization these days has turned into an essential part of our lives. Among all the available forms of plastic
one of the mostly utilized form is Polyethylene Terephthalate (PET) is a sort of plastic that is typically utilized in
669
manufacturing of plastic bottles [1]; because of its advantages such as non-reactive with substance, light in weight,
high temperature resistance excellent gas tapping quality for beverages[2] . The amount of consumption of PET
is increased drastically which increasing the problem of solid waste disposal. A most preferred way for disposal
to this land-fill but as PET is non-biodegradable material disposal by landfill means preserving hazardous material
forever because degradation of this waste is very slow [3]-[4]. They may block the drainage system of the city,
will not allow rainwater to percolate in the ground causing into the reduction of the groundwater table. Most of
the floating matters in sea contain major part is a plastic waste. The “Great Pacific Garbage Patch” consist of 90%
of plastic waste among all the floating matter. In 2006 UN environment programme estimated that the ocean
contains 46000 pieces of plastic in per square kilometer area [5]. Most of the sea animal dies every year because
of eating or entangled in plastic pieces due to this it is very harmful to aquatic life. In the recent Kerala flood, it
is observed that whatever floating matters major constitute of this is plastic. Incineration process for disposal of
this waste could be dangerous because it releases a various poisonous chemical in the environment which are
harmful for environment. At present use of plastic cannot be avoidable especially PET waste. Reusing of PET
bottles is dangerous for human life it would be one of the causes of a disease like cancer. Recycling of this waste
is one of the ways to relif the environment. The study has been carried out on the use of waste plastics as concrete
and mortar constituent. The PET waste is utilized as replacement of constitutes of concrete such as fine aggregate
and cement.
The inclusion of PET fibers in the concrete is advantageous to control of plastic shrinkage cracking. The plastic
shrinkage cracks are widely explicated in a mortar, which has a large surface of exposure [6]. When surfaces are
subjected to wetting and drying or freeze-thawing plastic shrinkage cracks are developed resulted into the rapid
ruin of structure.[7] PET fiber has a low elastic modulus which is sufficient to give high performance to restrain
cracks propagation in a mortar [8]. Fibers in mortar act as uniformly distributing reinforcing material against the
crack enlargement due to plastic shrinkage. Propagation of microcracks and macrocracks is a severed issue in
hardening mortar can be arrested by providing uniform distribution of PET [9]. PET fibers present in mix act as
a bridge and stop the propagation of microcracks. Free shrinkage and restrained shrinkage tests have evaluated
fibers capacity to reduce plastic shrinkage [10]-[11]. The diminish in free shrinkage does not necessarily indicate
a reduction in overall cracking tendency. The cracking sensitivity of harden mix is primarily controlled by
enhanced toughness owing to the incorporation of fibers and shrinkage deformation [12]. The toughness plays a
very important role during the performance in static, dynamic and fatigue actions [13]. The properties of concrete
and mortar get altered because of The use of latexes or resins as a polymer. Reduction in permeability, increase
in flexibility of concrete property are the advantages of polymer material. The dosage of the polymer is between
10% to 20% of the mass of cement. This resin could be utilized effectively at concentrating at paste and aggregate
interface.[14]
Concrete is one of the widely used construction material in all types of construction worldwide because
of its advantages such as ease of construction, strong, durable. Concrete constitutes cement, fine aggregate, and
coarse aggregate; 70 to 75% volume of concrete is occupied by coarse aggregate and fine aggregate. The easiest
available natural source for of fine aggregate is a river; due to rapid growth in urbanization, these natural resources
are depleting. Nowadays availability of fine aggregate is big question due to this construction sector is facing a
tremendous problem. On the other hand due to growth in industrialization more waste generated day by day.
Aluminum, Copper, and Steel are the most necessary materials in today’s world. Copper production process
670
generates the copper slag in a significant amount [15]. Nearly 13 million tonnes generated every year. Around
two-tonne copper slag is generated in production each tonne of copper which needs land to disposal. Using copper
slag as a substitute for fine aggregate could be the solution for the construction sector and environment. A lot of
research is carried out on copper slag as a substitute of copper slag as sand and cement. The pozzolanic activity
index for copper slag is 87% according to ASTM requirement 618 °C for portland cement mineral additives [16].
Excellent mechanical and physical properties can be achieved by incorporating the copper slag as fine aggregate.
Use of silica fumes and copper slag has shown better strength [17]. The workability of high performance concrete
for water cement ratio of 0.35 with super plasticizer increases as an increase in copper slag substitute for sand.
The compressive strength at 28 days curing goes on increasing up to 50% replacement then starts decreasing [18].
Up to 15% of copper slag weight of cement and 1.5% hydrated lime as an initiator for pozzolanic reactions can
be used in portland cement replacement [19]-[20]. Addition of copper slag as cement substitute has shown
reduction in Permeability, chloride ion permeability, alkali-silica reaction, and increase sulfate resistance [21].
Copper slag has a smooth, glossy surface texture, low water absorption; excellent compressibility improves the
workability and dynamic behavior. Presence of excess water at higher fineness adversely affect on properties of
concrete[22]. The researcher has shown that copper slag can be used as fine aggregate.
In this research work, PET resin is prepared with PET to Glycol ratio (30:70) and is used for substitute
of cement in 5%,10% and 15% by mass of cement and copper slag is added 20%, 40% and 60% as a substitute
of fine aggregate. The effect of PET resin on consistency, soundness, initial setting, final setting, is studied. The
effect of copper slag and cement mortar flow test, compressive strength, tensile strength, and abrasion resistance
is studied. Authors have attempted to check the feasibility of utilizing copper slag and PET in cement mortar.
2 MATERIALS
2.1 Cement
Ultra Tech 53 grades Ordinary Portland cement is used for this study the specific gravity of cement is 3.15
2.2 Fine aggregate
Fine aggregate used in this research work is procured from Tapi river Saranghkheda, Maharashtra, India. Physical
properties are shown in Table 1
2.3 Copper Slag
Copper slag used in this research work is produced from Birla Copper Hindalco Industries, Dahej, Gujrat, India.
Physical properties are shown in Table 1and Chemical properties shown in Table 2
Table 1 Physical properties of Fine Aggregate
Sample Specific gravity Water absorption Zone as per IS383
Fine aggregate 2.62 1.10 Zone I
Copper Slag 3.40 0.10 Zone I
671
2.4 Water
Potable water is used in casting and curing.
Table 2 Chemical Composition of copper slag
Chemical Composition Copper Slag %

SiO2 + Al2O3 + Fe2O3 90.09
SiO2 33.85
Al2O3 2.79
Fe2O3 53.45
CaO 6.06
MgO 1.61
SO3 1.89
Na2O3 0.28
TiO2 0
Mn2O3 0.06
CI 0.01
Loss on ignition 0
2.5 PET Resin
PET Resin was obtained by glycolysis process with 30:70 PET to glycol ratio in DEG (Diethylene Glycol).
Glycolysis of waste PET was done using DEG having 100 molecular weight [23] along with 0.5% zinc acetate as
a catalyst. A three-necked glass flask with condenser and stirrer was used to conduct the reaction. The ingredients
were heated at 180°C for an hour; the temperature was further elevated till disappearance of all the solids at 210°C
[24].
2.6 Mix Proportion
Mix proportion and replacement of material for this work is ash shown in table
Table 3 Mix proportion for mortar (1:3) in kg/m3
PET resin% Weight of PET Resin Fine Pond ash % Pond ash weight
Cement
Aggregate
788 0 - 1965 0 -
748 5 12.5 1768 20 510
714 10 25.0 1014 40 1020
888 60 1530
674 15 37.5
24 80 2040
672
2.7 Experimental Programme
Standard consistency, Initial setting time (IST) and final setting time (FST) is determined.as per IS 4031-1968
and IS 269-1976 Soundness is determined as per IS 4031-1968 and IS 269-1976. The mix ratio for normal cement
mix is 1:3 (Cement: Sand) and water of quantity ( + 3) percent of the combined weight of cement and sand mix
chosen for determination of compressive and tensile strength and abrasion resistance. PET resin is added in normal
cement mix with 5%, 10%, and 15% by weight of cement for 30:70 PET to glycol ratio and fine aggregate (FA)
is replaced with copper slag from 0%, 20%, 40%, 60%, and 80%. The flow table test is carried out according to
IS 4031 part IV 2005. The compressive strength of cement mortar is determined as per IS 4031 part IV 2005. The
tensile strength of mortar is determined as per IS 269 2013, ASTM C307 and IS 4031 part 8. Styrene monomer is
added in PET resin with ratio 1:1 to reduce the viscosity and Methyle ethyle ketone peroxide (MEKPo) and Cobalt
napthlate is used as an initiator for PET resin [24]. Abrasion test is carried out for mortar on abrasion testing
machine for this test the specimen of size 70.7mm x 70.7mm x 20mm were casted and tested after 28 days of
curing. In this research work for compression test 135 specimen, for tensile strength 135 specimen and for
abrasion test 45 specimen were casted an tested.
3 Result and discussion
3.1 Standard Consistency
The standard consistency is water required to complete the hydration process of cement paste. The water required
for standard consistency of the cement was slightly reduced with an increase in the addition of PET resin in the
cement paste. The water requirement is reduced up to 3.32%, 9.67%, and 12.90 % respectively for 5%, 10%, and
15% PET resin content as compared to normal cement content and same is shown in Table 4
Table 4 Cement Properties
Sr. No. Specimen IST in FST in Minutes Soundness in mm Consistency in %

Minutes
1 Cement 120 490 3 31
2 5% Resin 110 420 2 30
3 10% Resin 95 390 2 28
4 15% Resin 90 350 2 27
3.2 Soundness of Cement
It is essential that cement paste once set should not undergo a large change in volume if there is a change in
volume that will be affecting to structure. A small change in the expansion of the cement was found due to the
addition of PET resin 5%,10% and 15% PET resin content. as shown in Table 4
3.3 Setting Time
Addition of PET resin with curing agent affects the setting time of cement. Initial setting time (IST) is decreased
by 10, 25 min and 30 minutes respectively for 5 %, 10% and 15% PET resin content as compare to normal cement
673
paste. Reduction in initial setting time leads into the reduction of time for transportation of concrete and concreting
work. Final setting time (FST) is also decreased by 70, 100 and 140 minutes in comparison with normal cement
paste respectively for 5 %, 10% and 15% PET resin content same is shown in Table 4. This reduction in final
setting time will help for retrofitting pavement fast track maintenance work. Setting time is mainly caused by
hydration of C3A and C3S and accompanied by temperature rises in the cement paste initial set correspondence
rise of temperature final set correspondence to the peak temperature. Addition of PET resin along with curing
agent increases the temperature rise at the initial and peak level due to this there is a reduction in the initial and
final setting time.
3.4 Flow table test on mortar
The mortar flow test is important considering the workability of mortar especially when the nonconventional
material is used in mortar production. It was observed that there is an increase in slump flow of mortar with the
addition of PET resin and copper slag content as shown in Figure 1. Maximum slump flow was observed for 15%
PET resin content and 80 % of copper slag. Slump flow increases 33.75 % to 56.25 % , 36.25% to 58.75% ,
51.25% to 78.75% and 61.25% to 86.25 % for 5% ,10% and 15% PET resin content for 20% , 40%, 60% and
80% copper slag content respectively as compare to normal cement mortar. Slump flow increase because of the
copper slag particle shape and ball bearing effect of copper slag. Copper slag has very less water absorption it also
contributes in slump flow of mortar.
160
140
Slump Flow in mm
120
100
80
60
40
20
0
Cement 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15%
Resin Resin Resin Resin Resin Resin Resin Resin Resin Resin Resin Resin
0 20% 40% 60% 80%
PET resin and Copper slag in %
Figure 1Slump flow
3.5 Compressive strength
The effect on compressive strength for substitution of PET resin and copper slag as cement and fine aggregate
respectively over 1,7, and 28days curing age is as shown in Figure 2. The result has shown that there is increase
in compressive strength up to 60% of replacement of copper slag for 5%, 10%, and 15% PET resin content at all
curing ages. Among all the PET resin content 15% of PET resin dosage has shown good impact over compressive
strength. After one day curing compressive strength for 15% PET resin content is increased by 3.54, 3.89, and
4.01 times for 20 %, 40% and 60% copper slag respectively as compare to normal cement mortar. A similar trend
is observed for 5% and 10% PET resin content up to 60% of replacement of coppers slag. Compressive strength
674
at one day curing is nearly 90% of 28 days compressive strength for normal cement mortar. Therefore, the PET
mortar could be used as a repair material for maintenance work. Beyond 60% replacement of coppers, slag has
shown a reduction in compressive strength for 80% of coppers slag replacement the compressive strength is
increased by 3.09 times for 15% resin content as compare to normal cement mortar, but it is less by 22.88 % for
60% of copper slag replacement for 1 day curing.
After 28 days of curing compressive strength for 15% resin content is increased by 3.15%, 6.88%, and
9.56% for 20%, 40%, and 60% of copper slag replacement respectively in comparison with normal cement paste.
A small reduction in compressive strength is observed for 20% coppers slag 5.36% and 1.40% for 5% and 10%
PET resin content respectively. This reduction is because of insufficient quantity of binder availability. Among
all the PET resin content 15% resin content shows a good impact on the compressive strength. A notable reduction
in compressive strength is observed for 80% of copper slag replacement and all PET resin content. The copper
slag has a glossy surface texture less water absorption due to this there will excess water available in mix increases
the porosity and due to the separation of the particle, there is a reduction in compressive strength. Among all the
mixes 60% of copper slag and 15% PET resin content gives good strength compare to normal cement mix. PET
Addition MEKPo and cobalt napthalate as curing agent in PET resin form the ammonia curing agent which
contribute in the early age strength of cement mortar.
1 day 7 day 28 days

70.00
Compressive strngth in N/mm2
60.00
50.00
40.00
30.00
20.00
10.00
0.00
Cement 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15%
0 20 40 60 80
PET resin and copper slag in %
Figure 2 Compressive strength

3.6 Tensile Strength
The tensile strength of mortar is determined at 1,7, and 28 days of curing. Addition of PET resin in 5%, 10%, and
15% with copper slag up to 60% replacement as fine aggregate has shown very good improvement in tensile
strength of cement mortar in all curing ages as compare to normal mortar mix. Beyond 60% replacement of copper
slag for all PET resin content tensile strength of mortar is more in for 1 day and 7day curing but it is less for 28
days curing compare to normal cement mortar. Presence of PET particle in cement mix act as bridging between
two particles increases the tensile strength of mortar. Tensile strength at 1 day curing for 5%, 10% and 15% resin
675
content increases by 33.33 % to 186.67% ,146.67% to 260% , 226.67% to 365.33% and 179.33% to 218.67%
for 20% ,40% 60% and 80% replacement of copper slag respectively as compare to normal cement mortar same
is as shown in Figure 3. After 28 days curing tensile strength for 5%, 10%, and 15% PET resin content is increased
by 1.39% to 9.86%, 6.39% to 31.94%, and 14.44% to 45.69% for 20%, 40% and 60% copper slag replacement
respectively as compare to normal cement mortar. The percentage gain in tensile strength at 7 days and 28 days
strength is less than 1 day curing strength in comparison with a normal specimen. Reduction in tensile strength
for 80% copper slag replacement is 15.83%, 10.42%, and 4.72% for 5%, 10%, and 15% PET resin for 28 days
curing respectively in comparison with normal cement mortar. The strength variation for 1, 7 and 28 days curing
is very small for all specimens of copper slag and PET resin. The relation between compressive strength and
tensile strength at 28 days curing is as shown in Figure 4
1 day 7 day 28 days

Tensile Strength in N/mm2
6.00
5.00
4.00
3.00
2.00
1.00
0.00
5% Resin
10% Resin
15% Resin
5% Resin
10% Resin
15% Resin
5% Resin
10% Resin
15% Resin
5% Resin
10% Resin
15% Resin
Cement
0 20% 40% 60% 80%

PET resin and Copper slag in %
Figure 3 Tensile strength
20% 40% 60% 80% y = 3E-06x3.5054

Tensile Strength in N/mm2
6.00
R² = 0.8462
5.50
y = 0.000x3.379
5.00 R² = 0.882
4.50 y = 0.090x0.937
4.00 y = 0.095x - 1.3206 R² = 0.962
R² = 0.999
3.50
3.00
2.50
2.00
45.00 50.00 55.00 60.00
Compressive Strength in N/mm2
Figure 4 Relation between compressive strength and tensile strength

3.7 Abrasion test
Abrasion test on mortar is determined for 28 days of curing. Abrasion resistance goes on increasing as the
percentage of PET resin and up to 80% of copper slag replacement. The reduction abrasion resistance for 80%
676
copper slag is because of excess water present in the mix it will lead to more voids. Reduction in abrasion
resistance varies from 10.82% to 26.12 %, 18.66% to 28.36%, 38.06% to 52.24%, and 9.7% to 25% for 20%,
40% 60% and 80% of copper slag for 5%, 10%, and 15% PET resin content respectively as compare to normal
cement mortar the same is shown in Figure 5. The increase in abrasion resistance is because of PET resin mortar
has high early strength. The presence of copper slag also contributes to increasing the abrasion resistance because
copper slag particle is having a dense structure, fewer voids, sharp angular edges, and good compressibility.
3
Weight loss Loss in (%)
2.5
1.5
0.5
0
Cement 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15%
0 20% 40% 60% 80%
PET resin and copper slag in %
Figure 5 Abrasion test on mortar
4 Conclusion
The growth in the construction industry is unavoidable. The impact over the environment by construction
industry is a major concern and it increasing the interest in sustainable development. The finding of this
experimental work to utilize PET resin and copper slags are a substitute for cement and Fine aggregate
respectively. The finding of this work is as follows.
1) The PET resin can be used as binder substitute for cement as it does not alter the consistency,
soundness, IST, FST properties of cement. Substitute material for cement is much needed now
because cement is material which is majorly contributing to the carbon footprint.
2) Copper slag can be used as a substitute to fine aggregate as it acts as inert material in cement mortar.
Addition of copper slag up to 60% is possible beyond this there is a reduction in strength.
3) Addition of PET resin and MEKPo as curing agent gives early strength at 1 day curing;
approximately 90% of normal cement mortar strength at 28 days curing; which will be advantageous
in retrofitting works.
4) Change in compressive strength and tensile strength is very less for 7 days and 28 days curing in
comparison with 1 day curing. 15% of PET resin content shows good contribution in all replacement
of copper slag.
5) Mortar flow increases as % of PET resin and copper slag substitution goes on increasing.
677
6) Abrasion resistance of cement mortar increases as substitution % of PET resin and copper slag
increases.
7) Among all the % of PET resin and Copper slag content, 15% of PET resin and 60% of copper slag
has shown good contribution in all parameter. This will reduce the major load on the environment.
Especially in the location where availability of fine aggregate is a problem.
5 References
1. Bedi, R., Chandra, R., Singh, S.P.: Mechanical Properties of Polymer Concrete. 2013, Journal of
Composites (2013), dx.doi.org/10.1155/2013/948745
2. Yao, Z., Zhang, X., Ge, Z., Jin, Z., Han, J., Pan, X.: Mix Proportion Design and Mechanical Properties of
Recycled PET Concrete. J. Test. Eval. (2015). doi:10.1520/JTE20140059
3. Saikia, N., De Brito, J.: Mechanical properties and abrasion behaviour of concrete containing shredded
PET bottle waste as a partial substitution of natural aggregate. Constr. Build. Mater. (2014).
4. Janfeshan Araghi, H., Nikbin, I.M., Rahimi Reskati, S., Rahmani, E., Allahyari, H.: An experimental
investigation on the erosion resistance of concrete containing various PET particles percentages against
sulfuric acid attack. Constr. Build. Mater. (2015). doi:10.1016/j.conbuildmat.2014.12.037
5. Saikia, N., De Brito, J.: Use of plastic waste as aggregate in cement mortar and concrete preparation: A
review, (2012)
6. Pereira De Oliveira, L.A., Castro-Gomes, J.P.: Physical and mechanical behaviour of recycled PET fibre
reinforced mortar. Constr. Build. Mater. (2011). doi:10.1016/j.conbuildmat.2010.11.044
7. Agavriloaie, L., Oprea, S., Barbuta, M., Luca, F.: Characterisation of polymer concrete with epoxy
polyurethane acryl matrix. 37, 190–196 (2012). doi:10.1016/j.conbuildmat.2012.07.037
8. Journal, a C.I.M., Apr, M.: Mechanical Properties of Polyester Polymer Concrete Using Recycled
Polyethylene Terephthalate. Search. 3–5 (2010)
9. Marthong, C.: Effects of PET fiber arrangement and dimensions on mechanical properties of concrete.
IES J. Part A Civ. Struct. Eng. (2015). doi:10.1080/19373260.2015.1014304
10. Rahmani, E., Dehestani, M., Beygi, M.H.A., Allahyari, H., Nikbin, I.M.: On the mechanical properties of
concrete containing waste PET particles. Constr. Build. Mater. (2013).
11. Fraternali, F., Farina, I., Polzone, C., Pagliuca, E., Feo, L.: On the use of R-PET strips for the
reinforcement of cement mortars. Compos. Part B Eng. (2013). doi:10.1016/j.compositesb.2012.09.070
12. Dębska, B., Lichołai, L.: The selected mechanical properties of epoxy mortar containing PET waste.
Constr. Build. Mater. (2015). doi:10.1016/j.conbuildmat.2015.07.031
678
13. Reis, J.M.L., Carneiro, E.P.: Evaluation of PET waste aggregates in polymer mortars. Constr. Build.
Mater. (2012). doi:10.1016/j.conbuildmat.2011.08.020
14. Mahdi, F., Abbas, H., Khan, A.A.: Flexural, shear and bond strength of polymer concrete utilizing
recycled resin obtained from post consumer PET bottles. Constr. Build. Mater. (2013).
15. Alp, I., Deveci, H., Süngün, H.: Utilization of flotation wastes of copper slag as raw material in cement
production. J. Hazard. Mater. 159, 390–395 (2008). doi:10.1016/j.jhazmat.2008.02.056
16. Moura, W.A., Gonçalves, J.P., Lima, M.B.L.: Copper slag waste as a supplementary cementing material
to concrete. J. Mater. Sci. 42, 2226–2230 (2007). doi:10.1007/s10853-006-0997-4
17. Bhoi, A.M., Patil, Y.D., Patil, H.S., Kadam, M.P.: Feasibility Assessment of Incorporating Copper Slag
as a Sand Substitute to Attain Sustainable Production Perspective in Concrete. Adv. Mater. Sci. Eng. 2018,
(2018)
18. Al-Jabri, K.S., Al-Saidy, A.H., Taha, R.: Effect of copper slag as a fine aggregate on the properties of
cement mortars and concrete. Constr. Build. Mater. 25, 933–938 (2011).
19. Tixier, R., Devaguptapu, R., Mobasher, B.: The Effect of Copper Slag in the Hydration and Mechanical
Properties of Cementious Mixtures. Cem. Concr. Res. 27, 1569–1580 (1997). doi:10.1016/S0008-
8846(97)00166-X
20. Mobasher B, Devaguptapu R, A.A.: EFFECT OF COPPER SLAG on the HYDRATION of BLENDED
CEMENTITIOUS MIXTURES B. Mobasher 1 M. ASCE, and R. Devaguptapu 2 , A.M. Arino 2. Proc.
ASCE Mater. Eng. Conf. Mater. new Millenn. 1677–1686 (1996)
21. Najimi, M., Pourkhorshidi, A.R.: Properties of concrete containing copper slag waste. Mag. Concr. Res.
63, 605–615 (2011). doi:10.1680/macr.2011.63.8.605
22. Wu, W., Zhang, W., Ma, G.: Optimum content of copper slag as a fine aggregate in high strength concrete.
Mater. Des. 31, 2878–2883 (2010). doi:10.1016/j.matdes.2009.12.037
23. Yue, Q.F., Xiao, L.F., Zhang, M.L., Bai, X.F.: The Glycolysis of Poly(ethylene terephthalate) Waste:
Lewis Acidic Ionic Liquids as High Efficient Catalysts. 1258–1271 (2013). doi:10.3390/polym5041258
24. Purohit, J., Chawada, G., Choubisa, B., Patel, M., Dholakiya, B.: Polyester Polyol Derived From Waste
Poly ( Ethylene Terephthalate ) for Coating Application on Mild Steel Polyester Polyol Derived From
Waste Poly ( Ethylene Terephthalate ) for Coating Application on Mild Steel. 2012, (2012)
679
Paper ID- 136
FINITE ELEMNET ANALYSIS OF PROFILED DECK

COMPOSITE SLAB USING ANSYS
Aniket A.Shirgaonkar1,Yogesh D. Patil2,Hemant S. Patil2

1 S V National Institute of Technology, Surat, India
{Aniket Shirgaonkar, Yogesh Patil,Hemant Patil LNCS}@Springer.com
Abstract. Shear connectors between profiled deck sheet and concrete topping plays
vital role in load caring mechanism of composite slab. They distribute horizontal shear
force to the base element of composite slab and also creates tensile link into the
concrete. The degree of composite action in flexure is depends on the mechanical as
well as geometrical properties of the shear connectors. To study the characteristics of
shear connectors finite element method is one of the capable methods. This paper
describes a finite element analysis of profiled deck composite slab with different types
of shear connector using ANSYS. The main objective is to determine the deflection,
slip as well as failure characteristics of composite slab for 2 different types of shear
connectors: headed stud and double headed stud.
Keywords: Composite slab,Shear Connector,Slip,concrete.
1. Introduction
Profiled deck composite slab is mostly used as modern construction technique because
of its minimum erection time and least weight to depth ratio. Concrete topping is
provided over profiled deck steel sheet, here deck sheet has to play two major roles, it
acts as tensile reinforcement and also it acts as permanent formwork during construction
work[1]. Additional minimum reinforcement is still required to control cracking due to
temperature and shrinkage effect, nowadays welded wire mesh is also used as
secondary reinforcement. In composite construction, shear connector between Profiled
deck sheet and concrete plays major role as they provide shear connection for composite
action in bending and reduces slip[2–4].
Fig. 1 Details of composite slab
680
Shear connectors mainly resist the vertical uplifting forces and horizontal shear. Headed
stud is most commonly used shear connector, headed studs are connected to steel beams
by means of welding which requires special welding equipment and high voltage for its
operation[5–8]. Shear resistance offered by headed stud is circumscribed due to its low
contact area with concrete[5]. These ill effects of headed stud connector can be
eliminated by using new type of connector, namely double headed stud.
2. Finite Element Analysis

For analyzing behavior of composite slab under four-point loading Finite element
method is one of the capable method[9]. For determining multi-axis stresses for
concrete, William-Warnke’s fracture model were used as source. In ANSYS
Workbench non-liner behavior of model is analyzed by Newton-Raphson method.
2.1 Geometry
Two models were prepared using Solid work 3D drawing software and imported in
ANSYS WB environment to carry out further analysis. The geometry of the composite
slab having dimension of 2700 mm X 533.6 mm X 120 mm with 2 supports of
dimension 533.6 mm X 100 mm X 20 mm. Fig. 1 represents the composite slab
considered for the study. Appropriate material properties are assigned to the model
which are listed below (table 1- table 3).
Fig. 2 Dimensions of profiled steel sheet (mm)
Fig. 3 Double headed stud (60mm height)
681
Table 1 Material properties for Concrete
Young’s modulus 25000 Mpa

Bulk modulus 13021 Mpa
Shear modulus 10593 Mpa
Maximum tensile pressure 3.5 MPa
Table 2 Multilinear isotropic hardening properties for concrete
Plastic strain Stress (Mpa)
0 0
0.0003 7.33
0.0006 14.76
0.0012 22.05
0.0018 24.86
0.002 25
Table 3 Material properties for Profiled deck sheet
Density 7850 Kg/m3

Young’s modulus 2e+05 pa
Bulk modulus 1.6667e+11 pa
Shear modulus 7.6923e+10 pa
Bilinear isotropic hardening properties for Profiled deck sheet
Yield strength 250 Mpa
Tangent modulus 1450 Mpa
After assigning material properties to model next step is to simulate the four-point
bending test, analyses were carried out by applying the point load over the composite
slab onto 533.6 X 100 mm2 area in ANSYS WB. Hight of stud is kept constatnt in both
slab as 60mm. Explicit dynamics analysis system is adopted for analysis as it gives
relibale results as well as well defined crack pattern. In the full-scale model, simply
supported condition is created using boundary conditions. Left support is hinged and
right support is roller support. Full scale model with applied load and boundary
conditions are shown in Figures.In finite element analysis of composite slab contact
modeling between steel and concrete is major task, frictional connection with frictional
coefficient 0.5 is assigned to contact surface[10], here contact surface is of concrete
surface and target surface is steel sheet. Composite slab model is discretized using 20
mm mesh size, Incremental load up to 100 kN is applied and analysis is carried out.
682
Fig. 4 meshing by sizing method (20 mm)
3. Results
Nonlinear Analysis were carried out on both the models and slip at hinged support, total
deflection at the mid-span and failure mode of the composite slab were observed.
Composite slab is failed in flaxural and ductile behaviour is observd.Following are the
results obtained by finite element analysis of composite slab. Figure 5 shows the
deflection counters of composite slab with headed stud connector. Figure 6 and figure
7 shows the slip and crack formation in concrete.
Fig. 5 deflection contours
683
Fig. 6 Slip at roller support
Fig. 7 Crack formation near hinge support
Fig. 7 Load – deflection plot (double headed stud)
684
Fig. 8 Load – deflection graph (headed stud)
Fig. 9 Load – slip graph (double headed stud)
Fig. 10 Load – slip graph (headed stud)
685
4. Conclusion
In this finite element analysis deflection and slip characteristics of profiled deck
composite slab are studied, type of shear connector is the major parameter of study.
conventionally used stud connector is compared with new type of double headed
connector on the basis of shear resistance capacity. The behavior of double headed
connector is very favorable. Based on comparison of headed stud connector with double
headed stud connector, the double headed connectors performance is better than headed
stud connector and it can be a better alternative to it.
The main conclusions from finite element analysis of composite slabs with
profiled steel sheeting under static loading are:
• Both the models were failed due to critical loss of shear bond between concrete and
deck sheet.
• For model with double headed stud the load at which first crack observed is 88.2kN
whereas for model with headed stud it is 78.4 kN, hence double headed stud
connector shows more ductile behavior.
• Composite slab with double headed stud reduces deflection significantly (8.82%)
so it can active barter composite action than heeded stud.
• Shear resistance capacity of double headed stud is 30% more than headed stud due
to increase in contact area with concrete.
• Both the connectors were successful in resisting slip significantly.
• Based on comparison of headed stud connector with double headed stud connector,
the double headed connectors performance is better than headed stud connector
and it can be a better alternative.
5. Acknowledgments
The authors want to thank the Mr. Ashish Shete, Prescient Informatics Pvt. Ltd.
Kolhapur for their kind collaboration.
6. References
1. Porter, M.L., Jr, C.E.E.: Scholars ’ Mine Investigation of Cold-formed Steel-
deck-reinforced Concrete Floor Slabs. (1971)
2. Johnson, R.P., Shepherd, A.J.: Resistance to longitudinal shear of composite
slabs with longitudinal reinforcement. J. Constr. Steel Res. 82, 190–194 (2013).
doi:10.1016/j.jcsr.2012.12.005
3. Hedaoo, N.A., Gupta, L.M., Ronghe, G.N.: Design of composite slabs with
profiled steel decking: a comparison between experimental and analytical studies. Int.
J. Adv. Struct. Eng. 4, 1–15 (2012). doi:10.1186/2008-6695-3-1
4. Marimuthu, V., Seetharaman, S., Arul Jayachandran, S., Chellappan, A.,
Bandyopadhyay, T.K., Dutta, D.: Experimental studies on composite deck slabs to
determine the shear-bond characteristic (m - k) values of the embossed profiled sheet.
J. Constr. Steel Res. 63, 791–803 (2007). doi:10.1016/j.jcsr.2006.07.009
686
5. Merryfield, G., El-Ragaby, A., Ghrib, F.: New shear connector for Open Web
Steel Joist with metal deck and concrete slab floor system. Constr. Build. Mater. 125,
1–11 (2016). doi:10.1016/j.conbuildmat.2016.08.006
6. Ali Shariati: Various types of shear connectors in composite structures: A
review. Int. J. Phys. Sci. 7, 2876–2890 (2012). doi:10.5897/IJPSX11.004
7. Shariati, M., Ramli Sulong, N.H., Shariati, A., Khanouki, M.A.: Behavior of
V-shaped angle shear connectors: experimental and parametric study. Mater. Struct.
Constr. 49, 3909–3926 (2016). doi:10.1617/s11527-015-0762-8
8. Shariati, M., Sulong, N.H.R., Arabnejad, M.M.K.H., Mahoutian, M.: Shear
resistance of channel shear connectors in plain , reinforced and lightweight concrete. 6,
977–983 (2011). doi:10.5897/SRE10.1120
9. Abdullah, R., Kueh, A.B.H., Ibrahim, I.S., Easterling, W.S.: Characterization
of shear bond stress for design of composite slabs using an improved partial shear
connection method. J. Civ. Eng. Manag. 21, 720–732 (2015).
doi:10.3846/13923730.2014.893919
10. Sayin, B., Manisali, E.: An investigation of interfacial stresses in reinforced
concrete beams using FRP laminates. J. Adhes. 86, 1132–1157 (2010).
doi:10.1080/00218464.2010.519260
687
Paper ID- 137
A comparative study of Seismic Response of Structure

isolated with Triple Friction Pendulum Bearing and
Single Friction Pendulum Bearing under Different
Hazard Level of Earthquake
Ankit Sodha1*, Sandeep Vasanwala1, Devesh Soni2, Shailendra Kumar1

1
Applied Mechanics Department
Sardar Vallabhbhai National Institute of Technology
Ichchhanath Circle, Surat 395007, Gujarat, India
*aankitsodha@gmail.com
2
SardarVallabhbhai Patel Institute of Technology
Vasad 388 306, Anand District, Gujarat, India
Abstract
The Triple Friction Pendulum (TFP) system is a new generation sliding isolation
having four spherical sliding surfaces with three effective pendula. Due to multiple
sliding surfaces, TFP system shows highly adaptive behaviour under different hazard
level of earthquakes, despite being a passive system. The Report describes
mathematical model and seismic response of TFP system under three different
earthquake records consist of service level, design basis and maximum considered
earthquakes. By considering effective period and effective damping in combination
with desirable displacement capacity of TFP bearing designs are considered. A
comparative response of the building isolated by the FPS and TFP with same effective
damping, effective time period and displacement capacity under different hazard level
of earthquake. It shows that the performance of TFP is superior over the FPS. This is
due to the presence of multiple sliding surfaces which dissipate seismic energy and
distribute sliding displacement over the multiple surfaces. It is also found that the TFP
bearing shows stiffens at low input, softens with increasing input, and then stiffens
again at higher levels of input. Thus, it shows highly adaptive behaviour under
different hazard level of earthquake.
Key words: Seismic isolation, Triple Friction Pendulum system, Friction Pendulum
system, Multi hazard level earthquake.
1. Introduction
The sliding isolation system is based on the concept of sliding friction [Mostaghel
and Tanbakuchi, 1983]. The sliding type isolation system including various restoring
force devices such as rubber as used in Electric de France system [Gueraud et al.,
1985], resilient-friction base isolator system [Mostaghel and Khodaverdian, 1987],
and friction pendulum system [Zayas et al., 1990]. To divulge adaptive behavior of
688
friction pendulum system, a concept of multiple friction pendulum system has been
introduced.[Tsai et al., 2010]. Further Double Concave Friction Pendulum (DCFP)
system with two concave surfaces [Fenz and Constantinou, 2006] and Triple Friction
Pendulum (TFP) system with four sliding surfaces have been developed [Fenz and
Constantinou, 2008; Becker and Mahin, 2013; Dhankot and Soni, 2016, 2017]. The
DCFP and TFP offer the designer extra design parameters to optimize the
performance in terms of coefficients of friction and radius of curvature of difference
sliding surfaces.
2. TFP Bearing and its mathematical model

The TFP bearing is made up of four stainless steel concave surfaces with Teflon
coating nested by a slider as shown in Figure 1(a). The di, μi and Ri are the
displacement capacity, coefficient of friction and radius of curvature of the ith
spherical surface, respectively. By increasing magnitude of displacement, TFP
bearing exhibits multiple changes in stiffness and strength. Multiple sliding surfaces
of the TFP results in more complex force-deformation behaviour compared to
currently used isolation systems.
(a)
(b)
Figure 1 (a) Schematic Diagram and (b) Series Model of TFP

The TFP can be modeled as a series model [Fenz and Constantinou 2008] by
connecting three single concave FP elements in series as shown in Figure 1(b). It
contains linear elastic spring element with stiffness, 1/Reff, representing the restoring
force provided by the spherical sliding surface, friction element, μi, representing the
surface friction and gap element, di, representing the finite displacement capacity for
689
each FP element with a parallel arrangement. The effective radius of curvature and
coefficient of friction of the three FP element of series model are expressed as,
Reff 1 = Reff 2 + Reff 3
(1)
Reff 2 = Reff 1 − Reff 2 (2)
Reff 3 = Reff 4 − Reff 3 (3)
μ1 =μ 2 =μ3 (4)
μ2 =μ1 (5)
μ3 =μ 4 (6)
here, Reff,i=Ri-hi; The radial distance between the ith surface and pivot point of the
articulated slider is represented by hi.
3. Ground motions
The suites of time histories developed as part of SAC steel project representing a
range of seismic hazard levels from seismic zone 2 to zone 4 (Soil type SD) has been
used for the present study. Three hazard levels with various probability of occurrence,
SLE (50% in 50 years), DBE (10% in 50 years), and MCE (2% in 50 years), are
considered. A set of 3 time histories, 1 from each of the three bins: SLE, DBE and
MCE is presented in Table 1 [Somerville et al., 1998].
Table 1 Multiple hazard level ground motions
Record Name Amplitude Duration PGA

Lable Scale Factor (sec) (cm/sec2)
SLE
LA41 Coyote Lake, 1979 2.28 39.38 578.34
DBE
LA03 Imperial Valley, 1979 1.01 39.38 386.04
MCE
LA21 Northridge, 1994 1.15 59.98 1258.00
4. Governing equation of motion and its solution

For the N-story superstructure total dynamic Degree-of-freedom (DOF) considering
one DOF at each floor is N + 3. Therefore, the governing equations of motion for the
N story superstructure model are expressed in matrix form as,
[ M ]{u&&} + [ K ]{u&} + [C ]{u} = − [ M ]{r}(u&&b + u&&g ) (7)
where, [M] is N sized square matrices of the superstructure for mass, [K] is for
stiffness and [C] is for damping.
Newmark’s step-by-step method assuming linear variation of acceleration over a
small time interval of the order of 1 × 10–6 sec used to solve governing equation of
690
motions for the structure with isolation containing highly nonlinear force-deformation
behaviour. Further, Runge-Kutta method for incremental hysteretic displacement
components in hysteretic model for each time step are obtained by solving a
differential equation by a fourth-order.
5. Properties of the Isolation system

To compare FPS and TFP, the effective damping ratio, time period and design
displacement are kept same as ξeff = 10%, 15%, 20%, Teff = 4sec, 5sec, 6sec and D =
1m respectively, so that the energy dissipated in each cycle, Eloop remain same for all
the three systems. The effective period, Teff and damping ratio, ξeff are given by,
W
Teff = 2π (8)
K eff g
Eloop
ξ eff = 2
(9)
2π K eff D
where, Eloop is the total energy dissipated at each cycle of the isolator displacement; W
is the total weight on the isolator; Keff is the effective linear stiffness and D is the
maximum isolator displacement under the specified level of motion.
While designing the TFP system, it is assumed that the largest coefficient of
friction for the isolation system has been reached and no displacement capacities of
individual sliders have been reached.
For the FPS effective stiffness, Keff is obtained from the geometry of force-
deformation curve and the energy dissipated in each cycle, Eloop is calculated from the
area enclosed by the hysteresis loop (Refer Figure 2(a,c)) as given by,
1
μ+ ( D − Dy )
Reff
K eff = W (10)
D
 Dy 
Eloop = 4  μ −  ( D − D y )W (11)
 Reff
 
(a) FPS (b) TFP
691
(c) FPS (d) TFP

Figure 2 Schematic diagram of (a) FPS (b) TFP and hysteretic force
force-displacement
displacement
relationship of (c) FPS (d) TFP
Similarly, for the TFP the Keff and Eloop in each cycle are given by (Refer Figure
2(b,d)),
* ** * **
K D y + K 2 (u − D y ) + K3 (u − u ) + K 4 ( D − u )
K eff = 1 W (12)
D
( )
2
* *
Eloop = 4 u ( K 2 − K3 )W + 4 DD y ( K 2 − K1 )W + 4u D ( K3 − K 2 )W
(13)
(13
( )
2
** ** 2
+4u D ( K 4 − K3 )W + 4 u ( K 3 − K 4 )W + 4 D y ( K1 − K 2 )W
where,
1 1 1
K1 = ; K2 = ; K3 =
R eff 2 R +R
eff 2 eff 3 R
eff 3+R eff 1
(14)
(14
1
and K 4 =
Reff 1 + Reff 4
u * = (μ
μ1 − μ 2 ) Reff 2 + (μ1 − μ 3 ) Reff 3 (15)
(15
u = u + (μ
**
μ 4 − μ1 )( Reff 1 + Reff 3 )
*
(16)
(16
here, Dy is the isolator displacement at which motion on the sliding surface initiates.
The Dy iss takes as 0.025 cm in this study [[Constantinou et al., 1990].
For the FPS, radius of curvature and coefficient of friction araree calculated using
Equations (10) and (11). ). And, for the TFP, radius of curvature and coefficient of
friction of different sliding surfaces araree calculated using Equations (12) to (16),(16
respectively, through iterative process. The three TFP and FPS configurations
designed using above procedure and equations are presented in Table 2.
692
Table 2 Equivalent properties for series model of TFP and FPS
ξeff Teff D(m) Reff 1 Reff 2 Reff 3 μ1 μ2 μ3
FPS1 10 4 1 4.8 0.04

TFP1 10 4 1 0.6 2.7 2 0.01 0.05 0.03
FPS2 15 5 1 8.27 0.039
TFP2 15 5 1 0.9 5.05 0.45 0.01 0.038 0.058
FPS3 20 6 1 13 0.035
TFP3 20 6 1 1.2 6.4 5.4 0.01 0.035 0.045
6. Numerical Study
In this study, a five-storey shear building with diagonal mass matrix is

considered. The mass, stiffness and modal damping ratio of the each floor is kept
constant in each mode of vibration for simplicity in analysis. The time period of
superstructure can be kept fixed at the base by selecting such a value of stiffness. The
characterization of base isolated system done by the parameters like, the ratio of base
mass to the superstructure floor mass, mb/m, damping ratio of the superstructure, ξs
and the fundamental time period of the superstructure, Ts. The numerical values are
kept constant as Ts = 0.5 sec, ξs = 0.02, mb/m = 1 and msi/m = 0.0005 in this study. To
achieve desirable fundamental period of 0.5 sec, the storey stiffness, K=115,740
kN/m and floor mass, m=59,378 kg have been taken and along with bay width 3.0m
and storey height 5.0 m and 3.0 m. To study seismic response, all three response
quantities i.e., isolator displacement, ub, base shear normalized by total weight of the
building, Fb/W, and top floor absolute acceleration, A(g) are considered.
6.1 Behaviour of FPS and TFP under multi hazard level earthquakes
The Time variation of displacement of the three FP elements, x1, x2, x3 and total
isolator displacement, ub of TFP1 under LA21 (MCE) are shown in Figure 3. It is seen
that in SLE event, FP4 and FP5 elements remains more in sticking compare to MCE
due to its high friction and low level shaking while other element keeps on sliding due
to relatively lower friction. The hysteretic behaviour of all three FP elements and the
TFP system under LA 21 (MCE) is shown in Figure 4. It shows that with the increase
in earthquake hazard level, sliding displacement on each surface increases and for
MCE event the isolator deforms into sliding regime VIII and IX and the sliding
surfaces come in contact with the restrainers as exhibited by sudden increase in slope
of hysteretic loop.
693
Figure 3 Time variation of displacement of the three FP elements, x1, x2, x3 and total
isolator displacement, ub, of TFP1 under LA 21 earthquake
The isolation system designed for MCE is too stiff and so heavily damped that it
does not move during more probable, lower-level earthquake. This results in much
less isolation than promised causing damage of non-structural elements, equipment
and disturbance to occupants [Kelly, 1999]. To achieve desired performance for MCE
while maintaining good performance for SLE, the isolation system should be very
stiff at low input shaking SLE, softens with increasing input at the DBE, and then
stiffens again at higher levels of input MCE. An isolation system adopting changes in
isolator properties as demanded by input motion is called adaptive isolation system.
Figure 4 Hysteretic behavior of all the elements of TFP1 under LA 21 (MCE)
694
Figure 5 Hysteretic behavior and Time histories of response of TFP1 and FPS1
under LA 41 (SLE)
under LA 03 (DBE)
695
under LA 21 (MCE)
To study adaptive behaviour of TFP over a FPS, the example building was
subjected to three levels of ground excitations, LA 41 (SLE), LA 03 (DBE), LA 21
(MCE) for a three different design configurations as shown in Table 2. The hysteresis
behaviour histories of response of TFP and FPS for each level of earthquake is shown
in Figure 5 to Figure 7.
Here, the TFP bearing stiffens at low input (SLE), softens with increasing input
(DBE), and then stiffens again at higher levels of input (MCE). Thus, it shows
highly adaptive behaviour, despite being a passive system.
Peak values of Top floor acc, g , Bearing Disp., mm and base shear, Fb/W of three
different configurations for FPS and TFP under different hazard level of earthquakes
are as shown in Table 3.
Here for the effective damping ratio ξeff=10% time period Teff = 4sec, reduction in
a peak response of TFP compare to FPS for the top floor acceleration are 20%, 12%
and 50% under SLE, DBE and MCE respectively. for the reduction in a peak response
of base shear are 10%, 5% and 5% under SLE, DBE and MCE respectively.
For the effective damping ratio ξeff =15% time period Teff = 5sec, reduction in a
peak response of TFP compare to FPS for the top floor acceleration are 30%, 35%
of base shear are 10%, 5% and 8% under SLE, DBE and MCE respectively.
696
Table 3 Peak values of Top floor acc, g , Bearing Disp. and base shear, Fb/W
EQ Model Top floor Bearing Base shear Fb/W

acc, g Disp. (mm)
ξ eff = 10%, Teff = 4 sec
MCE FPS 0.424 489.38 0.142
(LA21) TFP 0.347 472.00 0.127
FPS 0.469 471.68 0.096
TFP 0.339 472.60 0.087
FPS 0.413 480.19 0.072
TFP 0.277 470.39 0.063
DBE FPS 0.321 714.86 0.189
(LA03) TFP 0.281 750.88 0.180
FPS 0.325 616.24 0.114
TFP 0.214 646.93 0.107
FPS 0.275 634.04 0.084
TFP 0.182 626.79 0.075
SLE FPS 0.378 122.63 0.066
(LA41) TFP 0.190 128.90 0.062
FPS 0.405 110.62 0.052
TFP 0.172 113.87 0.048
FPS 0.346 112.19 0.044
TFP 0.147 116.89 0.042
697
For the effective damping ratio ξeff = 20% time period Teff = 6sec, reduction in a
peak response of TFP compare to FPS for the top floor acceleration are 35%, 35%
of base shear are 15%, 10% and 5% under SLE, DBE and MCE respectively. Here no
significant change in bearing displacement.
7. Conclusions
Presented in this Report is the behaviour of TFP system compared with FPS using
nonlinear time history analysis with multi hazard level excitation and directivity
focusing earthquakes. The MDOF system is analyzed for three multi hazard
earthquake records consist each of SLE, DBE and MCE records to achieve a wide
range variation in frequency and intensities to understand the behavior
comprehensively. The following conclusions can be derived for the study:
1. The TFP gives greater flexibility for the selection various design parameters
to optimize the isolator performance compare to FPS, owing to three
spherical sliding surfaces, three effective pendula. thereby can be very useful
in controlling the response parameters of a seismically isolated building.
2. The TFP shows highly adaptive behaviour compare to FPS under different
hazard level of earthquake, despite being a passive system. As it stiffens at
SLE (low input), at DBE it get softens and again it get stiffens at MCE
(higher input).
3. As increase in effective damping ratio and time period leads to more
reduction in a peak response of TFP compare to FPS for peak response of top
floor acceleration and base shear. Here no significant change in bearing
displacement.
References
1. Mostaghel, N., & Tanbakuchi, J. [1983] “Response of sliding structures to

earthquake support motion,” Earthquake engineering & structural
dynamics, 11(6), 729-748.
2. Gueraud, R., Noel-Leroux, J. P., Livolant, M., & Michalopoulos, A. P. [1985]
“Seismic isolation using sliding-elastomer bearing pads,” Nuclear Engineering
and Design, 84(3), 363-377.
3. Mostaghel, N., & Khodaverdian, M. [1987] “Dynamics of resilient‐friction base
isolator (R‐FBI),” Earthquake engineering & structural dynamics, 15(3), 379-
390.
4. Zayas, V. A., Low, S. S., & Mahin, S. A. [1990] “A simple pendulum technique
for achieving seismic isolation,” Earthquake spectra, 6(2), 317-333.
5. Tsai, C. S., Lin, Y. C., & Su, H. C. [2010] “Characterization and modeling of
multiple friction pendulum isolation system with numerous sliding
interfaces,” Earthquake Engineering & Structural Dynamics, 39(13), 1463-1491.
6. Fenz, D. M., & Constantinou, M. C. [2006] “Behaviour of the double concave
friction pendulum bearing” Earthquake Engineering & Structural
Dynamics, 35(11), 1403-1424.
698
7. Fenz, D. M., & Constantinou, M. C. [2008] “Modeling triple friction pendulum

bearings for response-history analysis,” Earthquake Spectra, 24(4), 1011-1028.
8. Becker, T. C., & Mahin, S. A. [2013] “Approximating peak responses in
seismically isolated buildings using generalized modal analysis,” Earthquake
Engineering & Structural Dynamics, 42(12), 1807-1825.
9. Dhankot, M. A., & Soni, D. P. [2016] “Seismic response of triple friction
pendulum bearing under multi hazard level excitations,” International Journal of
Structural Engineering, 7(4), 412-431.
10. Dhankot, M. A., & Soni, D. P. (2017). “Behaviour of Triple Friction Pendulum
isolator under forward directivity and fling step effect,” KSCE Journal of Civil
Engineering, 21(3), 872-881.
11. Constantinou, M., Mokha, A., & Reinhorn, A. [1990] “Teflon bearings in base
isolation II: modeling,” Journal of Structural Engineering, 116(2), 455-474.
12. Somerville P, Anderson D, Sun J, Punyamurthula S, Smith N. [1998]
“Generation of ground motion time histories for performance-based seismic
engineering,” Proceedings 6th National Earthquake Engineering Conference.,
Seattle, Washington.
13. Kelly, J. M. [1999] “The role of damping in seismic isolation,” Earthquake
engineering & structural dynamics, 28(1), 3-20.
14. Shome N, Cornell CA. [1991] “Probabilistic Seismic Demand Analysis of
Nonlinear Structures,” Dept. of Civil Engineering, Stanford Univ., Stanford,
California.; Rep. No. RMS-35.
15. Panchal, V. R., & Jangid, R. S. [2008] “Variable friction pendulum system for
seismic isolation of liquid storage tanks, ” Nuclear Engineering and
Design, 238(6), 1304-1315
699
Paper ID- 138
Assessment of Important Parameters for Seismic Analysis

and Design of Confined Masonry Buildings
Bonisha Borah 1[0000-0002-1441-6555], Vaibhav Singhal 2[0000-0001-6116-1586] and Hemant B.

Kaushik 1[0000-0001-5896-6543]
1
Department of Civil Engineering, Indian Institute of Technology Guwahati, Assam, India
2 Department of Civil and Environmental Engineering, Indian Institute of Technology Patna,
Bihar, India
bonisha@iitg.ac.in, singhal@iitp.ac.in, hemantbk@iitg.ac.in
Abstract. Confined masonry (CM) building is emerging as a popular building

construction system in many earthquake prone countries including India. This
building type has performed really well in several destructive earthquakes, even
though it started as an informal construction. A finished CM building looks like
a reinforced concrete (RC) infilled structure, however, the structural behaviour
of both the systems is quite different. Apart from excellent performance in seis-
mic events, it is very cost effective and easier to construct. But the design meth-
odology for CM buildings is yet to be standardized for Indian conditions due to
lack of research. Therefore, it is important to study the various factors affecting
the performance of CM building in comparison to other building types. The ob-
jective of this study is to identify the parameters influencing the seismic perfor-
mance of CM buildings. Dependency of seismic response of CM buildings on
different key parameters such as material and geometric properties, gravity loads,
amount of longitudinal steel present in tie columns, and aspect ratio of the wall
is discussed in this paper.
Keywords: Confined masonry, Seismic response, Seismic analysis, Design pa-

rameters.
1 Introduction
A systematic and methodical structural analysis and design to construct structure or

building ensures a safe, stable and durable structure capable of resisting all the probable
loads during its intended life. Buildings should be adequately designed, built and main-
tained to minimize the seismic risks associated with it. However, embracing of new
technologies to build structures is a sine-qua-non in today’s age and hence these tech-
nologies have been and are being adopted worldwide. Especially in seismic prone re-
gions, different construction methodologies have emerged over the last few decades as
alternatives to seismically vulnerable unreinforced masonry (URM) building and im-
properly designed and constructed infilled reinforced concrete (RC) frame building.
Among them confined masonry (CM) is an attractive building construction technology
for low-tech construction in seismic regions. Though this building system started as an
700
informal construction, it showed very good resistance to past seismic events. As a re-
sult, CM structure has attracted considerable interest in the research field. Several
guidelines and country codes were developed in the past years for the construction of a
CM structure, however, most of these guidelines were determined subjectively on the
basis of experience, and are primarily applicable for only one to two-storeyed buildings.
Therefore, a systematic research is required for developing a detailed engineered design
instead of a prescriptive approach. In this regard it is very important to know the main
governing design parameters which affect the response of this building system under
any lateral loads. In the next section, the behaviour of CM structure under gravity and
lateral earthquake loads is discussed.
2 Load transfer mechanism in CM building
Confined masonry is a composite construction system, where masonry structural

walls are first constructed and after that small-sized RC tie-elements (tie-columns and
tie-beams) are built at the walls’ perimeter and some other salient locations. The RC
tie-elements are not intended to carry either vertical or horizontal loads, and are conse-
quently not designed to behave as moment-resisting frame. These are mainly provided
to confine the masonry wall to engage it under gravity and lateral loading and provide
ductility. Even though, at a glance, CM building looks similar to an infilled RC frame
building, its behaviour under gravity and lateral loading and the load distribution mech-
anism is quite different as shown in Fig. 1 [1]. In infilled RC frame, usually there exists
a small gap between beam soffit and masonry panel and the frame is relatively very
stiffer compared to flexible confining frame of CM wall (Fig. 1a). Therefore under
gravity loading, the load transferred to the infill panel is almost negligible, whereas in
CM wall as the beam is cast after the construction of the panel, continuity in gravity
load transfer occurs from tie-beam to CM panel as shown in Fig. 1b and Fig. 2.
Gap No Gap
(a) Infilled RC Frame (b) Confined Masonry
Fig. 1. Comparison of behaviour of (a) infilled RC frame wall and (b) CM wall
From the investigation of experimental test studies and past earthquake damage re-
ports, the general modes of failure of CM structures can be broadly categorized into in-
plane failure, out-of-plane failure, failure of diaphragm, connection and non-structural
elements [2]. Because of the monolithic action developed between the masonry panel
and confining frame, the out-of-plane strength of CM structure is better as compared to
701
that of URM and infilled RC frame structure. Therefore, the in-plane load-displacement
behaviour of CM wall has attracted considerable interest in the seismic research. Under
in-plane lateral loading, masonry may fail under diagonal compression mode or by bed
joint sliding or diagonal stepped cracking. In case of CM wall, generally diagonal
stepped cracking is the most common type of failure. Therefore, under lateral loading,
masonry walls act as diagonal struts subjected to compression, while RC confining
members act in tension and/or compression, depending on the direction of lateral earth-
quake forces. The load transfer mechanism of a CM wall subjected to gravity and lateral
load is shown in Fig. 2. The detailed in-plane behaviour of CM wall under lateral load-
ing and the effects of different parameters are discussed in the next section.
= +
Gravity load transfer Lateral load transfer
Fig. 2. Load transfer mechanism in confined masonry building
3 Important parameters for seismic analysis and design of CM

building
A particular structural system greatly influences the analysis and design principle.
However, there are some basic parameters which should be considered for design of
seismic resistant building system and these are - seismic zone, soil condition, number
of stories, building configuration etc. Building configuration, i.e., size and shape deter-
mines the way of seismic force distribution within the structure. Generally irregular
configurations are avoided for their problematic stress concentration and torsion. Sym-
metrical arrangement of mass and balanced stiffness against either direction is neces-
sary to keep torsion within a manageable range. Seismic design of structures highly
depends on their characteristic force-deformation relationship, which is defined as a
function of three most important design parameters - strength, stiffness and deforma-
bility or ductility. For adequate seismic performance, strength and deformation capac-
ities of a structure must be greater than the demand imposed by a design earthquake.
Strength of a structure assures the ability to resist applied forces within a safe limit.
Stiffness of a structure determines the degree of resistance to deflection or drift. Duc-
tility assures that the structure is able to deform beyond the elastic range and dissipate
seismic energy through plastic deformations. Structural and architectural detailing and
construction quality control is very important to ensure ductility and natural damping
of the structure and to keep damages to a limited and repairable range.
An idealized load-deformation envelope curve developed in a past study [1] for a
CM wall subjected to lateral load is described in Fig. 3. Until the first visible diagonal
702
cracks, the behaviour can be assumed linearly elastic (point A). At higher drift excur-
sion level, the inclined shear cracks extend towards the tie-columns and result in sig-
nificant decrease in the wall stiffness (point B). This post-cracking behaviour is directly
influenced by the friction and brick interlock in the masonry panel and by the shear
resistance of tie-columns. With further loading (point C), the confined wall experiences
strength deterioration due to crushing of bricks and shearing of tie-column ends. The
post-peak behaviour (descending branch B-C) and stiffness of the panel at large defor-
mation are mainly governed by the cross-section and reinforcement details of tie-col-
umns [1, 3]. Therefore, a properly designed and constructed CM wall can attain signif-
icant lateral deformation and ductility before the failure.
20
Maximum strength
B Point A
15
A
C
Load
10 Point B
Slope = Stiffness
5 Ductility
Point C
0
0 1 2 3 4 Displacement
5 6 7 8 9 10 11 12 13 14
Fig. 3. Idealized load-deformation envelope curve for a CM wall subjected to lateral load
The parameters – stiffness, strength and deformation capacity – of a CM building

are functions of different variables which are generally related to the material proper-
ties, geometric characteristics, reinforcement detailing, opening area, loading condi-
tion, overburden pressure etc. In the following section, some important parameters for
seismic analysis and design of CM structure are discussed and their effects on the lateral
load-displacement behaviour are studied.
3.1 Wall density

Wall density is a ratio of cross sectional area of all walls in one direction to the total
floor area. It is an important factor for deciding the required adequate wall area of a
CM building. The lateral load resistance at a floor level can be determined as the sum
of the wall resistances projected in the direction where seismic loading is being consid-
ered. The required wall density in each orthogonal direction of a building to resist the
given seismic base shear force can be determined by applying simplified method, which
is based on Mexican practice [1, 4]. As recommended in the guideline, for buildings
with different types of masonry, soil condition and storey level, the minimum value of
wall density for each direction of the building plan should range from 1% - 2% for low
hazard level (PGA ≤ 0.08g), 1% - 5% for medium hazard level (PGA ≤ 0.25g) and 1.5%
- 9.5% for high hazard level (PGA ≤ 0.4g) [1]. After deciding the wall density and
thereby the wall area, the thickness and length should be finalized for the individual
703
wall to resist the probable gravity and lateral forces. Individual wall capacity depends
on many parameters which are discussed below.
3.2 Masonry properties

Masonry is a nonlinear, nonhomogeneous and anisotropic material consisting of two
primary materials, masonry units such as bricks, blocks etc., and mortar, which can be
cement or lime based with sand, soil and water. Both materials are different in their
behavior with quite different properties. Compressive strength and elastic modulus of
unit and mortar are the basic indicators of material quality. Bricks are relatively stiffer
than mortar with much higher compressive strength as compared to mortar. Hence,
generally it is believed that strength and stiffness of masonry would lie within mortar
and unit. Depending upon the availability of material, cost, climatic condition, living
habits and tradition, different types of unit and mortar combinations are adopted in dif-
ferent parts of the world. However as per the guidelines, natural stone masonry or adobe
masonry and masonry units with horizontal perforations are not recommended for CM
construction. The minimum compressive strength of masonry unit should be 4 MPa for
hand-made clay bricks, 5 MPa for solid and hollow concrete blocks and 10 MPa for
hollow clay units, machine-made and multi-perforated clay bricks [1]. Furthermore,
CM walls built using unreinforced low-strength hollow concrete blocks are generally
more prone to brittle failures as compared to the walls built using solid concrete and
clay units [1]. Yáñez et al. tested sixteen specimens of solid and perforated CM walls
built with concrete masonry and hollow clay brick [5]. The stiffness and strength ob-
tained for the solid CM walls were 49 kN/mm and 130 kN using concrete block ma-
sonry units, whereas, large strength and stiffness, i.e., 82 kN/mm and 199 kN were
obtained using hollow clay brick masonry units as shown in Fig. 4. By observing the
behaviour of all the test specimens, it was concluded that the hollow clay brick masonry
unit specimens present larger strength and stiffness degradation than the concrete ma-
sonry unit specimens, although the former specimens provide larger maximum hori-
zontal load and energy dissipation capacity.
240
200
Capacity (kN)
160
120
80
40 Concrete masonry CM wall

Hollow clay brick CM wall
0
0 2 4 6 8 10 12 14
Drift (%)
Fig. 4. Effect of masonry on lateral load-deformation behaviour of soild CM wall [5]
704
3.3 Confinement effect

The deficiencies in URM in seismic events are improved by providing RC elements
at the outer periphery of the masonry panel in the form of tie beam and tie column.
Masonry is a composite element with many joint interfaces. It performs fairly well un-
der compression but is weak in tension. Tensile damage is observed even at small dis-
placement levels causing premature failure of wall. The confinement prevents disinte-
gration of masonry which reduces stiffness degradation and enhances ductility charac-
teristic of wall. This further improves energy dissipation characteristics and capacity of
the wall [6]. The effectiveness of confinement depends on various factors such as size,
grade of concrete, grade of steel, percentage reinforcement, location etc.
Confining elements used in CM masonry are relatively small in cross section com-
pared to a moment resisting RC frame. Even joints are not treated as fully rigid joints
which do not provide effective frame action. Minimum dimension is generally recom-
mended as 150 mm  150 mm with 15 MPa as minimum compressive strength of con-
crete [1]. Even though the role of confining element is to confine masonry wall to avoid
premature failure, certain minimum reinforcement is required to avoid sudden failure
of members. Use of four 10 mm dia. deformed steel bars with yield strength of 400 MPa
is recommended as longitudinal reinforcement in tie-elements [1]. Also, low grade re-
inforcement can be used with suitable modifications. During a seismic event, maximum
damage concentrates at ground floor level due to high seismic load [7, 8]. It is obvious
that increase in percentage reinforcement will result in increasing lateral capacity sub-
stantially as shown in Figs. 5 and 6 [9, 10], therefore it is a wise choice to provide more
reinforcement in the confining elements of the ground floor. From the experimental
study of Quiroz et al. [10], it was observed that specimens with high tie-column rein-
forcement, (Wall Id: A1-1 and A1-2) exhibited higher lateral strength as shown in Fig.
5.
300
250
Axial Reinforcement in
200 Wall Id
Capacity (kN)
Tie-column Tie-beam
150 A1-1 0.734 % 0.473 %
100 A1-1 A1-2 0.734 % 0.845 %
50 A1-2 A2-1 0.411 % 0.473 %
A2-1
0
0 0.0025 0.005 0.0075 0.01
Drift (mm/mm)
Fig. 5. Effect of tie-column and tie-beam reinforcement on lateral load-deformation behaviour

of CM wall [10]
Although axial reinforcement increases the lateral load carrying capacity, the tie col-
umns and tie beams are also required to be provided with lateral ties to prevent brittle
shear failure mode. Therefore, lateral ties also play a vital role in the behavior of CM
705
wall. After the initial diagonal crack, when load increases the crack gets propagated to
tie column leading to shear concentration at the ends of the tie column. Adequate trans-
verse reinforcement is needed to achieve desired deformation and energy dissipation
characteristics [9]. It is evident from Fig. 6 that high axial reinforcement increases lat-
eral capacity and high shear reinforcement enhances ductility of wall by preventing
early shear failures. Mild steel bars of minimum 6 mm diameter at maximum spacing
of 200 mm are recommended with reduced spacing at column ends for the regions of
high seismicity [1].
270
180 Reinforcement in tie-column

Wall
90 Id Axial Shear
reinforcement reinforcement
Capacity (kN)
0 A 3.8% (Rich) 1.28% (Rich)

A B 3.8% (Rich) 0.3% (Poor)
-90
B C 0.99% (Poor) 1.28% (Rich)
-180 C D 0.99% (Poor) 0.3% (Poor)
D
-270
-30 -10 10 30
Displacement (mm)
Fig. 6. Effect of tie-column reinforcement on lateral load-deformation behaviour of CM wall

[9]
3.4 Interface between wall edge and tie-column

The bonding between a masonry wall and RC tie-elements is important for satisfac-
tory earthquake performance and for delaying undesirable cracking and separation at
the wall-to-tie-column interface. Generally, masonry units are staggered using toothing
or shear-key at the tie-column locations to achieve good bonding between a masonry
wall and adjacent RC tie-columns. Good bonding can also be achieved by providing
dowels anchored into RC tie-columns. A comparison study was performed by Ma-
tošević et al. [11] to study the in-plane response of CM walls with different connection
details - toothed connection, U-shaped dowel connection and no connection. The test
results showed that the connection details did not influence the initial stiffness or the
maximum lateral resistance significantly, but they did improve the nonlinear wall be-
haviour and hysteretic energy dissipation. Wall with toothed or U-shaped dowel con-
nectors exhibited more ductile behaviour compared to the one with no connection (Fig.
7a). The connection not only improves in-plane behaviour but also improves the out-
of-plane stability of wall. Singhal and Rai [12] tested half-scaled CM wall specimens
with different density of toothed connection and considered the successive applications
of out-of-plane and in-plane loading. The seismic performance of CM wall in compar-
ison with that of a typical infilled RC frame wall was also studied. It was observed that
the CM walls with or without toothing exhibited improved in-plane and out-of-plane
responses in comparison to infill masonry panels. The increased density of toothing
706
caused significant improvement in post-peak behavior under in-plane loads as shown

in Fig. 7b, however, it didn’t have significant effect on out-of-plane behavior [12].
180 120
(a) (b)
150
90
Capacity (kN)
Capacity (kN)
120
90 60
60 Infilled RC
No connection 30 CM-No toothing
30 Toothed connection CM-Fine toothing
U-shaped dowel connection CM-Coarse toothing
0 0
0 0.2 0.4 0.6 0.8 1 0 0.5 1 1.5 2
Drift (%) Drift (%)
Fig. 7. Effect of wall to tie-column interface on lateral load behaviour of CM wall: (a) speci-
mens tested by [11], (b) specimens tested by [12]
3.5 Aspect ratio
Depending on the aspect ratio (AR), i.e., height to length ratio of masonry panel,
different failure modes can occur in CM wall. The walls with low aspect ratio (length
of wall is larger than the height) mainly fail in shear mode, however, the walls with
large aspect ratio (length of wall is much less than the height) are more susceptible to
flexural mode of failure due to excessive bending [2, 4]. Therefore, squat CM walls
exhibit a mixed shear failure mode with combination of diagonal tension and sliding
along one or more horizontal joints [13]. For slender walls, the typical damage includes
horizontal tensile cracks in mortar and tie-columns at the tension end of the wall and
crushing of bricks in the compression zones [2, 14]. Though shear strength of CM wall
increases with decrease in aspect ratio, the drift corresponding to ultimate load de-
creases. Therefore, a sudden and brittle failure is observed in squat walls as shown in
Fig. 8 [13].
350
AR = 2.13
300
AR = 1.48
Capacity (kN)
250 AR = 1.18
200 AR = 0.96
150 AR = 0.58
100
50
0
0 0.003 0.006 0.009 0.012 0.015
Drift (mm/mm)
Fig. 8. Effect of aspect ratio on lateral load-deformation behaviour of CM wall [13]
707
3.6 Overburden pressure

It is one of the influencing parameters on the lateral strength of CM wall. When a
masonry panel is subjected to lateral load, layers of masonry units tend to slide over
one another. The resistance offered against sliding is developed by friction and adhe-
sion between bricks and mortar. As frictional resistance is directly related to normal
stress applied, the axial load improves lateral capacity and energy dissipation charac-
teristics of CM wall. Fig. 9 shows the load-deformation behaviour of CM walls (with
no wall reinforcement and low height of repeated lateral forces) tested by Yoshimura
et al. [15]. From the figure, it can be concluded that shear strengths of the CM walls
increases with increase in vertical axial load, though the failure occurs at a lower drift.
350
280
Capacity (kN)
210
140
High compression
70 Low compression
Vertical tension
0
0 0.002 0.004 0.006
Storey drift (rad)
Fig. 9. Effect of overburden pressure on lateral load-deformation behaviour of CM wall [15]
3.7 Wall reinforcement
Though the placing of horizontal wall reinforcement within the mortar joints is an
option to improve the seismic response of CM wall, it is not a much popular choice. As
masonry is weak in tension, when a wall panel is subjected to lateral load diagonal
cracks get developed in the direction perpendicular to the principle tensile stress. This
cracking governs the lateral capacity, stiffness degradation and energy dissipation char-
acteristics of wall. With provision of horizontal wall reinforcement, the process of crack
initiation and propagation gets delayed as reinforcements help in resisting tensile
stresses. Moreover, improved lateral load, deformation and energy dissipation capacity
is achieved with more uniform distribution of inclined cracking [14, 16, 17]. Fig. 9
shows the behaviour of single paneled CM wall with and without horizontal reinforce-
ment. For the CM walls with horizontal reinforcement, 2-high strength deformed wires
at a spacing of 280 mm c/c with reinforcement ratio 0.071% and 0.190% are used.
3.8 Opening
The opening presents in typical masonry wall has negative effects on the capacity
when subjected to seismic load [5, 12, 18, 19]. Under the action of lateral loading, stress
708
10
No horizontal reinforcement
0.8 0.071 % horizontal reinforcement
0.190 % horizontal reinforcement
Capacity (MPa)
0.6
0.4
0.2
0
0 0.005 0.01 0.015 0.02
Drift (mm/mm)
Fig. 9. Effect of horizontal reinforcement on lateral load-deformation behaviour of CM wall

[16]
concentration is observed at the corners of openings causing shear cracks, which makes
wall panel unstable and leads to failure of CM wall. For a CM wall with opening size
up to 10% of wall area have same in-plane behavior under lateral load as a solid wall
[1]. Beyond this opening size, strength and stiffness reduces drastically.
The deficiencies due to openings can be overcome by provision of confining ele-
ments around the openings [19]. Confining elements prevent shear cracks at corners of
openings and maintain stability of wall. It also allows strut formation in the wall, which
improves strength and ductility significantly. Many standards and manuals suggest dif-
ferent location and arrangement of confining elements around opening. There are some
disagreements on detailing and arrangement of confining elements. Experiments
showed that the provision of continuous sill and lintel beam is a better alternative in
terms of ductility of CM wall panel as shown in Fig. 10 [12]. In-plane capacity of con-
tinuous tie-column and continuous sill and lintel beam arrangement is almost same but
the later system is more ductile as the beam divides the wall into smaller panels with
low aspect ratio ensuring well distributed diagonal shear cracks throughout the wall.
120
90
Capacity (kN)
60
30 No confinement
Vertical confinement
Horizontal confinement
0
0 1 2 3
Drift (%)
Fig. 10. Effect of confinement around opening on the lateral load-deformation behaviour a CM
wall [12]
709
11
4 Conclusion
In the age of RC constructions, around 80% of buildings in India are unreinforced

masonry structures. With almost 54% of geographical area vulnerable to earthquakes,
unreinforced masonry buildings impose great risks because of their poor seismic per-
formance. Confined masonry is one of the most suitable and economical building ty-
pologies for a low-to-medium rise low-cost construction practices. The article identifies
some of the important parameters which affect the seismic performance of CM building
and these are – wall density, masonry wall properties, confining element properties,
interface between wall-to-confining frame, aspect ratio of wall, overburden pressure,
wall reinforcement, opening area etc. Among them, parameters like masonry strength,
aspect ratio etc. are the strength dependent parameters, whereas, tie-column reinforce-
ment ratio, wall reinforcement etc. are mainly ductility dependent parameters. Salient
conclusion which can be drawn from this study are briefly discussed in the following:
 Masonry characteristics are mainly responsible for the strength and stiffness
of CM structure.
 Confining frame does not have significant contribution to lateral strength,
however, it improves in-plane ductility and reduces stiffness degradation of
wall.
 Wall-to-tie-column interface in the form of toothing or U shaped dowels im-
proves out of plane stability.
 Wall with confined opening in the form of continuous horizontal beam at lintel
and sill performs better than other alternative confining schemes.
 Aspect ratio, though not widely studied, controls failure modes and initial
stiffness and strength of CM wall.
 Wall reinforcement delays crack initiation and improves ductility of wall.
It is observed that a lot of attention has been given in the literature to tie-column and
masonry properties in understanding in-plane behavior of CM wall, whereas less atten-
tion has been given to study the behavior of wall with openings. Interface connection
between wall and tie-columns and wall and tie-beams also need to be investigated to
understand the out-of-plane behavior of CM wall. Systematic experimental studies are
required be conducted to understand the influence of various parameters on lateral load
response of CM buildings to prepare a unified design protocol for such buildings.
References
1. Meli, R., Brzev, S., Astroza, M., Boen, T., Crisafulli, F., Dai, J., Farsi, M., Hart, T., Mebarki,
A., Moghadam, A. S., Quinn, D., Tomazevic, M., Yamin L.: Seismic Design Guide for Low-
Rise Confined Masonry Buildings. World Housing Encyclopedia: EERI-WHE (2011).
2. Matthews, T., Riahi, Z., Centeno, J., Charlet, A., Garcia, H. J., Hoffman, C., Safaie, S., El-
wood, K.: Evaluation of Confined Masonry Guidelines for Earthquake-Resistant Housing.
University of British Columbia (UBC) EERI Student Chapter Committee on Confined Ma-
sonry Construction (2007).
710
12
3. Ishibashi, K., Meli, R., Alcocer, S. M., Leon, F., San- chez, T. A.: Experimental Study on
Earthquake-Resistant Design of Confined Masonry Structures. In: 10th World Conference
on Earthquake Engineering, pp. 3469-3474, Madrid (1992).
4. Brzev, S.: Earthquake-Resistant Confined Masonry Construction. National Information
Center for Earthquake Engineering, Kanpur, India (2008).
5. Yáñez, F., Astorza, M., Holmberg, A., Ogaz, O.: Behavior of confined masonry shear walls
with large openings. In: Proc. 13th World Conference of Earthquake Engineering, paper no.
3438. Vancouver, Canada (2004).
6. Tomaževič, M., Klemenc, I.: Seismic behavior of confined masonry walls. In: Earthquake
Engineering and Structural Dynamics, 26(10), 1059–1071 (1997a).
7. Alcocer, S., Arias, J.G., Vazquez, A.: Response Assessment of Mexican Confined Masonry
Structures Through Shaking Table Tests. In: Proceedings of the 13th World Conference on
Earthquake Engineering, paper no. 2130, Vancouver, Canada (2004a).
8. Tomaževič M., Bosiljkov V., Weiss P.: Structural behaviour factor for masonry structures.
In: 13th World Conference on Earthquake Engineering (2004).
9. Kato, H., Goto, T., Mizuno, H.: Cyclic loading tests of confined masonry wall elements for
structural design development of apartment houses in the third world. In: Proc. 10th World
Conference on Earthquake Engineering, Balkema, Rotterdam (1992).
10. Quiroz L. G., Maruyama Y., Zavala C.: Experimental assessment of the cyclic behavior of
Peruvian confined masonry walls and numerical modeling using genetic algorithms. In: 16th
World Conference on Earthquake Engineering (2017).
11. Matošević, Ɖ., Sigmund, V., Guljaš, I.: Cyclic testing of single bay confined masonry walls
with various connection details. In: Bulletin of earthquake engineering, 13(2), pp. 565-586
(2015).
12. Singhal, V., Rai, D. C.: In-plane and out-of‐plane behavior of confined masonry walls for
various toothing and openings details and prediction of their strength and stiffness. Earth-
quake Engineering & Structural Dynamics, 45(15), 2551-2569 (2016).
13. Gavilán, Pérez J. J., Flores, L. E., Alcocer, S. M.: An experimental study of confined ma-
sonry walls with varying aspect ratios. Earthquake Spectra, 31(2), 945-968 (2015).
14. Zabala, F., Bustos, L. J., Masanet, A., Santalucia, J.: Experimental Behavior of Masonry
Structural Walls Used In Argentina. In: 13th world conference on earthquake engineering,
paper no. 1093, Vancouver, B. C., Canada (2004).
15. Yoshimura, K., Kikuchi, K., Kuroki, M., Liu, L., Ma, L.: Effect of Wall Reinforcement,
Applied Lateral Forces and Vertical Axial Loads on Seismic Behavior of Confined Concrete
Masonry Walls. In: 12th World Conference on Earthquake Engineering, paper no. 0984
(2000).
16. Aguilar, G., Meli, R., Diaz, R., Vazquez-del-Mercado, R.: Influence of Horizontal Rein-
forcement on the Behaviour of Confined Masonry Walls. In: 11th World Conference on
Earthquake Engineering, paper no. 1380, Acapulco (1996).
17. Kumazawa, F., Ohkubo, M.: Nonlinear Characteristics of confined Masonry Wall with Lat-
eral Reinforcement in Mortar Joints. In: Proc. 12th World Conference on Earthquake Engi-
neering, paper no. 0743, Auckland, New Zealand (2000).
18. Aguilar, G., Meli, R., Diaz, R., Vazquez-del-Mercado, R.: Influence of Horizontal Rein-
forcement on the Behaviour of Confined Masonry Walls. In: 11th World Conference on
Earthquake Engineering, paper no. 1380, Acapulco (1996).
19. Kuroki, M., Kikuchi, K., Nonaka, H., Shimosako, M.: Experimental Study on Reinforcing
Methods Using Extra RC Elements for Confined Masonry Walls with Openings. In: Proc.,
15th World Conference on Earthquake Engineering, paper no. 4967, Lisbon, Portugal
(2012).
711
Paper ID- 139
Design and Performance Criteria for Fire Resistant

Design of Structures – An Overview
Nitant Upasani1* (0000-0002-0729-254X), Mansi Bansal1(0000-0001-7014-4745),

Ashirbad Satapathy1(0000-0003-2062-2977), Sanket Rawat1(0000-0002-1985-7579),
G. Muthukumar1(0000-0002-4836-0975)
1Department of Civil Engineering, BITS Pilani, Pilani Campus, Rajasthan, India - 333031
*Corresponding author, Email: nitantupasani@gmail.com
Abstract. Concrete, in spite of being inherently fire resistant, cannot be

considered as a fire-proof material. It undergoes substantial variation in its
characteristics during exposure to elevated temperatures. These variations may
become hazardous for structural stability and serviceability depending upon the
type and extent of exposure. Moreover, high strength concrete, which is
commonly used in tall buildings, may perform poorly against fire due to its high
binder content and very low permeability. In spite of all the aforementioned
factors, the performance under fire has been given very restricted consideration
in the current Indian practice corresponding to the design provisions of various
structural elements. Although, several standards provide guidelines to achieve
fire safety in structures, still the provisions for high strength concrete and spalling
prevention are not specifically available in the major international and national
standards. This paper attempts to compare the provisions corresponding to fire
resistant design in Indian standard with the respective provisions of other country
standards e.g. ACI 216 (2007), NZS 3101(2006), EN 1991-2 (2004) etc.
Comparison parameters primarily include the design requirements for various
structural members to improve their fire resistance. Furthermore, suitable
recommendations for Indian Codal provisions are attempted in the later part of
the paper to achieve superior performance under elevated temperatures.
Keywords: Elevated temperature, Reinforcement cover, Fire resistance rating,

Spalling.
1 Introduction
In the recent decades, India has encountered remarkable urban development. Population
explosion and migration to urban areas in the country has put a colossal pressure on the
housing and infrastructure needs of the nation. As a result, development of substantial
number of multi-storeyed buildings in the urban areas has occurred and is on the ascent.
Despite the fact that there has been such incredible infrastructural development in the
nation in the recent times, the fire safety has not been given proper consideration.
Unavailability of proper guidelines and ignorance have caused horrific accidents in the
past that took the lives of many and also caused significant monetary losses. Although
various Indian standards provide guidelines to enhance structural performance during
fire, the provisions available are too elementary and do not seem to cover all the
significant parameters. The provisions available in many international standards are
712
also similar and they also restrict the consideration of significance of fire performance
in design. In general, all the national and international standards have specified
minimum size of member and the cover to the reinforcement corresponding to different
fire resistance rating (FRR). FRR for structural elements can be defined as the time, in
minutes or hours, that a particular structural member or element can withstand a
standard fire exposure. It is generally considered to range from 30 to 240 minutes by
several countries’ standards. The effect of various parameters such as the types of
aggregates to be used, cover requirements, design principles for various structural
elements such as beams, columns, and slabs whose integrity forms the prime
importance for the safety of the building, with respect to fire resistance has also been
studied extensively. Also, although the performance of high strength concrete is poor
when exposed to fire, its use for construction of multi-storeyed structures is essential
and cannot be avoided. Hence, it is extremely crucial that special care be taken for
provision of enhancement of fire resistance of multi-storeyed buildings. Moreover,
spalling which is one of the chief causes of deterioration of concrete at elevated
temperatures, has not been considered by the standards of several countries and no
specific provision has been recommended to prevent this phenomenon.
This paper aims at providing a comparative study of the standards of various countries
and suggesting additional measures for better design approaches and improved
performance of reinforced and prestressed concrete structures under fire exposure.
Certain parameters which have been found to have effect on the fire performance of
concrete by various researchers have also been discussed.
2 Indian Specifications for Fire Resistant Design
Codal specifications for concrete structures in India mainly distinguish between the
design for different structural members such as columns, beams, slabs etc. on the basis
of fire resistance rating and size of the members. The major emphasis in Indian
standards for the design of reinforced concrete members when exposed to fire, is to
ensure that the structural members have sufficient thickness or width and cover over
the reinforcement. Besides, occurrence of the spalling phenomenon in high strength
concrete due to fire exposure, has also been accounted for in terms of extra cover to
reinforcement. On the other hand, Indian codes lack in suggesting any other provisions
for improved fire resistant design of concrete structures. Salient specifications related
to fire resistance of concrete structures are discussed in the following passages.
2.1 IS 1642:1989 Fire safety of buildings (General) – Details of construction
Indian standard IS 1642 [1] specifies minimum dimension and cover to the
reinforcement corresponding to different fire resistance period (Table 1). Separate
guidelines are provided for beams, columns and slabs depending upon the support
condition and degree of exposure.
713
Table 1: Minimum dimension of structural elements for different fire resistance ratings
as per IS 1642:1989
Minimum Beams Slabs Columns

Dimension
(mm), Reinforce Reinforced Reinforce Reinforced Fully exposed 50 % exposed One face exposed
Excluding any concrete - concrete - concrete - concrete -
finish for a fire Simply Continuous Simply Continuous
resistance of supported supported
(hours)-
Width Cover Width Cover Width Cover Width Cover Width Cover Width Cover Width Cover
0.5 h 80 20 80 20 75 15 75 15 150 20 125 20 100 20

1h 120 30 80 20 95 20 95 20 200 25 160 25 120 25
1.5 h 150 40 120 35 110 25 110 20 250 30 200 25 140 25
2h 200 60 150 50 125 35 125 25 300 35 200 25 160 25
3h 240 70 200 60 150 45 150 35 400 35 300 30 200 25
4h 280 80 240 70 170 55 170 45 450 35 350 35 240 25
2.2 IS 456:2000 Indian Standard-Plain and reinforced concrete -code of practice
Indian code IS 456 [2] is widely followed for the construction of concrete structures in
India. To incorporate the benefit of inherent fire resistance of concrete, few
specifications are highlighted in the standard. Hence, the innate fire resistance of
concrete is assured by providing minimum overall member dimensions and nominal
concrete cover to the steel reinforcement in design, similar to the earlier standard.
Clause 26.4.3 of the code, assures the fire resistance by providing minimum nominal
cover to the normal weight aggregate concrete. Also, the code has special provisions
for increase in nominal cover in case of structural members with more risk of spalling.
Dependency of fire resistance of concrete on type of aggregates i.e. lightweight
aggregates or normal weight aggregates, has also been mentioned. However, minimum
dimension and cover to the reinforcement stipulations for various concrete structural
members have only been provided for normal weight aggregates. The code also
emphasises on detailing of reinforcement being such that both individual and whole
structure remain integrated throughout the exposure period. Additional measures to
avoid the risk of spalling, such as, use of fire resistant finishes for insulation and
provision of fire resistant false ceilings increase in cover to reinforcements, and
provision of sacrificial steel in tensile zone have also been recommended by the code.
Specifications based on minimum cover and member dimensions for normal-weight
aggregate concrete members for required fire resistance are specified as shown in the
Table 2 for different structural members.
714
Table 2: Minimum dimension of structural elements for different fire resistance ratings
as per IS 456:2000
Minimum Beams Slabs Columns
Dimension
Cover Cover Column width/diameter
(mm),
Excluding Minimum Minimum
Nominal
any finish Reinforced Reinforced width of Reinforced Reinforced thickness of Fully exposed 50 % One face
Cover
for a fire concrete concrete beam concrete concrete slab exposed exposed
resistance of Simply Continuous Simply Continuous
(hours)- supported supported
0.5 h 20 20 200 20 20 75 150 125 100 40
1h 20 20 200 20 20 95 200 160 120 40
1.5 h 20 20 200 25 20 110 250 200 140 40
2h 40* 30 200 35* 25 125 300 200 160 40
3h 60 40* 240 45 35* 150 400 300 200 40
4h 70 50 280 55 45 170 450 350 240 40
*Cases that are star marked require attention to the additional measures necessary
to reduce the risks of spalling.
3 International Provisions
3.1 American Standard (ACI 216.1-07/ TMS 0216-07)
The American Standard for fire resistance of reinforced concrete structures provides
minimum dimension and cover to the reinforcement specifications, similar to the Indian
codes. In addition, the type of aggregate used is also taken into account in the American
code, according to which the minimum size of a member is provided depending on the
fire-rating of the building. Member size and cover thickness are given for restrained or
unrestrained beams and slabs, for prestressed or passively reinforced members.
Continuous unrestrained members have longer fire endurance than simply supported
members for the reason that the former can redistribute moments, eventually resulting
in failure of negative reinforcement over the supports. In concrete design, for ambient
conditions, the amount of negative reinforcement is limited to avoid compressive,
brittle failure of the member. It specifies the advantage of better spalling resistance of
carbonate aggregates than the siliceous aggregates. The specifications are provided
differently for different structural members. For columns, ACI 216 [3] distinguishes
concrete columns first, on the basis of compressive strength with the threshold for
differentiation being 12000 psi. It is important to note that the American code is more
conservative and insists on providing an FRR of at least 1 hour for each structural
715
member, unlike other codes. The minimum dimension for a particular FRR of columns
having compressive strength less than or equal to 12000 psi has been specified in the
code. For columns with compressive strength greater than 12000 psi, the least
dimension for any type of aggregates should be 24 inches, according to the code.
Furthermore, the minimum cover to the reinforcement specified in the standard does
not depend upon the aggregate type and compressive strength and hence taken as 2
inches, or 1-inch times the required FRR; whichever is less. For beams and slabs, the
American Standard provides specific guidelines for reinforcement detailing to attain a
certain rating of fire resistance in addition to the recommendations provided for the
minimum cover requirements and dimensions for continuous beams and slabs.
Redistributed maximum positive moment must be the primary basis to determine the
fire resistance of the bending members. When the continuous beams and slabs are
exposed to fire at the bottom, the temperature increases at the bottom and a temperature
gradient is established between the top and the bottom surface which causes the
expansion of the bottom surface. Consequently, the lift at the ends causes the reduction
in positive moments and increase in negative moments at interior supports. Hence, it is
recommended to increase the negative moment carrying capacity of the members.
Furthermore, the maximum reinforcement that has to be provided in the negative
moment region is given by,
𝜔=𝜌𝑓𝑦/𝑓𝑐′≤0.30 (1)
where,
𝜌=𝐴𝑠/𝑏𝑑, to limit the compressive failure in this region.
𝐴𝑠 is the area of steel reinforcement of yield strength 𝑓𝑦,
𝑓𝑐′ is the concrete compressive strength,
𝑏 and 𝑑 are the width and depth of the beam or slab strip.
When floors and roofs are made of one layer of normal weight concrete and one layer
of lightweight concrete, it’s fire resistance can be formulated by numerically by
empirical solutions given below:
• If the layer exposed to fire is of normal weight concrete
R = 0.057(2t2 – dt + 6/t) (2)
• If the layer exposed to fire is of lightweight concrete
R = 0.063(t2 + 2dt – d2 + 4/t) (3)
where, R is the fire resistance (in hours), t is total slab thickness (in inches) and d is
thickness of the layer exposed to fire. Formulation for fire resistance rating when the
slab comprises of more number of layers
When the slab is made up of more number of layers of different types of materials, fire
resistance may be calculated as:
FRR = (FRR10.59 + FRR20.59 +......+ FRRn0.59 + A1 + A2 +......+ An)1.7 (4)
where, FRRn is the fire resistance rating of nth layer and An represents the air factor (0.3)
for each continuous air gap.
716
3.2 New Zealand Standard (NZS 3101: Part 1: 2006 Concrete Structures
Standard)
New Zealand standard mainly recommends to design a structural member with fire
resistance rating greater than the required fire resistance and the parameters to judge
the fire resistance are defined as – adequacy, integrity and insulation. The integrity
criteria are considered to be satisfied if the member meets the requirement of the rest
two criterions. The insulation criteria can be made as per the requirements if the surface
is covered by a layer of insulating material or if the cover of the reinforcement be
increased by providing a layer of insulating material. NZS 3101 [4] has taken into
account a factor different from other standard for fire resistance design of structural
members which is “Axis-distance”. Axis distance is defined as the weighted average
distance of a group of longitudinal bars from the axis of the bars to the nearest exposed
surface.
To achieve the integrity, adequacy and insulation criteria for slabs, FRR is prescribed
from half hour to 4 hours, similar to Indian standards and corresponding to each
duration, minimum dimension and axis distance have been detailed. Provisions are
specified for columns and walls similar to beams and slabs. In case of columns and
walls, an additional parameter has been defined named as “Load Value”, and an
expression is provided for its calculation which depends on axial load capacities at
normal and elevated temperatures. This expression is given by:
𝜂fi = Nf*/Nu (5)

where,
𝜂fi → load value.
Nf* → factored design axial load on the column in fire condition.
Nu → axial load capacity of column at room temperature.
Load value is either taken as 0.7 or is calculated as per the given formula. Hence, the
axis distance and minimum dimensions are stated for columns depending upon fire
resistance rating, exposure percentage and load value. Furthermore, different provisions
are stated for fire resistance of simply supported and continuous beams and also the
effect of percentage exposure of beam to fire is also indicated in terms of change in
dimensions and cover accordingly in separate clauses. NZS 3101 standard insists on
different fire resistance rating for different types of slabs i.e. solid slabs, flat slabs and
ribbed slabs. Additionally, the axis distance has been listed for different categories of
solid slabs (one way, two-way and continuous) for each fire durations.
Also, subjected to the insulation parameter, an alternative is to use lightweight
aggregates. They generally have better fire resisting capacity in dry building fires
because they have thermal expansion almost equal to the cement paste as a result of
which they have good bonding between aggregates and concrete and are stable at high
temperatures.
3.3 Australian Standards AS 3600 - 2009 for concrete structures
The Australian code defines cover to reinforcement in terms of “Average Axis

Distance, am” and provides specifications to calculate the same for the presence of
717
single layered or multi-layered reinforcements of same or different size and

characteristic strength, prestressing tendons and combinations of these. AS 3600 [5]
specifies minimum dimensions and cover to the reinforcement for concrete sections to
achieve a given fire resistance to a standard fire exposure similar to other standards.
The code classifies beams on the basis of support conditions and provides combinations
of Avg. Axis Distance (am) & Width of Beam (b), in tabular form as well as its graphical
correlation which signifies that for the same FRR requirement, with increase in width
of the beam (b), the requirement of Avg. Axis Distance (a m) decreases.
AS 3600 [5] has also laid down conditions for beams exposed to fire on all sides.
Fig. 1. Structural adequacy requisites for fire resistance of beams: (A) simply
supported beams (B) continuous beams [5]
AS 3600 [5] has prescribed values of dimensions corresponding to various FRRs for
satisfying Insulation as well as Adequacy criteria separately in case of slabs. For
adequacy of slabs, the code mentions prescriptive criteria and distinguishes slabs on the
basis of core (solid/hollow), ribbed, support condition as well as on the basis of one-
way or two-way behaviour of slabs. However, the Indian code doesn’t specifically
mention any difference in values of thickness or cover for one-way and two-way
behaviour of slabs.
For columns, the Australian Code provides two tabular methods viz. Restricted Tabular
Method and General Tabular Method for determination of dimensions and cover
parameters (structural adequacy), both of which are prescribed for braced columns. The
restricted tabular method mentions the Avg. Axis Distance (as) and smaller cross-
sectional dimension (b) requirements with respect to a varying parameter called load
level (µfi) which can be calculated as follows:
𝑁𝑓∗
µ𝑓𝑖 = (6)
𝑁𝑢
718
where,
Nf∗ = design axial load in the fire situation
Nu = ultimate strength in compression, or tension, at a cross-section of an eccentrically
loaded compression or tension member respectively
The general tabular method further prescribes combinations of Avg. Axis Distance (as)
and smaller cross-sectional dimension (b) with respect to the variation of two more
parameters viz.
ω = 1.3 As fsy /Ac fc′ (7)

denotes the mechanical reinforcement ratio at normal temperature conditions and ‘ɳ’
which is given by:
N∗f
ɳ= A f ′ As fsy
(8)
0.7[( c c)+( )]
1.5 1.15
where,
As = cross-sectional area of reinforcement
Ac = smallest cross-sectional area of the concrete strut at any point along its length and
measured normal to the line of action of the strut
fsy = characteristic yield strength of reinforcement
fc′ = characteristic compressive (cylinder) strength of concrete at 28 days
3.4 Canadian Code NBC 2010
National Building Code of Canada [6] and the Canadian Commission on Building and
Fire Codes (CCBFC) state that low density aggregate concretes generally perform
better under fire action than natural stone aggregate concretes. Furthermore, it
segregates the types of concrete on the basis of the nature of the aggregate used which
is described as follows:
• Type S concrete: Coarse Aggregate is granite, quartzite, siliceous gravel / dense
materials containing at least 30% quartz, chert or flint.
• Type N concrete: Coarse Aggregate is cinders, broken brick, blast furnace slag,
limestone, calcareous gravel, trap rock, sandstone / similar dense material
containing not more than 30% of quartz, chert or flint.
• Type L concrete: All aggregates are expanded slag, expanded clay, expanded shale
or pumice.
• Type L1 concrete: All aggregates are expanded shale.
• Type L2 concrete: All aggregates are expanded slag, expanded clay or pumice.
• Type L40S concrete: The fine portion of the aggregate is sand and low density
aggregate; sand does not exceed 40% of the total volume of all aggregates
• Type L120S & Type L220S concretes: The fine portion of the aggregate is sand
and low-density aggregate; sand does not exceed 20% of the total volume of all
aggregates.
The Canadian code prescribes the same minimum cover to principal reinforcement for
different types of aggregate concretes in case of reinforced concrete beams. However,
719
for prestressed concrete beams, the minimum thickness of concrete cover over steel
tendons depends on the types of concrete as well as on the cross-sectional area. For
minimum dimension criteria, it only stipulates a minimum width of at least 100 mm for
reinforced concrete beams.
For reinforced and prestressed concrete slabs, the code prescribes minimum thickness
and minimum cover over reinforcement for the different types of concrete (Types S, N
& L). In case of reinforced concrete columns, the code mentions a term ‘Overdesign
Factor’ which can be defined as the ratio of the calculated load carrying capacity of the
column to the strength of column required to carry a specified load. For various types
of aggregate concrete, the minimum dimension (t) of rectangular columns depends
upon the fire-resistance rating required (R), the area of vertical reinforcement as
percentage of the column area (p) and a factor ‘f’ which varies with the overdesign
factor, the effective length factor (k) and the unsupported length of the column (h). The
diameter requirement of a circular column can be calculated by adding an increment of
20% to the minimum dimension as calculated above for a given type of concrete. For
the calculation of minimum thickness of concrete cover over the vertical steel
reinforcement, the code prescribes empirical formula depending upon the number of
hours of fire resistance required.
3.5 Eurocode 2
EN 1992-1-2 [7,8] incorporates the effect of aggregate on fire resistance of reinforced

concrete structural sections by means of providing a factor K to allow for the decrease
in the characteristic compressive strength for different types of aggregates, as shown in
Figure 1. The standard primarily specifies this factor for carbonate and siliceous
aggregates in the both graphical and tabulated form. Eurocode specifies fire resistance
of reinforced concrete columns, similar to Australian standard which depends upon
exposure level and gives minimum dimension as well as concrete cover or axis distance.
In addition to these factors, Eurocode takes into consideration another parameter named
as “Degree of utilisation in fire situation” which also plays a significant role in
describing minimum member dimension and axis distance. This factor is almost similar
to the Load value as specified in New-Zealand standard and is given by
𝑁𝐸𝑑.𝑓𝑖
𝜇𝑓𝑖 = (9)
𝑁𝑅𝑑
where,
μfi → Degree of utilisation in fire situation.
NEd.fi → Design axial load in fire situation
NRd → Design resistance of column at normal temperature.
Additionally, a table provided in this standard for different fire resistance rating of
concrete is valid for only certain conditions, unlike all other standards where there is
no mention of the validity of the table. The conditions are –
Effective length of the column under fire conditions ≤ 3 m
- first order eccentricity under fire conditions: e ≤emax
- amount of reinforcement: As < 0.04 Ac
720
Fig. 3. Variation in coefficient k (ϴ) with increasing temperature for normal weight
c
concrete with siliceous (solid line) and calcareous aggregates (dashed line)
Web dimension of the beam depending upon the class of the section as WA, WB and
WC is also presented in Eurocode. It is interesting to note that Eurocode covers the
salient provisions in case of high strength concrete. Unlike every other standard, it has
specified the criteria for regulating the spalling phenomenon during fire exposure.
Furthermore, with increase in chances of spalling on use of high strength concrete, the
Euro code takes into account special provisions to take care of the spalling. For high
strength concrete, reduction in strength is applied and this reduction has been made
varying upon the class of concrete. In general, the methodology applied for higher
strength concrete is almost same as that for normal strength concrete, with the additional
inclusion of a factor k.
Spalling provisions: Additionally, Eurocode also suggests to limit the moisture content
of concrete to control explosive spalling. The recommended limiting moisture content
by mass is mc = 3 %. If the moisture content exceeds mc, spalling effect on load bearing
may be measured with assumption of local loss of cover to one reinforcing bar or bundle
of bars in the cross section, then calculating the subsequent increase in temperature and
the reduced load bearing function R.
When the cover requirements exceed 40mm to 50mm, the following measures are
suggested to prevent the spalling phenomenon.
• Vermiculite being a thermally stable and resistant material, its use for
plastering can be beneficial.
• To prevent the fire spread to the upper floors, false ceiling made of fire
resistant materials can be utilized.
• Use of lightweight aggregates that perform better at elevated temperatures.
• Use of sacrificial steel.
It is interesting to note that, the use of supplementary cementitious material in concrete
is highlighted in terms of fire performance. If the fly ash content > 6 %, and for 80/95
< C grade < 90/105, at least one of the following measures must be provided:
721
A. A reinforcement mesh with:

• Minimal cover of 15 mm,
• Wire diameter greater than 2 mm
• Maximum pitch of 50 mm × 50 mm.
• The nominal cover to the main reinforcement should not be less than 40
mm.
B. Concrete used must be confirmed that it does not spall when exposed to
elevated temperatures using various tests.
C. Proven protective layers that exhibit the prevention of spalling when subjected
to fire conditions must be used.
D. At least 2 kg/m3 of monofilament polypropylene can be mixed with concrete
to improve its fire resistance.
Dependency of the frame on the individual member for its stability aids in deciding the
provision of spalling protection for that member. Spalling prevention design is usually
for the members impinged on flames, or when the temperature of the cover
reinforcement is in excess of 100℃.
4 Discussion
It can be analysed from the above discussion that the guidelines provided by different
countries, including India, lack somewhere in taking into account number of important
parameters for the construction of fire-resistant structures. Table 4 provides a concise
comparison of the adopted parameters and the corresponding guidelines in the different
codes mentioned above, for the enhancement of performance against fire exposure.
Table 4. Comparison of parameters specified in different standards related to fire

performance of concrete
Codes IS 1642 ACI 216 NZS 3101 AS 3600 [5] NBCC [6] Eurocode 2
[1] and IS [3] [4] [7,8]
456 [2]
Measures Minimum Minimum Minimum Minimum Minimum Minimum
dimension dimension, dimension dimension, dimension, dimension,
, concrete , Axis concrete reinforcemen Axis
concrete cover distance cover(axis t cover distance
cover distance) Spalling
Provision
Depending Support Aggregate Exposure Exposure, Type of Aggregate
Parameter condition, type, (%), Load Support concrete type,
s Degree of Restrained/ value condition (depending support
exposure unrestraine on type and condition
d beams, volume of exposure,
Exposure aggregates strength of
used), concrete,
Overdesign degree of
Factor utilisation
Special FRR 1 hr - Graphical Overdesign Spalling
features 4 hr Correlation Factor provisions,
s cementitiou
s materials
722
4.1 Type of Aggregates

The type of aggregates exceptionally influences thermal performance as it constitutes
70-80 % of volume of concrete. Studies have found that silica, the major constituent of
siliceous aggregates, causes significant loss in the strength of concrete beyond
600ᵒC. According to Ma et al. [9], use of siliceous aggregates, such as granite, in
concrete demonstrates unfavourable mechanical properties at high temperatures as
compared to the performance of calcareous aggregates in concrete, such as dolomite
and limestone. As observed by Cheng et al. [10], concrete made of calcareous
aggregates attains larger ultimate strain than that made from siliceous aggregate, thus
rendering better ductility to concrete at elevated temperatures. Moreover, calcareous
aggregates decompose at a higher temperature than siliceous aggregates. However, both
of these aggregates portray greater resistance to fire than usual river aggregates. On the
other hand, several codes have also discussed the suitability of use of lightweight
aggregates in concrete and have found that they perform better under fire exposure
capacity [1,3-5] as their coefficient of thermal expansion is almost equal to that of
cement paste which results in effective bonding and stability at high temperatures [4].
They are generally the by-product of incineration or volcanic eruption and therefore
have high resistance to heat and low heat conductivity. Thus, concrete manufactured
using lightweight aggregates shows better mechanical as well as insulating properties
under fire as compared to normal weight aggregates.
4.2 Structural Elements - Hollow Steel Columns

It has been observed that the formation of cracks in a normal concrete column triggers
at a temperature of 300ᵒC and the phenomenon of spalling is observed at around a
temperature of 500ᵒC causing the concrete to relinquish its compressive strength by
70% to 85% [11]. Concrete with bars of smaller diameter offers greater resistance to
fire than that with bars of larger diameter. Whenever steel hollow columns are used as
structural members, the fire resistance can be improved significantly by filling them
with concrete [12]. Further, the hollow steel columns filled with normal strength
concrete offers greater resistance to fire than the one filled with high strength concrete.
4.3 Spalling
The availability of free water and moisture in the concrete forms the primary driver for
the phenomenon of explosive spalling when concrete is subjected to higher
temperatures. Occurrence of this phenomenon is not possible without the moisture
availability. If the concrete member is dehydrated to a distance of 200 mm to 300 mm
from the surface that is exposed to fire, the member becomes less vulnerable to spalling
action. Moreover, by the addition of fibres and improving tie configuration, spalling
can be abated to a significant extent and fire performance of concrete can be improved
723
[13]. Existing literature has suggested the use of polypropylene fibres as one of the most
effective method to improve the fire resistance. The melting of polypropylene inside
concrete takes place at around 170ᵒC. This melting of fibres facilitates the formation of
discontinuous reservoirs or a network of pores in concrete. Hence, the steam generated
inside the material vaporizes off through these networks of pores. This process makes
steam evacuation easier, thus contributing to the reduction of internal pore pressure
which is one of the causes of spalling. Therefore, similar special provisions may be
developed which mainly specify the techniques to reduce the spalling on fire exposure
[14].
4.4 Supplementary cementitious materials

Several studies have been conducted on the possibility of improvement in fire resistance
of concrete on addition of supplementary cementitious materials. Research suggests
that use of clinoptilolite, zeolite, fly ash materials has improved the fire performance
by an appreciable amount [15-16]. Temuujin et al. [17] suggested that metakaolin based
geopolymer type coatings prepared using industrial sodium silicate solution when
applied on steel substrate show excellent thermal resistance. This is possibly due to the
fact that these coatings have high adhesive strength to the substrate which is required
for better fire resistance properties. On the other hand, geopolymers are emerging as
alternative binding materials. According to Sarker et al. [18] of all the base materials
available, low-calcium fly ash has been extensively used and found to be the most
effective source material suitable for concrete applications. They also studied the
influence of fire exposure on low-calcium fly ash based geopolymer concrete and
observed that the heat transfer is faster in the geopolymer concrete than in ordinary
Portland Cement (OPC) concrete. This leads to a reduced temperature gradient in fly
ash based geopolymer concrete, thus resulting in only minor surface cracking and
improved spalling resistance at around temperatures of 800℃ to 1000℃. Therefore,
suitable guidelines may also be developed to facilitate the use of various supplementary
cementitious materials.
5. Recommendations for Improvement in Indian standard

Although the Indian codes have discussed about the dependency of performance of
concrete under fire exposure on the types of aggregates viz. light weight aggregates and
normal weight aggregates, it has laid down provisions for normal weight aggregate
concretes only. Therefore, specific provisions and recommendations may be developed
to assimilate the behaviour of aggregates on the basis of density i.e. light weight or
normal weight and nature of material i.e. siliceous or calcareous aggregates.
IS 456 [2] has only highlighted the role of cover to reinforcement for enhancing the
resistance against spalling. It is essential to incorporate the addition of certain materials
such as polypropylene fibres, basalt fibres etc. which can enhance the resistance of
concrete to spalling, as discussed in the previous sections. Detailed and specific
724
guidelines pertaining to prevention of spalling must be laid down especially for high
strength concrete used in multi-storeyed buildings, similar to Eurocode which includes
methods for spalling prevention; strength reduction factor in high-strength concrete at
elevated temperatures, Characteristic strength reduction as per aggregates type etc.
Additional recommendations may also be provided to monitor the use of supplementary
cementitious materials such as zeolite, fly ash, vermiculite for plastering and for
additional coatings and finishes to enhance the fire resistance of concrete.
6. Conclusions
Indian, USA, New-Zealand, Australian, Canadian and European standards related to
fire have been briefly studied in this paper and the difference in their criterions are
properly indicated. It has been observed that all the standards for structural design
during fire have a prescriptive method, in which minimum dimension of the member
and cover to the reinforcement are prescribed for a given fire resistance rating.
Redistribution of moments in continuous members, superior requirements in design for
high strength concrete and spalling resistance are found to be amongst the primary
concerns which have not been extensively addressed in several country standards.
Besides, some of the standards also make use of special parameters e.g. load value,
degree of utilisation, aggregate effect etc. for defining minimum member dimension
and cover to the reinforcement. IS 1642 does not provide guidance on high strength
concrete or design to resist spalling, whereas IS 456 recommends to increase in cover
in certain fire situations to avoid the risk of spalling. Therefore, it is suggested that
Indian code provisions for spalling of concrete, especially high strength concrete needs
to be developed and the provided recommendations may be followed as preliminary
guidelines for the same.
7. References
1. IS 1642 (1989), Indian standard- Fire safety of buildings (General) – Details of

construction, Bureau of Indian standard, New Delhi, India.
2. IS 456 (2000), Indian standard-plain and reinforced concrete -code of practice,
Bureau of Indian standard, New Delhi, India.
3. ACI 216.1-07/ TMS 0216-07 (2007), Standard specifications for fire resistance of
concrete and masonry assemblies, American concrete Institute, MI, USA.
4. NZS 3101: Part 1 (2006), New Zealand standard concrete structures standard Part
1 - The Design of Concrete Structures, Wellington, New-Zealand.
5. AS-3600 (2001), Australian standard- Concrete structures, Standards Australia
international, Sydney, Australia.
6. National Research Council Canada (2010), National Building Code of Canada
2010 (NBCC), Division B, Appendix D, Fire performance rating, Institute of
Research in Construction, Ottawa, ON, Canada.
725
7. EN 1991-1-2 (2002), Eurocode 1: Actions on structures - Part 1-2: General actions

-Actions on structures exposed to fire, European committee for standardization,
Brussels, Belgium.
8. EN, 1992-1-2 (2004), design of concrete structures. Part 1-2: general rules—
structural fire design, Eurocode 2, European Committee for Standardization,
Brussels, Belgium.
9. Q. Ma, R. Guo, Z. Zhao, Z. Lin, K. He, Mechanical properties of concrete at high
temperature—A review, Constr. Build. Mater. 93 (2015) 371–383.
10. F. Cheng, V.K. Kodur, T.C. Wang, Stress-strain cures for high-strength concrete
at elevated temperatures, J. Mater. Civ. Eng. 16 (1) (2004) 84–94.
11. Romero ML, Moliner V, Espinos A, Ibañez C, Hospitaler A (2011), Fire behavior
of axially loaded slender high strength concrete-filled tubular columns, J Constr
Steel Res 2011;67:1953–65. doi: 10.1016/j.jcsr.2011.06.012
12. Kodur V. R. (1999), Fibre-reinforced concrete for enhancing structural fire
resistance of columns, Fibre-Structural Applications of Fibre-Reinforced Concrete,
ACI SP-182, pp. 215–234, 1999.
13. Shorter GW, Harmathy TZ (1961), Discussion on the Fire Resistance of
Prestressed Concrete Beams, Proceedings of the Institution of Civil Engineers
1961;20:313
14. Naik T.R. (1983), Temperature effects on concrete, A symposium sponsored by
ASTM committee C-9 on concrete and concrete aggregates, ASTM special
technical publication, Kansas City, MO.
15. Kubilay Akçaözog˘lu, Mustafa Fener, Semiha Akçaözog, Recep Öcal (2014),
Microstructural examination of the effect of elevated temperature on the concrete
containing clinoptilolite, Construction and Building Materials 72 (2014) 316–325.
doi: 10.1016/j.conbuildmat.2014.09.023
16. Canpolat F, Yilmaz K, Köse MM, Sümer M, Yurdusev MA (2004) Use of zeolite,
coal bottom ash and fly ash as replacement materials in cement production. Cem
Concr Res 34:731-735
17. Temuujin J, Van Riessen A, Williams R. Influence of calcium compounds on the
mechanical properties of fly ash geopolymer pastes. J Hazard Mater 2009;167:82–
8 doi: 10.1016/j.jnoncrysol.2010.09.063
18. Sarker, P.K., Kelly, S., Yao, Z., 2014. Effect of fire exposure on cracking, spalling
and residual strength of fly ash geopolymer concrete. Mater. Des. 63 584-592
https://doi.org/10.1016/j.matdes.2014.06.059
726
Paper ID- 140
Performance of Metakaolin Concrete under the Acidic Environment
Tripti sonker1, Rakesh Kumar2, Anupam Rawat3

1
triptiknit@gmail.com, .2rkpat@mnnit.ac.in, 3anupam@mnnit.ac.in
Department of Civil Engineering, MNNIT Allahabad, U. P. India.
Abstract:
The concrete is most widely used consuming material in the present era. The performance of
concrete is highly dependent on their constituent’s material that is water cement ratio and curing
condition. In this study concrete prepared with ordinary Portland cement mix and 5%, 7.5%,
10% and12.5% inclusion with Metakaolin. Concrete cubes were casted and checked their
compressive strength under normal curing water and acidic environment at age of 7, 28, 56 and
90 days.
Keyword: Normal concrete, Metakaolin concrete, compressive strength and sulphuric acid
Introduction: Now-a-days humans dependency towards concrete is higher, they are heavenly
surrounded by the concrete in all respect either infrastructure, transport facility etc. so in order to
sustain the environment is serious issue. The tones of production of cement are not environment
friendly work, so to overcome these problems and finding some replacement to incorporate with
cement and enhancing their properties. The mineral admixtures are plays vital role to reduces
carbon emission during production and also curtail the disposal of solid waste by products.
Metakaolin differs from other supplementary cementitious materials (SCMs), like fly ash, silica
fume, and slag, in that it is not a byproduct of an industrial process; it is manufactured for a
specific purpose under carefully controlled conditions. Metakaolin is produced by heating kaolin,
one of the most abundant natural clay minerals, to temperatures of 650-900°C. This heat
treatment or calcination, serves to break down the structure of kaolin. Bound hydroxyl ions are
removed and resulting disorder among alumina and silica layers yields a highly reactive,
amorphous material with pozzolanic and latent hydraulic reactivity, suitable for use in cementing
applications. When used as a partial replacement for Portland cement, metakaolin may improve
both the mechanical properties and the durability of concrete[1]. Form the survey of different
mineral admixtures; authors’ conclude that higher the dose of alumina will reduce the rate of
penetration inside the concrete that means decrease the permeability of concrete[2].Metakaolin is
a manufactured pozzolanic mineral admixture, which significantly enhances many performance
727
characteristics of cement-based mortars, concretes and related products. The study has been
made to evaluate the effect on mechanical and durability properties of M25 grade concrete made
with replacement of cement with Metakaolin, (0%, 2.5%, 5.0%, 7.5%, 10.0% and 12.5%) by
weight.
Replacement of metakaolin at 20% enhanced the compressive strength and split tensile
strength, and reduced the drying shrinkage and made concrete more imperious/ resistance against
the environment. The kaolin clay blended with ordinary Portland cement having variation in dose
of metakaolin 0%, 10%, 15%, 20%, 25% and 30% with different combinations of water to binder
ratio and inclusion of super-plasticizer. To optimize the dose of metakaolin in order to maintain
other properties like workability, compressive strength and flexural strength, with respect to
durability aspects magnesium sulphate curing results have been studied as Vietnam having
coastal areas considering low concentrated magnesium sulphate environment 20% by weight
metakaolin is optimized.[3].
The effect of metakaolin replacement by cement with 0,5,10 and15% author find out that
mixing of metakaolin will enhance their characteristics by improvising their compressive
strength, reduce its permeability hence increase its durability. They performed no of experiments
for testing of compressive strength, water absorption study, GWT water permeability test, water
penetration depth test on standards cubes. Split tensile test, modulus of elasticity test and rapid
chloride penetration test were tested through cylinders[4]. A study was done on the fiber
reinforced concrete by replacing cement with quarry dust (0%, 10%, 15%, 20%, 25% and 30%)
and metakaolin (0%, 2.5%, 5%, 7.5%,10% and 12.5%) by weight and addition of steel fibers and
glass fibers in different percentages (0%,0.5%,1%,1.5% and 2%). To conforming each material
properties a set of experiments like compression test for cube, flexure test for beams and Split
tensile test for cylinders. Optimizing all the materials with their specific values, at 25% quarry
dust, 10% metakaolin, 0.5% glass fiber and 1.5% steel fiber. After that acid durability test and
acid attack have been via 5% H2SO4 and HCL which cause loss in strength of concrete.[5].
Author studied the resistance against acidic environment by metakaolin concrete (replacement of
dose from 0%, 10%, 20% and 30%). After that concrete is immersed in different concentration of
0.25%, 0.5%. 1% and 2% of sulphuric acid environment to check the loss of mass in concrete
and finds metakaolin shows positive results.[6]. To analyze the corrosion behavior with change
728
in binder content by replacement of OPC with 15% metakaolin by weight and by Portland slag
cement (PSC) with varying water-to-binder ratio (w/b) and exposed to NaCl curing condition.
Author observed form the experimental results that metakaolin concrete exhibit better results
than Portland slag cement[7].
Materials:
Cement: Ordinary Portland cement (43 grades) of MP Birla was used. It was tested as per Indian
Standards Specifications IS: 8112-1989[8]. Its specific gravity are 3.01, normal consistency of
27%, fineness of 1.1%, initial setting time is 56 min and final setting time is 400 min.
S.No. Test Requirements as per Results

IS :8112-1989
1 Consistency test 26
26.5 56.5
2 Specific gravity 3.15 3.01
3 Initial setting time 30 minutes (min.) 56 min
4 Final setting time 600 minutes (max.) 400 minutes
5 Fineness 10% 1.1%
6 Soundness 10 mm 2.5 mm
7 Compressive strength test Not less than 23MPa (3 days) 23.3MPa (3 days)
Not less than 33MPa ( 7 days) 34.1MPa (7days)
Not less than 43MPa(28 days) 45 MPa (28 days)
Fine aggregate: The locally available natural river and was used as fine aggregate. It was tested
as per Indian Standards Specifications IS: 383-1970. Its fineness modulus is 2.69.
729
S.No. Properties Materials

Natural coarse aggregates Natural
Of 10 mm & 20 mm Sand
1 Specific gravity --- 2.65
2 Water absorption % 1% 0.9% 1.76
3 Bulk density, kg/m3 1580 1540 1680
4 Fineness modulus 6.7 7.1 2.7
5 Impact value % 15 15 ------
6 Crushing value % 25 25 -----
Metakaolin: Metakaolin (purchased from Kaolin Techniques Pvt. Ltd., Kutch, Gujarat. Average
particle size is 1.5-2.5.
Experiment and results:
BIS10262-2017 has all the guidelines for normal mix design of concrete, in which M25 grade of
concrete normal mix design have decided. For optimizing the cement content trials have done
with different mix proportions of coarse and fine aggregate with cement and water. The water
cement ratio will varying also and on the basis of compressive strength testing the mix
proportion is finalize for further replacement of cement to maintain the environment friendly
construction. Here 5 trials have done to minimize the cement content as per IS-456 2000[9]; to
economize the construction and optimize compressive strength.
S. No. w/c Cement Water Coarse Aggregate Fine

3 3 3
ratio (Kg/m ) (Kg/m ) (Kg/m ) Aggregate
3
20 mm 10 mm (Kg/m )
Trial 1 .43 361 164 736.2 490.8 695.22
Trial 2 .43 343.8 158.8 741 494 686.5
Trial 3 .40 383.16 162.96 749.98 499.9 690.1
Trial 4 .40 369 157.68 758.4 506.1 698
Trial 5 .40 350 149.93 766.23 510.8 712.95
730
Method of concrete mix design (BIS: 10262-2009[10]) for Normal concrete:
Cement 43 Grades
Concrete M25
Exposure condition Moderate
Air content 2%
Shape of coarse aggregate Angular
Maximum size of aggregate 20mm
Specific gravity of cement 3.15
Specific gravity of coarse aggregates 2.7
Specific gravity of fine aggregates 2.6
Water absorption for coarse aggregate 0.5 %
Water absorption for fine aggregates 1%
Fine aggregates zone II
Slump required 100 mm
Mineral admixtures used in place of cement due to its high emission of CO2, during its
production time. To become the sustainable development of environment, addition of mineral
admixtures reduces the greenhouse gases. So in order to contribute in the sustainable
development of the world, to improvising the traditional concrete characteristics the mineral
admixture is used to make better concrete with respect to durability and strength. Metakaolin is
pozzolanic material which is thermally activated aluminosilicate, and it is getting from the
calcination of kaolin clay at temperature between 650°C to 800°C. Addition of Nano metakaolin
instead of metakaolin having ultrafine characteristics: replacement of cement with 1%, 3%, 5%,
7%, 9% and 10%. The comparative study of workability and compressive strength of 3, 7 and 28
days strength have done by normal OPC and metakaolin replaced concrete with 1, 3, 5, 7, 9 &
10% to assess the behaviour of Ultra high performance concrete[11]. The incorporation of
metakaolin at 0%, 5%.10% and 15% will affect the strength, sorptivity and carbonation
731
resistance against with two different water to binder ratio; for better command author studied the
microstructures and Nano-mechanical properties[12]. The replacement of Metakaolin with
cement content with 2.5%, 5%, 7.5%, 10%, 12.5% and 15% by weight have been done in order
to check the workability and strength of metakaolin concrete. Workability is decreasing by
increasing the metakaolin content in cement which is checked by the slump test. The
compressive strength testing have done on set standards given in IS 516(1963)[13] for 7 and 28
days strength will be taken from the different samples curing at normal tap water. The results
obtained from the samples are listed below for comparison point of view and depicted through
the bar chart. From the graph, the quantity of metakaolin is optimized on the ground of durability
prospective.
S.No. Cement Metakaolin Water Slump Compressive

(Kg/m3) (Kg/m3) (Kg/m3) value Strength (MPa)
7 Days 28 Days
Mk 0 350 0 149.93 22.6 32.67
Mk 5 332.5 17.5 149.93 84 23.5 33.10
Mk 7.5 323.75 26.25 149.93 80 25 37.33
Mk 10 315 35 149.93 73 24.3 35.2
Mk12.5 306.25 43.75 149.93 70 24.2 35
Mk 15 297.5 52.5 149.93 65 22.4 33.5
100
Slump in mm
50
0
SLUMP
Metakaolin replacement
732
The workability of samples obtained from slump test. The workability of concrete
corresponds to the ease of handling the concrete. The compressive strength of metakaolin
replaced concrete of 7 and 28 days.
40
compressive 30
strength 20
N/mm2
10 7 DAYS
0
28 DAYS
Metakaolin replacement % by weight of cement
After optimizing the metakaolin at 7.5% by weight in cement while considering the workability,
compressive strength test.
Concrete type Compressive strength test N/mm2

7 days (normal curing) 28 days(normal curing) 56 days 90 days
Normal concrete 22.73 35.66 40.34 44.46
Metakaolin 25.76 39.11 43.44 48.78
concrete
60
50
40
30 OPC
MK
20
10
0
733
Change in compressive strength: 150mm cubic specimens were prepared for to examine for
change in compressive strength of concrete cubes as per guide lines. The loss in strength of
samples to normal tap water curing and acid curing After 28 days of normal water curing
samples are deep in aggressive environment that is sulphuric acid.
Concrete type % change on strength loss

After 56 days After 90 days
OPC 11.26 9.33
Metakaolin concrete 12.33 10.89
The interest of use of metakaolin in cement mortar and concrete is of high. Use of metakaolin
in concrete is of 7.5% in replacement of cement gave better strength results and durability
improvement. Water permeability, absorption was much improved in use of metakaolin which
leads to increase in density of concrete. Use of metakaolin in preparing acid resistance concrete
such as chloride permeability, sulphate resistance showed good results.
Conclusion:
1. The optimum replacement of metakaolin by 7.5% weight to cement is found; to give
maximum change in compressive strength as compared with normal OPC.
2. The workability of metakaolin concrete is low as compared to normal conventional concrete;
because the surface area of the specimen increased hence increase the de.
3. For the durability point of view, metakaolin reduce the permeability and absorption of the
concrete; due to filler effect of metakaolin particle which reduced the porosity and
permeability of the concrete.
References:
[1] R. Siddique and J. Klaus, “Influence of metakaolin on the properties of mortar and
concrete: A review,” Appl. Clay Sci., vol. 43, no. 3–4, pp. 392–400, 2009.
[2] T. Ayub, N. Shafiq, and S. U. Khan, “Durability of Concrete with Different Mineral
Admixtures : A Review,” Int. J. Civ. Environ. Eng., vol. 7, no. 8, pp. 601–612, 2013.
734
[3] D. D. Vu, P. Stroeven, and V. B. Bui, “Strength and durability aspects of calcined kaolin-
blended Portland cement mortar and concrete,” Cem. Concr. Compos., vol. 23, no. 2001,
pp. 471–478, 2001.
[4] P. Dinakar, P. K. Sahoo, and G. Sriram, “Effect of Metakaolin Content on the Properties
of High Strength Concrete,” Int. J. Concr. Struct. Mater., vol. 7, no. 3, pp. 215–223, 2013.
[5] A. Pradesh, “A STUDY ON STRENGTH OF FIBRE REINFORCED CONCRETE WITH
PARTIAL REPLACEMENT OF CEMENT WITH QUARRY DUST AND
METAKAOLIN PROPERTIES,” NTERNATIONAL J. Prof. Eng. Stud., vol. 9, no. 5, pp.
121–133, 2018.
[6] R. Pavel and T. Jiří, “Resistance of Concrete with Metakaolin Addition to Acid
Environment,” Key Eng. Mater., vol. 677, pp. 144–149, 2016.
[7] A. N. Borade and B. Kondraivendhan, “Effect of Metakaolin and Slag blended Cement on
Corrosion Behaviour of Concrete,” IOP Conf. Ser. Mater. Sci. Eng., vol. 216, p. 12022,
2017.
[8] “IS 8112 (1989)_ Specification for 43 grade ordinary Portland cement.pdf.” .
[9] M. Kisan, S. Sangathan, J. Nehru, and S. G. Pitroda, “IS.456 : 2000 Indian Standard
PLAIN AND REINFORCED CONCRETE - CODE OF PRACTICE,” 2000.
[10] “IS-10262-2009-CONCRETE MIX DESIGN - INDIAN STANDARD METHOD -
CivilBlog.” .
[11] M. S. M. Norhasri, M. S. Hamidah, A. M. Fadzil, and O. Megawati, “Inclusion of nano
metakaolin as additive in ultra high performance concrete ( UHPC ),” Constr. Build.
Mater., vol. 127, pp. 167–175, 2016.
[12] S. Barbhuiya, P. Chow, and S. Memon, “Microstructure , hydration and nanomechanical
properties of concrete containing metakaolin,” Constr. Build. Mater., vol. 95, pp. 696–
702, 2015.
[13] M. Kisan, S. Sangathan, J. Nehru, and S. G. Pitroda, “IS: 516 -1959 Indian stndard
METHODS OF TESTS FOR STRENGTH OF CONCRETE,” 1959.
`
735
Paper ID- 141
Static Analysis of Skew Laminated Composite Plates using Isogeometric

Finite Element Method
Nidhi Devi1*, A. Bhar1, and R. Pandey1
1
Department of Applied Mechanics, MNNIT Allahabad, U.P., PIN - 211004, India
* ram1601@mnnit.ac.in
Abstract. Isogeometric analysis (IGA) bridges the gap between the Computer Aided Design (CAD) and Finite
Element Method (FEM). IGA uses Non-uniform Rational B-Splines (NURBS) as the interpolation function. This
paper is an effort to study the accuracy of results from IGA with respect to FEM for analysis of skew laminated
composite plate. Mathematical formulation is developed using Semi-Refined Higher Order Theory (SRHSDT7) and
computer implementation is done accordingly. Isogeometric Results are compared with Finite element result and
analytical results using first order shear deformation theory (FSDT) and SRHSDT7.
Keywords: IGA, NURBS, Laminated Composite plate, Semi-refined higher order shear deformation theory
1 Introduction
From the recent years classical finite element method (FEM) is used for the analysis of many structural and
mechanical elements in engineering applications. In the finite element method shape functions (polynomial
functions) that the only approximation of geometry are used. On the other side we have well known functions like
Beziers, B-spline and non-uniform rational B-Spline (NURBS) functions etc. in computer aided design (CAD).
Combining both features the new concept introduced by Hughes et al. [1] is called isogeometric method (IGA). This
method fills the gap between the two more important area of analysis and modelling in engineering field.
Skew plate often finds its application in civil, aerospace, naval, mechanical engineering structures. They are
mostly used in construction of bridges for distributing compound alignment requirements mainly in civil
engineering fields. For the analysis of skew plate we can use rectangular meshing but in case of rectangular meshing
complexity increases at the edges of plate. So for the convenience we use skew element for the meshing of plate as
in Fig. 1 and further this element is mapped to rectangular domain.
Y a

X
Fig. 1: Computational Domain of Isotropic Skew Plate
736
After introduction of the basic general concept, IGA was demonstrated to be applied for simple structural
mechanics problems on rods, beams etc. by Cottrell et al. [2]. There are many works seen in the literature on
isotropic skew plate and laminated skew plate with analytical method by as well as classical finite element method.
The effect of skew angle is studied on flexural response of laminated composite plate by Upadhyay et al. [4].
Behaviour of skew plate using finite element method is observed by Srinivasan et al. [5] for thermo-mechanical
loading.
In this paper the effect of skew angle by newly developed method isogeometric analysis using first order shear
deformation theory as well as semi-refined higher order shear deformation theory is presented. Present results are
compared with analytical results as well as finite element method. Effect of skew angle and different parameter like
plate aspect ratio, loading parameter etc. is presented.
2 Theoretical Formulation
For the analysis of skew laminated composite plate using semi-refined higher order shear deformation theory
(SRHSDT7) following assumptions are made:
 The laminate has perfectly bonded layers means there is no slip between the adjacent layers.
 Each lamina can be isotropic, orthotropic or transversely isotropic.
 The reference plane for the laminate is the middle plane that contains X-Y axis of the reference coordinate
system and Z axis is perpendicular to this XY plane.
2.1 NURBS as basis for analysis:
Knot vectors: Non uniform rational B-Splines, here the knot vectors are not uniform. A knot vector is one
dimensional non-decreasing set of coordinates in parameter space.
  1 ,  2 ,....... n  p 1 (1)
where, n is the number of basis functions and p is the polynomial order used to construct B-Splines. The
univariate basis functions are defined by Cox-De Boor recursive formula [7, 8].
  i  
N Ib ( )  N i , p 1 ( )  i  p 1 N ( ) (2)
i  p  i i  p 1  i 1 i 1, p 1
For order p  0 :
1      
Ni, p   i  (3)
0 otherwise 
The basis functions N I ( , ) , used in eqn (9) are defined by tensor product of univariate basis functions Ni , p ( ) and
M j , q ( ) of order p and q in parametric directions  and  respectively [9], as
N Ib ( , )  Ni , p ( ) M j , q ( ) (4)
NURBS function is written as;
N Ib wI
N I ( , )  (5)

m n
A
N Ab ( , ) wA
To understand the isogeometric process, consider the open knot vector in both directions are {1,2 ,8} and
{1,2 ,8} is 0 0 0 1 2 3 3 3 as first and last values of knots are p+1 times. As from Fig. 2 index space form
from knot vectors and parameter space from by unique knot vectors that are 0 1 2 3 for both directions. In the
737
NURBS based isogeometric analysis there is two types of mesh that are control mesh and physical mesh. Control
points describe the control mesh and control mesh interpolates the control points. Control mesh is not the actual
geometry it is only control the geometry. Physical mesh is the actual decomposition of geometry. Physical mesh has
two elements patch and knot span. The index space of a patch uniquely recognizes each knot and discriminates
among knots having multiplicity greater than one. When we consider multiplicity m and p is the degree of knots,
then knot functions have C p  m continuity
𝞰
8
𝞰
7
𝞰
6
3
𝞰
5
2
𝞰
4
𝞰 1
3
𝞰 0
2 0 1 2 3
𝞰
1 (b)
𝞷 𝞷 𝞷 𝞷 𝞷 𝞷 𝞷 𝞷
1 2 3 4 5 6 7 8
𝝶
(a)
-1 𝞷
-1 1
(c)
Fig. 2: Single Patch Surface Model: (a) Index Space, (b) Parameter Space, (c) Parent Element
2.2 Displacement Field

A typical laminated plate is shown in Fig. 3, along with the fibre orientation in a ply with reference to X-axis,
where 1-2-3 represents the material axes. The Spatial displacements of the plate as functions of the coordinate and
time variables, in terms of reference plane displacements as per the SRHSDT7 [6], are given by
U ( x, y, z, t )  u ( x, y, t )  z x ( x, y, t )  z 3 x* ( x, y, t )
V ( x, y, z, t )  v( x, y, t )  z y ( x, y, t )  z 3 y* ( x, y, t )
(6)
W ( x, y, z , t )  w( x, y, t )
738
Further u, v, w are the inplane displacements  x ,  y are the rotations of the mid-plane around y-axis and x-axis,
respectively and  x* and  y* are the higher order terms.
3 1
2
Z
h
X
Fig. 3: A Typical Four layer Composite Plate

Mechanical Strains:
In-plane strain components are written as;
 xx  u, x  uo, x  z x , x  z 3 x*, x
 yy  v, y  vo , y  z y , y  z 3 y*, y
 xy  u, y  v, x  uo, y  vo, x  z ( x , y   x , y )  z 3 ( x*, y   y*, x )
Transverse shear strains are written as;
 xz  u, z  w, x   x  wo , x  3z 2 x*  x  z 2x*
 yz  v, z  w, y   y  wo , y  3z 2 y*   y  z 2 y*
x   x  wo , x ;x*  3 x*
where;
 y   y  wo , y ; y*  3 y*
Plate constitutive Relations:
From the Hook’s law plane stress written as;
 xx  0   xx 
k k
 Q11 Q12 Q16 0
  Q  
 yy   21 Q22 Q26 0 0   yy 
   
  xx   Q61 Q62 Q66 0 0   xx  (7)
   
 0 0 0 Q55 Q54   xz 
 xz   
 yx   0 Q45 Q44   yx 
 0 0
739
E11 E22  E
Q11  , Q22  , Q12  21 11
where; 1   12 21 1   12 21 1   12 21
Q44  G23 ,, Q55  G13 , Q66  G12
Where E11, E22 are the Young modulus of elasticity in the x and y directions respectively, G12 , G23 , G13 are the shear
modulus in the x-y, y-z, z-x planes respectively.
Stress-strain relations for the kth lamina with arbitrary fiber orientation in global reference system;
 xx 
k  Q11 Q12 Q16 0 0   k
    xx
Q 21 Q 22 Q 26 0 0   yy 
 yy     
 
  xx   Q 61 Q 62 Q 66 0 0   xx  (8)
    
 xz   0 0 0 Q 55 Q54   xz 

 yx   0  
 0 0 Q 45 Q 44   yx 
2.3 Isogeometric Formulation

In the isogeometric analysis method, the displacement components at a generic point (x, y) in the reference
plane, as in eqn (6), are interpolated with those at the control points, for a given element, as compared to nodes in
case of FEM. For example, the inplane displacement component in X-direction u is interpolated between its values
at the control points for e-th element, as
n m
u e ( , )  NI
I ( , )u I (9)
where u I is the degree of freedom associated with control point I and n  m is the number of control points and
basis functions. Further, and are the local coordinates for given element in parametric domain, mapped suitably
from the physical domain (x, y) of the reference plane. Similarly all other displacement components are interpolated
using the control point variables.
2.4 Governing Equation:

The governing equations of motion for the skew plate can be derived for a general undamped dynamic problem,
using well known Hamilton’s principle given by
t2
 t (T  U  W ) dt  0 (10)
1
where, T is the kinetic energy, U is the strain energy and W is the potential energy of external load. After
substituting these values in the eqn (10) and due operations, rearrangement and simplification, one obtains the
general instantaneous equation of motion in standard discretized form as
 M d    K d    P (11)
where, [M] is the global mass matrix, [K] is the global stiffness matrix and {P} is the global load vector.
Further, d  and {d} are the unknown acceleration and displacement vectors defined at the control points. Now,
specifically for a static (time independent) problem, eq n (11) is reduced to solution of which yields the control point
displacements.
 K d  P (12)
Required displacements at any point of the plate may then be found out using due interpolation mentioned already.
740
3 Numerical Results:
In this section present results are compared with analytical solution and existing FEM solutions. As IGA is used
exact geometry that defined by the NURBS basis functions, so that the main governing factor is knot vectors.
Convergence of results can be obtained by increasing the control points.
3.1 Isotropic Plate:
For the convergence, consider a square plate with dimension (100 mm×100 mm), thickness h =1.0 mm with
uniform loading q = 6.894×10-4 N/mm2. Skew angles 80⁰, 60⁰ and 40⁰ are considered. Material properties for the
plate are young modulus of elasticity (E = 117.21 GPa), modulus of rigidity (G = 45.08 GPa) and Poisson’s ratio ( 
= 0.3). Consider uniform open knot vectors in both directions. Non-dimensional deflection parameter is defined as
103WD Eh3
W  where; D  .
qa 4 12(1   2 )
0
80
1.2 0
60
0
40
Non-Dimenstional Deflection Parameter
0.8
0.4
0 100 200 300 400 500
Number of Control Points

Fig. 4: Convergence of clamped edge isotropic plate of different skew angle using FSDT
1.5
0
80
0
60
Non-Dimenstional Deflection Parameter
0
40
1.0
0.5
0.0
0 100 200 300 400 500 600
Number of Control Points
Fig. 5: Convergence of clamped edge isotropic plate of different skew angle using SRHSDT7
741
From the Fig. 4 and 5, it can be observed that results are converging as control points are increases. It is also
seen for more skewness results are converse with less number of control points. Here we consider only uniform knot
vectors and more accuracy we can consider multiplicity of knots also.
Non-dimensional central deflection parameter in Table 1 of isotropic square skew plate with material
properties: young modulus of elasticity (E = 117.21 GPa), modulus of rigidity (G = 45.08 GPa) and Poisson’s ratio (
 = 0.3), a/b = 1.0, a/h = 100.0 and uniform loading q = 6.894×10-4 N/mm2 is computed. Present isogeometric
analysis result and finite element result are compared with analytical results [3] using first order shear deformation
theory as well as semi-refined shear deformation theory.
Table 1: Non-Dimensional Deflection Parameter for clamped Edge isotropic skew Plate of different skew angle
Skew Angle Analytical Result FSDT SRHSDT7

(Morley,1964) Present FEM Present IGA Present FEM Present IGA
800 1.200 1.202931 1.158002 1.202905 1.150675
750 1.123 1.125371 1.084752 1.125344 1.078003
700 1.021 1.023378 0.988251 1.023350 0.982255
600 0.769 0.771039 0.748552 0.771009 0.744351
500 0.500 0.501878 0.491071 0.501843 0.488629
450 0.377 0.378296 0.372036 0.378254 0.370327
400 0.270 0.269485 0.266651 0.269435 0.268526
3.2 Laminate Composite Plate:
Consider a symmetric and anti-symmetric cross-ply laminated composite skew plate with skew angle = 60⁰, a/b
= 1.0 and a/h = 20.0. Material properties for the laminate plate are following:
E1
 10,12  0.22, G12  G13  0.33E2 , G23  0.2 E2
E2
where, E2 = 8.0 Non-dimensional deflection parameter is defined as w  w h and load parameter is defined as
4
Q  qa .
E2 h 4
Table 2: Non-Dimensional Deflection Parameter for clamped Edge symmetric and anti-symmetric skew Plate using FSDT
Q = 200 Q = 500
Lamination
Analytical Present Present Analytical Present IGA Present
Scheme
Result [3] IGA FEM Result [3] FEM
[0/90/90/0] 0.5635 0.699812 0.688012 0.9827 1.749533 1.720032
[0/90/0/90] 0.5368 0.640624 0.628476 0.9620 1.601562 1.571191
Table 3: Non-Dimensional Deflection Parameter for clamped Edge symmetric and anti-symmetric skew Plate using SRHSDT7
Lamination Q = 200 Q = 500

Scheme
Analytical Present Present Analytical Present Present FEM
Result [3] IGA FEM Result [3] IGA
[0/90/90/0] 0.5635 0.712550 0.701792 0.9827 1.781375 1.72003
[0/90/0/90] 0.5368 0.641961 0.630708 0.9620 1.604904 1.57677
742
Non-dimensional deflection parameter for symmetric and anti-symmetric cross-ply skew Plate are compared
with analytical results [4] using first order shear deformation theory in Table 2 and semi-refined shear deformation
theory in Table 3. Variation of non-dimension deflection parameter with load parameter is shown in Fig. 6 and 7 for
symmetric [45/-45/-45/45] and anti-symmetric [45/-45/45/-45] angle-ply laminated composite plate with skew angle
30⁰. The plate considered in this case is moderately thick (a/h = 20.0) and clamped boundary conditions for all
edges.
Analytical Result [3]

0.4 Present IGA-FSDT
Present FEM_FSDT
Non-Dimensional Deflection Parameter
Present IGA_SRHSDT7
Present FEM_SRHADT7
0.3
0.2
0.1
0 200 400 600

Load Parameter
Fig. 6: Variation of non-dimensional deflection parameter with load parameter for symmetric [45/-45/-45/45] angle ply
clamped laminated composite skew (   30 ) plate
Analytical Result [3]

Present IGA_FSDT
0.4 Present FEM_FSDT
Non-Dimensional Deflectional Parameter
Present IGA_SRHSDT7
Present FEM_SRHSDT7
0.3
0.2
0.1
0.0
0 200 400 600
Load Parameter
Fig. 7: Variation of non-dimensional deflection parameter with load parameter for anti-symmetric [45/-45/45/-45] angle
ply clamped laminated composite skew (   30 ) plate
743
Variation of non-dimensional deflection parameter with a/b ratio is seen in Fig. [8,10] and Fig. [9,11] for first
order shear deformation theory and semi-refined shear deformation theory respectively. A and B represents anti-
symmetric cross-ply laminate [0/90/0/90] and symmetric cross-ply laminate [0/90/90/0] respectively. C and D
represent anti-symmetric angle-ply laminate [45/-45/45/-45] and symmetric angle-ply laminate [45/-45/-45/45]
respectively.
0
A_90
0
A_60
0
A_30
Non Dimensional-Deflection Parameter
3
0
B_90
0
B_60
0
B_30
2
0 1 2 3
a/b Ratio
Fig. 8: Variation of non-dimensional deflection parameter with a/b Ratio for anti-symmetric [A = 0/90/0/90] and symmetric
[B = 0/90/90/0] clamped laminated composite skew plate using FSDT
0
A_90
0
A_60
0
A_30
Non-Dimnesional Deflection Parameter
3
0
B_90
0
B_60
0
B_30
2
0 1 2 3
a/b Ratio
Fig. 9: Variation of non-dimensional deflection parameter with a/b Ratio for anti-symmetric [A = 0/90/0/90] and symmetric
[B = 0/90/90/0] clamped laminated composite skew plate using SRHSDT7
744
3 0
C_90
0
C_60
0
C_30

0
D_90
0
D_60
2 0
D_30
0
0 1 2 3
a/b Ratio
Fig. 10: Variation of non-dimensional deflection parameter with a/b Ratio for anti-symmetric [C = 45/-45/45/-45] and
symmetric [D = 45/-45/-45/45] clamped laminated composite skew plate using FSDT
3
0
C_90
0
C_60
0
C_30
0
D_90
0
2 D_60
0
D_30
0
0 1 2 3
a/b Ratio
Fig. 11: Variation of non-dimensional deflection parameter with a/b Ratio for anti-symmetric [C = 45/-45/45/-45] and
symmetric [D = 45/-45/-45/45] clamped laminated composite skew plate using SRHSDT7
It is observed from Fig. [8-11] that variation in non-dimension deflection parameter is more in symmetric and
anti-symmetric cross-ply laminated composites than the angle-ply composites. Effect of skewness is also observed
with a/b Ratio in Fig. [8-11] and it is seen that non-dimensional deflection parameter is decreases as skew angle
decreases at particular a/b ratio. It is seen that present isogeometric result with semi refined higher order theory are
745
closed to analytical results by Upadhyay et al. [4]. This is because of as we considering higher order terms in
displacement fields results are improved.
4 Conclusion:
In this paper result are compared with first order shear deformation theory and semi-refined shear deformation
theory. Non-dimensional deflection parameter is compared with analytical and finite element result for different
loading parameter. For the present case we consider only rectangular plate so the effect of isogeometric method is
not considerably visual here as rectangular/square plate form can be represented accurately by standard equation.
This work can be further extended to complex geometries like circular, hyperboloid etc. that can be easily
represented by B-spline NURBS and isogeometric method is visible.
Nomenclature
a Dimension of plate in X-direction
b Dimension of plate in Y-direction
 Skew angle
 Fiber angle
Greek Symbols
 , Parametric Directions
Ni , p NURBS Basis Function
Subscripts
p Order of Basis Function in X-direction
q Order of Basis Function in Y-direction
References
1. Hughes T.J.R., Cottrell J.A., Bazilevs Y.: Isogeometric analysis: CAD finite elements NURBS exact geometry
and mesh refinement. Comput. Method Appl. Mech Eng. 194, 4135–4195 (2005).
doi:10.1016/j.cma.2004.10.008
2. Cottrell J.A., Reali A., Bazilevs Y., Hughes T.J.R.: Isogeometric analysis of structural vibrations. Comput.
Methods Appl. Mech. Engrg. 195, 5257–5296 (2006). doi:10.1016/j.cma.2005.09.027
3. Morley, L.S.D.: Bending of clamped rectangular plate. Quart. J. Appl. Math. 3, 293–317 (1964).
4. Upadhyay A.K., Shukla K.K.: Large deformation flexural behavior of laminated composite skew plates: An
analytical approach. Composite Structures 94, 3722–3755 (2012).
http://dx.doi.org/10.1016/j.compstruct.2012.06.015
5. Singh S.K. and Chakrabarti A., Static, vibration and buckling analysis of skew composite and sandwich plates
under thermo mechanical loading, Int. J. of Applied Mechanics and Engineering 18(3), 887–898 (2013). doi:
10.2478/ijame-2013-0053
6. Bhar A., Phoenix S.S., Satsangi S.K.: Finite element analysis of laminated composite stiffened plates using
FSDT and HSDT: A comparative perspective. Composite Structures 92, 312–321 (2010).
doi:10.1016/j.compstruct.2009.08.002
7. Piegl Les and Tiller Wayne: The NURBS Book. Second Edition, Verlag Berlin Heidelberg: Springer (1997).
8. Rogers David F.: An introduction to NURBS with historical perspective. Elsevier, Academic Press, United
States of America (2001).
746
Paper ID- 143
Fabrication and Mechanical properties of Marble Dust Filled Aluminum

Metal Matrix structural Composite
Sandeep Kashyap1, Hariom Tripathi2,⌂
1
Mechanical Engineering Department, MNNIT Allahabad, Prayagraj-211004, INDIA
2
Material Science and Engineering, MNNIT Allahabad, Prayagraj -211004, INDIA
ABSTRACT
The use of Aluminum and its alloys in the aerospace, transportation, automobile and power transmission are
growing in these days because it has good corrosion resistance, lightweight, and high specific strength. But
aluminum and its alloys have some drawbacks like low absolute strength at higher temperature, low impact
resistance, and low hardness etc. In this research, the aluminum metal matrix composite has been manufactured with
the help of the stir casting technique, for improving the mechanical properties of the aluminum alloy. Where Al
6063 has been taken as the matrix material and the Graphite (Gr)(2%wt), glass fiber (2% wt), Boron Carbide
(B4C)(4% wt) and marble dust (2%,4%,6%wt) in powder form has been taken as the reinforcement materials. The
present research work has been undertaken with an objective to explore the use of marble dust as a reinforcing
material as a low-cost option for improving the mechanical properties.
Keywords: Aluminum Metal Matrix structural Composite, Stir casting technique, Graphite, Glass fiber, Boron
Carbide, Marble dust.
1. Introduction researchers due to its capability to change the

physical properties (thermal expansion, density
Composite materials are the special system of and thermal diffusivity etc) and mechanical
materials composed of two or more materials behaviour (ultimate strength, fatigue strength,
popularly known as matrix and reinforcement creep strength, corrosion and wear resistance
[1]. Matrixes are the basic part of the composite etc.)[4] of matrix materials by changing the
system whose properties can be altered by different reinforcement phase. Also the increase
adding the secondary phases [] (reinforcement in the demand in the field of aerospace,
materials). The main functions of matrix structural components and light weight
materials are to transfer the stresses which are automotive applications like break disk, engine
generated due to the application of the load to piston led to rapid growth in the field of
the reinforcement particles or fibers for the MMC’s[5]. Aluminium MC’s are more popular
protection of matrix materials [2]. in the field of MMC’s due to its light weight,
Reinforcement phases are generally in the form corrosion resistance, good electrical conductivity
of particles or fiber, which have been dispersed and high specific strength.
in a controlled amount within the matrix for In general composites of aluminium can be
getting the desired properties. The reinforcement fabricated by semisolid processing (Friction stir
particles generally improve the mechanical processing) [6], liquid state processing (squeeze
properties like hardness, wear resistance, casting, stir casting etc.) and power metallurgy
ultimate strength and stiffness etc[3]. method. Generally hard particles like ceramics
On the basis of chemical nature of the matrix (boron carbide, alumina, graphite, silicon
materials composites are divided into three carbides etc.) are used reinforcements in case of
groups as polymer matrix (PMC), metal matrix AMC’s.
(MMC) and ceramic matrix composites (CMC). On the application of the externally applied load
Nowadays MMC’s are quite popular among the to the AMC’s, aluminum matrix transfer loads to
⌂
Hariom Tripathi, MNNIT, Allahabad
Email id: hariom.tripathi7@gmail.com
Contact No.:8005385469
747
disperse reinforcement particles and the applied 2.2. AMC’s Preparation

external load is shared by both reinforcement
and matrix, this is the main cause for the Marble dust reinforced hybrid AMC’s were
strengthening of the matrix material. One more produced by stir casting technique in the
thing, the strong interface in between the matrix electrical resistance furnace as shown in Fig. 1.
and the dispersed reinforcement particles is Matrix material (aluminium6063) was placed
required to obtained higher mechanical property inside the graphite crucible and was melted
of composites. The interface bonds are created inside the electrical furnace in the inert
by the mutual dissolution throughout the casting atmosphere and when the temperature of liquid
that is why superior wetting is required for the metal reached at 770-790°C, then magnesium
reinforcement particles. powder was added for enhancing the bonding
The objective of this work is to fabrication of an between the aluminum and reinforcement
advanced series of Al6063 based MMC’s particles.
reinforced with and without marble dust [7]
particles and observe the effect of reinforcement
particles on the mechanical properties.
2. Experimental Procedure
2.1. Selection of Matrix and Reinforced

materials
In this present research work Al 6063 alloy

having density 2.7gm/cc and melting point
6550C was used as the matrix material and glass
fiber (2%)(.005 - .01mm dia.), graphite(2%),
B4C (4%)and varying % of marble dust powder
(0, 2,4,6%) were added as the reinforcement
particles for the preparation of hybrid composite.
Marble dust having density 2.8gm/cc is an
industrial waste which is utilized in this research
work; it contains approximate 28.35% SiO2,
MgO 16.25%, CaO 40.45%, and Fe2O3 9.7%.
Graphite powders of size 25-35μm, boron Fig.1: Stir Casting Setup
carbide particles of size 20-30μm and marble
dust of size 30-50μm were used for this research As, by the addition of magnesium particles in
work. the melted aluminium the viscosity and surface
During the time of casting, 1-1.1wt% of tension of the molten matrix metal decreases.
magnesium was also added in the molten metal After that preheated reinforced particles (glass
to enhance the wettability of reinforcement fiber (2%), graphite (2%), B4C (4%) and varying
particles with the molten matrix metal % of marble dust powder (0, 2, 4 and 6%
(aluminum 6063). The chemical compositions of weight)) were added through funnel and the
al 6063 which is used as the matrix in the mixture of melted metal and reinforcement
composite are shown in table 1. particles were stirred at 450-550 rpm for 15-20
Table1: Chemical Composition of Al6063. minutes. After that, the liquid mixture was
poured into the sand mold (as shown in Fig.2)
Elements Al Mg Si Fe Others and the melted composites were allowed to cool
Wt.% Max 0.45- 0.2- Max ~0.65
97.5 0.90 0.6 0.35
⌂
748
accurate hardness value 5 readings were taken

for each sample.
Instron UTM was used for testing the tensile
behavior of different composites. The test was
conducted at room temperature and using
1.5mm/min as the strain rate. Samples for tensile
testing were prepared with the help of wire
EDM as per the standard (ASTM E8).
Tinius olsen UTM was used for the flexural
(bending) strength test as shown in Fig. 3. For
Fig.2: Sand mold with casted product flexural test samples were prepared according to
standard (ASTM-E290-14) and flexural test
And by this way, four different grades (0, 2, 4 were performed at 1.5 mm/min.
and 6weight% of marble dust) of AMC’s were
prepared. The details of prepared different
aluminium matrix composites composition was
presented in Table 2.
Table 2: Designation of Samples
Designation Composition (wt %)
A0 Al 6063+ glass fiber (2%) +

graphite (2%) + B4C (4%)
A1 Al 6063+glass fiber (2%) +
graphite (2%) +B4C (4%) +
Marble dust (2%)
A2 Al 6063+ glass fiber (2%) + Fig. 3: Flexural strength testing machine
graphite (2%) + B4C (4%) +
marble dust (4%) 3. Results and discussion
A3 Al 6063+glass fiber (2%) +
graphite (2%) + B4C (4%) + 3.1. Micro-hardness
marble dust (6%)
Hardness of composite is defined as the
2.3. Mechanical Testing resistance of the surface of composite against the
indentation. It is a surface property of material.
In order to find out the change in the micro The micro hardness of all the four composites is
hardness, tensile strength and flexural strength shown in Fig.4. The strength of bonding
of different AMC’s, Test specimens were cut (interfacial strength) in between the aluminium
according to ASTM rule. matrix and the different reinforcement particles
Samples were polished like mirror for which are present in the composite tell us about
performing micro hardness test. Micro hardness the micro hardness. Fig. 3 shows that the
of different composites was determined by the addition of varying % of marble dust in Al-
help of Vickers micro hardness tester as per the based composite improve the hardness of
guidelines of ASTM E384.Mirror polished composite (76 Hv to 88 Hv). The maximum
samples of the composite were placed on hardness (88 Hv) was observed for the 6 wt %
Bakelite so that hardness sample could not slide marble dust reinforced composite and minimum
on the application of load. In the hardness test, a hardness (76 Hv) was observed for unreinforced
load of 5 kg was applied by the help of diamond marble dust composite. This shows that hardness
indenter for 15-20 seconds and for getting of AMC can be improved by the use of varying
⌂
749
% of marble dust because well interfacial But it is observed that for the case of 6%
bonded hard particle resists the motion of reinforced marble dust the value of tensile
dislocations. strength comes to be equal to 242 MPa this
value is less as compared to 4% marble dust
reinforced composite this is because in the case
of 6% reinforced system agglomeration of
marble dust particle take place.
With the help of 3 points bend test, flexural

strength has been obtained. The behavior of
bend test for different % of marble dust has
shown in Fig.5. Marble dust reinforced
composites shows higher flexural strength as
compared to unreinforced marble dust
composite. The maximum value of flexural
strength (262 MPa) is for 4% reinforced marble
dust and the minimum value of flexural strength
Fig.3: Vickers Micro hardness Results
(210 MPa) for unreinforced marble dust
composite.
3.2 Tensile Strength
The behaviors of tensile strength with varying

wt % of the marble dust have shown in Fig. 4.
Fig. 5: Flexural Behavior of Al Composites
This increase in flexural property occurs due to

the good interface bonding between the hard
Fig. 4: Tensile Behavior of Al Composites particles and the matrix material. But it is
observed that for the case of 6% reinforced
It is observed from the data of tensile test that marble particles the flexural strength comes to
the strength (tensile) of the marble reinforced be equal to 251 MPa this is less as compared to
composites has higher tensile strength as the % 4% marble reinforced system this is because for
of marble dust particles increases from 0 to 4% 6% reinforced system agglomeration of marble
(180 MPa to 249 MPa) this is due to fact that the dust particle take place.
load is transferred to the reinforcement
particles which are strongly bonded with the 4. Conclusion
matrix of aluminium and due to pinning effect. In this research, it is concluded that after the
addition of marble reinforcement
⌂
750
1. The addition of varying % of marble dust in 9. ASTM Standard E8/E8M-16a, Standard

Al-based composite improves the hardness of Test Methods for Tension Testing of
composite (76 Hv to 88 Hv) and the maximum Metallic Materials.
hardness (88 Hv) was observed for the 6 wt % 10. Park BG, Crosky AG, Hellier AK (2001)
Journal of Materials Science : Material
and minimum for unreinforced.
characterization and mechanical properties
2. The tensile strength of the marble reinforced of Al2O3-Al metal matrix composites : 2417
composites has higher tensile and flexure – 2426 doi:10.1023/A:1017921813503
strength as the % of marble dust particles 11. Melgarejo HZ, Suarez OM, Kumar S (2008)
increases from 0 to 4% (180 MPa to 249 MPa Composites Part A: Applied Science and
for tensile and 210MPa to 262 MPa for flexure). Manufacturing: Microstructure and
3. It is observed that for the case of 6% Properties of Functionally Graded Al-Mg-B
reinforced marble particles the tensile and Composites Fabricated by Centrifugal
flexural strength comes to be equal to 242 MPa Casting: 1150-1158.
and 251 MPa respectively, these values are less doi:10.1016/j.compositesa.2008.04.002
12. Kumar GBV, Rao CSP, Selvaraj N,
as compared to 4% marble reinforced system
Bhagyashekhar (2010) Journal of Minerals
this is because agglomeration of marble dust & Materials Characterization & Engineering
particle take place for 6% reinforced system. : Studies on Al6061-SiC and Al7075-Al2O3
REFERENCES Metal Matrix Composites: 43-55.
doi:10.4236/jmmce.2010.91004
1. Mathews FL, Rawlings RD (1994) 13. Kennedy AR (2002) Journal of materials
Composite Materials; Engineering and science: The Microstructure and Mechanical
Science. doi: 10.1016/C2013-0-17714-8. Properties of Al-Si-B4C Metal Matrix
2. Singh AA, Afrin S, Karim Z (2017) Green Composites: 317-323.
Composites: Versatile Material for Future. doi:10.1023/A:1013600328599
Green Energy and Technology. Doi: 14. Sharifi EM, Karimzadeh F, Enayati MH,
10.1007/978-3-319-49382-4_2. (2011) Materials & Design: Fabrication and
3. Bauri R, Yadav D (2018) Introduction to Evaluation of Mechanical and Tribological
Metal Matrix Composites. Metal Matrix Properties of Boron Carbide Reinforced
Comp. by Friction Stir Processing: 1-16. Aluminum Matrix Nanocomposites: 3263-
doi:10.1016/B978-0-12-813729-1.00001-2. 3271. doi:10.1016/j.matdes.2011.02.033.
4. Haghshenas M (2016) Metal–Matrix 15. Mishra SK, Satapathy A (2014) Materials
Composites. Reference Module in Materials Science and Technology: Ceramic
Science and Materials Engineering. doi: Particulate Filled ZA-27 Metal Matrix
10.1016/B978-0-12-803581-8.03950-3. Composites:Comparative Analysis:1495-
5. Mallick PK (2012) Advanced materials for 1499.doi:10.1179/1743284713Y.000000045
automotive applications: an overview. 3
Advanced materials for automotive 16. Bandyopadhyay NR, Ghosh S, Basumallick
applications: an overview: 5-27. A (2007) Materials and manufacturing
Doi:10.1533/9780857095466.5 processes: New Generation Metal Matrix
6. Bharti A, Tripathi H (2019) Enhancement of Composites: 679-682
Fatigue Life of TIG-Welded Joint by doi:10.1080/10426910701384872
Friction Stir Processing. Renewable Energy 17. Rohatgi PK (2013) Defence Science
and its Innovative Technologies1: 51-59. Journal: Metal Matrix Composites: 323-349.
Doi: 10.1007/978-981-13-2116-0_5. doi:10.14429/dsj.43.4336
7. Xavier L.F. and Suresh P (2016) Wear 18. Behera R, Das S, Chatterjee D, Sutradhar G
Behavior of Aluminium Metal Matrix (2011) Journal of Minerals and Materials
Composite Prepared from Industrial Waste. Characterization and Engineering:
Sci. World J.ID 6538345: 1-8. doi: Forgeability and Machinability of Stir Cast
10.1155/2016/6538345. Aluminum Alloy Metal Matrix Composites:
8. ASTM Standard E384, 2011e1, Standard 923-927. doi:10.4236/jmmce.2011.1010072
Test Method for Knoop and Vickers
Hardness of Materials, ASTM International.
⌂
751
Paper ID- 144
Properties of sustainable concrete using copper slag at varying cement

content.
A. M. Bhoi1* (https://orcid.org/0000-0002-4349-2485), and Dr. Y. D. Patil1 (https://orcid.org/0000-0002-4080-

3188)
1
Applied Mechanics Department, S. V. National Institute of
Technology, Surat, Gujrat.
*Corresponding author
Abstract
Concrete can deliver substantial benefits during its service life, but it also has some marked negative impact. The
exploitation of natural resources as a raw material for construction and emission of CO 2 during the production of
cement has a negative impact on the environment. Sustainable development demands not only reduction in the
use of natural resources but also a reduction in the amount of CO2 generated, without compromising the attributes
of concrete. It is envisaged to investigate the attributes of concrete made using copper slag as a substitute for
natural sand at reduced cement content. Concrete mixtures were prepared substituting 60% natural sand with
copper slag. The cement content in the mixtures was gradually reduced for each mixture, at a decrement of 2%.
The properties of concrete mixtures with reduced cement content were compared with properties of conventional
concrete. The mixtures were evaluated for density, compressive strength, split tensile strength, and flexural
strength. Present investigation infers up to 4% cement reduction is possible when 60% fine aggregate is substituted
by copper slag, without compromising the attributes of concrete. The utilization of copper slag will not only
relieve the landfill burden but also substantially reduce carbon footprint making concrete sustainable construction
material.
Keywords: Sustainability, Copper slag, Cement reduction, Compressive strength, Split tensile strength, Flexural
Strength
Contact details of the corresponding author.
● Akshaykumar M. Bhoi,
● Telephone: +91-9168527700,
● fax: +91-253-257-1439,
E-mail address: akshaybhoi@gmail.com
752
1 Introduction
The remarkable versatility, flexibility, stress redistribution ability and ease of construction makes concrete a
significant construction material. Concrete can deliver substantial benefits during its service life, but it also has
some marked negative impact. The exploitation of natural resources as a raw material for construction and
emission of CO2 during the production of cement has a negative impact on the environment. Sustainable
development demands not only reduction in the use of natural resources but also a reduction in the amount of CO2
generated, without compromising the attributes of concrete.
Driven by the rising environmental concern researchers round the globe have focused on the utilization
of the industrial waste materials for the production of the concrete. Researchers have undertaken various studies
to investigate the effect of utilizing industrial waste materials as a substitute for either aggregate or binding
material in concrete. To name a few are: fly ash (Kadkade and Mishra 2011; Ismail et al. 2013; Chousidis et al.
2015; Lavanya et al. 2015), imperial smelting slag (Tripathi and Chaudhary 2016), glass waste (Chidiac and
Mihaljevic 2011; Kim et al. 2017), bottom ash (Ghafoori and Bucholc 1996; Andrade et al. 2007; Kurama and
Kaya 2008; Thandavamoorthy 2015), GGBFS (Robins et al. 1992; Nazari and Riahi 2011; Tüfekçi and Çakır
2017) and so on.
The present investigation is the part of a long-term study carried out to investigate the suitability of using
copper slag as a partial or full substitute of the natural river sand in concrete. A part of the study on the use of
copper slag as a substitute for natural sand at constant and varying water content is already published (Bhoi et al.
2018). The previous investigation by the authors infers that 60% substitution of the natural sand with copper slag
results in the enhanced strength attributes of the concrete. It is envisaged to investigate the attributes of concrete
made using copper slag as a substitute for natural sand at reduced cement content. Concrete mixture were prepared
substituting 60% natural sand with copper slag. The cement content in the mixtures was gradually reduced for
each mixture, at a decrement of 2%. The properties of concrete mixtures with reduced cement content were
compared with properties of conventional concrete. The mixtures were evaluated for density, workability,
compressive strength, split tensile strength, and flexural strength.
2 Experimental program
2.1 Materials
Cement
53 grade, Ordinary Portland cement conforming to the specifications laid down in IS 12269-2013 (BIS: 12269-
2013 2013) was used for the preparation of mortar mix.
Fine aggregate
Control concrete mix (Mix CC) was prepared using natural river sand brought from Tapi river, Maharashtra. The
remaining 5 mixes (RC (reference concrete), and M1 to M4) were prepared using copper slag procured from Birla
copper Hindalco Industries, Gujrat, (comprising 60% copper slag as a sand substitute. The sand, as well as copper
slag used in the present study, is categorized within zone-I, as per IS 383-1970 R 1997 (BIS: 383-1970 1997).
The river sand and copper slag obtained had a specific gravity of 2.67 and 3.44 respectively, which was determined
753
by the pycnometer method as per IS 2386- part III-1963, R 2002 (BIS:2386 part III-1963 2002). Water absorption
of sand and copper slag was determined to be 0.98% and 0.017%, respectively, following the procedure depicted
in IS 2386- part III-1963, R 2002 (BIS:2386 part III-1963 2002).
The chemical composition of copper slag
The chemical composition of copper slag was determined using an X-ray fluorescence spectrometer. Copper slag
consists of approximately 55% Fe2O3, 33% SiO2, 6% CaO, 3% Al2O3 along with traces of MgO, SO3, Na2O,
TiO2, Mn2O3, and CuO. The exceptionally high amount of Fe2O3 is the reason for its high specific gravity. Low
amount of lime present makes it almost unlikely to be used as the cementitious material.
2.2 Methods
Mix preparation
Mix proportion of M40 was obtained following the guideline laid down in BIS: 10262-2009 2009 at the water to
cement ratio of 0.44. The quantities of the ingredients used per m3 of concrete are shown in Table 1. Conventional
concrete mix was prepared using natural river sand as fine aggregate in concrete (mix CC) using cement content
as per the mix design for M40 grade concrete. Mix RC (reference concrete) was prepared by substituting 60% of
natural sand with copper slag in the mix CC keeping same binder content. Remaining four mixes were prepared
by reducing the cement content in the mix RC by 2% at each level of reduction (2%, 4%, ^%, and 8%).
Preparation and testing of the specimen
Cube shaped specimen with 150 mm side were used for compressive strength and density as per BIS: 516 -1959
2004. Cylindrical specimens of size 150 mm diameter and 300 mm height were evaluated for split tensile strength
as per BIS:5816-1999 2004. Flexural strength was determined referring BIS: 516 -1959 2004.
Water was slowly added to the properly dry mixed ingredients of concrete and mixing was continued to
get proper mix. Slump test was conducted on the fresh concrete mix to determine the workability of concrete.
Molds were then filled in layers and were properly compacted over the vibrating table.
Table 1 Details of mix preparation for various concrete mixes-quantities in kg/m3.
Mix designation Cement Sand copper slag 20 mm 10 mm Water

kg/m3 kg/m3 kg/m3 kg/m3 kg/m3 kg/m3
Control Concrete (CC) 448 668 0 564 569 197
Reference Concrete
448 262 514 675 450 179
(RC)
M 1 (2% less cement) 439 263 517 677 451 176
M 2 (4% less cement) 430 265 520 679 452 172
M 3 (6% less cement) 421 266 523 682 455 168
M 4 (8% less cement) 412 267 524 685 457 165
754
3.1 Workability and density
Density Slump
2595 180
170
2575
Density of Concrete (kg/m3)
160
150
2555
140
Slump (mm)
2535 130
120
2515
110
100
2495
90
2475 80
CC RC M1 M2 M3 M4
Concrete mix
Figure 1 Slump and density of concrete
The variation in the workability of concrete measured in terms of slump of fresh concrete is depicted in Figure 1.
It is apparent from the Figure 1 that the workability of concrete dramatically increased when the copper slag
substituted 60% natural sand. The variation in the cement content had a minor effect on the workability of
concrete. The workability of mix RC was 73% higher than that of the mic CC. The workability all concrete
containing copper slag was more or less similar and was higher than the workability of control concrete. The
almost zero water absorption of copper slag along with its glassy surface texture justify such behavior of the mix.
Figure 1 also illustrates the density of concrete mixes. Figure 1 infers that the substation of the natural sand with
copper slag substantially affects the density of the concrete mix. The high specific gravity of the copper slag in
comparison with that of natural sand is an obvious reason for the variation in the density of the concrete mix. Even
though the density of the concrete containing copper slag was considerably higher than that of concrete with
natural river sand, the densities of the concrete mixture M1 to M4 was comparable with that of mix RC.
755
60 7 DAYS 28 DAYS 56 DAYS

55
Compressive strength (N/mm2)
50
45
40
35
30
25
20
CC RC M1 M2 M3 M4
Concrete mix
Figure 2 Compressive strength of concrete mixtures with age.
Figure 2 depicts the results of compressive strength test carried on the concrete mixes. Concrete mix RC shows a
noticeable improvement in the compressive strength in comparison with concrete mix CC. The mix RC achieved
a compressive strength 10% higher than the compressive strength of concrete mix CC at the age of 28 days. In
comparison with mix CC mix M1 also attained 2.92% higher strength at the age of 28 days. The 28 days
compressive strength of concrete mix M2 was marginally less (0.43%) than that of concrete mix CC. Lastly, mix
M3 and M3 could attain compressive strength respectively, 4.28 and 18.17% less than that of mix CC.
At the age of 7 days, concrete mix CC, RC, M1, M2, M3, and M4, respectively attained 65.93, 63.02,
60.22, 60.69, 61.29, and 61.17% of their 28 days compressive strength. Further at the age of 56 days the
compressive strength of CC, RC, M1, M2, M3, and M4, respectively enhanced by 3.42, 3.37, 1.91, 1.77, 1.82,
and 1.75% of their respective compressive strength at the age of 28 days. It can be seen from figure 2 that the
compressive strength of concrete mixes decreased as the cement content in the mix was reduced. At 2% reduction
in the cement content (mix M1) the compressive strength at 28 days age was higher than that of mix CC, while
that at 4% reduction in cement (mix M2) content was insignificantly less than that of mix CC. However, further
reduction in the cement content resulted into considerable decrease in the compressive strength of the concrete.
The better compressibility of copper slag particles along with the improved cohesion owing to angular,
sharp edges (Wu et al. 2010a, b) is the reason for the superior performance of the concrete containing copper slag
as a sand substitute. However, reduction in the amount of binding material caused a decrease in the strength
attainment in the mixes with reduced cement content.
756
3.3 Split tensile strength
7 7 DAYS 28 DAYS 56 DAYS

Split tensile strength (N/mm2)
0
CC RC M1 M2 M3 M4
Concrete mix
Figure 3 Split tensile strength of concrete mixtures with age.
Figure 3 illustrated the influence of reduction in the cement content on the split tensile strength of the concrete
mixes. At the similar cement content, the split tensile strength of concrete with 60% copper slag (mix RC) was
7.14% higher than that of mix CC at the age of 28 days. Even when the cement content in the mix with copper
slag was reduced by 2% (mix M1) and 4% (mix M2), the split tensile strength was, respectively, 3.36 and 1.47%
higher than that of mix CC. Further reduction in the cement content resulted in the reduction in the compressive
strength by 7.14 and 13.23%, respectively for mix M3 (6% reduced cement) and M4 (8% reduced cement).
Author ACI 318-99 (1999)

5.5
Split tensile strength (N/mm2)
5
R² = 0.928
4.5
3.5
3
35 40 45 50 55
Figure 4 Relationship between compressive strength and split tensile strength of concrete mixtures at 28 days.
Figure 4 depicts the relationship between the compressive strength and the split tensile strength of the
1
concrete mixes at the age of 28 days. The theoretical value of split tensile strength is considered as
1.8
√𝑓𝑐𝑘 ,
757
(ACI-318 1999) (where, fck is the compressive strength of the concrete at the age of 28 days),. All the concrete
mixes in the present investigation attained a split tensile strength higher than that given by the theoretical value
given by the formula.
The results of the flexural test on the various concrete specimens are illustrated in figure 5. Similar to the
compressive strength and split tensile strength the flexural strength of the concrete mix RC was higher than that
of mix CC. The concrete mix RC and M1 (2% less cement) attained flexural strength, respectively 3.94 and
1.42% higher than that of mix CC. Moreover, the compressive strength of mix M2 (4% less cement) was
marginally less (0.47%) than that of mix CC. The flexural strength of mixes decreased with reduction in cement
content. However, the results infer that the amount of the cement in the mix can be reduced up to 4%. Further
reduction in the cement content resulted into flexural strength less than that of mix CC.
8
7 DAYS 28 DAYS 56 DAYS
7
Flexural Strength (N/mm2)
0
CC RC M1 M2 M3 M4
Concrete mix
Figure 5 Flexural strength of concrete mixtures with age.
The relationship between the compressive strength and flexural strength at the age of 28 days is shown
in figure 6. Similar to the split tensile strength of the concrete, the flexural strength for all mixes was higher than
the theoretical value given by the relation 0.7 √𝑓𝑐𝑘 (BIS: 456-2000 2005).
758
Author BIS 456-2000 R 2005

Flexural strength (N/mm2) 7
6.5
R² = 0.9104
6
5.5
4.5
4
35 40 45 50 55
Figure 6 Relationship between compressive strength and flexural strength of concrete mixtures at 28 days.
4 Conclusion
The present experimental investigation was conducted to access the suitability producing copper slag concrete at
reduced cement content, to enhance the sustainability credential of concrete construction. Experimental study was
conducted to investigate the attributes of copper slag concrete at reduced cement content. The findings of the
present investigation indicate that cement content in the concrete mix can be reduced if copper slag is used as 60%
substitute for the natural river sand in the production of concrete. The utilization of copper slag will not only
relieve the landfill burden but also substantially reduce carbon footprint by reducing the cement content, making
concrete sustainable construction material.
Following conclusions can be drawn based on the present investigation.
1) The substitution of natural sand by copper slag increases the workability of concrete mix at the same
water content.
2) With the use of copper slag the density of concrete increases.
3) At 60% substitution of natural sand by copper slag, 2% cement reduction results in better attributes of
concrete in comparison with conventional concrete.
4) 4% cement reduction along with 60% sand substituted by copper slag results in marginally lowered
attributes of concrete.
5) Use if copper slag along with reduced cement can produce sustainable concrete without losing the desired
strength, therefore it is recommended to use 60% copper slag as a sand substitute along with 4% cement
reduction.
5 References
ACI-318 (1999) Building Code Requirements for Structural Concrete ( Aci 318M-99 ) and
Commentary ( Aci 318Rm-99
Andrade LB, Rocha JC, Cheriaf M (2007) Evaluation of concrete incorporating bottom ash as
759
a natural aggregates replacement. Waste Manag 27:1190–1199. doi:

10.1016/j.wasman.2006.07.020
Bhoi AM, Patil YD, Patil HS, Kadam MP (2018) Feasibility Assessment of Incorporating
Copper Slag as a Sand Substitute to Attain Sustainable Production Perspective in
Concrete. Adv Mater Sci Eng 2018:
BIS: 10262-2009 (2009) Indian concrete mix design guide lines. Bur Indian Stand Dehli
BIS: 12269-2013 (2013) Ordinary Po rtland Cement 53 grade- Specification. Bur Indian Stand
Dehli
BIS: 383-1970 R (1997) Specification for Coarse and Fine Aggregate From Natural Sources
for Concrete. Bur Indian Stand Dehli
BIS: 456-2000 R 2005 (2005) Plain and Reinforced Concrete. Bur Indian Stand Dehli. doi:
624.1834 TAY
BIS: 516 -1959 R 2004 (2004) Method of tests for strength of concrete. Bur Indian Stand Dehli
BIS:2386 part III-1963 R 2002 (2002) Method of Test for aggregate for concrete, Part III
Specific gravity, density, voids, absorption and bulking. Bur Indian Stand Dehli
BIS:5816-1999 R 2004 (2004) Splitting tensile strength of concrete. Bur Indian Stand Dehli
Chidiac SE, Mihaljevic SN (2011) Performance of dry cast concrete blocks containing waste
glass powder or polyethylene aggregates. Cem Concr Compos 33:855–863. doi:
10.1016/j.cemconcomp.2011.05.004
Chousidis N, Rakanta E, Ioannou I, Batis G (2015) Mechanical properties and durability
performance of reinforced concrete containing fly ash. Constr Build Mater 101:810–817.
Ghafoori N, Bucholc J (1996) Investigation of Lignite-Based Bottom Ash for Structural
Concrete. J Mater Civ Eng 8:128–137. doi: 10.14359/289
Ismail I, Bernal SA, Provis JL, et al (2013) Influence of fly ash on the water and chloride
permeability of alkali-activated slag mortars and concretes. Constr Build Mater 48:1187–
1201. doi: 10.1016/j.conbuildmat.2013.07.106
Kadkade DG, Mishra PK (2011) Fly ash blended cement for concrete pavements. 17–24
Kim SK, Kang ST, Kim JK, Jang IY (2017) Effects of Particle Size and Cement Replacement
of LCD Glass Powder in Concrete. Adv Mater Sci Eng 2017:3928047. doi:
10.1155/2017/3928047
Kurama H, Kaya M (2008) Usage of coal combustion bottom ash in concrete mixture. Constr
Build Mater 22:1922–1928. doi: 10.1016/j.conbuildmat.2007.07.008
Lavanya G, Jegan J, Lavanya G, Jegan J (2015) Durability Study on High Calcium Fly Ash
760
Based Geopolymer Concrete. Adv Mater Sci Eng 2015:1–7. doi: 10.1155/2015/731056
Nazari A, Riahi S (2011) Effects of Al2O3 nanoparticles on properties of self compacting
concrete with ground granulated blast furnace slag (GGBFS) as binder. Sci China Technol
Sci 54:2327–2338. doi: 10.1007/s11431-011-4440-y
Robins PJ, Austin SA, Issaad A (1992) Suitability of GGBFS as a cement replacement for
concrete in hot arid climates. Mater Struct 25:598–612. doi: 10.1007/BF02472228
Thandavamoorthy TS (2015) Feasibility of making concrete using lignite coal bottom ash as
fine aggregate. Arch Civ Eng 61:19–30. doi: 10.1515/ace-2015-0022
Tripathi B, Chaudhary S (2016) Performance based evaluation of ISF slag as a substitute of
natural sand in concrete. J Clean Prod 112:672–683. doi: 10.1016/j.jclepro.2015.07.120
Tüfekçi MM, Çakır Ö (2017) An Investigation on Mechanical and Physical Properties of
Recycled Coarse Aggregate (RCA) Concrete with GGBFS. Int J Civ Eng 15:549–563.
doi: 10.1007/s40999-017-0167-x
Wu W, Zhang W, Ma G (2010a) Mechanical properties of copper slag reinforced concrete
under dynamic compression. Constr Build Mater 24:910–917. doi:
10.1016/j.conbuildmat.2009.12.001
Wu W, Zhang W, Ma G (2010b) Optimum content of copper slag as a fine aggregate in high
strength concrete. Mater Des 31:2878–2883. doi: 10.1016/j.matdes.2009.12.037
761
Paper ID- 145
WEAR BEHAVIOR OF MARBLE DUST FILLED ALUMINIUM METAL

MATRIX STRUCTURAL COMPOSITE
Hariom Tripathi1⌂, Sandeep Kashyap2
1
Material Science and Engineering, MNNIT Allahabad, Prayagraj-211004, INDIA
2
Mechanical Engineering Department, MNNIT Allahabad, Prayagraj-211004, INDIA
ABSTRACT
In the present work, the stir-casting method for the preparation of Al-6063 composites was adopted. Samples were
prepared with the reinforcing different weight % of marble dust (2%, 4%, 6%wt) and fixed wt % of Graphite
(Gr)(2%wt), glass fiber (2% wt), and Boron Carbide (B4C)(4% wt) by, Stir Casting Setup.
The present work has provided the descriptions of the wear characteristics of the marble dust filled Al-6063
composites. A series of erosion experiments were conducted on the composites under various test conditions using
an Air jet erosion machine. It was seen that with the incorporation of marble dust powder, the wear resistance
property of composites was significantly improved.
Keywords: Aluminum Metal Matrix structural Composite, Stir casting technique, Erosion wear,
Graphite, Glass fiber, Boron Carbide, Marble dust.
1. Introduction
The manufacturing of materials which are used one reinforcement particles are present are
in engineering applications should be durable, known as the hybrid composites [9,17]. Now a
high strength and light in weight [1-3]. These day’s hybrid aluminium based composites are
types of materials are more popular for the getting popularity in many engineering field
purpose of research and industries. Aluminium applications because they can alter the
based composites (AMCs) fulfill these mechanical as well as the physical properties
conditions because they have low density, (thermal conductivity, friction coefficient,
design flexibility better specific strength and density, wear resistance etc) which are desired
good wear properties etc [4-5]. AMCs consists for the composite system. The main reason for
of two parts first one is aluminium base and incorporate the more than one reinforcement
second part consists of different reinforcement particles is to increase the wettability between
materials that are used for the enhancing the the matrix and reinforcement medium,
properties of the base material [3,5]. The enhancing the mechanical properties and wear
reinforcements are generally those materials resistance etc [9, 17].
which have superior properties then the matrix The addition of graphite as a reinforcement
materials and they are embedded in the Al particle in this hybrid composite system reduces
matrix in Al based composite system. It has the heat buildup because coefficient of friction
been found that the AMCs demonstrate better decreases with the addition of graphite [11-16].
design flexibility, mechanical, physical and But the addition of graphite decreases the
tribological properties as compare to without mechanical properties like hardness, tensile
reinforced aluminium base materials [2-6]. strength [11-16] so for enhancing the
The performance of Al based composites can be mechanical properties hard particles like boron
further enhanced by embedding more than one carbides (B4C)(4%) are added in the matrix
types of reinforcement particles [7-10], these material .The B4C has lower density then other
type of composites in which there are more than available commercial ceramic materials(TiC,
⌂
762
SiC, Al2O3 and TiB2 etc), higher specific time of casting, 1-1.1weight percentage of mg
strength, high hardness, good chemical inertness powder was also added in the molten aluminium
and good thermal stability etc that is why boron to increase the wettability between the different
carbide is the ideal candidate as the reinforcement particles with the aluminum 6063
reinforcement in aluminium matrix[7-11] .But as matrix. The compositions of alloy which is used
the cost of B4C is more so for economical in this research as the matrix in the composite
purpose in this research for further increase in are shown in table 1. The melting point of
the mechanical properties composites are Al6063 is 650-6600C.
prepared with the reinforcing different weight % Table1: Chemical Composition of Al6063.
of marble dust (2%, 4%, 6%wt) and fixed wt %
of Graphite (Gr)(2%wt), glass fiber (2% wt), and Elements Al Mg Si Fe Others
Boron Carbide (B4C)(4% wt) by, Stir Casting Wt.% Max 0.45- 0.2- Max ~0.65
Setup. 97.5 0.90 0.6 0.35
The most important factor for the preparation of 2.2. Casting of AMC’s
particles reinforced composites are the
wettability of particles with the matrix material. To produce marble dust reinforced hybrid
In particle reinforced Al based hybrid composite AMC’s, the billet of al6063 alloys were melted
for enhancing the wettability 1-1.1% of in the electrical resistance furnace and stir
magnesium powder is added in the composite casting technique was used for the preparation of
during the casting. hybrid composite (shown in Fig. 1).
In general hybrid composites of aluminium are
fabricated by power metallurgy method,
semisolid processing (Friction stir processing)
[18] and liquid state processing (squeeze casting,
stir casting etc.). Among all the fabrication
techniques casting is very popular for fabricating
composite in bulk.
The objective of this research work is to

fabricate an advanced series of Al6063 based
MMC’s reinforced with and without marble dust
particles and observe the effect of reinforcement
particles on the erosion wear properties.
Fig.1: Stir Casting Setup
2. Experiments
Matrix material billet (aluminium6063) was
2.1. Matrix and Reinforced materials
placed inside the graphite crucible and was
In this research work 6063 series of aluminium melted inside the electrical furnace in the inert
alloy having specific gravity 2.7 was used as the atmosphere and when the temperature of liquid
matrix material and varying % of marble dust metal reached at 770-790°C (heating rate 25
powder (0, 2, 4, 6%) and fixed % of glass fiber °C/min), then magnesium powder was added for
(2%) (.005-.01mm dia.), graphite powder (2%), enhancing the bonding between the aluminum
and B4C (4%) were added as the reinforcement and reinforcement particles
in the matrix material for the development of After that preheated (350-400°C) reinforced
aluminium based hybrid composite. Marble dust particles (glass fiber (2%), graphite (2%), B4C
having specific gravity 2.8 is an industrial waste (4%) and varying % of marble dust powder (0,
which is utilized in this research work; it 2, 4 and 6% weight)) were added and the
contains oxides of silicon, magnesium, calcium mixture of melted metal and reinforcement
and iron. Graphite powders of avg. size 30μm, particles were stirred at 450-550 rpm for 15-20
minutes for uniform distribution of particles in
boron carbide particles of avg. size 25μm and
marble dust of avg. size 40μm were used. At the the matrix. After that, the liquid mixture of
763
hybrid composite was poured into the sand mold

(as shown in Fig.2) and the melted composites 2.3. Particle Erosion Test (Wear Test)
were allowed to cool. And by this way, four In order to find out the wear resistance of
different grades (0, 2, 4 and 6weight% of marble different AMC’s, Test specimens were
dust) of AMC’s were prepared. prepared according to ASTM rule.
Samples were polished like mirror for
performing erosion wear test. The set up
used in this test was proficient of generating
duplicatable erosive condition for
performing wear resistance capability of
prepared composites. The pictorial view of
the Air-Jet Erosion Test machine (at
Mechanical System Design Lab, Mechanical
Engineering Department, MNNIT,
Allahabad) is shown in Fig.3. The erosion
wear test rig contains air dryer unit, particle
mixer, particle feeder, air compressor and
accelerating chamber units.
Fig.2: Sand mold with casted wear test

specimen
Table2: Designation of Samples
Code Composition (wt %)
A0 Al 6063+ GF (2%) + Gr (2%) + B4C

(4%) Fig.3: Erosion Testing Machine
A1 Al 6063+GF(2%) + Gr (2%) +B4C
(4%) + MD (2%) In this research work, preheated sand (silica) of
A2 Al 6063+ GF(2%) + Gr(2%) + B4C different particle sizes 75μm, 125μm, 150μm
(4%) + MD(4%) and 250μm were utilized as the erodent
A3 Al 6063+GF(2%) + Gr(2%) + B4C medium. The compressed and dried air mixed
(4%) + MD(6%) with the sand particles in the mixing chamber
(feed rate of erodent 5gm/min). The mixture of
air with silica particles was accelerated by the
764
help of nozzle (convergent tungsten carbide) of

dia. 1.5 mm. During the time of impact testing
by the help of erodent, the samples were held at
different angles w.r.t. erodent flow direction.
The stand-off distance kept constant (10mm)
between specimen and nozzle. The sample size
for erosion test specimen (25mm X 25mm X
10mm) was well polished by using emery paper
of 2000 grades and cloth polishing. Erosion test
specimens were weighed before and after the Fig.5: Surface of Composite after the Wear
test (particle erosion test) using the electronic Test on the Variation of Impingement Angles
balance (accuracy ± 0.01 mg). The losses of
weight were recorded and by the help of It was observed that the erosion rates were
subsequent calculation, erosion rate was maximum at an angle of 60º for all the samples
measured. irrespective of particulate content. This showed
3. Results and Discussion the semi – ductile erosion nature of the metal
matrix composite. It is evident from Fig. 4 that
The erosion wear rates of marble dust filled Al 5% boron carbide and 0% marble dust
6063 metal matrix composites under the various reinforced in Al 6063 metal matrix
conditions were investigated. composite(A0) exhibits maximum erosion rate
and 5% boron carbide + 6% marble dust
3.1. Influence of Impingement Angle reinforced in Al 6063 metal matrix composite
In this type of testing, erosion tests were (A3) shows the minimum erosion rate
performed at different impingement angle while irrespective of the impingement angle. It is
the other parameters remained constant (Impact because of the presence of marble dust, and the
velocity = 60 m/sec, and erodent size = 250μm). proper mixing of particulate and matrix i.e. the
The results were plotted in between the wear matrix covers the whole filler materials in the
rate (mg/kg) and the impingement angle (in composites and it exhibits along the bonding
degree) as shown in Fig.4. and Fig.5 shows the between among all of them. The Fig. 5 shows
surface of composite after the wear test at the the wear pattern of metal matrix irrespective of
different impingement angles. different impingement angle. When the silica
sand particles were strikes on the surface of
composites at various angles, then they primarily
interact with the matrix of the composites. After
sufficient loss of materials, they interact on the
reinforced particulates. As the hardness value of
B4C and marble dust is much higher than the
aluminium matrix material that is why the rate
of erosion decreased Due to the presence of hard
particles in the aluminium metal matrix.
Fig.5 also indicates the erosion

propagation relevant to each of the impingement
angle. When solid particles impact at an angle,
the hard particle penetrate, the surface lead to
removal of materials mostly by micro –
ploughing. When the removal of particles starts,
then there is the formation of the passage from
the place of removed particles on the surface of
the base material, resulting in the more efficient
Fig.4: Wear Rate Vs Impingement angle
removal of base material. In case of impact at an
765
angle 30º the penetration range of solid particle It is quite cleared from fig.6 that erosion rate of
into the matrix was very minimum and passing the aluminium based metal matrix composites
passage very high, resulting less removal of the increases with the increase in the velocity of
matrix material. impact. It was also observed that the composite
As in the further increased of impingement (A3) shows minimum erosion rate followed by
angle up to 60º, the penetration range of the hard the composites (A0, A1, A2). If the velocity is
particle into matrix material slightly increases very low (31.58 m/s) then the stresses induced
with sufficient passage for flow out the eroded due to impact are not sufficient (for the
particles. In other terms, the impact angle also occurrence of the plastic deformation) and wear
determines the magnitude and direction of proceeds due to the surface fatigue.
impact velocity. It mainly resolves the impact
velocity into two main components. Firstly the However, with the increase in velocity of impact
normal component to the surface, determine how (45 to 62.07 m/s) the erosion rate significantly
much the impact and the load. Secondly the increases because at higher impact velocity the
component of tangential velocity determines erosion rate occurs due to plastic deformation
how much the sliding take places and shear and the increase in penetration of erodent
loading (addition to the normal load) to the particles on impact. These effect causes
surface. From above, it is signified that the dissipation of more amount of energy to the
change in impingement angle leads towards the composite surface. This leads to increase in the
change in magnitude and nature of stress. crack growth and the erosion rate increases.
3.2. Impact Velocity Influence 4. Concluding Remarks

The erosion rate variation with respect to impact
velocity is shown in Fig. 6. Erosion test were This chapter has provided the experimental
performed at four different velocities and all investigation into the erosion behavior of marble
other variables were kept constant (Impingement dust filled Al 6063 composites leading to the
angle = 60º, and erodent size = 250μm). following major conclusions:
1. An industrial waste like marble dust
successfully utilized to produce low cost
metal matrix composites.
2. Erosion resistance of the developed
composites is increasing with filler
content and A3 (6 wt% marble dust)
composite has shown least erosion rate
than others.
3. With the increase in velocity of impact
(45 to 62.07 m/s) the erosion rate
significantly increases.
REFERENCES
1. Mathews, F.L. and Rawlings, R.D.,
Composite Materials; Engineering and
Science, Chapman and Hall, 1994.
2. Park, B. G., “Material characterization and
mechanical properties of Al2O3-Al metal
matrix composites”, Journal of Materials
Science 36, 2001, pp. 2417 – 2426.
Fig.6: Impact Velocity Influence on 3. Stolarski, T. A, “Tribology in Machine
Erosion Rates on different AMCs Design”, 1990, Heiman Newnes, UK.
4. Päivi Kivikytö-Reponen, “Correlation of
Material Characteristics and Wear of
Powder Metallurgical Metal Matrix
766
Composites”, Doctoral Theses in Materials During Orthogonal Cutting”, International

and Earth Sciences, Helsinki University of Journal of Machine Tools and Manufacture,
Technology, Laboratory of Materials Vol.47 (10), 2007, pp. 1497-1506.
Science, Espoo”, 2006. 15. Shoukry, S. N. , Prucz, J. C. ,
5. Budinski, K. G, “Surface Engineering for Shankaranarayana, P. G. and William, G.
Wear Resistance”, 1998, Prentice Hall, New W., “Microstructure Modeling of Particulate
Jersey. Reinforced Metal Matrix Composites”,
6. Robinowicz, E, “Friction and wear of Mechanics of Advanced Materials and
materials”, 1965, John Willey, New York, Structures, Vol.14 (6), 2007, pp. 499- 510.
USA. 16. Jayamathy, M., et al., “Influence of
7. Kosel, T. H., “Solid Particle Erosion”, 1992, Reinforcement and Processing on the Wear
ASM Handbook, ASM International, 18, Response of Two Magnesium Alloys”,
199-213. Materials and manufacturing processes,
8. Ranganath, G., Sharma, S. C., and Krishna, Vol.20 (2), 2005, pp. 255-271.
M., “Dry Sliding Wear of Garnet Reinforced 17. Melgarejo, Humberto, Z., Marcelo O.S., and
Zinc/Aluminium Metal Matrix Composites”, Kumar S., “Microstructure and Properties of
Wear Vol. 251 (1), 2001, 1408-1413. Functionally Graded Al-Mg-B Composites
9. Kumar, M. P., “Dry Sliding Wear Behaviour Fabricated by Centrifugal Casting”,
of Garnet Particles Reinforced Zinc- Composites Part A: Applied Science and
Aluminium Alloy Metal Matrix Manufacturing, Vol.39 (7), 2008, pp. 1150-
Composites”, Materials Science- 1158.
Medziagotyra, Vol.12 (3), 2006, 1392-1420. 18. Bharti A, Tripathi H (2019) Enhancement of
10. Rohatgi, Pradeep K., “Metal Matrix Fatigue Life of TIG-Welded Joint by Friction
Composites”, Defence Science Journal, Stir Processing. Renewable Energy and its
Vol.43 (4), 2013, 323-349. Innovative Technologies1: 51-59. Doi:
11. Bandyopadhyay, N. R., Ghosh S., and A. 10.1007/978-981-13-2116-0_5.
Basumallick., “New Generation Metal
Matrix Composites”, Materials and
manufacturing processes, Vol.22 (6), 2007 :
679-682.
12. Chou, T. W., Kelly A., and Okura A.,
“Fibre-Reinforced Metal-Matrix
Composites," Composites, Vol.16 (3)1985,
pp. 187-206.
13. Avila, Antonio F., and Kumar, K. T.,
“Analysis of Laminate Metal Matrix
Composites", Journal of thermal stresses,
Vol.21 (9), 1998, pp. 897-917.
14. Pramanik, Alokesh, Zhang, L. C., and
Arsecularatne J. A. , “An FEM Investigation
into the Behavior of Metal Matrix
Composites: Tool–Particle Interaction
.
767
prshntbnsd@gmail.com Paper ID- 148
Seismic Response of Overhead Water Tank with Different Staging System

Prashant Bansode1*, Chandrasen Rajemahadik2
1*
PG Student, PVPIT Budhgaon, Sangli, Maharashtra, India
2
Lecturer, Sanjay Ghodawat Polytechnic, Atigre, Kolhapur, Maharashtra, India
Corresponding Author – Prashant Bansode (email id – prshntbnsd@gmail.com)
Abstract
RC Overhead water tanks are considered as a part of critical and key services in many towns and cities.
Safety as well as working of such structures is considered as very crucial during earthquakes, since they put up for
necessities like drinking water, fire fighting during fire accidents, etc. Such structures should stay in operating
condition even after several earthquakes. This RC overhead water tanks involves huge water mass supported on the
top of rangy supporting system called as staging. This investigation is basically studious to discover the seismic
behavior of RC overhead water tanks. It is noticed that, RC overhead water tanks shows effective results with frame
staging, rather than shaft staging. In this study, a FEM based model is used for discovering the nonlinear seismic
performance of RC overhead water tanks with several staging configuration. The staging configuration involved in
this study consists of beams, columns and bracing beams in lateral and vertical direction, which relatively increases
the strength and stiffness of tank supporting system. A fusion of the most frequently constructed tank size, tank
capacity, staging heights, staging system; is developed and observed for different parameters such as base shear,
base moment, lateral displacement and time period of vibration. The commercial software STAAD.Pro is used for
structural analysis and design. Since the seismic response of structure depend upon its dynamic properties and
frequency of ground motion. This seismic analysis of water tank structure cannot be simply carried out on the basis
of maximum value of ground acceleration. To surmount the above cumbersome, earthquake response spectrum
analysis is widely used for earthquake analysis of water tank structures. While using this method, smooth design
spectra is used to determine the utmost value of displacement and forces in members at every mode of vibration.
Finally it has been concluded that, among all bracing considered in this study Water tank with lateral cross bracing
and vertical X - bracing performs well during earthquake.
Keywords: Seismic Response, Spectrum, Finite Element Method, STADD Pro.
1 Introduction
RC overhead elevated water tanks are the supreme city services in many flat region, and subsequently, their
proper functioning during and after strong earthquakes is an important point of interest [1]. These water tanks are
unsteady and very crucial structures; which are prone to have damage during strong ground motions called as
earthquake. Damage to such structures due to earthquakes may disturb drinking water supply system, waste water
treatment plants, failure in preventing large fires and also causes other economic loss [2]. Unlike other structures
like bridges, towers, tunnels; the weight of water storage tanks varies in time because of variable water level. The
failure of such water tank structures not only disturbs the working of water supply system, but also causes fires or
other environmental contamination due to leakage of flammable materials or hazardous chemicals stored in it [3].
Study of several earthquakes in past have shown that, overhead water tanks have very poor and occasionally
disastrous seismic performance. Different types of damages are observed in water tank ranging from circumferential
cracks to complete crumble and failure of water tank with RC pedestals. Different examples showing failure of
overhead water tanks are reported in past during strong ground motions such as Chile Earthquake 1960 [4]; Alaska
1964 [5]; Manjil-Roudbar 1990 [6], Jabalpur 1997 [7], and Bhuj 2001 [8].
There are several reasons for collapse of overhead water tanks, among which unstable design of supporting
elements such as beams, columns and braces, low strength and stiffness of supporting element, etc; is highlighted in
this paper. Beside this, the interaction of soil-structure is also accountable for collapse of water tanks during
earthquake [9]. Among different staging systems, frame staging is widely used than shaft staging. This frame staging
include columns, beams and braces, which acts like a path between water container and foundation for transferring
loads. RC overhead water tanks comprises huge mass of water supported on the peak of slender supporting system,
768
prshntbnsd@gmail.com
which is most crucial concern for the collapse of water tank during earthquakes. Due to the lack of knowledge of
design of supporting system some of the water tanks have shown failure/collapse during earthquake. So it is required
to stay focused on seismic security of this lifeline structure with respect to different staging system which will be
safe during earthquake [10].
The purpose of this work is to illustrate the seismic performance of overhead water container with different
staging systems. The RC overhead water tank behaves like an inverted pendulum type structure as shown in Fig.2,
which resist the lateral seismic forces by means of strength and stiffness of vertical staging system. Subsequently,
the main purpose of this work is to know the effectiveness of tank supporting staging system with different
configuration. Finite element method was used in modeling of RC overhead water tank and its behavior was
considered as nonlinear in nature [1]. Within the limits of Lagrangian-Lagrangian approach, the displacement
parameter is taken as unknown variable for both solid and fluid domains in finite element mesh. Newmark method is
employed to solve the dynamic equation system, for time discretisation algorithm,
[M]Ü + [C] Ů + [K]Ս = F(t) (1)
where; M is mass matrix, C is damping ratio matrix, K is stiffness matrix, Ü is vectors representing accelerations, Ů
is vectors representing velocities, Ս is vectors representing displacements and F(t) is vectors representing external
load on the system [11]. The record of actual time history analysis of a particular location is required, to carry out
the seismic investigation of a structure. But it is very difficult to have such records at each and every location.
Hence, to overcome this hassle earthquake response spectrum method can be used in seismic investigation of
structures. Earthquake response spectrum uses the smooth design spectra and enables us to calculate peak values of
force in members and displacements in all modes of vibration [2].
2 Response Spectrum Method

The foundation of earthquake engineering is to express how a structure will perform under complex ground
acceleration. This performance of structure is determined by distribution pattern of its mass and stiffness. For
example, a bulky structure will experience low accelerations with respect to ground, where as a slender structure
will accelerate away from ground motions, which will result into low absolute accelerations [12].
Maurice Anthony Biot has introduced response spectrum method (RSM) for the first time in 1932 at
Caltech. This method is used to determine the response of structures exposed to seismic forces by using waves or
vibration mode shapes. While introducing the response spectrum method Biot stated that, a structure has certain
number normal modes of vibration, and each mode corresponds to a certain frequency [12].
In order to perform the seismic analysis and design of a structure to be built at a specific location, actual
time history record is required, which is not possible to have at each and every location. In the past, there were two
major disadvantages of actual time history approach. Firstly, this approach generates large amount of output data
which require computational effort to analyze and; secondly, the analysis is required to repeat for respective
earthquake motions. Since the response of structure depends upon its own dynamic properties and the frequency
content of ground, the seismic force cannot be calculated simply on the basis of maximum values of ground
acceleration. Hence, earthquake response spectrum method is the perfect tool for seismic analysis of structures. This
method is advantageous to calculate the forces in members and displacements, as it comprises the estimation of only
maximum values by employing smooth design spectra [13].
3 STADD.Pro V8i
STAAD or (STAAD.Pro) is a commercial software which is widely used for analysis and design of
structures. This software was initially launched at Yorba Linda, CA in 1997 by Research Engineers International
and later on it was sold to Bentley Systems in 2005. This software authenticates different codes based on steel,
timber, concrete, etc. This software can perform several analysis like geometric non-linear analysis, 1st order static
analysis, 2nd order p-delta analysis, Pushover analysis (Static-Non Linear Analysis), cable analysis, imperfection
analysis, buckling analysis; dynamic analysis like modal extraction, time history analysis and response spectrum
analysis.
769
3.1 Assumptions in the Analysis

For complete analysis of structure, the required matrix is generated on the basis of following assumptions;
1. The structure consists of beams, plates, and solid elements which are collectively joined, loaded and
reacted by reactions in specific direction at their nodes [15].
2. These beams and plates are acting as longitudinal members having somewhat doubly symmetrical cross
section along its overall length. Beams are subjected to a longitudinal axial force along with bending and torsion in
mutually perpendicular directions [15].
3. Equilibrium is maintained at each node and each node comprises of six degree of freedom system to
define the torsion and bending properties of any member in the generation of relevant matrices [15].
4. The required matrices are formed by using two types of co-ordinate systems, viz., local system and
global systems [15].
4 Model Provision
When a water tank is under static condition, water inside the tank applies static pressure on tank walls. This
pressure is called as hydrostatic pressure. But when a water tank is under seismic activity, horizontal acceleration is
exerted on both tank wall and water both, which produces additional hydrodynamic pressure on tank base and tank
wall. Now the complete mass of water will split Impulsive mass and Convective mass [16]. The term impulsive
mass is defined as the mass of water in lower portion of water tank, which behaves as if it is rigidly attached to tank
wall. This impulsive mass accelerates along with tank wall and a impulsive hydrodynamic pressure is applied on
tank base and tank wall. Now the remaining water mass in the upper portion of water tank undergoes waving effect
or sloshing effect. This sloshing effect is defined as periodic motion of free water surface in a half filled water tank.
If this water mass is permitted to slosh freely, then it will exert additional hydrodynamic pressure on tank base and
tank wall; and the mass of water associated with this pressure is called as convective mass [17].
A RC overhead tank can be glamorized into single lumped mass model or two mass model or as spring
mass model etc. [10]. Spring mass model of elevated water tank is shown in Fig.1, where mc represents convective
mass attached rigidly to tank wall by a spring with stiffness (Kc); and mi represents impulsive mass which is rigidly
attached to tank wall. For RC overhead tanks, the water tank model can also be idealized into two-mass model,
which subsists of two degrees of freedom system and it is closer to reality. Housner in 1963 has proposed the theory
of two-mass model for overhead water tank as shown in Fig.2. where, mi, mc, Kc, Ks etc. are the parameters of spring
mass model [18].
Fig. 1. Spring mass model of Elevated water tank (Source: IITK GSDMA, p. 21)
770
Fig. 2. Two mass idealization of elevated water tank (Source: IITK GSDMA, p. 21)
5 Problem Formulation
The present study is carried out to observe the seismic response of RC Overhead Water Tank with different
staging configuration system. The height of water container is 6 meter, which is supported by a staging system of 20
meter height (including the depth of foundation). The results obtained are compared for tank full and tank empty
condition with different staging configuration system for various parameters such as base shear, base moment storey
displacement and time period of vibration.
Table 1. Geometrical properties of water tank
Tank vessel property Tanks staging property
Dimension/ Dimension/
Parameter Parameter
Value Value
3
Capacity of the tank vessel (m ) 900 Height of staging (m) 20.00
Diameter of Tank (m) 16.00 Number of Columns 12
Free Board (m) 0.30 Size of Columns (m) 1.00 X 0.50
Height of Cylindrical Wall (m) 4.00 Size of Bracing (m) 0.40 X 0.40
Thickness of Cylindrical Wall (m) 0.20 Depth of Foundation (m) 2.00
Rise of Top Cone (m) 1.00
Thickness of Top Conical slab (m) 0.20 Seismic Parameters
Size of Top Ring Beam (m) 0.40 X 0.40 Parameter Valve
Rise of Bottom Conical Dome (m) 1.00 Type of Structure/Frame SMRF
Thickness of Conical Dome (m) 0.20 Seismic Zone Zone – III
Size of Bottom Ring Beam (m) 0.40 X 0.40 Type of Soil Hard Soil
Thickness of Bottom Slab (m) 0.30 Response Reduction Factor R = 5.0
Size of Circular Ring Beam (m) 0.50 X 0.50 Importance Factor I = 1.5
771
Table 2. Material Properties

E 27386 M.Pa
Concrete Fc 30 M.Pa
Weight of volume unit 25 kN/m3
E 2.1 x 105 M.Pa
Steel Fc 500 M.Pa and 415 M.Pa
Weight of volume unit 78.5 kN/m3
Desnity 10 kN/m3
Water
Bulk Module 2.10 G.Pa
The two mass model parameters derived by Housner depend upon the geometry of water tank, which are
slightly different from those given in ACI 350.3. These parameters are applicable for water containers with rigid
walls. Basically, RC overhead water tanks are having rigid wall; whereas steel tanks are having flexible wall. Since
the present study is carried out on RC overhead water tank, parameters corresponding to water tanks with rigid wall
are recommended and listed in table 3. These parameters are not depending upon flexibility of elastic pads between
base and wall. And it should be noted that, for a particular values of h/D ratio, the summation of impulsive and
convective water mass will not be equal to total mass of water. However, the difference is usually small (2 to 5 %)
and in our case it is found to be 4.12%. This difference is associated with presumptions and calculations made in the
derivation of these quantities. We should also know that for shallow water tanks, values of (hi*) and (hc*) can be
greater than the value of (h) due to chief contribution of hydrodynamic thrust on base.
Table 3. Parameters of Spring Mass Model

𝑫
𝒎𝒊 𝐭𝐚𝐧𝐡 𝟎. 𝟖𝟔𝟔
= 𝒉
𝒎 𝑫
𝟎. 𝟖𝟔𝟔
𝒉
𝒉𝒊
= 𝟎. 𝟑𝟕𝟓 For h/D ≤ 0.75
𝒉
𝒉𝒊 𝟎. 𝟎𝟗𝟑𝟕𝟓
= 𝟎. 𝟓𝟎 − For h/D > 0.75
𝒉 𝒉/𝑫
𝑫
𝒉𝒊∗ 𝟎. 𝟖𝟔𝟔
= 𝒉 𝟎. 𝟏𝟐𝟓 For h/D ≤ 1.33
𝒉 𝑫
𝟐 𝐭𝐚𝐧𝐡 𝟎. 𝟖𝟔𝟔
𝒉
𝒉𝒊∗
= 𝟎. 𝟒𝟓 For h/D > 1.33
𝒉
𝒉
𝒎𝒄 𝐭𝐚𝐧𝐡 𝟑. 𝟔𝟖
= 𝟎. 𝟐𝟑 𝑫
𝒎 𝒉
𝑫
𝒉
𝒎𝒄 𝐭𝐚𝐧𝐡 𝟑. 𝟔𝟖
= 𝟎. 𝟐𝟑 𝑫
𝒎 𝒉
𝑫
𝒉
𝒉𝒄 𝐜𝐨𝐬𝐡 𝟑. 𝟔𝟖 − 𝟏. 𝟎
=𝟏− 𝑫
𝒉 𝒉 𝒉
𝟑. 𝟔𝟖 𝐬𝐢𝐧𝐡 𝟑. 𝟔𝟖
𝑫 𝑫
𝒉
𝒉𝒄∗ 𝐜𝐨𝐬𝐡 𝟑. 𝟔𝟖 − 𝟐. 𝟎𝟏
= 𝟏− 𝑫
𝒉 𝒉 𝑫
𝟑. 𝟔𝟖 𝐜𝐨𝐬𝐡 𝟑. 𝟔𝟖
𝑫 𝒉
𝒎𝒈 𝟐
𝒉
𝑲𝒄 = 𝟎. 𝟖𝟑𝟔 𝐭𝐚𝐧𝐡 𝟑. 𝟔𝟖
𝒉 𝑫
772
6 Design Criteria
For the purpose of determining the design seismic forces, ACI 350.3, 2001 and IS 1893 (Part – 1) : 2002 is
prominently used. The value of horizontal design seismic force coefficient Ah for a structure should be calculated by
[19-21],
𝑍 I Sa
Ah = xR x (1)
2 g
Further, the design base shear (VB) along respective direction of the structure is divide into impulsive and
convective mode, above the top of footing is given as follows;
in impulsive mode;
VBi = (Ah)i × (mi + ms) × g (2)
and in convective mode is given by;
VBc = (Ah)c × mc × g (3)
where; (Ah)i and (Ah)c are the horizontal seismic forces coefficient in Impulsive and convective mode, mi and mc are
the seismic weight of the structure in impulsive and convective mode, ms is Mass of container and one-third mass of
staging and g is acceleration due to gravity
Finally, the total base shear VB, is obtained by summing the value of base shear in impulsive and
convective mode by using Square root of Sum of Squares (SRSS) rule and is given as follows;
VB = VBi 2 + VBc 2 (4)
7 Modeling of Elevated Water Tank

Finite element method is applied to the following overhead water tank systems. The behavior of concrete
material was considered as nonlinear. Even the tank water and supporting system is modeled by using finite element
as shown in figures from Fig. 3 to Fig. 5. More details are as follows;
(a) No vertical Bracing (b)Vertical Diagonal Bracing (c)Vertical X Bracing

Fig. 3. Intze Tank with no Lateral Bracing and different Vertical Bracing
Table 4. Results obtained from FEM based STADD Pro

with Vertical Diagonal
without Vertical Bracing with Vertical X Bracing
Intze tank with no Lateral Bracing
Bracing and Tank Tank Tank
Tank Full Tank Full Tank Full
Empty Empty Empty
Base Shear (KN) 783.87 452.42 809.48 478.03 835.09 503.65
Base Moment (KN.m) 15915.45 9299.47 16251.32 9544.1 16599.73 9867.58
Storey Displacement (mm) 63.599 35.688 8.277 4.837 5.816 4.682
Time Period of Vibration (s) 3.057 2.1741 1.0606 0.7921 0.9624 0.7338
773
No vertical Bracing Vertical Diagonal Bracing Vertical X Bracing

Fig. 4. Intze Tank with Lateral Cross Bracing and different Vertical Bracing

Intze tank with Lateral Cross Bracing
Empty Empty Empty
Base Shear (KN) 804.78 473.33 830.4 498.95 856.01 525.31
Base Moment (KN.m) 15501.84 8973.23 16547.07 9824.3 16879.73 10147.43
No vertical Bracing Vertical Diagonal Bracing Vertical X Bracing

Fig. 5. Intze Tank with Lateral Radial Bracing and different Vertical Bracing

Intze tank with Lateral Radial Bracing
Empty Empty Empty
Base Shear (KN) 822.35 490.91 847.97 516.52 873.57 542.14
Base Moment (KN.m) 15754.43 9133.78 16784.62 10049.8 17099.9 10342.83
774
On accomplishing the software analysis, the results are represented in table number 4 to 6. These results
can be observed precisely to get the following prominent observations;
1. The base shear in all the cases of water tanks increases in the range of 3 to 4 percent for tank with vertical
diagonal bracing under ‘tank full’ condition and the same increases between the range of 5 to 6 percent under
‘tank empty’ condition.
2. For the tanks with X – bracing the base shear increases in the range of 6 to 7 percent under ‘tank full’ condition
and 11 to 13 percent for ‘tank empty’ condition.
3. For the base moment parameter consistent results with base shear are obtained.
4. Significant improvement is observed in terms of storey displacement, it reduces drastically in the range of 7 to 8
times.
5. Similar behaviour of the tank is observed in terms of time period of vibration, it reduces in the neighbourhood
of 3 times.
The logical reasoning for this output can be put as follows;
The base shear increases because, with the addition of bracing members the mass of the structure increases
and hence base shear. The same thing holds good for base moment. The storey displacement decreases drastically
because, with the provision of bracing members the structure becomes stiff and results into reduction of storey
displacement which is beneficial for stability of structure during earthquake. Thus, the results obtained above states
that, Intze tank with Lateral Radial Bracing and Vertical X Bracing proves to be more seismically stable, since it
contributes maximum Base shear and comprehensive amount of storey displacement.
8 Procedure for Calculation of Seismic Responses by using IS: 1893-2002 (Part 2)

Complete analysis of Intze tank with lateral radial bracing and vertical X bracing as shown if Fig. 5(c)
consists following procedure,
1) Weight calculations
2) Modeling of liquid
3) Lateral stiffness of staging
4) Time period calculations
5) Design horizontal Seismic coefficient
6) Base shear and Base moment
7) Hydrodynamic pressure & Sloshing wave height
8) Analysis for tank empty condition
8.1 Weight Calculations

Mass of water tank consists of mass of roof slab, tank wall, mass of water, gallery (if any), mass of ring
beams, floor slab, floor beams and column with footing, etc. In figure b, ms is structural mass, which includes mass
of water container and one third mass of supporting system and mi and mc is mass of water in impulsive and
convective mode.
Table 7 Weight of different components of water tank
Component Weight (KN)
Top Dome 994.5
Top Ring Beam 312.57
Cylindrical Wall 1389.51
Bottom Ring Beam 312.57
Circular Ring Beam 230.85
Bottom Dome 483.16
Conical Dome 848.23
Water 8829
Columns and Footing 2261.25
Braces 1744.32
775
As per table number 7,

Weight of empty container = 994.5 + 312.57 + 1389.51 + 312.57 + 230.85 + 483.16 + 848.23
= 4571.39 KN
Weight of staging = 2261.25 + 1744.32
= 4005.57 KN
Weight of Structure = Weight of empty container + one third weight of staging
= 4571.39 + 1/3 × 4005.57
= 5906.58 KN
8.2 Parameters of Spring mass model

Total weight of water = 8,829 KN = 88,29,000 N.
Volume of water, V = 900 m3
Mass of water, m = 9,00,000 Kg.
Diameter of tank, D = 15.8 m.
For obtaining parameters of spring mass model, an equivalent circular container of same volume and
diameter equal to diameter of tank at top level of liquid will be considered.
Let h be the height of equivalent circular cylinder,
𝐷 2
π h = 900 (5)
2
15.8 2
h = 900 / [π × ]= 4.59 m
2
ℎ 4.59
Hence, = = 0.29.
𝐷 15.8
Therefore, from table no. 3:
mi / m = 0.334;
mi = 0.334 ×9,00,000 = 3,00,600 Kg
mc /m = 0.625;
mc = 0.625 × 9,00,000 = 5,62,284 Kg
hi / h = 0.375;
hi = 0.375 × 4.59 = 1.72 m
hi*/ h = 0.1753;
hi* = 0.1853 × 4.59 = 0.85 m
hc / h = 0.543;
hc = 0.543 × 4.59 = 2.49 m
hc*/ h = 1.28;
hc* = 1.28 × 4.59 = 5.87 m
About 33 % of total mass of water is excited in impulsive mode, whereas 63% of total mass of water
excites in convective mode. The summation of mass of water in impulsive and convective mode is 8,62,884 kg
which is 4 % less than the total mass of water.
8.3 Lateral Stiffness of Staging

The force attracted by the supporting system acting at CG of water tank so as to produce unit deflection is
called as Lateral Stiffness of supporting system. As per Section 8.1, centre of gravity of water tank is the combined
CG of empty water tank and mass of water in impulsive mode. FEM is employed to model the supporting system
and hence the displacement of CG of water tank due to an arbitrary force of 10 KN is obtained as 16.65 X 10-5 m.
Thus, lateral stiffness of supporting system,
10
Ks = = 593,489,710 Kg/m (06)
1.685 X 10 −8
776
8.4 Analytical calculations of Time Period of vibration

Time period of vibration is calculated as per IS 1893-2002 part 2, in case of tank full condition for intze
tank with lateral radial bracing and vertical X bracing;
Time Period of vibration of impulsive mode;
𝑚𝑖+ 𝑚𝑠
Ti = 2π = 0.245 s (07)
𝐾𝑠
Time Period of vibration of convective mode

𝐷
Tc = Cc = 4.67 s (08)
𝑔
8.5 Analytical calculations of Base Shear

Base shear at the bottom of staging in impulsive mode is given by equation (1) and (2);
VBi = Ahi × (mi + ms) × g
Z I Sa
Ahi =
2
× R
× g
i
Sa 1
g
i = Ti
= 4.081
Therefore;
Ahi = 0.0979
And, VBi = 866.95 KN
Base shear in convective mode is given by equation (1) and (3);
VBc = Ahc × mc × g
𝑍 I Sa
Ahc =
2
× R
× g c
𝑆𝑎 1
c = = 0.2141
𝑔 Tc
Therefore;
Ahc = 0.00514
And, VBc = 28.35 KN
Hence, the total base shear at the bottom of staging is given by equation (4),
V = 866.952 + 28.352
V = 867.41 KN
8.6 Sloshing wave height

dmax = Ahc × R × D/2 = 0.203 m (9)
9 Conclusions
Seismic analysis and behavior of RC overhead water tank have been successfully performed and prominent
results are observed for frame type of staging system. Usually, during strong ground motion most of the overhead
water tanks fall due to failure of their supporting system. Since, the tank supporting system is very vital and helpless
under lateral forces due to an earthquake, their susceptibility and functioning is important to understand. Hence,
modeling and statistical analysis of RC overhead water tank is performed by using STAAD PRO software.
Moreover, the seismic performance of RC overhead water tank with different staging configuration is observed by
using two mass model method. From previous literature it is observed that, RC overhead water tanks perform well
with frame staging system by using draft code IS: 1893 (Part-2) guidelines due to the following characteristics.
1. Base Shear increases as the level of bracing increases from no vertical bracing to vertical diagonal bracing and
from vertical diagonal bracing to vertical X – bracing under tank full and tank empty condition.
777
2. Base Shear also increases as the lateral bracing level increases from no lateral bracing to lateral cross bracing
and from lateral cross bracing to lateral radial bracing under both tank full and tank empty condition.
3. It is also observed that base shear is maximum for all intze tank under full condition than intze tank under
empty condition.
4. Base Shear is maximum for Intze tank with lateral radial bracing and vertical X – bracing among other bracing
systems and it is maximum for tank full condition than tank empty for all considered bracing.
5. Bending Moment at the top of foundation increases as level of bracing increases for all considered bracing types
and it is maximum for tank full condition than for tank empty condition.
6. In tank full and tank empty condition, base shear and base moment both have been increased by 23 % when the
staging configuration system was upgraded by one level.
7. Since total base shear and base moment in tank full condition are more than those in tank empty condition,
seismic design will be hold by tank full condition.
8. The percentage increase in maximum bending moment at the base of water tank for tank empty condition is
more than those for tank full condition. Hence, the empty water tanks are more vital during earthquake. The
reason behind such behavior is centre of gravity, that shifts slightly upward for empty tank and hence the
distance between seismic forces induced at the base to the centre of gravity of empty tank increases. Thus, the
maximum bending moment calculated at the base of empty water tank increases.
9. While calculating the lateral stiffness of tank supporting system, flexibility of bracing beams is clearly included
in IS: 1893-2002 (Part-2) draft code, which was not included in IS: 1893-1984 (Part-1).
10. The height of sloshing wave (dmax.) is proportional to inner diameter (D) of container and design seismic
coefficient for convective (Ahc) mode. Height of sloshing wave should be always less than the free board
provided, if height of sloshing wave is more than free board, then the sloshing waves produced will apply extra
pressure on the roof slab. This study is carried out in zone-III and it is observed that height of sloshing wave is
less than the free board provided.
11. Storey displacement in lateral direction goes on reducing as level of bracing increases. It is noted that, bracing
systems increases the stiffness of water tank supporting system, which significantly results into reduction of
storey displacement.
12. Whereas it is seen that, storey displacement is minimum for intze tank with lateral radial bracing and vertical X
– bracing.
13. Fundamental time period of vibration is more for tank full condition than tank empty condition for all
considered bracing types; and it is minimum for intze tank with lateral cross bracing and vertical X – bracing.
14. Among all bracing considered in this study Intze tank with Lateral Cross bracing and Vertical X Bracing
performs well during earthquake.
10 References
1. F. Omidinasab, H. Shakib.: Seismic Response Evaluation of the RC Elevated Water Tank with Fluid-Structure
Interaction and Earthquake Ensemble. KSCE Journal of Civil Engineering, 16(3) pp. 366-376 (2012). doi:
10.1007/s12205-011-1104-1.
2. Prashant A Bansode, V. P. Datye.: Seismic Analysis of Elevated Water Tank with Different Staging
Configuration. Journal of Geotechnical Studies Volume 3 Issue 1.
3. V. R. Panchal and D. P. Soni.: Seismic Behaviour of Isolated Fluid Storage Tanks: A-state-of-the-art Review.
KSCE Journal of Civil Engineering, 18(4) pp. 1097-1104 (2014). Copyright ⓒ2014, Korean Society of Civil
Engineers. doi: 10.1007/s12205-014-0153-7.
4. Soheil Soroushnia, Sh. Tavousi Tafreshi, F. Omidinasab , N. Beheshtian , Sajad Soroushnia.: Seismic
Performance of RC Elevated Water Tanks with Frame Staging and Exhibition Damage Pattern. The Twelfth
East Asia-Pacific Conference on Structural Engineering and Construction. Procedia Engineering, 14 pp . 3076–
3087 (2011). doi: 10.1016/j.proeng.2011.07.387.
778
5. T. W. Cooper.: A Study of the Performance of Petroleum Storage Tanks during Earthquakes, 1933-1995. US
National Institute of Standards and Technology, Gaithersburg, Maryland 20899 (1997).
6. A.M. Memari, M.M. Ahmadi, B. Rezaee.L.: Behaviour of reinforced concrete water towers during Manjil-
Roudbar earthquake of June 1990. Proceedings of the 10th world conference on earthquake engineering;
Madrid, Spain, vol. 9 pp. 4953-4960 (1992).
7. Jain S.K., Murty C.V.R., Arlekar J., Sinha R., Goyal A., and Jain C.K.: Some Observations on Engineering
Aspects of the Jabalpur Earthquake of 22 May 1997. EERI Special Earthquake Report, EERI Newsletter, Vol.
31, No. 8 (1997).
8. Jain S.K., Lettis W.R., Ballantyne D., Chaubey S.K., Dayal U., Goel R., Goyal A., Hengesh J., Malhotra P.,
Murty C.V.R., Narula P.L., Saikia C.K., Singh M.P., Vatsa K.: Preliminary Observations on the Origin and
Effects of the January 26, 2001 Bhuj (Gujarat, India) Earthquake. EERI Special Earthquake Report, EERI
Newsletter, Vol. 35, No. 4 (2001).
9. Swathi C. Naik, M.S. Bhandiwad.: Seismic Analysis and Optimization of a Rectangular Elevated Water Tank.
Bonfring International Journal of Man Machine Interface, Vol. 4, Special Issue (2016).
10. Dona Rose K J, Sreekumar M, Anumod A S.: A Study of Overhead Water Tanks Subjected to Dynamic Loads.
International Journal of Engineering Trends and Technology, Vol. 28, No. 7 (2015).
11. L. Kalani Sarokolayi, B. Navayi Neya, H. R. Tavakoli, J. Vaseghi Amiri.: Dynamic Analysis of Elevated Water
Storage Tanks due to Ground Motions Rotational and Translational Components. Arabian Journal of Science
and Engineering, Vol. 39, Issue 6, pp. 4391–4403 (2014). doi: 10.1007/s13369-014-1042-6.
12. Rebecca L. Johnson.: Theory of Response Spectrum Analysis. Department of Earth and Environmental Science,
New Mexico Tech.
13. National Programme on Technology Enhanced Learning. https://nptel.ac.in/courses/105101004/4. Last Visited
11:55 Am, 14/10/2018.
14. Mr. Manoj Nallanathel, Mr. B. Ramesh, A.B. Pavan Kumar Raju.: Stability Analysis of Concrete Gravity Dam
Using Staad Pro. International Journal of Pure and Applied Mathematics, Vol. 119, No. 17, pp. 297-310 (2018),
ISSN: 1314-3395.
15. Ali Ajwad, Liaqat A. Qureshi , M. Shoaib Ashraf, Abdul Mannan, Noshin Zaib and Sadia Kalsoom.:
Comparison of Computer Aided Analysis and Design of Multistorey Hospital Building using Etabs 9.5 and
Staad Pro.2005. Sci. Int. (Lahore), Vol. 27, No. 5, pp. 4083-4089 (2015), ISSN 1013-5316; CODEN: SINTE 8.
16. Syed Saif Uddin.: Seismic Analysis of Liquid Storage Tanks. International Journal of Advanced Trends in
Computer Science and Engineering, Vol. 2 , No.1, pp. 357-362 (2013), ISSN 2278-3091.
17. D. K. Landge1, Dr. P. B. Murnal.: Structural Assessment of Circular Overhead Water Tank Based on Shaft
Staging Subjected to Seismic Loading. International Journal of Emerging Technology and Advanced
Engineering, Vol. 4, Issue 6, pp. 145-151 (2014), ISSN 2250-2459.
18. Gloria Terenzi, Elena Rossi.: Seismic analysis and retrofit of the oldest R/C elevated water tank in Florence.
Bull Earthquake Engg. Vol. 16, pp. 3081-3102 (2018). doi: 10.1007/s10518-017-0306-4.
19. ACI 350.3, 2001.: Seismic design of liquid containing concrete structures. American Concrete Institute,
Farmington Hill, MI, USA.
20. IS Code 1893:2002 (Part II).: Criteria for Earthquake Resistant Design of structures. Bureau of Indian
Standards, New Delhi.
21. IITK-GSDMA.: Guidelines for seismic design of liquid storage tanks (2005).
779
Paper ID- 152
A Relational Approach to Quantify Sustainable Performance of Concrete

using Preference Selection Index (PSI) Method
A. Suchith Reddy1, P. Rathish Kumar2, P. Anand Raj3
1
Department of Civil Engineering, NIT Warangal, Telangana, India, 506004;
E-mail: asr.nitwarangal@gmail.com
2
E-mail: rateeshp@gmail.com
3
E-mail: a_raj_p@yahoo.co.in
Abstract
Selecting a preferred material among a pool of alternatives which perform as desired and promotes
sustainability is a crucial task. Evaluation of material performance considering subjectivity and
conflicting criteria is the major concern of construction stakeholders and is a Multi-Criteria Decision
problem. Investigating the sustainable performance of Supplementary Cementitious Material (SCM)
considering multi-dimensional and interdependent target-based criteria is the major concern in decision
making. The present study utilizes and explores the concept of Preference Selection Index (PSI) in
selecting the optimum sustainable cementitious material considering fresh and hardened properties,
durability and Life-Cycle Cost (LCC) aspects. A case study has been considered to illustrate the
approach by varying the percentages of Flyash (FA) and Ground Granulated Blast Furnace Slag
(GGBS) replacement dosages in Self Compacting Concrete (SCC). This simple and lucid approach will
facilitate designer and engineers in choosing the right cementitious material to achieve sustainability.
Keywords: Supplementary Cementitious Material; Sustainability; Preference Selection Index (PSI);

MCDM; Flyash; GGBS.
1 Introduction
One of the major challenges in engineering design is to choose the right material from a number
of alternatives available in the market. In the present scenario, development of new and innovative
material is necessitated to replace the traditional materials as they have either become extinct or causing
a huge burden on sustainability. Selection of the right material is important for functional use, design,
performance, and practicability [1]. Different materials may perform differently with respect to a single
attribute. To choose an optimal material and achieve desired results, the requirements should be robust
enough to achieve the required performance. For example, in the case of concrete, the cost should be
reasonable, should be durable and shall also obey sustainable design principles.
Concrete is the second maximum consumed material next to the water. It is expected that the
demand for concrete will rise up to 16 billion ton per year by 2050 [2]. The use of concrete cannot be
avoided due to its unique features and advantages, but the impacts caused by concrete can be reduced
by producing concrete by focusing more on sustainable aspects like use of right materials, methods, and
technologies [3]. The implementation of these sustainable aspects in the construction industry has been
awarded points by the building assessment tools like Building Research Establishment Assessment
Methodology (BREEAM) and Leadership in Energy and Environment Design (LEED), related to
sustainable material criteria [4]. These tools assess the level of sustainability achieved by buildings.
1.1 Use of Supplementary Cementitious Material (SCM)

Concrete is a multi-functional material and cement is an integral part of it. The structural
elements are designed to serve the desired application. This is achieved by using different types of
concrete. The use of different types of mineral admixtures or Supplementary Cementitious Materials
1
780
(SCM’s) with varying percentage of cement replacement, makes it difficult to find the optimum material
for the desired application. Use of various SCM’s like Flyash (FA), Ground Granulated Blast Furnace
Slag (GGBS), Metakaolin (M), Silica Fume (SF) etc., not only reduces the consumption of cement but
also reduces the carbon footprint and waste leading to promoting sustainability [5]. The annual world
production of GGBS and FA are 530 MT and 430 MT respectively. In India, it is around 215 MT in
2017 and expected to increase to 400 MT by 2020 [6].
1.2 Challenges of Concrete Optimization
Selection of a suitable material for the design of concrete elements involves various attributes
like physical properties, workability, strength, environmental performance, fire resistance, durability
aspects, cost etc. All these have to be considered concurrently instead of taking single attribute at a
time. In addition to this, a single attribute cannot judge the performance of concrete with distinctive
properties and satisfy the desired properties [7]. Hence, it is a challenge to select such a material which
fits into the attributes discussed above.
1.3 Preference Selection Index (PSI)
Decision making is complex and challenging when a wide range of alternatives and conflicting
criteria are involved. Methods available in assisting decision making taking into account the number of
criteria/attributes and their interrelationship are either cumbersome or difficult to understand while the
weight of criteria/attributes plays a crucial role in optimizing the alternatives. In evaluating concrete
performance, the towards sustainability, all the criteria involved are equally important. Preference
Selection Index (PSI) method is the right choice to suit the requirement of a decision problem. This
method is usually preferred when there is no requirement of individual criteria weights that have high
conflict in providing the relative weight of criteria/attributes [8], [9].
2 Need for the Study
To choose an optimal solution and achieve desired results, the requirements should be strong
enough to evaluate the performance. With the variation in replacement level of SCM’s, there will be a
reduction in the material and labor cost. Also, there will be a benefit from the environment and ecology
point of view. Of the various techniques to reduce the emissions and waste use of SCM’s like FA and
GGBS as partial replacement of cement in concrete in SCC is considered a good possibility [10].
Utilization of these materials not only reduces the total lifecycle cost but also benefits the environment
[11]. According to the studies conducted by Flower & Sanjayan, (2007), FA and GGBS are found to
reduce the greenhouse gases by 15% and 22% respectively in a typical concrete mix. Though FA and
GGBS may be slow in attaining mechanical strength, the replacement levels of these materials will
hence change the performance of concrete for various parametric measures. To assess any material
performance, it is necessary to understand the influencing criteria. Various criteria are utilized to assess
the performance of the concrete. Most of the studies available in the literature have considered criteria
related to environment and economic aspects and there is not much mentioned about the influence of
technical aspects to assess the quantitative concrete sustainability [12], [13]. Also, studies on strength,
thermal resistance characteristics and other durability criteria related to socio-economic aspects are
important [14]. The present study deals with the strength as well as service life criterion of concrete
mixes. To prioritize the mixes developed from a set of SCC/combination of FA and GGBS, the
sustainable performance is evaluated based on Preference Selection Index (PSI) method considering the
seven influencing criteria described earlier. The methodology thus developed will bring out both the
qualitative method of assessing and optimize the mixes on the basis of their performance.
3 Methodology
Assessment of concrete performance is based on numerous factors related to strength, durability
and service life. In addition to these, to incorporate sustainability aspects of concrete performance, the
Environment, Social, Economic and Technological aspects. The other important aspect which is now
2
781
gaining attention in achieving sustainable construction is the technological aspect [15] needs to be
considered, wherein the concept of 3R’s (Reuse, Recycle, Replace) can be integrated using new
innovative technologies and methodologies. Reuse and recycling of materials reduce the carbon dioxide
and comes under Technological aspect and reduction in CO2 is related to an Environmental aspect.
Also, preserving the virgin material and reducing cost relates to Social and Economic aspects. Thus,
the assessment of the performance of a certain material involves Quadra-Bottom Line (QBL) approach
incorporating Social, Environmental, Economic and Technological (SEET) aspects.
3.1 Performance evaluation Criteria for optimizing sustainable concrete
The performance of concrete is based on quantitative data considering fresh properties, hardened
properties, and durability properties. The present work proposes 7 influencing criteria for evaluating
the sustainable performance of concrete. Compressive Strength (CS), Segregation Ratio (SR), Binder
Intensity (BI), Carbonation (C), Chloride Diffusion Coefficient (CDC), Service Life (SL) and Life-
Cycle Cost (LCC). These are the evaluating criteria, which are interdependent on each other and the
alternative can be optimized using the PSI method. The reason behind considering criteria C, CDC, and
SL is to evaluate the resistance of concrete to degradation which is directly associated with the service
life of concrete.
3.2 Definitions:
3.2.1 Compressive Strength (CS)
The load-bearing capacity of any type of concrete is assessed based on compressive strength. It is a
primary criterion to compare concretes made of different recipes, to assess the hardened properties.
3.2.2 Segregation Ratio (SR)
The Segregation Ratio is one of the best methods to evaluate the stability of self-compacting concrete
under fresh properties. According to Wet Sieving Stability Test (GTM Screen Stability Test), SR is the
ratio of the mass of concrete poured onto a 5mm sieve of 350mm diameter to the mass of mortar in the
sieve pan [16]. As per ERNARC specifications [17], the range of segregation ratio should be in between
5 and 15% for acceptable segregation resistance.
3.2.3 Binder Intensity (BI)
To reduce the CO2 emissions from cement production, the amount of utilization has to be reduced.
Incorporating SCM’s by replacing cement in concrete shall meet the potential demand for cement
utilization and minimize waste. Use of mineral admixtures in concrete contributes to sustainable
construction. Reduction in emissions will depend upon on the amount of replacement of SCM’s in
concrete, at the same time the concrete should perform as desired. Binder Intensity is the measure of
the quantity of binder (Kg/m3) required to produce strength of 1 MPa concrete at a given age [18]. It
expresses the binder significance for developing the strength of concrete for various replacement levels.
3.2.4 Carbonation
A concrete structure should withstand and maintain the expected performance during its service life.
The durability of concrete can be increased by avoiding the entry of aggressive elements into the
concrete. Carbonation is the phenomenon of the reaction of calcium hydroxide present in the concrete
with atmospheric carbonate ions to form calcium carbonate, due to which the pH value concrete
decreases, further leading to steel de-passivation and corrosion. The main factor for carbonation to occur
is a porous matrix, which directly relates to the water to cement ratio [14].
3.2.5 Chloride Diffusion Coefficient
It is a very important property of concrete that affects the durability and service life. The diffusion of
chloride is the main reason for the deterioration of concrete leading to corrosion of reinforcement. The
reason for the penetration of chlorides is due to diffusion and capillary absorption, which is governed
by porosity and permeability of concrete [19]. Introduction of SCM’s will reduce the chloride ingress
by dropping the number of chlorides required to initialize the process.
3
782
3.2.6 Service Life

The concrete that is able to maintain the expected performance during its service life is a durable
concrete. During the service life phase of structure, the durability performance of concrete plays a major
role in sustainability. Enhancing the service life enables a reduction in energy consumption and improve
occupant health leading to sustainable construction [20].
3.2.7 Life- Cycle Cost
The Lifecycle cost is considered one of the major criteria in a decision problem [21]. This includes the
cost of material and manpower, transportation, manufacturing/production for each cubic meter of
concrete. The present study adopted mix proportions from Zhao et al. [16] and Service Life estimations
of Life 365 from Vieira et al. [22] and explored the application of PSI method for assessing the
performance of FA and GGBS based SCC for 20%, 30% and 40% replacements of cement. The mix
proportions of FA series and GGBS series concretes are shown in Table 1.
3.3 Procedure involved in PSI method
The following steps are involved in PSI method for solving the decision problem.
Step 1: Identification of goal, material alternatives (Ai), selection criteria (Cj) and ( ) is the
performance evaluation of alternative with respect to criteria to form Decision Matrix.
1 2 ….
⋯
⋯
=[ ]=
⋮ ⋮ ⋮ ⋮
..
…
where is the performance measure i = 1,2, 3, .., n and j = 1,2,3,…m

Step 2: Normalization of Decision. Conversion of different data units to the non-dimensional unit.
Matrix is transformed to values between 0 and 1.
If the criteria are beneficial, then bigger the better = (For Benefit) (1)
If the criteria are non-beneficial, then smaller is better = (For Cost) (2)
Step 3: By the principle of sample variance analogy, the computation of Preference Variation Value
(PVj) can be obtained as
=∑ − 2
(3)
∑
where =
is mean of normalized values of jth criteria.
Step 4: The computation of Overall Preference Value ( j) for jth criteria is obtained from deviation
(∅ ) ∅ =1− (4)
∅
Normalization of Overall Preference Value: =∑ ∅
(∑ = 1) (5)
Step 5: Using the Equation 6, the Preference Selection Index ( ) for each alternative (Ai) is obtained
=∑ ∗ (6)
The higher the value of Preference Selection Index (Ii), better is the priority of selection for a sustainable
concrete.
4
783
4 Result and Discussion (Illustration of an example)

It can be noted from the results in Table 1 that for identical properties, GGBS based SCC mixes
performed better than the FA-based mixes in terms of strength and service life (Durability) criteria. It
also indicates the importance of all the four criteria viz: Social, Environment, Economic and
Technological sustainability. The selected criteria consider properties from fresh to hardened properties
and durability in terms of service life. Table 1 shows the mix proportions and parametric measures of
SCC with varying Flyash (FA) and Ground Granulated Blast Furnace Slag (GGBS) percentages [16],
[22].
Table 1 Mix proportions and parametric measures of SCC
Binder Materials (kg/m3 ) Chloride

Super- Service
Fine Coarse Compressive diffusion
plasticizer Water Life
Mix Aggregates Aggregates Strength Segregation coeffi
OPC Flyash GGBS dosage (kg/m3) (in
(kg/m3) (kg/m3) (MPa) Ratio (mm^2/s
(ml) years)
x 106)
C- Mix 460 0 0 693.81 1085.2 5.76 161 75.2 9.2 5.16 15.7
GGBS 20 368 0 92 693.81 1085.2 5.76 161 77.3 9.4 3.08 33.1
GGBS 30 322 0 138 693.81 1085.2 5.76 161 76.4 9.7 2.21 54.8
GGBS 40 276 0 184 693.81 1085.2 5.76 161 75.1 9.8 1.42 105.2
FA 20 368 92 0 693.81 1085.2 5.76 161 74.5 9.3 3.61 35.9
FA 30 322 138 0 693.81 1085.2 5.76 161 73.7 9.5 2.57 66.2
FA 40 276 184 0 693.81 1085.2 5.76 161 71.5 9.7 4.75 57.2
Based on the PSI values obtained from the procedure stated in section 3.3 and results from Table 1 for
all the alternatives, it was observed that the binder intensity of FA and GGBS based SCC increased
linearly with increase in the percentage of replacement corresponding to the strength. FA series-based
SCC mix has binder intensity values of 1.23, 1.87 and 2.57, which are higher compared to GGBS series-
based SCC mix values of 1.19, 1.81 and 2.45 respectively. As the binder intensity values of GGBS
series decreases, the PSI values are found to be decreasing, but in the case of FA, the PSI value of FA
30 is higher than FA 40 (Fig 1). This signifies the importance of binder intensity at the higher
replacement of FA. Though the binder intensity of FA 40 is observed to be higher among FA series-
based SCC, the overall PSI value is noted to be less than FA 30. This is due to the influence of other
criteria which results in PSI considering all criteria. The optimum performance alternative (SCC mix)
with respect to binder intensity is found to be GGBS 40 based SCC. With the increase in replacement
levels of FA and GGBS in SCC, the carbonation depth has enhanced compared to conventional SCC
mixes. The main reason for these findings could be based on two factors: Decrease in calcium hydroxide
content with an increase in the addition of FA and GGBS and an increase in porosity of the concrete
[16]. It was also noted that FA series-based SCC is predominant in carbonation than the GGBS series-
based SCC for the same replacement level (Fig 2). Similarly, the segregation resistance is slightly more
than the conventional SCC. However, all the mixes are in the range of 5% and 15% which is considered
to be satisfactory as per EFNARC specifications [17].
It was also observed with increase in carbonation depth, the PSI values also increase, which is not a
good sign for concrete performance. This unexpected trend of carbonation and PSI may be due to higher
porosity of the concrete. Even for concretes made with supplementary cementitious materials exhibited
higher carbonation. In particular, concretes with higher percentages of FA are more susceptible to
carbonation [23]. The optimum alternative (SCC mix) with respect to carbonation is GGBS 40 based
SCC which achieved a PSI value of 0.893. From Fig 3, it is also clear that as the replacement levels of
FA and GGBS increased, the Segregation Ratio (SR) also increased.
5
784
3.0 8 7.35
2.57
Carbonation Depth (mm)

2.45
2.5 7 6.25
5.85
1.87 6 5.25
Binder Index
2.0 1.81
5
3.65 3.75
1.5 1.19 1.23 4
1.0 3
2 1.35
0.5
1
0.00
0.0 0
C- Mix GGBS GGBS GGBS FA 20 FA 30 FA 40 C- Mix GGBS GGBS GGBS FA 20 FA 30 FA 40
20 30 40 20 30 40
Figure 1 Binder Intensity with respect to FA and GGBS Figure 2 Carbonation depth of FA and GGBS series-
series-based SCC based SCC
The conventional SCC mix has the least SR of 9.2 and corresponding PSI value of 0.685, whereas for
FA 40 and GGBS 40 based SCC the SR is 9.7 and 9.8 and corresponding PSI values are 0.759 and
0.893 respectively. Thus, the PSI method is adopted for minimizing the impacts and maximizing the
performance of concrete considering all the factors.
Compressive strength is one of the key characteristic properties of concrete for assessing the
technological aspect of sustainability. The results of the present work show that the strength has
decreased in FA and GGBS based SCC (Fig 4). Though the FA and GGBS series-based SCC has
attained lesser strengths than conventional SCC, there is an increase in the durability. It can be noted
that the PSI method balances the strength and durability aspects in optimizing the sustainable
alternative. Among the various alternatives, GGBS 40 based SCC possessed the highest value of PSI
with 0.893.
9.9 9.8 78 77.3

9.8
Compressive Strength (MPa)
9.7 9.7 77 76.4

9.7 76 75.2
Segregation Ratio
75.1
9.6 9.5 75 74.5
9.5 73.7
9.4 74
9.4 9.3 73
9.3 9.2 72 71.5
9.2 71
9.1 70
9 69
8.9 68
GGBS
GGBS
GGBS
Mix
FA 20
FA 30
FA 40
Mix
FA 20
FA 30
FA 40
GGB
GGB
GGB
C-
S 20
S 30
S 40
C-
20
30
40
Figure 3 Segregation ratio FA and GGBS series-based Figure 4 Compressive Strength w.r.t. FA and GGBS
SCC series-based SCC
With the use of FA and GGBS in SCC, the chloride penetration reduced compared to conventional SCC
mix. It was noted that with the increase in replacement levels, there is a decrease in the Chloride
Diffusion Coefficient (CDC). At a given replacement level, GGBS series-based SCC attained a lower
CDC than those of FA series. However, FA series-based SCC had less CDC than the conventional SCC.
Also, with the drop in CDC values, there was an increase in PSI values (Fig 5), which further contributes
to the sustainable performance of concrete. The least value of CDC of 1.42 is for GGBS 40 based SCC
which has the highest PSI value of 0.893. In addition to the mechanical properties, the durability of
6
785
concrete affects the overall performance of concrete. The service life of FA and GGBS based SCC was
estimated using Life 365 software [22]. From the results (Fig 6), it can be noticed that with the increase
in replacement, the service life of GGBS series-based SCC increased linearly, but in the case of FA
series-based SCC, it is not linear. Also, the service life of FA 40 based SCC is 57.2 years which is less
than the FA 30 based SCC of 66.2 years. However, FA 40 based SCC has attained higher service life
than the conventional SCC. The optimum sustainable alternative is considered as GGBS 40 based SCC
with an SL of 105.2 years and a PSI value of 0.893.
120 6
105.2 5.16
4.75
100 5
Service Life (in years)
Chloride Diffusion Coefficient (mm2/s

80 4 3.61
66.2 3.08
54.8 57.2
60 3 2.57
2.21
33.1 35.9
40 2 1.42
x106)
15.7
20 1
0 0
C- Mix
GGBS 20
GGBS 30
GGBS 40
FA 20
FA 30
FA 40
C- Mix
GGBS
GGBS
GGBS
FA 20
FA 30
FA 40
20
30
40
Figure 6 Service Life of FA and GGBS series-based SCC Figure 5 CDC of FA and GGBS series-based SCC
The Life Cycle Cost (LCC) is measured per cubic meter of concrete in terms of material cost, labor
cost, transportation cost, manufacturing/production cost. The LCC is based on location and time
constraint and shall vary by location and time, and mainly depend on the various factors like availability,
demand, and supply. From Fig 7, LCC of conventional SCC is observed to be high compared to FA and
GGBS series-based SCC.
12000
10000
Life Cycle Cost
8000
6000
4000
2000
0
C- Mix GGBS 20 GGBS 30 GGBS 40 FA 20 FA 30 FA 40
Figure 7 Life Cycle Cost w.r.t FA and GGBS series-based SCC
In the present study, the prevailing rates of material, labor and transportation, and manufacturing in
South India have been considered. With the change in the percentage replacement of cement by FA and
GGBS, the cost of material changes. Though the least LCC is attained by FA 40 based SCC, the
optimum PSI value of 0.893 was achieved by GGBS 40 based SCC. Fig 8 represents the normalized
values of the seven criteria with respect to the seven alternatives, where GGBS 40 based SCC has come
out as an optimized mix with a PSI value of 0.893. The following steps are involved in determining
the PSI and arrive at the ranking to optimize the material.
Step 1: The possible alternatives (Ai), and selection criteria (Cj) for the problem is decided. In
the present work, seven criteria have been considered for optimizing the SCC mix. Table 2
shows the decision matrix outlining the various material alternatives and criteria.
7
786
Table 2 Decision Matrix
CS SL (in LCC CDC

Mix SR BI C (mm)
(MPa) years) (INR) (mm^2/s x 106 )
C-MIX 9.2 75.2 0.00 15.7 9750 5.16 1.35
GGBS 20 9.4 77.3 1.19 33.1 8750 3.08 3.65
GGBS 30 9.7 76.4 1.81 54.8 8410 2.21 5.25
GGBS 40 9.8 75.1 2.45 105.2 7950 1.42 6.25
FA 20 9.3 74.5 1.23 35.9 8250 3.61 3.75
FA 30 9.5 73.7 1.87 66.2 7860 2.57 5.85
FA 40 9.7 71.5 2.57 57.2 7240 4.75 7.35
Note: SR-Segregation Ratio, CS- Compressive Strength, BI-Binder Intensity, SL-Service Life LCC-Life Cycle
Cost, CDC-Chloride Diffusion Coefficient, C- Carbonation Depth
A multicriteria decision-making process is required to optimize the mix (alternative) considering the
criteria related to fresh properties, strength and durability properties.
Step 2: Normalization of decision matrix to form a non-dimensional unit. The criterion SR, CS, BI,
SL are beneficial criteria (To be maximized) and LCC, C, and CDC are cost criteria (To be minimized)
using Equations (1 and 2). The same is shown in Table 3.
Table 3 Normalized Decision Matrix
Mix SR CS BI SL LCC CDC C
C- Mix 0.94 0.97 0.00 0.15 0.74 0.28 1.00
GGBS 20 0.96 1.00 0.46 0.31 0.83 0.46 0.37
GGBS 30 0.99 0.99 0.70 0.52 0.86 0.64 0.26
GGBS 40 1.00 0.97 0.95 1.00 0.91 1.00 0.22
FA 20 0.95 0.96 0.48 0.34 0.88 0.39 0.36
FA 30 0.97 0.95 0.73 0.63 0.92 0.55 0.23
FA 40 0.99 0.92 1.00 0.54 1.00 0.30 0.18
Step 3: Calculation of the Preference Variation ( ) value using Equation 3.
Table 4 Preference Variation value Matrix
SR CS BI SL LCC CDC C
C- Mix 0.001 0.000 0.382 0.123 0.018 0.059 0.392
GGBS 20 0.000 0.001 0.024 0.034 0.002 0.003 0.000
GGBS 30 0.000 0.000 0.007 0.000 0.000 0.016 0.014
GGBS 40 0.001 0.000 0.112 0.250 0.001 0.233 0.025
FA 20 0.000 0.000 0.019 0.025 0.000 0.015 0.000
FA 30 0.000 0.000 0.012 0.017 0.002 0.001 0.020
FA 40 0.000 0.002 0.146 0.002 0.015 0.048 0.036
Step 4: Computation of the Overall Preference (∅ ) value using Equation 4 (Table 5)
Table 5 Deviation and Overall Preference Values of criteria
SR CS BI SL LCC CDC C
Deviation (∅ ) 0.997 0.996 0.298 0.548 0.961 0.625 0.513
Normalized Overall
0.2018 0.2018 0.0604 0.1110 0.1946 0.1266 0.1038
Preference Value ( )
Step 5: Computation of Preference Selection Index for each alternative using Equation 6.
The alternatives are ranked based on the Preference Selection Index ( ) value, Higher the value, better
is the rank of a certain mix. The details of the values obtained from step 5 is shown in Table 6.
Table 6 PSI and Ranking of FA and GGBS based SCC
Mix C- Mix GGBS 20 GGBS 30 GGBS 40 FA 20 FA 30 FA 40
PSI ( ) 0.6855 0.7160 0.7750 0.8926 0.7108 0.7749 0.7587

Sustainable
0.768 0.802 0.868 1.000 0.796 0.868 0.850
Performance
Rank 7.00 5.00 3.00 1.00 6.00 2.00 4.00
8
787
It can be concluded from Table 6 that the sustainable performance of GGBS series-based SCC is better
than the FA series-based SCC and conventional SCC. The merit of the PSI method is that it balances
the criteria by minimizing the negative impacts and maximizing the performance of concrete from a
series of alternative cementitious material alternatives and optimizes a sustainable alternative.
0.300
0.250
0.200
0.150
0.100
0.050
0.000
PSI SR CS BI SL LCC CDC C
C-Mix GGBS 20 GGBS 30 GGBS 40 FA 20 FA 30 FA 40
Figure 8 Normalized values of criteria with respect to PSI
5 Conclusions
Assessing the sustainable performance of concrete involving multi-dimensional and interdependent

target-based criteria is a decision-making problem. It is difficult to understand and assign the weights
to conflicting and relative criteria. The present paper highlights the approach in evaluating and
optimizing the sustainable performance of concrete by utilizing Preference Selection Index (PSI)
method, without the requirement of criteria weights. Keeping in view Social, Environmental, Economic
and Technological aspects of sustainability, the study proposed seven influencing criteria for evaluating
sustainable performance of concrete viz., Compressive Strength (CS), Segregation Ratio (SR), Binder
Intensity (BI), Carbonation (C), Chloride Diffusion Coefficient (CDC), Service Life (SL) and Life-
Cycle Cost (LCC), which are conflicting and interdependent on each other. In particular, the approach
not only considers the strength and durability but also considers criteria like service life of concrete and
Life-cycle Cost, to prioritize the concrete based on performance and cost.
Specific conclusion:
 The study explored the application of PSI through a case study in optimizing the alternative for
Self-Compacting Concrete (SCC) mix by varying Flyash (FA) and Ground Granulated Blast
Furnace Slag (GGBS) percentages (20%, 30%, and 40%).
 The alternative GGBS 40% based SCC has achieved the highest PSI value of 0.893 and is ranked
first whereas SCC conventional mix has achieved 0.685 and is ranked last among the seven
alternatives and this turns out to be 30.2% less sustainable.
 The alternative FA30 based SCC is observed to be better than FA 40 based SCC in terms of
segregation ratio, compressive strength, carbonation, chloride diffusion coefficient, and service life
and secured the second rank with PSI value of 0.774.
 The standard deviation of PSI values was found to be 0.068 and the PSI of GGBS 40 (i.e.,0.893) is
higher than the mean PSI value of 0. 759. The GGBS 30 and FA 30 has almost attained similar PSI
values of 0.7749 and 0.7748 respectively.
9
788
 Based on the seven criteria, the sustainable performance priority of FA and GGBS series-based
SCC are optimized as GGBS 40 > FA 30 > GGBS 30 > FA 40 > GGBS 20 > FA 20 > C-Mix.
The developed approach facilitated to consider the characteristic properties of concrete from fresh to
hardened and durability aspects in quantifying the sustainable performance of concrete, which will help
designer and engineers in choosing a right cementitious material for sustainable performance.
References
[1] Rao, R. Venkata. "A decision-making methodology for material selection using an improved compromise ranking
method." Materials & Design 29, no. 10 (2008): 1949-1954.
[2] Chen, Pi-Cheng, Kun-Hsing Liu, and Hwong-wen Ma. "Resource and waste-stream modeling and visualization as
decision support tools for sustainable materials management." Journal of Cleaner Production 150 (2017): 16-25.
[3] Proske, Tilo, Moien Rezvani, Sebastian Palm, Christoph Müller, and Carl-Alexander Graubner. "Concretes made of
efficient multi-composite cement with slag and limestone." Cement and Concrete Composites 89 (2018): 107-119.
[4] Park, Jiyoung, Jungwon Yoon, and Kwang-Hyun Kim. "Critical review of the material criteria of building
sustainability assessment tools." Sustainability 9, no. 2 (2017): 186.
[5] Tian, Guangdong, Honghao Zhang, Yixiong Feng, Danqi Wang, Yong Peng, and Hongfei Jia. "Green decoration
materials selection under interior environment characteristics: A grey-correlation based hybrid MCDM
method." Renewable and Sustainable Energy Reviews 81 (2018): 682-692.
[6] Surabi, “Fly ash in India : Generation vis-à-vis Utilization and Global perspective,” Int. J. Appl. Chem., vol. 13, no.
1, pp. 29–52, 2017.
[7] Ahmad, Tayyab, and Muhammad Jamaluddin Thaheem. "Developing a residential building-related social
sustainability assessment framework and its implications for BIM." Sustainable cities and society 28 (2017): 1-15.
[8] Maniya, Kalpesh, and M. G. Bhatt. "A selection of material using a novel type decision-making method: Preference
selection index method." Materials & Design 31, no. 4 (2010): 1785-1789.
[9] A. Suchith Reddy, P. Anand Raj, and P. Rathish Kumar, “Developing a Sustainable Building Assessment Tool
(SBAT) for Developing Countries—Case of India”, Urbanization Challenges in Emerging Economies: Resilience and
Sustainability of Infrastructure. 2018. https://doi.org/10.1061/9780784482032.015
[10] Morbi, A., S. Cangiano, and E. Borgarello. "Cement-Based Materials for Sustainable Development." In Second
International Conference on Sustainable Construction Materials and Technologies, Coventry University and the
University of Wisconsin Milwaukee Center for by-products Utilization. 2010.
[11] Ondova, Marcela, Nadezda Stevulova, and Edita Zelenakova. "Energy savings and environmental benefits of fly ash
utilization as a partial cement replacement in the process of pavement building." Chemical Engineering
Transactions 25 (2011): 297-302.
[12] Flower, David JM, and Jay G. Sanjayan. "Green house gas emissions due to concrete manufacture." The International
Journal of life cycle assessment 12, no. 5 (2007): 282.
[13] A Suchith Reddy, P. Rathish Kumar, P. Anand Raj, “Sustainable performance indicators in the built environment for
developing countries a.suchith reddy,” in Second International Conference on Advances in Concrete, Structural and
Geotechnical Engineering (ACSGE 2018), 2018.
[14] Rao, Ravipudi Venkata. Decision making in the manufacturing environment: using graph theory and fuzzy multiple
attribute decision making methods. Springer Science & Business Media, 2007.
[15] Khatib, Jamal, ed. Sustainability of construction materials. Woodhead Publishing, 2016.
[16] Illankoon, IM Chethana S., Vivian WY Tam, and Khoa N. Le. "Environmental, economic, and social parameters in
international green building rating tools." Journal of Professional Issues in Engineering Education and Practice143,
no. 2 (2016): 05016010.
[17] Zhao, Hui, Wei Sun, Xiaoming Wu, and Bo Gao. "The properties of the self-compacting concrete with fly ash and
ground granulated blast furnace slag mineral admixtures." Journal of Cleaner Production 95 (2015): 66-74.
[18] EFNARC, “Specification and Guidelines for Self-Compacting Concrete,” Rep. from EFNARC, vol. 44, no. February,
p. 32, 2002.
[19] Elchalakani, Mohamed, Tarek Aly, and Emad Abu-Aisheh. "Sustainable concrete with high volume GGBFS to build
Masdar City in the UAE." Case Studies in Construction Materials 1 (2014): 10-24.
[20] Widyawati, Ratna, Jun Takahashi, Hisao Emoto, and Ayaho Miyamoto. "Remaining life prediction of an aged bridge
based on concrete core test." Procedia Engineering 95 (2014): 88-99.
[21] J. R. Clifton and D. J. Naus, “Service- Life Prediction Reported by ACI Committee 365. ACI 365.1R-00,” p. 44,
2000.
[22] Srivastava, Raghav. "Sustainable Development: Towards A New Paradigm For India." European Journal of
Sustainable Development 5, no. 4 (2016): 51-60.
[23] Vieira, Darli Rodrigues, João Luiz Calmon, Robson Zulcão, and Felipe Zanellato Coelho. "Consideration of strength
and service life in cradle-to-gate life cycle assessment of self-compacting concrete in a maritime area: a study in the
Brazilian context." Environment, Development, and Sustainability (2017): 1-23.
[24] Faella, Ciro, Carmine Lima, Enzo Martinelli, Marco Pepe, and Roberto Realfonzo. "Mechanical and durability
performance of sustainable structural concretes an experimental study." Cement and Concrete Composites 71 (2016):
85-96.
10
789
Paper ID- 154
Quantifying uncertainty in structural responses of polymer sandwich

composites: A comparative analysis of neural networks
R. R. Kumara*, T. Mukhopadhyayb, K. M. Pandeya, S. Deya

a
Department of Mechanical Engineering, National Institute of Technology Silchar, India
b
Department of Engineering Science, University of Oxford, Oxford, United Kingdom
*Corresponding author’s e-mail: ravinits2014@gmail.com
Abstract
The manufacturing and fabrication of complex polymer sandwich composite plates involve
various processes and parameters, and the lack of control over them causes uncertain system
parameters. It is essential to consider randomness in varying parameters to analyse polymer
sandwich composite plates. The present study portrays uncertainty quantification in structural
responses of polymer sandwich composite plates using the surrogate model. The comparative
predictability of artificial neural network (ANN) and polynomial neural network (PNN) for
uncertain structural responses (such as natural frequencies) of the sandwich plate is
presented. The proposed ANN, as well as PNN algorithm, is found to be convergent for
uncertain vibration response, which is verified and validated with direct Monte Carlo
simulation (MCS). The predictability of PNN is observed to be more efficient than that of
ANN. Typical material properties, skew angle, fibre orientation angle, number of laminate,
core thickness is randomly varied to quantify the uncertainties. Both the PNN and ANN saves
the computational time and cost significantly compared to traditional MCS techniques.
Keywords: Polymer sandwich plate, Artificial neural network, Polynomial neural network,
Monte Carlo simulation, Randomness.
Introduction
Sandwich plates are widely used in cost sensitive than that of weight sensitive applications
with respect to monolithic panels [1]. The cost-effective sandwich plate requires sandwich
core of low cost material which exhibits better weight sensitiveness as well. The development
and automation in production processes make possible the production of low cost sandwich
panels. The inevitable effect of uncertainty can affect the vibrational response of such
structures significantly. Therefore, it is essential to consider the material and geometric
uncertainties in order to include the variability and inaccuracies occurring during
manufacturing and operating conditions. In general, a surrogate based uncertainty
quantification approach [2-13] can be adopted in case of problems where efficient solutions
[14-21] are not available. In this paper, we aim to present a comparative performance of
artificial neural network [22-24] and polynomial neural network [25] for uncertainty
quantification of sandwich plates.
Governing equation for natural frequency
For structural deformation, strain-displacement relation [26] can be expressed as
 U ( ) V ( ) W ( ) U ( ) V ( ) U ( ) W ( ) V ( ) W ( ) 
{ ( )}     
x 
(1)
 x y z x y z x z
i.e. { ( )}  [ A( )]{ ( )}

790
where, [A] is unit step function. The generalized displacement vector at any point can be
expressed as
n
{S ( )}    i ( ) Si ( ) (2)
k 1
where, {s}  {u0v0 w0 x yuuvu wuul vl wl }T By using equation (1), strain vector can be expressed
as
{ ( )}  [ A( )]{S ( )} (3)
where, [A] is strain-displacement matrix.

Hamilton’s principle based dynamic equilibrium equation (for free vibration analysis) can be
expressed as
[ R( )]{S }  2[M ( )]{S } (4)
where, [ R ( )] = Stochastic free vibration frequency and

[ M ( )] = Global mass matrix which can be expressed as
nu  nl
[ M ( )]     ( )[ N ] [ J ] [ N ][ J ]dxdydz   [ N ] [K ( )][ N ]dxdy
k 1
k
T T T
(5)
where,  k ( ) is random mass density of k-th order, [J] is of the order of 3X11 containing Z
terms and some constant and [N] is shape function matrix. The stiffness matrix [ K ( )] can
also be expressed as
nu  nl
[ K ( )]  
kl
k ( )[ J ]T [ J ]dz (6)
The natural frequencies can be obtained by solving an eigenvalue problem of the mass and
stiffness matrices described above [26].
Artificial neural network
An artificial neuron, similar to the biological neuron is the basic processing unit of ANN
which receives input from other sources and outputs the final result by combining and
performing a non-linear operation on the result [23-24]. ANN is having good predictability to
recognize and reproduce the cause-effect relationships. This is performed by training multiple
input-output systems which enables ANN an efficient surrogate model to represent the
complex system. It consists of the Input layer, hidden layer and output layer as depicted in
figure 1. The performance of ANN is statistically measured by the root mean squared error
(RMSE), the coefficient of determination (R2) and the absolute average deviation (AAD),
which can be expressed as follows
1 m
RMSE =  ( X j  X jd )2
m j 1
(7)
791

m
j 1
( X j  X jd ) 2
2
R =1- (8)

m
j 1
( X jd  X n ) 2
 1 m ( X j  X jd ) 
AAD =     100 (9)
 m j 1 X jd 
Where m = Number of points,

Xj = Predicted value,
Xjd = Actual value and
Xn = Average of the actual values
Fig. 1 Architecture of ANN
Polynomial neural network
An input-output data in optimal PNN structure [24-25, 27] can be expressed as

(Xi, Yi) = (x1i ,x2i ,x3i ,...xni, yi) (10)
where i = 1,2,3 … n
Let Y is the output of the system which required to model. Polynomial regression equation is
computed for each pair of input variable xi and xj and output Y
Y = A+ Bxi + Cxj + Dx2i + Ex2j + Fxixj (11)
where i, j = 1, 2, 3...n
where co-efficient of polynomial equation is represented as A, B, C, D, E, F
The estimated output Yî can be expressed as
n n n n n n
Yî = fˆ ( x1 , x2 , x3 ,....xn )  A0   Bi xi    Cij xi x j    D x x j xk  ...
ijk i (12)
i 1 i 1 j 1 i 1 j 1 k 1
where i,j,k = 1,2,3...n

X (x1, x2,...,xn )= input variable vector
P(A0,Bi,Cij ,Dijk ,...) = vector of coefficients or weight of lvakhnenko polynomials.
792
The components of the input vector X can be independent variables, functional forms or finite
difference terms. This algorithm provides model structure and model system output on the
value of most noteworthy inputs of the system. The basic taxonomy for the architecture of
PNN structure is given in figure 2. The methodology of stochastic natural frequency analysis
of sandwich plate by using a surrogate model is presented in figure 3.
Fig. 2 Taxonomy for architecture of PNN
Fig. 3 Flow diagram of stochastic analysis using surrogate Model
Results and discussion
The present study portrays the comparative analysis for predictability of ANN and PNN
model for uncertainty quantification of first five natural frequencies of simply supported
skewed (30°) sandwich plate considering a combined variation of all input parameters until
otherwise mentioned. The sandwich plate is considered to be of 10 cm length (l), 10 cm width
(b) and 1 cm thickness (t). The ratio of core and face-sheet thickness is considered ten. Both
upper and lower face-sheet is having 4 laminate of 90°/0°/90°/0° orientation angle.
Stochasticity in skew angle is taken as Δ =5% whereas Δ =10% is taken for ply-orientation
angle, core thickness, face sheet thickness, core material properties and face sheet material
properties. The deterministic FE code is validated by the authors in a previous study [26] and
the same code is used for the current analysis. The material properties considered in this
793
(a) (f)
(b) (g)
(c) (h)
(d) (i)
(e) (j)
Fig. 4 Probability density function of MCS and ANN for (a) 64 (b) 128 (c) 256 (d) 512, MCS and PNN for (a)
64 (b) 128 (c) 256 (d) 512 sample run, of simply supported composite sandwich plates considering combined
variation
paper are [28]: face-sheet (E1 = 131 GPa, E2 = E3= 10.34 GPa, G12= G13=6.895 GPa,
G23=6.205 GPa, , , ρ =1627 kg/m3) and core (E1 = E2 = E3 = 0.00689 GPa, G12= G23= G31=
0.00345 GPa, 12  13  23  0  21  31  32 , ρ = 97 kg/m3). To check the predictability
of surrogate model probability density function (PDF) plot, scatter plot and mean percentage
794
(a) (e)
(b) (f)
(c) (g)
(d) (h)
Fig. 5 Scatter plot of ANN model and original FE model for (a) 64 (b) 128 (c) 256 (d) 512, PNN model and
original FE model for (e) 64 (f) 128 (g) 256 (h) 512, sample run of simply supported composite sandwich plates
considering combined variation
(a) (b)
Fig. 6 Percentage error in mean value of first five natural frequencies between (a) MCS and ANN results, (b)
MCS and PNN results with respect to different sample size considering combined variation.
795
error plot is presented. The predictability of ANN and PNN model is verified in figure 4 by
using PDF plot of the first five natural frequencies for different sample size. Thereafter
negligible deviation of scatter plot from the diagonal line confirms the prediction ability of
the surrogate model. However, the efficiency of the model depends upon the required number
of sample size for surrogate model formation as presented in figure 5. Mean percentage error
is calculated for the surrogate result obtained from ANN and PNN model by using 64, 128,
256, 512 sample run with respect to original MCS result. The sample size of 512 is observed
to be optimum for ANN model whereas sample size of 128 is found to be optimum in case of
PNN model on considering first five natural frequencies as depicted in figure 6. Both the
model gives the efficient result to that of the result obtained through direct MCS as it requires
only 128 (in case of PNN) and 512 (in case of ANN) sample sized Monte Carlo Simulation to
get result equivalent to 10000 sample Monte Carlo Simulation.
Conclusion
The novelty of the present study includes the assessment of comparative prediction ability of
artificial neural network and polynomial neural network for first five uncertain natural
frequencies of sandwich plate considering the combined variation of all random input
parameters. To check the predictability of ANN and PNN as a surrogate model, error analysis
is carried out in addition to probability density function. The scatter plot of the result is also
obtained through ANN as well as PNN model with respect to the original finite element
model. PNN is observed to deliver better prediction than ANN compared to conventional
MCS. In general, it is observed that natural frequencies remarkably deviate from their
respective deterministic values. Hence it is required to consider the effect of stochasticity of
input parameters in the analysis of sandwich plates for a practical design.
References
[1] Pflug J, Verpoest I. Sandwich materials selection charts. Journal of Sandwich Structures & Materials.
(2006);8(5):407-21.
[2] Mukhopadhyay T., Chowdhury R., Chakrabarti A. (2016) Structural damage identification: A random
sampling-high dimensional model representation approach, Advances in Structural Engineering, 19(6)
908–927
[3] Dey T.K., Mukhopadhyay T., Chakrabarti A., Sharma U.K.(2015) Efficient lightweight design of FRP
bridge deck, Proceedings of the Institution of Civil Engineers - Structures and Buildings, 168 (10) 697 –
707
[4] Mahata A., Mukhopadhyay T., Adhikari S. (2016) A polynomial chaos expansion based molecular
dynamics study for probabilistic strength analysis of nano-twinned copper, Materials Research Express, 3
036501
[5] Mukhopadhyay T., Mahata A., Dey S., Adhikari S. (2016) Probabilistic analysis and design of HCP
nanowires: an efficient surrogate based molecular dynamics simulation approach, Journal of Materials
Science & Technology, 32(12) 1345–1351
[6] Metya S., Mukhopadhyay T., Adhikari S., Bhattacharya G. (2017) System Reliability Analysis of Soil
Slopes with General Slip Surfaces Using Multivariate Adaptive Regression Splines, Computers and
Geotechnics, 87 212–228
[7] Dey S., Mukhopadhyay T., Sahu S. K., Adhikari S. (2018) Stochastic dynamic stability analysis of
composite curved panels subjected to non-uniform partial edge loading, European Journal of Mechanics /
A Solids, 67 108–122
[8] Dey S., Mukhopadhyay T., Khodaparast H. H., Adhikari S. (2016) A response surface modelling approach
for resonance driven reliability based optimization of composite shells, Periodica Polytechnica - Civil
Engineering, 60 (1) 103–111
[9] Karsh P. K., Mukhopadhyay T., Dey S. (2018) Spatial vulnerability analysis for the first ply failure
strength of composite laminates including effect of delamination, Composite Structures, 184 554–567
[10] Naskar S., Mukhopadhyay T., Sriramula S. (2018) Probabilistic micromechanical spatial variability
quantification in laminated composites, Composites Part B: Engineering, 151 291-325
[11] Maharshi K., Mukhopadhyay T., Roy B., Roy L., Dey S. (2018) Stochastic dynamic behaviour of
hydrodynamic journal bearings including the effect of surface roughness, International Journal of
Mechanical Sciences, 142–143 370–383
796
[12] Naskar S., Mukhopadhyay T., Sriramula S., Adhikari S. (2017) Stochastic natural frequency analysis of
damaged thin-walled laminated composite beams with uncertainty in micromechanical properties,
Composite Structures, 160 312–334
[13] Mukhopadhyay T., A multivariate adaptive regression splines based damage identification methodology
for web core composite bridges including the effect of noise, Journal of Sandwich Structures & Materials,
DOI: 10.1177/1099636216682533
[14] Mukhopadhyay T., Adhikari S. (2017) Stochastic mechanics of metamaterials, Composite Structures, 162
85–97
[15] Mukhopadhyay T., Adhikari S. (2016) Free vibration analysis of sandwich panels with randomly irregular
honeycomb core, Journal of Engineering Mechanics, 142(11) 06016008
[16] Mukhopadhyay T., Adhikari S. (2017) Effective in-plane elastic moduli of quasi-random spatially
irregular hexagonal lattices, International Journal of Engineering Science, 119 142–179
[17] Mukhopadhyay T., Mahata A., Adhikari S., Asle Zaeem M. (2018) Probing the shear modulus of two-
dimensional multiplanar nanostructures and heterostructures, Nanoscale, 10 5280 – 5294
[18] Mukhopadhyay T., Mahata A., Adhikari S., Asle Zaeem M. (2017) Effective mechanical properties of
multilayer nano-heterostructures, Nature Scientific Reports, 7 15818
[19] Mukhopadhyay T., Mahata A., Adhikari S., Asle Zaeem M. (2017) Effective elastic properties of two
dimensional multiplanar hexagonal nano-structures, 2D Materials, 4 025006
[20] Mahata A., Mukhopadhyay T., Probing the chirality-dependent elastic properties and crack propagation
behavior of single and bilayer stanene, Physical Chemistry Chemical Physics, DOI: 10.1039/C8CP03892A
[21] Mukhopadhyay T., Adhikari S., Batou A., Frequency domain homogenization for the viscoelastic
properties of spatially correlated quasi-periodic lattices, International Journal of Mechanical Sciences,
DOI: 10.1016/j.ijmecsci.2017.09.004
[22] Karsh P. K., Mukhopadhyay T., Dey S. (2018) Stochastic dynamic analysis of twisted functionally graded
plates, Composites Part B: Engineering, 147 259-278
[23] Dey S., Mukhopadhyay T., Spickenheuer A., Gohs U., Adhikari S. (2016) Uncertainty quantification in
natural frequency of composite plates - An Artificial neural network based approach, Advanced
Composites Letters, 25(2) 43–48
[24] Dey S., Mukhopadhyay T., Adhikari S. (2018) Uncertainty quantification in laminated composites: A
meta-model based approach, CRC Press, ISBN 9781498784450
[25] Dey S., Naskar S., Mukhopadhyay T., Gohs U., Sriramula S., Adhikari S., Heinrich G. (2016) Uncertain
natural frequency analysis of composite plates including effect of noise – A polynomial neural network
approach, Composite Structures, 143 130–142
[26] Dey, S., Dey, S., Mukhopadhyay, T., Naskar, S., Dey, T. K., Chalak, H.D., Adhikari, S."Probabilistic
characterisation for dynamics and stability of laminated soft core sandwich plates." Journal of Sandwich
Structures & Materials, DOI: 10.1177/1099636217694229
[27] Dey, S., T. Mukhopadhyay, and S. Adhikari. "Metamodel based high-fidelity stochastic analysis of
composite laminates: A concise review with critical comparative assessment." Composite
Structures (2017); 171: 227-250.
[28] Chalak HD, Chakrabarti A, Iqbal MA, Sheikh AH. Free vibration analysis of laminated soft core
sandwich plates. Journal of Vibration and Acoustics. (2013);135(1):011013.
797
Santosh Kumar Paper ID- 155
MNIT Jaipur
santk1986@gmail.com
Optimization of Cantilever Retaining Wall using Particle Swarm Optimization
Y. P. Mathur1* and Santosh Kumar1

1
Department of Civil Engineering, Malaviya National Institute of Technology Jaipur, India
yogeshpmathur@gmail.com
santk1986@gmail.com
Abstract. Earth retaining structures are ubiquitous in the man-made environment. Reinforced concrete cantilever retaining walls are
considered to be more economical as compared to their traditional mass concrete or masonry gravity wall counterparts. Design of
cantilever retaining walls involves stability checks for overturning, sliding and bearing in addition to stress limitations of concrete and
steel. A trial section is generally assumed and calculations are made for these checks. The important parameters involved are unit
weight of backfill soil, soil-concrete friction factor and bearing capacity of the soil. The base length and its components namely toe
and heel lengths, and base thickness are highly dependent upon these factors and vary with stem height and surcharge. To adopt a
suitable economical trial section is sometimes a tedious job. A window based computer program has been developed for designing
such walls. This paper is concerned with the cost optimisation of the cantilever retaining wall with the help of computer program using
MATLAB. PSO algorithm has been applied to the design of cantilever retaining wall for the minimization of the cost objective
function. The problem selected is example 14.9 from Reinforced Concrete Design Book (Third Edition) by S Unnikrishna Pillai and
Devdas Menon. The design cost of solution of the book was found out to be Rs. 36247 per meter while the cost achieved after
optimization is Rs. 23951 per meter length.
Keywords: Cantilever Retaining Wall, Optimization, Particle Swarm Optimization, MATLAB
1 Introduction
Retaining wall, as the name implies, is used to hold the loose earth or backfill in position. It should be capable of retaining earth mass
without overturning and sliding under the action of active earth pressure generated due to retained soil and surcharge if any. In these
two stability cases the weight of soil and self-weight of wall act as stabilizing forces. The third important stability consideration in the
design of wall is to spread the resulting forces over sufficient area so that actual pressure does not exceed the allowable bearing
capacity of soil.
798
The design for strength or determination of steel to resist the bending and shear in three slabs i.e., stem, toe and heel is the last step
and it is possible only once adequate factors of safety have been ensured against overturning, sliding and bearing. The design for
strength is made only after satisfying all the three stability checks.
It may be concluded from the foregoing introduction that design of retaining wall is a tedious job, which requires several trials before
reaching a suitable section that can be designed for strength. The iterative nature of the problem makes it well suited for a computer
solution.
The objective of this paper is to design a cantilever retaining wall and optimize it to find a cost-effective solution that minimizes
capital investment whilst ensuring a good system performance under specific design criteria. This paper basically provides a detailed
method of implementing an optimization technique, called Particle Swarm Optimization on cantilever retaining wall.
The two main costs of cantilever retaining wall design are steel and concrete, which often create contradictory objectives in the design.
Any reduction in concrete section is likely to result in an increase in steel requirements and consequently, in the steel costs. On the
other hand, reducing steel requires bigger concrete section and hence increased concrete cost. Therefore, finding an economical design
for cantilever retaining wall requires an optimal trade-off between concrete and steel costs, which cannot be achieved by an
engineering judgment but can be easily done with the help of optimization techniques.
2 Literature Review
Over the last decade, various kinds of modern bionic optimization algorithms with their special advantages have been created. The
methods such as enumeration, linear programming, nonlinear programming, dynamic programming and mixed integer programming
are usually used to solve optimal design problems. This paper has been prepared with an objective to provide a detailed description of
the application of PSO algorithm to cantilever retaining wall. Several optimization methods have been proposed to minmize the total
cost of cantilever retaining wall. A brief review of many of the optimization techniques has been presented below.
SED-PSO (Stochastic Evolutionary Dynamics) has been used by Zhi-jie et.al [3] to speed up the researching process of thr
particles.Each particle in the swarm associates with the process of reproduction which is controlled by a stochastic process having
frequency dependent fitness. The result shows that SED-PSO performs better in the convrgence property over traditional PSO in terms
of iteration with only a slight precision dropdown.
To minimize the total cost of the retaining wall, Genetic algorithms and tabu search techniques were implemented to solve this
difficult optimization problem. An adaptive rule and a dynamic search strategy were developed to assist the search procedures to find
better solutions [4].
799
Sequential unconstrained minimization technique along with Powell’s algorithm for multidimensional search and quadratic
interpolation technique for one-dimensional search has been adopted for the cost minimization of cantilever retaining wall [5]. The
result showed that on varying the stem top width from 20 to 30 cm, the percentage increase in the cost ranges from 9 to 15%.
3 Particle Swarm Optimization
Particle swarm optimization (PSO) is a population based stochastic optimization technique developed by Dr. Eberhart and Dr.
Kennedy in 1995. It is inspired by social behavior of organisms such as bird flocking, fish schooling and herds of animals to adapt to
their environment, find the sources of food, and avoid predators by implementing an "information sharing" approach, hence, develop
an evolutionary advantage. In particle Swarm Optimization, each particle flies in N- dimensional space on its own with the help of
previous experiences of its own and the neighbours. The birds represented by particles move or behave differently for different
sources of food. So we can say that the particles move stochastically. Each particle keeps track of its coordinates as well as the best
position of all the particles called as gbest. The best position achieved by a particle is called as pbest. The particle moves in the search
of food so its position is regularly updated which is expressed in terms of iteration. The postion is identified by the location and the
velocity of the particle. At the end of iteration the particle has a velocity in the direction of current gbest value.
The position of 𝑖𝑡ℎ particle is represented by 𝑋𝑖 and its velocity as 𝑣𝑖 . The new position and the new velocity of the particle is achieved
by the following expression.
𝑣𝑖𝑛+1 = 𝜔𝑣𝑖𝑛 + 𝑐1 𝑟𝑎𝑛𝑑1 (𝑝𝑏𝑒𝑠𝑡𝑖 − 𝑋𝑖𝑛 ) + 𝑐2 𝑟𝑎𝑛𝑑2 (𝑔𝑏𝑒𝑠𝑡 − 𝑋𝑖𝑛 ) (1)
Where, 𝑛 = iteration number; 𝑟𝑎𝑛𝑑1 , 𝑟𝑎𝑛𝑑2 = random numbers uniformly distributed between 0 and 1; 𝑐1 , 𝑐2 = two positive
constants, called cognitive and social parameter respectively and 𝑋𝑖𝑛 = position of 𝑖𝑡ℎ particle in 𝑛𝑡ℎ iteration. The new position is
given by the expression as follows.
𝑋𝑖𝑛+1 = 𝑋𝑖𝑛 + 𝑣𝑖𝑛+1 (2)
Where, 𝑋𝑖𝑛+1 = new position of the 𝑖𝑡ℎ particle for 𝑛 + 1𝑡ℎ iteration and 𝑣𝑖𝑛+1 = new velocity as calculated from above equation.
The process is repeated until the process reaches its termination point. When the velocities are calculated it has no limits because the
search space of the particle sometimes is too large and sometimes too short. When a particle achieves its best position and calculates
the new position it may go very far from the optimal solution in the search space. Hence there should be some limits on the velocity.
To avoid for the solutions to be trapped in local solutions range of the velocity is decided which is again problem based.
The role inertia weight ω is also crucial as it regulates the tradeoff between the global and the local exploration abilities of the swarm.
A suitable value of inertia weight results in the reduction of iterations required to locate the optimum solution. Similarly, the values of
c1 and c2 also help in the final convergence of the solution. The values of random numbers maintain the diversity of population.
800
4 Objective Function
The design of cantilever retaining wall is a mixture of discrete as well as continuous variables. The weight of steel is being calculated
by the available diameters present in the market and the quantity of concrete is a continuous variable. Hence in this study eight
decision variables have been taken i.e. stem, heel, toe and base slab dimensions for calculating the quantity of concrete and the weight
of steel. These decision variables are taken such as not to violate the constraints which are listed as follows:
Stem Top width: 200 mm to 400 mm
Base slab width: b= 0.4H to 0.6H, and 0.6H to 0.75H for surcharged wall
Base slab thickness: H/10 to H/14
Toe projection: (1/3 to1/4) of Base width
Area of steel: (i) Maximum -4% of gross c/s area (ii) Minimum-0.15% of gross c/s area
In this particular case, if any of the components of the position vector takes a negative value, it will be set to zero. On other hand, if it
is greater than the maximum value for that component it will be saturated to this maximum value.
The main objective of the study here is to minimize the overall cost of the structure which includes the cantilever retaining wall cost
and the penalty cost.
𝑇𝑜𝑡𝑎𝑙 𝐶𝑜𝑠𝑡 = 𝐶𝐶 ∗ 𝑉𝐶 + 𝐶𝑆 ∗ 𝑊𝑆 + 𝑃𝑒𝑛𝑎𝑙𝑡𝑦 (3)
Where 𝑪𝑪 =Cost of Concrete per cubic meter
𝑽𝑪 =Volume of Concrete in cubic meter
𝑪𝑺 =Cost of Steel per kg
𝑾𝑺 =Weight of steel in kg
𝑷𝒆𝒏𝒂𝒍𝒕𝒚 =Penalty cost when constraints are violated
In the present study concrete cost per cubic meter is taken as Rs. 7000 and cost of steel as Rs.72 per kg. The penalty cost is imposed
when the constraints are violated. Penalty imposed in all the cases = Rs. 1000000
Penalty Function
Penalty functions, a technique used in solving constrained optimization problems, are used to restrict the solution search to designs
that meet all criteria. The unconstrained problems are formed by adding a term to the objective function that consists of a penalty
801
parameter and a measure of violation of the constraints. The measure of violation is nonzero when the constraints are violated and is
zero in the region where constraints are not violated. The penalty functions are described below:
Penalty due to Moment of resistant: If the moment of resistance of any section (heel, toe, stem) is less than the max moment coming to
these sections respectively in that case penalty is to be imposed on that solution.
Penalty due to SBC: If SBC < Max soil pressure at slab base or
Min soil pressure < 0
Penalty due to shear stress: If shear stress at any of the section (heel, toe, stem) > allowable shear stress according to IS 456 then
penalty is to be imposed on that solution.
Penalty due to Factor of Safety of sliding and overturning: If FOS of sliding and overturning <1.55 then penalty is to be imposed on
that solution.
Algorithm for Particle Swarm Optimization
Start
Initialize the positions and velocities of each
particle
Evaluate the fitness value of each particle
Find and update pbest and gbest
Update velocities and positions of each particle
Evaluate the fitness value of each particle
802
Are termination No
criteria met?
Yes
Stop
Fig. 1. Algorithm of PSO
The termination condition may be stated either in the terms of a maximum number of iterations or in the event that value of the fitness
function is not being changed by a small percentage.
The values of c1, c2 are taken in the computer after performing and observing lots of iterations and their results accordingly, so that
they can bring about change in the section dimensions and area of steel.
5 Present Approach
In the present work, a computer program coded in MATLAB is used to solve the problem of optimal retaining wall design. The
method requires the interior feasible initial solution, thus a program was developed based on concept of feasible dimensions of the
sections and area of steel set. The applied algorithm, considers stem width, heel and toe thickness and area of steel as design variable.
The values taken as input for diameter of reinforcement bar correspond to the commercially available diameters. The given data as per
example 14.9 referred from Reinforced Concrete Design Book (Third Edition) by S Unnikrishna Pillai and Devdas Menon are
tabulated below in the table no. 1.
Table no.1: Input data as per example 14.9 referred from Reinforced Concrete Design Book (Third Edition) by S Unnikrishna Pillai and Devdas
Menon
Input Data
Height of Soil to be retained (m) 4
Soil below ground (m) 1.25
Total height of stem (m) (H) 5.25
Unit weight of soil (kN/m3) 16
Friction angle of the soil 30
Bearing capacity of the soil (kN/m2) 160
803
Friction between soil and concrete 0.5
Surcharge (kN/m2) 40
Equivalent height of earth (hs) 2.5
H + hs 7.75
Effective cover of slab 0.09
Load factor 1.5
fck 25.0
fy 500
Bar dia slab 16
Bar dia stem 20
Effective cover of stem 0.06
FOS Stability against overturning 1.40
FOS Stability against sliding 1.40
Algorithm was successfully developed for an optimal design of cantilever retaining wall and can be employed to solve several
hypothetical as well as real life problems. The parameters of PSO which gives the best results are finalized after many trails. The
values of these parameters are listed in the table no. 2 and the upper and lower bounds of the decision variables are tabulated in table
no. 3. The section dimensions and area of steel calculated in this study have been compared with that of the results of the example
14.9 referred from Reinforced Concrete Design Book (Third Edition) by S Unnikrishna Pillai and Devdas Menon. The design cost as
per the example 14.9 referred from Reinforced Concrete Design Book (Third Edition) by S Unnikrishna Pillai and Devdas Menon
calculated was Rs. 36247 per meter (approximately) while the cost achieved after optimization shows better results along with
agreeable safety compliances. Comparision of the results obtained after optimization are done in the table no. 4. Result exhibit a total
cost of Rs. 23951 per meter length of wall (longitudinal).
Table No. 2: List of parameters of PSO and their values
Parameters Values
C1 25
C2 25
Npop 200
Maxit 1000
ω 1
Table No.3: Lower and upper bounds of decision variables
Variable Minimum Maximum Description

Var1 2.2 3.5 Heel Length
Var2 1.5 2.5 Toe Length
Var3 0.5 2.0 Slab thickness
Var4 0.3 1.5 Stem thickness at bottom
804
Var5 500 2000 Steel in heel
Var6 500 2000 Steel in toe
Var7 500 2000 Steel in stem
Var8 0.2 0.8 Stem thickness at top
Best Solution
Table No. 4: Comparision of solutions obtained
Decision Variable Values as per referred problem Optimized Values

Heel Length 2.6 2.2
Toe Length 1.3 1.64
Slab thickness 0.62 0.5
Stem thickness at bottom 0.65 0.3
Steel in heel 1074 1214.245
Steel in toe 806 670.206
Steel in stem 1741 1182.717
Stem thickness at top 0.2 0.2
Cost (Rs.) 36247 23951
7 Conclusion
Optimization technique adopted in this work proved to be successful in optimal designing of the cantilever retaining wall. In this
paper, the Particle swarm optimization method was applied for finding optimal concrete volume and area of steel for the conjunctive
least-cost design and structure of cantilever retaining wall. Hence after this study we can say that PSO is a very good alternative for
the optimization of a cantilever retaining wall. It can be broadly applied in the optimization of other concrete structures as well.
Future Scope of Study
Particle swarm optimization for cantilever retaining wall is in its early stages of exploitation and there is a lot of scope for further
work in this field. Certain aspects of the algorithms presented in this paper could be analyzed in more detail. Furthermore, more design
parameters can be varied which will also have an impact in deciding the prime factors and hence generate the optimum cost. The
direct as well as indirect cost can be included in the study. Also the optimization of various parameters of PSO may give good results.
Hence there is a lot of scope for researchers to explore the space of optimization through Particle Swarm Optimization.
References
1. Reinforced Concrete Design - Third Edition by Devdas Menon (Author), S. Pillai (Author)
2. Engelbrecht A.P., Fundamentals of Computational Swarm Intelligence, John Wiley and Sons, ltd (2005)
805
3. Zhi-jie, L., Xiang-dong, L., Xiang-dong, D.: A particle swarm algorithm based on stochastic evolutionary dynamics. Fourth
International Conference on Natural Computation, (2008). doi:10.1109/ICNC.2008.103
4. Liang, L. Y., Thompson, R. G., Young M. D.: Optimizing the design of sewer networks using genetic algorithms and tabu
search. Engineering, Construction and Architectural Management, Vol 11 Iss:2, pp.101-112, (2004).
5. Basudhar, P. K., Lakshman, B., Dey A.: Optimal cost design of cantilever retaining walls. IGC 2006.
6. IS-456:2000, Code of Practice for Plain and Reinforced Concrete, Bureau of Indian Standards, New Delhi, 1981.
806
Paper ID- 156
Impact Analysis of Ultra High Performance Fiber Reinforced Concrete slabs

Suruchi Mishra1, Rizwanullah2, H. K. Sharma3
1,2,3
Civil Engineering Department, National Institute of Technology, Kurukshetra, Haryana,
India-136119
1
E-mail: suruchi_6160058@nitkkr.ac.in
Abstract
An investigation into the Ultra high performance fiber reinforced concrete (UHPFRC) slabs under drop weight
impact loading with varying fiber volume content (0-10%) has been conducted and presented in this study. The
main focus of the work is to develop UHPFRC of M100 and above grade UHPFRC and study the failure pattern and
energy abortion capacity of UHPFRC slabs under impact loading. A 50 kg of impactor with drop height 500 mm is
used.
The experimental results revealed that impact resistance is directly proportional to compressive strength of material,
and the UHPFRC slab panels corresponding to M5 with 8% fibre content exhibit maximum impact capacity.
Keywords: UHPFRC, Steel fibers, Drop weight impact load, Energy absorption capacity.
1 Introduction
In recent years, terrorist attacks are becoming increasingly more frequent throughout the world. As anyone who
gives attention to history and current events knows, buildings are often targets of explosive threats and terrorist
attacks. Attacks that targeted on buildings not only cause structural failure but also lead to massive loss of property
and life. Impact loads can cause a significant threat to our buildings and civil infrastructure with potentially
dangerous consequences. The performance of structures under extreme loading conditions has attracted much
attention from researchers and engineers in recent decades. The material behavior under a dynamic load is different
from that under a static load, so it is important to investigate the dynamic strength, dynamic damage, and dynamic
stress-strain relationship of engineering materials.
UHPFRC is a relatively new generation of fiber cementitious composites which have been developed to give
significantly higher material performance than other concrete classes. Such superior properties have been achieved
through the use of an optimized combination of its ingredients like cement, fine sand, micro silica, high range water
reducing agent (HRWRA), a very low water contents and fibers.
2 Experimental Programme
2.1 Materials
The properties of the material used for making UHPFRC mix are determined in the laboratory as per specified in the
relevant code of practice. The different material used in the present study was Portland cement (OPC 53 grade as per
IS 12269: 2013), silica fume, fine sand (size less than 600 micro meter), high range water reducing agent (HRWRA)
(Super plasticizer), steel fiber and water.
807
2.2 Mix Proportion and Procedure
The production of the mixture and the optimization of the mixing procedure were performed based on reported
techniques for similar patented material .The optimized derived mixture proportions for UHPFRC investigated in
this study are presented in Table 1. For UHPFRC total seven no. of mix were prepared with different fibre ratio (0%,
2%, 3%, 4%, 6%, 8% and 10%).
Table 1 Mix proportion of UHPFRC ((Lachance et al. 2016)
Cement Silica fume Water Quartz Sand HRWRA Fibre volume Water/binder
(kg/m3) (kg/m3) (kg/m3) (kg/m3) (kg/m3) content (%) ratio
1015 254 211 609 49.5 0- 10 % 0.20
All the dry materials which have been weighted were added systematically into the pan mixer, starting with the
cement, followed by the silica fume and silica sand. The materials were then dry-mixed for one minute to ensure a
thoroughly blended mixture. Following this, the super plasticizer together with the clean water was added into the
mixture and the mixing continued for another 10-12 minutes. Finally, steel fibers were carefully spread into the pan
and mixed for another 2-3 minutes. This technique ensured a proper dispersion of the steel fibers in the cement
paste. The casting of all UHPFRC specimens used in this material characterization study was completed within 20
minutes after the completion of mixing.
2.3 Casting and De-Moulding
Necessary steps were taken to ensure the effective production of the specimens as the results depend on the output of
the casted specimens. Before casting foam oil was applied to the moulds to release the friction and get the good
surface at the sides and the bottom. After properly mixing the cement mortar was filled into the required moulds.
UHPFRC is a self-compacting concrete and there was no need of vibration. The surface of the mortar was finished it
self and leveled with the top of the molds. The finished specimens were left to harden in the air for 48 hours. The
specimens were removed from the moulds after 48 hours of casting and were placed in potable water at room
temperature in the laboratory curing tank.
2.4 Test Setup and Procedure
2.4.1 Compression Test
One of the most commonly specified and measured properties of concrete is compression strength. In compliance
with the IS: 516 – 1959 standard, the compressive strength test was performed on the cubic samples with an array
dimension of 70.6 x 70.6 x 70.6 mm. The arithmetic mean of three samples for each mix was calculated to evaluate
the results of the compressive strength tests conducted after 28 th day. The compressive load was monotonically
applied using a compression testing machine (CTM) with a loading rate of 3 kN/s and maximum load capacity of
1000 kN. The average compressive strength was achieved 131Mpa after 28 days.
2.4.2 Impact Resistance Test
For impact test, the entire slab specimen, a special set up was prepared by installing a drop weight rope and pulley
mechanism. A smooth pulley was carefully fixed to the top of girder of a movable crane. Slab specimen was
supported on two I-beam at two edges to avoid displacing due to jerks. The 0.5 kN drop weight consisted of a
concrete block cast in a vegetable oil tin and suitable reinforced was used. A 120 mm diameter steel ball was
808
embedded in the block at its base so that the impacting force was a point load applied at the center of the slab. This
impactor was dropped on the specimens from 500 mm height, to check the impact energy absorption capacity of the
slabs. A schematic diagram of the drop-weight impact test setup and supporting condition is illustrated in Fig.1(b).
The slabs are tested under the same loading and supporting conditions. The slabs are subjected to hard impact at
their midpoint and simply supported at their two corners. The corresponding number of blows with first crack and
spalling of concrete were noted.
Fig. 1(a) casting of slab specimen and Fig. 1 (b) Test setup for drop weight impact.

Experimental investigations were conducted to address the advantage of using UHPFRC in low-velocity impact
resistance structures, and to investigate the effect of fiber volume content on the impact behaviour of UHPFRC slab
panels.
A drop weight impact loading test was performed on square slabs with dimensions 500 x 500 x 50 mm. All slabs are
tested under same loading and supporting conditions as well as subjected to uniform testing termination criteria.
Impact was simulated as a hit of 0.50 kN impactor, with drop distance to the slab as 0.5 meters and the average
muzzle velocity was kept as 3.132 m/s with an average impact energy of 245.23 J. It is pertinent to note that most of
the researchers have conducted studies on UHPFRC/NSC/UHPC slab panels under impact loading under these
conditions only, Present studies have also been conducted in the same environment , keeping drop weight, drop
distance and slab dimensions same . However, there is few available data in literature related to the behaviour of
UHPFRC members subjected to low-velocity impact load with velocities up to 10 m/s, and hence present
investigations have been conducted .The test results are analyzed and discussed under the following heads.
3.1 Energy Absorption Capacity
UHPFRC slab panels were tested for number of the blows required to create first crack and fracture in the specimens
, and no. of blows were recorded corresponding to these stages. Total energy absorption of the specimens computed
for various specimens is tabulated in Table- 4.1and 4.2. Energy absorption was computed using the following
equation:
Energy absorption = Number of Blows x Weight of impactor x Height of fall
809
Where, weight and height of the impactor were 50kg and 500 mm, respectively.
Table 2 summarizes the energy absorption capacities of tested slabs with respect to first crack and fracture under
various no. of blows. It is evident from Table 2 that the use of UHPFRC material enhances the impact energy
absorption capacity significantly. Comparing the capacity of UHPFRC, it is in range of 2 to 3 times the capacity of
the UHPFRC slabs without fibers. The increased capacities of UHPFRC slabs are correlated to the steel fiber
content. Increasing fiber content from 2 to 4% has a significant effect on the energy absorption capacity compared to
increasing the fiber content from 4 to 8%. Additionally, the steel fiber contents are found to have significant
influence on the impact energy absorption capacity.
Table 2- Energy absorbed at first crack and at failure.
Sr. Mix Fibre No. of Blows Energy No. of Blows Energy Absorbed
No. Content Required at First Absorbed Required at at Failure
(%) Crack at First Crack Failure (J)
(J)
1 M0 0.0 1.0 245.23 2.0 490.69
2 M1 2.0 1.0 245.23 3.0 735.69
3 M2 3.0 2.0 490.23 4.0 980.92
4 M3 4.0 3.0 735.69 5.0 1226.15
5 M4 6.0 4.0 980.92 7.0 1716.61
6 M5 8.0 5.0 1226.15 9.0 2207.07
7 M6 10.0 4.0 980.92 7.0 1716.61
3.2 Crack Patterns and Failure Modes
It was shown that plain UHPC specimen failed in the brittle manner which illustrated that the slabs split into several
pieces. No cracks were observed on the upper surface of the UHPFRC specimens during visual inspections after
each drop. All specimens showed very fine multiple cracks in the central portion of the lower surface, which are
typically observed for strain hardening materials. The UHPFRC slabs exhibited superior performance with regard to
damage control characteristics under low-velocity impact loading conditions. No spalling, scabbing, and/nor
significant large fragmentations are observed. Additionally, the fragments are in the form of fine powder. The final
fracture of the specimen occurred by fiber pull out in one localized crack which exhibited a very rough fracture
surface. The failure mode for all UHPFRC specimens was in bending and identical to the quasi-static fracture mode.
It has been demonstrated that fibers play an important role in limiting the extent of damage level. Increasing the
fiber content led to an increase in number of cracks at the bottom surface of slabs and an overall reduction in the
width of cracks formed.
810
Figure- 2 Various modes of failure and failure pattern of UHPFRC slabs under drop weight impact.
811
Conclusions
Impact resistance is directly proportional to compressive strength of material, and the UHPFRC slab panels
corresponding to M5 with 8% fiber content exhibit maximum impact capacity. The favorable performance of the
slab panels tested under impact loading suggests that traditional continuous steel reinforcement can be eliminated in
case of UHPFRC designed for specific use under impact loading situations. This is because the impulse load is able
to be handled mostly by the material strength. UHPFRC , being greater resistant to impact loading as compared to
NSC in terms of spalling and scabbing, may result in manufacturing of highly resistant concrete elements such as
cladding panels and walls to be used in modern protective structures while decreasing their thickness and
appearance. The experimental results revealed that UHPC slabs are much more fragile than their NSC counterparts
due to their higher compressive strength and brillteness. Multiple macro cracks propagated through the UHPC
specimen under the drop weight impact which caused that the slab to split into several small pieces.
References
1. Bindiganavile V, B. N. (2002). Impact response of ultra-high-strength fibre reinforced cement composite.
ACI Mater. J, 543-548.
2. IS: 516 – 1959 (Reaffirmed 2004), Methods of Tests or Strength of Concrete. NEW DELHI: Bureau of
Indian Standards.
3. Jianzhong Lai, W. S. (2013). Dynamic Properties of Reactive Powder Concrete Subjected to Repeated
Impacts. ACI Materials Journal, 463-472.
4. Lachance, F. C. (2016). Development of Precast Bridge Slabs in High-Performance Fiber-Reinforced
Concrete and Ultra-High-Performance Fiber-Reinforced Concrete. ACI Structural Journal.
5. R. Yu, L. v. (2016). Impact resistance of a sustainable Ultra-High Performance Fibre Reinforced Concrete
(UHPFRC) under pendulum impact loadings. Construction and Building Materials, 203-215.
6. Richard P, C. M. (1994). Reactive powder concrete with high ductility and 200–800 MPa compressive
strength. . ACI , 507-518.
7. Richard, P. C. (1995). Composition of Reactive Powder Concretes. Cement and Concrete Research, 1501-
1511.
812
Paper ID- 157
Development of Low Carbon Binder for Sustainable

Construction as an Alternative to Conventional Cement
Amer Hassan1, Mohammed Arif2 and Mohd Shariq3

1,2,3
Department of Civil Engineering, Aligarh Muslim University, Aligarh, 202002 (India)
1
ameralburay@gmail.com
2
marifamu@gmail.com
3
mshariqdce@gmail.com
Abstract. The present study focuses on the development of a new low carbon
binder as an eco-friendly construction material which would replace the con-
ventional ordinary Portland cement. The reactions between alkali sources with
an alumina and silica-containing solid precursor obtained from industrial waste
materials such as fly ash, ground granulated blast furnace slag (GGBFS), me-
takaolin etc. is called alkali activation or geopolymerisation. The experimental
investigations have been undertaken to examine the compressive strength of
geopolymer concrete using fly ash containing alumina-silicate material and al-
kaline solution prepared by combining sodium hydroxide (NaOH) and sodium
silicate (Na2SO3) in different molarities. The experimental tests revealed that
the amount of water has a significant effect on the compressive strength of geo-
polymer concrete. Further, the geopolymer concrete mixture requires proper
mix proportion and temperature-controlled curing conditions to accomplish
good results.
Keywords: Geopolymer concrete, Fly ash, Alkaline solution, Sodium Silicates,

Sodium hydroxide, Compressive strength.
1 Introduction
The manufacture of Ordinary Portland Cement (OPC) contribute about 5% of CO2

emission to the environment and because of a big demand for construction in various
sector of industries such as, buildings, transportation, dams, tunnels and sewage etc.,
consequently, there is an urgent need for an alternate material binder which can re-
place OPC for a cleaner and sustainable construction [1–6]. Fly ash based Geopoly-
mer concrete is a new material that can be used for green construction purposes [7–9].
There are two main constituents of geopolymer concrete, namely the source material
with rich Alumina-Silica containing material such as (fly ash, metakaolin or GGBFS),
and alkaline solution which comprises of sodium hydroxide (NaOH) & sodium sili-
cate (Na2SiO3) solution [10–13]. The source material which is fly ash in this study
reacts with alkaline solution to produce alumina silicate gel [14–16]. This gel binds
aggregates and other unreacted materials to produce geopolymer concrete (GPC) [17].
813
The geopolymer concrete is considered as an ideal choice not only because it is a

sustainable material, but because it utilizes waste materials from industries [9, 18].
The geopolymer concrete has an excellent mechanical property, high early strength,
freeze thaw resistance, low chloride diffusion rate, resistance thermal stability and fire
resistance, which make it superior than OPC concrete [12, 19]. The unique perfor-
mance of geopolymer concrete especially in term of chemical resistance and durabil-
ity, may be attributed to the lower calcium content in the mixture of GPC [20–23].
Although use of geopolymer concrete is still in the beginning stage, however, re-
cently many of structures have been successfully constructed utilizing geopolymer
concrete. Some of these interesting structures include reinforced box culverts, con-
crete pipes, pavements, structural elements, pre-cast concrete like railway sleeper and
electric power pole and marine structures reinforced box culverts, precast railway
sleepers, precast footway panels, airport in southeast Queensland, Australia etc. [24,
25]
2 Objective of the Study
The main objective of the study is to develop a mixture proportioning process for
fly ash based geopolymer concrete and to investigate the mechanical behaviour of
geopolymer concrete as an alternative to ordinary Portland cement (OPC) concrete.
3 Source Materials, Mixture Proportions and Curing Procedure
3.1 Materials used

In this experimental work, fly ash is utilized as the source material to produce geo-
polymer concrete. The effect and role of coarse and fine aggregates are considered to
be similar as in the ordinary Portland cement concrete. The mass of coarse and fine
aggregates has been taken from 75% to 80% of the total mass of the Geopolymer
concrete matrix.
Sodium hydroxide (NaOH) activators have been used in the study because these
are superior to Potassium hydroxide activators (KOH) on account of performance as
well as cost. Sodium hydroxide used in the experimental work was in flake form
(Fig.1). The sodium hydroxide solution is mixed with sodium silicate solution to pro-
duce alkaline solution maintaining a ratio of 2.5 between Na2SiO3 solution and NaOH
solution. It is recommended that the alkaline solution be prepared by mixing both the
solutions together one day prior to use. The mass of Sodium hydroxide solids in the
solution varies as per the molarity (M) of the solution. The concentration of sodium
hydroxide solution (NaOH) liquid is measured in terms of Molarity (M). NaOH solu-
tion with a concentration of 8M and 14 M is prepared. In case of 8 molarity, it con-
sists of 8×40 = 320 grams of NaOH solids (in flakes form) per liter of the solution,
where 40 is the molecular weight of NaOH. In order to improve the workability, su-
per-plasticizer was added to the mixture. The compressive strength and workability of
814
geopolymer concrete are significantly influenced by the proportions and properties of

the constituent materials that make the geopolymer paste.
Fig. 1. Sodium hydroxide in form of flakes (NaOH)
3.2 Mixture Proportions and mixing procedure

Two mixes of geopolymer concrete (mixture1 & mixture 2) were prepared as
shown in Table 1. The chemical composition of the materials used in the experimental
work are given in Table 2. The EDX analysis of fly ash is shown in Fig. 2, and EDX
analysis of sodium hydroxide NaOH and sodium silicate Na2SiO3 are shown in Fig. 3
and Fig. 4 respectively.
For mix design of geopolymer concrete, in the first step the alkaline solution acti-
vator was prepared by dissolving sodium hydroxide flakes in water and then the sodi-
um silicate solution Na2SiO3 was added to NaOH solution and left for 24 hours. The
fine and coarse aggregates were mixed with fly ash for 3 minutes. The super-
plasticizer was added to the alkaline solution activator 1 minute before the addition of
the alkaline solution to the dry materials and then added to the dry materials and
mixed for 5-6 minutes in the mixer. When extra water was required, it was added to
the wet mix at least 1 minute after the mixing of the alkaline solution with the dry
materials. Fig. 5 shows the dry materials and alkaline solution activator before mixing
them together. The fresh geopolymer concrete in moulds after casting is shown in Fig.
6.
3.3 Curing procedure

The curing of the Geopolymer concrete is different from that of the Portland ce-
ment concrete OPC. The specimens were cured in the oven for 24 hours at 75°C. As
can be seen in Fig. 7.
815
Table 1. Mix proportion of geopolymer concrete (kg/m3)

No. of Coarse Fine Fly NaOH Na2SiO3 Super- Extra
Mixture agg. agg. ash solution solution plasticizer water
Mix-1 1294 554 408 41 (8M) 103 6 -
Mix-2 1294 554 408 41 (14M) 103 6 22.5
Table 2. Chemical composition of fly ash, sodium silicate and sodium hydroxide (by mass)
Fly ash Sodium silicate (Na2SiO3) Sodium hydroxide (NaOH)
SiO2 51.86 % SiO2 83.40 % SiO2 46.47 %
Al2O3 9.15 % Na2O 16.60 % Na2O 53.53 %
Fe2O3 33.98 %
CaO 2.26 %
MgO 1.66 %
K2O 1.09 %
Total 100 % Total 100 % Total 100 %
Fig. 2. EDX analysis of fly ash
816
Fig. 3. EDX analysis of NaOH
Fig. 4. EDX analysis of Na2SiO3
817
Fig. 5. Mixture of geopolymer concrete
Fig. 6. Casting of geopolymer concrete
818
Fig. 7. Curing of geopolymer concrete at oven
4 Results and discussion
The test results of compressive strength of geopolymer concrete for mixture-1 and
mixture-2 are given in Table 3.
Fig. 8 shows the variation of compressive strength in the three specimens of mix-
ture-1 and mixture-2. It can be seen from Table 3 and Fig. 8 that the mixture-1 of
GPC performed better than mixture-2 in term of compressive strength, which may be
attributed to extra water which has been added in the mixture-2. The cracked geopol-
ymer concrete cube specimen under compressive strength test is shown in Fig. 9.
The compressive strengths increased with the increase in NaOH concentration. At

high NaOH concentrations, the leaching of alumina and silica was enhanced and this
resulted in increased geopolymerization, and thus the strength increased [26]. In this
paper, it can be seen that the compressive strength is decreased when the added water
increased to the mixture regardless of concertation of NaOH solution used. However,
for the mix series with sodium hydroxide ratio of 1.0, the high NaOH concentration
resulted in a increment of strength in case of no added water used in the mixtures [27–
31].
Table 3. Test results of compressive strength of geopolymer concrete (MPa)

Designation Sample 1 Sample 2 Sample 3 Mean
No.
Mixture-1 25 21.8 20.4 22.4
Mixture-2 22.6 19.5 21.36 21.2
819
Fig. 8. Compressive strength of GPC at 28 days
Fig. 9. Crack pattern of GPC Cube under compression test
820
5 Conclusions
The experimental investigation validates the suitability of the utilizing Low-calcium

fly ash (LCFA) as a binder material to produce geopolymer concrete. The resulting
mix possesses comparable structural grade compressive strength properties and can be
thought of as an alternative eco-friendly and sustainable construction material.
The investigation further reveals that geopolymer concrete attains its target strength
much faster under heat cured condition compared to ambient cured condition. The
mechanical behavior of geopolymer concrete is almost similar to that of OPC con-
crete and the cracking behavior of geopolymer concrete is also similar to OPC con-
crete. It has also been observed that the change in the molarity of the alkaline solution
has significant effect on the final compressive strength of GPC.
References
1. Habert G, D’Espinose De Lacaillerie JB, Roussel N (2011) An environmental
evaluation of geopolymer based concrete production: Reviewing current
research trends. J Clean Prod. doi: 10.1016/j.jclepro.2011.03.012
2. McLellan BC, Williams RP, Lay J, et al (2011) Costs and carbon emissions
for geopolymer pastes in comparison to ordinary portland cement. J Clean
Prod. doi: 10.1016/j.jclepro.2011.02.010
3. Zhang Z, Provis JL, Reid A, Wang H (2014) Geopolymer foam concrete: An
emerging material for sustainable construction. Constr. Build. Mater.
4. Part WK, Ramli M, Cheah CB (2016) An Overview on the Influence of
Various Factors on the Properties of Geopolymer Concrete Derived From
Industrial Byproducts. In: Handbook of Low Carbon Concrete
5. Kong DLY, Sanjayan JG (2010) Effect of elevated temperatures on
geopolymer paste, mortar and concrete. Cem Concr Res. doi:
10.1016/j.cemconres.2009.10.017
6. Wallah SE, Rangan B V. (2006) Low-Calcium Fly Ash-Based Geopolymer
Concrete: Long-Term Properties. Concrete. doi:
10.1017/CBO9781107415324.004
7. Rangan BV (2010) Fly Ash-Based Geopolymer Concrete. Proc Int Work
Geopolymer Cem Concr December 2010:68–106 . doi: 10.1007/s10853-006-
0523-8
8. Zhuang XY, Chen L, Komarneni S, et al (2016) Fly ash-based geopolymer:
Clean production, properties and applications. J. Clean. Prod.
9. June J, Politecnica U, Lloyd N a., et al (2010) Geopolymer Concrete with Fly
Ash. Second Int Conf Sustain Constr Mater Technol
10. Hardjito D, Wallah SE, Sumajouw DMJ, Rangan BV (2004) On the
development of fly ash-based geopolymer concrete. ACI Mater J. doi:
10.14359/13485
11. Joseph B, Mathew G (2012) Influence of aggregate content on the behavior of
fly ash based geopolymer concrete. Sci Iran. doi:
10.1016/j.scient.2012.07.006
12. Olivia M, Nikraz H (2012) Properties of fly ash geopolymer concrete
821
designed by Taguchi method. Mater Des. doi: 10.1016/j.matdes.2011.10.036

13. Chindaprasirt P, Chareerat T, Sirivivatnanon V (2007) Workability and
strength of coarse high calcium fly ash geopolymer. Cem Concr Compos. doi:
10.1016/j.cemconcomp.2006.11.002
14. Allen AJ, Thomas JJ, Jennings HM (2007) Composition and density of
nanoscale calcium-silicate-hydrate in cement. Nat Mater. doi:
10.1038/nmat1871
15. Hajimohammadi A, Provis JL, Van Deventer JSJ (2011) The effect of silica
availability on the mechanism of geopolymerisation. Cem Concr Res. doi:
10.1016/j.cemconres.2011.02.001
16. Ismail I, Bernal SA, Provis JL, et al (2014) Modification of phase evolution in
alkali-activated blast furnace slag by the incorporation of fly ash. Cem Concr
Compos. doi: 10.1016/j.cemconcomp.2013.09.006
17. Hamouda AA, Amiri HAA (2014) Factors affecting alkaline sodium silicate
gelation for in-depth reservoir profile modification. Energies. doi:
10.3390/en7020568
18. Singh B, Ishwarya G, Gupta M, Bhattacharyya SK (2015) Geopolymer
concrete: A review of some recent developments. Constr Build Mater. doi:
10.1016/j.conbuildmat.2015.03.036
19. Ariffin MAM, Bhutta MAR, Hussin MW, et al (2013) Sulfuric acid resistance
of blended ash geopolymer concrete. Constr Build Mater. doi:
10.1016/j.conbuildmat.2013.01.018
20. Collins F, Sanjayan JG (1999) Strength and shrinkage properties of alkali-
activated slag concrete placed into a large column. Cem Concr Res. doi:
10.1016/S0008-8846(99)00011-3
21. Albitar, Ali M, Visintin, Drechsler (2017) Durability evaluation of
geopolymer and conventional concretes. Constr Build Mater 136:374–385 .
22. Chindaprasirt P, Homwuttiwong S, Jaturapitakkul C (2007) Strength and
water permeability of concrete containing palm oil fuel ash and rice husk-bark
ash. Constr Build Mater. doi: 10.1016/j.conbuildmat.2006.06.015
23. Bakharev T, Sanjayan JG, Cheng YB (2003) Resistance of alkali-activated
slag concrete to acid attack. Cem Concr Res. doi: 10.1016/S0008-
8846(03)00125-X
24. Davidovits PJ (2002) 30 Years of Successes and Failures in Geopolymer
Applications . Market Trends and Potential Breakthroughs . Geopolymer
2002 Conf. doi: 10.1017/CBO9781107415324.004
25. Van Deventer JSJ, Provis JL, Duxson P (2012) Technical and commercial
progress in the adoption of geopolymer cement. Miner Eng. doi:
10.1016/j.mineng.2011.09.009
26. Nuaklong P, Sata V, Chindaprasirt P (2018) Properties of metakaolin-high
calcium fly ash geopolymer concrete containing recycled aggregate from
crushed concrete specimens. Constr Build Mater 161:365–373 . doi:
10.1016/j.conbuildmat.2017.11.152
27. Hardjito D, Rangan BV (2005) Development and properties of low-calcium
fly ash-based geopolymer concrete. Res Rep GC 94
28. Hardjito D, Wallah SE, Sumajouw DMJ RB (2004) The development of fly
822
ash-based geopolymer concrete. ACI Mater J 101:467–72

29. Davidovits J (1996) From Ancient CONCRETE TO GEOPOLYMERS. Arts
Métiers No 180:1–6
30. Diaz-Loya EI, Allouche EN VS (2011) Mechanical properties of fly ash-
based geopolymer concrete. ACI Mater J 108:300–6
31. Ramujee K, Potharaju M (2017) Mechanical Properties of Geopolymer
Concrete Composites. Mater Today Proc 4:2937–2945 . doi:
10.1016/j.matpr.2017.02.175
823
Paper ID- 161
Structural Health Monitoring Using Wavelet Transform

Dr. Asif Husain
asifjmi@gmail.com, & ahusain2@jmi.ac.in
Professor, Civil Engineering Department, Faculty of Engineering & Technology,
Jamia Millia Islamia, New Delhi, India
ABSTRACT
Engineering structure damage detection has gained increasing attention from the engineering community since
unpredicted major hazards, most with human losses, have been reported. Aircraft crashes and the catastrophic
bridge failures are some examples. Security and economy aspects are the important motivations for increasing
research on structural engineering health monitoring. Since damage alters the dynamic characteristics of a
structure, namely its dynamic properties (natural frequencies, Eigen value, modal damping and modes of
vibration), several techniques based on experimental modal analysis have been developed in recent years. The
use of dynamic response in damage identification has been gaining considerable attention over the past few
decades. The aim of these methods is to detect the presence of a defect or a crack in components or structures.
This study focuses on using modal properties for the damage detection of single crack in simply supported
beam, with varying defect size and checking whether wavelets are able to detect crack of small size or not. The
direct use of modal properties such as natural frequencies and mode shapes is investigated for simply supported
beam, these properties are then used as a input for wavelet transform. The amplitude of wavelet coefficients at a
predefined damage location is found to be related to crack size; therefore sudden fluctuation in the amplitude
gives damage location.
Keywords: Engineering, Structures, Design, Damage detections, structural performance, structural

health monitoring (SHM)
1. INTRODUCTION
The Structural Health Monitoring consists to give, at every moment during the Life of a structure, a diagnosis on
the state of the constitutive materials, of the different parts, and of the full assembly of these parts constituting
the structure as a whole. The state of the structure must remain in a nominal domain resulting from the design. It
can be altered by the normal aging due to the usage and the action of the environment, and by accidental events.
Thanks to the time-dimension of monitoring, which allows considering the full history data base of the structure,
and to the help of the Usage Monitoring, it can also allow a prognosis (evolution of damage, residual life…).
Considering just the first function, the diagnosis, we could estimate that Structural Health Monitoring is a new
and improved way to make Non Destructive Evaluation. This is partially true, but Structural Health Monitoring
is much more. It involves the integration of sensors, possibly smart materials, data transmission, computational
power, and processing ability inside the structures. It allows reconsidering the design of the structure and the full
management of the structure itself and of the structure considered as a part of wider systems.
Structural damage detection has gained increasing attention from the scientific community since unpredicted
major hazards, most with human losses, have been reported. A method that covers the four steps of the process
of damage detection existence, localization, extent and prediction has not yet been recognized or reported. The
frequency-response function (FRF) curvature method encompasses the first three referred steps being based on
only the measured data without the need for any modal identification.
Higher operational loads, greater complexity of design and longer lifetime periods imposed to civil, mechanical
and aerospace structures, make it increasingly important to monitor the health of these structures. A wide variety
of highly selective non-destructive methods, using strain gauges, penetrating liquids, ultra-sound, visual
inspection, etc., are currently available for the detection of defects. Unfortunately, they are all localized
techniques, implying long and expensive inspection time; often, structural components are not inspected just
because of their inaccessibility and damage can propagate to critical levels between the inspection intervals.
824
The drawbacks of current inspection techniques have led engineers to investigate new methods for continuous
monitoring and global condition assessment of structures. That is the case for methods based on vibration
responses that allow one to obtain meaningful time and/or frequency domain data and calculate changes in the
structural and modal properties, such as resonance frequencies, modal damping and mode shapes, and use them
with the objective of developing reliable techniques to detect, locate and quantify damage. When damage occurs
in a structure the consequence is a change in one or more of its dynamic properties, namely the stiffness, mass
or damping. If one is referring to cracks, then it may be assumed that there will be a detectable change in
stiffness, the mass remaining unchanged and change in damping may be disregarded.
The previous decade has seen significant developments in the field of wavelet transform (WT). Today its
applications range from signal processing to image compression, from denoising to matrix multiplication etc. In
this work we are mainly concerned with the signal analyzing property of WT, and primarily its ability to extract
modal parameters (natural frequency, damping ratio and mode shapes) of a multi degree of freedom (MDOF)
system. By decomposing the signal in the time-scale plane, WT provides a bridge between the time domain and
the frequency domain representation of signals. Scale is a non-dimensional parameter inversely proportional to
the frequency in the signal. Traditionally, the data that was collected could either be represented in the time (or
space etc.) domain, or in the frequency domain by the use of Fourier series, or the discrete Fourier transform
(DFT). For many purposes these representations are enough. But there are applications when one would like to
go beyond the Fourier analysis. For example, the DFT plot of a signal which comprises of several sinusoidal
components would identify the frequencies of the sinusoids correctly but would not give any information of the
individual components stretch in time. This information, however, can be obtained using the time-frequency
representation of the signal.
A complete SHM approach consists of four basic steps which are required to be resolved sequentially.
(1) Identification of damage occurrence in the structure, if any,
(2) Identification of single or multiple damage locations
(3) Quantification of the level of damage, and
(4) Evaluation of structural performance and its useful remaining life. The idea is to do this reliably and
continuously, if possible, in a cost efficient manner. This study is concerned with the first two steps of
SHM, focusing on damage detection using wavelet transform.
1.1 Damage detection
 The need of damage detection and localization at the earliest possible stage becomes an important issue
throughout the aerospace, mechanical or civil engineering communities.
 The existence of damage in a structure results in changes of global dynamic characteristics.
 Therefore, relatively simple vibration measurements of a structural response and extraction of
information on natural frequencies, damping or mode shapes, make damage detection feasible.
 A relatively recent area of research in damage detection and localization is based on the wavelet
transform applied to mode shapes or static deflection data.
 The wavelet transform acts as a differential operator and can be applied effectively even for noisy
signals.
It is believed that the SHM problem is fundamentally one of statistical pattern recognition. This paradigm
(damage detection) can be described as a four-part process:
(1) Operational Evaluation,

(2) Data Acquisition, Fusion and Cleansing,
(3) Feature Extraction and Information Condensation, and
(4) Statistical Model Development for Feature Discrimination.
2. Literature Review
Cawley [1] not only specifies the type of damage, but he also quantifies the extent of detectable damage in terms
of the pipe diameter and wall thickness. Staszewski, et al. [2] demonstrate that temperature and ambient
vibrations can affect the performance of piezoelectric sensors employed in composite plate tests. Bartelds [3]
provides an example of a study where economic and life safety issues have been addressed. He states that the
direct costs of carrying out preventive inspections and the indirect costs associated with interrupted service
provide a strong stimulus for developing a SHM system for aircraft.
825
Sohn, et al. [4] summarizes a procedure for such normalization when direct measures of the varying input are
not available. When environmental or operating condition variability is an issue, the need can arise to normalize
the data in some temporal fashion to facilitate the comparison of data measured at similar times of an
environmental or operational cycle. As an example, Doebling and Farrar [5, 6] measured the temperature
differential across the deck of a bridge at 2 hr increments during a 24 hr cycle and correlate these measurements
with the change in the bridge’s natural frequencies.
In the context of civil engineering infrastructure, structural health monitoring (SHM) involves the development
and implementation of damage diagnosis and prognosis algorithms (Rytter, 1993; Farrar and Worden, 2007).
Recent research in wireless structural health monitoring (Straser and Kiremidjian, 1998; Lynch et al., 2004) has
shown that low cost micro eletro- mechanical (MEMS) sensors can be fabricated for measuring structural
response measurement thus allowing for a dense network of sensors to be deployed in structures.
The use of statistical signal processing and pattern classification techniques for damage detection in wireless
structural health monitoring has increased during the last decade (Sohnet al., 2001; Nair et al., 2006). In a recent
study performed, it has been shown that Haar wavelet coefficients of vibration signals are dependent on the
physical parameters of the system (Nair and Kiremidjian, 2009). However, it was observed that Haar wavelet
coefficients were not sensitive to minor damage [7].
Previous work in wavelet based system identification of non-linear structures was performed by Staszewski
(1998), Ghanem and Romeo (2000) and Kijewski and Kareem (2003). From a signal processing viewpoint,
initial work done was by Hou et al. (2000), where the discrete wavelet transform is used to study the transient
phenomenon when the stiffness of the structure is abruptly changed [8, 10].
Salawu, O.S., (1997) The use of natural frequency as a diagnostic parameter in structural assessment procedures
using vibration monitoring is discussed in the paper. The approach is based on the fact that natural frequencies
are sensitive indicators of structural integrity. In this paper, existing structural damage identification methods
based on dynamic characteristics of structures are examined [11].
Okafor A.C., Dutta A., (2000) In this paper the use of a laser-based optical system and wavelet transforms is
explored for the detection of changes in the properties of cantilevered aluminium beams as a result of damage.
The beams were modelled using the ANSYS 5.3 finite-element method and the first six mode shapes for the
damaged and the undamaged cases obtained. Damage was simulated by a reduction in the stiffness of one
element. Gaussian white noise was added externally to simulate field conditions. The results show that a
spatially-localized abnormality in the mode shape could be represented uniquely by a small set of wavelet
coefficients while the white noise was uniformly spread throughout the wavelet space [12].
Pai P. F., Young L.G., (2001) In this study, a WT damage detection method using Operational Deflection
Shape (ODS) difference is presented. A two-dimensional Continuous Wavelet Transform (CWT) is utilized to
decompose the difference between the ODS of the intact and damaged structure to detect and locate damage. To
demonstrate the ability of this technique, a numerical model of a steel plate is applied [13].
Nair, K.K. (2007) In this thesis, a damage detection algorithm based on the Morlet wavelet transform of the
vibration signal is presented and expression for the energies of wavelet coefficients (damage sensitive feature
vectors) using the Morlet wavelet basis is derived for a multi-degree of freedom system [14].
Ricardo Panciroli (2016) This work assesses the accuracy of a structural health monitoring (SHM) technique
previously presented in the literature. The SHM technique under exam relies on a modal decomposition method
and is intended to reconstruct the distributed deflection and strain fields from point-wise strain measurements at
known locations. The objective of this work is to assess the reliability of such algorithm to variations of sensors
number and their location. To this aim, the problem of a simply supported beam subjected to an impulsive
loading is solved analytically and the results are used as synthetic input data for the SHM algorithm [15].
3.0 Principle of Wavelet Transform

The Continuous Wavelet Transform is given by :
X = ∫ x(t) φ dt (1)
| |
826
Where:
 τ is the translation parameter.

 s is the scale parameter.
 φ is the Mother Wavelet.
Here are some of the most popular mother wavelets:
The Discrete Wavelet Transform (DWT) is given by:
g(t) = ∑ ∈ ∑ ∈ x, φ , .φ , (t) (2)
Where, φ , (t) = S φ(S t − nτ )
3.1 Wavelet Analysis
 A wavelet is a waveform of effectively limited duration that has an average value of zero.
 Wavelet analysis represents the next logical step: a windowing technique with variable sized regions.
 One major advantage afforded by wavelets is the ability to perform local analysis that is, to analyze a
localized area of a larger signal.
3.2 Wavelet Approach
 Many a time engineers are given with tasks to build structures which are required to support loads which
are dynamic in nature.
 For the analysis of such structures wavelet analysis to be used.
 The response of such structures recorded and plotted against time is generally known as time history.
 This particular signal will contain information about the structure and loads acting upon it in form.
This will be based on the post-processing of measurement data to detect anomalies from measurements (wavelet
decomposition, or any other transformation). After reaching to the conclusion (i.e. after ascertaining the best
algorithm from the entire developed algorithm), a comprehensive approach will be developed for:-
(1) Predicting the residual life of the structure.
(2) Risk assessment using structural health monitoring process.
(3) Using SHM during construction phase.
(4) Identifying the critical component of the structure, and suggest the retrofitting technique based on the
severity of the damage in that component.
3.3 Methodology
Whole work can be approached in three different phases.
(1) Developing a finite element model of any civil engineering infrastructure using any finite element
software (such as SAP2000, ABAQUS, ANSYS), then generating and acquiring the different responses
of the structure for further processing.
827
(2) Developing different algorithm for different techniques, and executing that algorithm over the response
generated by FEM software, in MATLAB.
(3) Experimentally verifying the developed algorithm.
4.0 SIMPLY SUPPORTED BEAM

The following studies on the simply supported beam modeled and are presented.
1. Damage detection of single crack in simply supported beam using any wavelet (damage size 10%, 30%
and 50%).
2. Damage detection of single crack in simply supported beam using mode shape curvature (damage size
10%, 30% and 50%). The Fig. 1 shows the model undamaged beam. The Fig. 2 represented the model
50% damaged beam at the center. Similarly 10% damaged beam and 30% damaged beam are FEM
modeled for the investigation.
3. Comparison of results obtained by the two methods.
Method Adopted and Result

1. The simply supported beams are modelled in SAP2000; The detail description of the beam data are
presented in Table 1 and Table 2, these beams are divided into 400 elements along its length.
2. Acquisition of modal parameter from SAP2000.
3. These modal parameters are then imported to MATLAB for further analysis.
4. Using Wavelet toolbox in MATLAB we generate 1D plot for variation of wavelet coefficient.
5. The comparison of 10%, 30% and 50% damaged beam with that of undamaged beam are presented in
terms of bar chart shown in Fig. 3 – Fig. 5.
Table 1 General Description of Models of Undamaged Beam and Description of 10% Damaged Beam
Description Undamaged Beam 10% Damaged Beam

Length of the beam 4m 4m
Cross-section of the beam 300mmx500mm 300mmx500mm
Material of the beam M30 and Fe415 M30 and Fe415
Loading on the beam 3kN/m 3kN/m
Damage in the beam No damage Damage length=100mm and

Cross-section of beam at
damaged
location=300mmx450mm
Fig. 1 FEM Model of Undamaged Simply supported Beam
828
Table 2 Description of 30% Damaged Beam and of 50% Damaged Beam
Description 30% Damaged Beam 50% Damaged Beam
Length of the beam 4m 4m
Cross-section of the beam 300mmx500mm 300mmx500mm
Material of the beam M30 and Fe415 M30 and Fe415
Loading on the beam 3kN/m 3kN/m
Damage in the beam Damage length=100mm and Cross- Damage length=100mm and Cross-section
section of beam at damaged of beam at damaged
location=300mmx350mm location=250mmx300mm
Fig. 2 FEM Model of 50% Damaged simply supported Beam
Table 3 Results obtained from FE analysis of different mode of damaged and undamaged beams
Mode No. Natural Frequency
Undamaged Damaged Beam

Beam
10% 30% 50%
1 45.642 45.613 45.467 45.205
2 173.85 173.85 173.83 173.8
3 363.93 363.76 362.5 361.35
829
Comparison Between 10% damaged Beam and Undamaged

Natural Frequency 400 Beam
300
200
100
0
1 2 3
Mode No.
Fig. 3 Comparison between 10% damaged and undamaged beam
Comparison Between 30% damaged Beam and Undamaged

400 Beam
350
300
Frequency
250
200
150
100
50
0
1 2 3
Mode No.
Comparison Between 50% damaged Beam and Undamaged Beam
400
350
300
Frequency
250
200
150
100
50
0
1 2 3
Mode No.
830
The different mode shape (First mode Shape) the simply supported beam are also obtained by finite element
analysis are shown in Fig. 6 to Fig. 8 for 10% damaged beam, 30% damaged beam and 50% damaged beam
respectively. By recording the sudden change in the wavelet coefficient in 1D plot we will predict the location
of damage. The Variation of wavelet coefficients along the length of 10 % damaged beam by using Haar
wavelet are shown in Fig. 9. Similar Variation of wavelet coefficients along the length of 30 % damaged beam
by using Haar wavelet and 50 % damaged beam by using Haar wavelet are observed. The Curvature Diagram
of 1st mode shape of 50% Damaged Beam is shown in Fig. 10. The curvature diagram of of 1st mode shape of
10% Damaged Beam and 30% Damaged beam are also obtained.
Fig. 6 1st Mode shape of 10% damaged beam
831
Fig. 9 Variation of wavelet coefficients along the length of 10 % damaged beam by using Haar wavelet
Fig. 10 Curvature Diagram of 1st mode shape of 50% Damaged Beam
5.0 Conclusion
Smart materials or sensors are a new development with enormous potential for structural health monitoring of
civil engineering structures. Some of them are currently being applied in the field, while others are being
evaluated under laboratory conditions. Although several methods for damage detection and localization have
been developed, they all have their own advantages, limitations, and scope of application. Most techniques
are based on a comparison between undamaged and damaged states of the materials. In the present study
following points are concluded.
 From the wavelet coefficient diagram (Fig.9) we observed that a sudden change occurs in the wavelet
coefficient at node no. 200 i.e. 2 m apart from the left support, which is the location of damage.
 It is found that mode shape curvature curve (Fig.10) is discontinuous at 2 m from the left support this
discontinuity in curvature curve defines the location of damage.
 Wavelets are very effective damage detecting tool. All types of cracks whether it is a minor crack i.e.10%
or a major crack i.e.50% can be detected by wavelets.
 Mode shape curvature method can detect the damages only when the variation in the curvature is high.
Sometimes variation in curvature is so small that discontinuity cannot be judged. Hence Mode shape
curvature method proves to be a good tool in detecting damages under higher modes.
 As we go for higher modes change in natural frequencies of undamged and damaged beam increases this
shows that damage can be detected very easily under higher modes. Wavelet analysis represents the next
logical step; a windowing technique with variable sized regions. The major advantage afforded by wavelets
is the ability to perform local analysis that is, to analyze a localized area of a larger signal.
832
6.0 References
[1] Cawley, P. (1997) “Long Range Inspection of Structures Using Low Frequency Ultrasound,” in Structural
Damage Assessment Using Advanced Signal Processing Procedures Proceedings of DAMAS ‘97, Univ. of
Sheffield, UK, 1-17.
[2] Staszewski, W.J., Biemans, C., Boller, C., and Tomlinson, G.R., (1999) “Impact Damage Detection in
Composite Structures-Recent Advances,” Structural Health Monitoring 2000, Stanford University, Palo Alto,
CA, 754-763.
[3] Bartelds, G., (1997) “Aircraft Structural Health Monitoring, Prospects for Smart Solutions from a European
Viewpoint” Structural Health Monitoring, Current Status and Perspectives, Stanford University, Palo Alto, CA,
293-300.
[4] Sohn, H., C. R. Farrar, N. Hunter and K. Worden, (2001) “Applying the LANL Statistical Pattern
Recognition Paradigm for Structural Health Monitoring to Data from a Surface-Effect Fast Patrol SOHN,
FARRAR, HEMEZ and CZARNECKI 7 Boat,” Los Alamos National Laboratory Report Los Alamos National
Laboratory report LA-13761- MS.
[5] Doebling, S. W. and C. R. Farrar, (1997) “Using Statistical Analysis to Enhance Modal-Based Damage
Identification,” in Structural Damage Assessment Using Advanced Signal Processing Procedures Proceedings of
DAMAS ‘97, Univ. of Sheffield, UK, pp. 199-210.
[6] Farrar, C. R and Worden, K. (2007). An introduction to structural health monitoring. Philosophical
Transactions of the Royal Society A, 365, 303-315.
[7] Ghanem, R. and Romeo, F. (2000). A wavelet based approach for identification of linear time varying
dynamical systems. Journal of Sound and Vibration, 234(5), 555-576.
[8] Nair, K. K., Kiremidjian, A. S. and Law, K. H. (2006). Time series based damage detection and localization
algorithm with application to the ASCE benchmark structure. Journal of Sound and Vibration, 291 (2), 349-368.
[9] Nair, K. K. and Kiremidjian, A. S. (2007). Time series based structural damage detection algorithm using
Gaussian mixtures modeling. ASME Journal of Dynamic Systems, Measurement and Control, 129(3), 285-293.
[10] Akbari, Ziarati, K.A multilevel evolutionary algorithm for optimizing numerical functions. International
Journal of Industrial Engineering Computations (2010) 2:419–430.
[11] Salawu, O.S., 1997, “Detection of structural damages through changes in frequency: A Review”,
Engineering Structures, Vol. 19, pp.718-723.
[12] Okafor A.C., Dutta A., (2000): Structural Damage Detection in Beams by Wavelet transforms, Smart
Materials and Structures, 9, 906-917.
[13] Pai P. F., Young L.G., (2001): Damage Detection of Beams using operational deflection shapes,
International Journal of Solid Structures.
[14] Nair, K.K. (2007):. Damage Diagnosis Algorithms for Wireless Structural Health Monitoring, Ph.D.
Dissertation, Department of Civil and Environmental Engineering, Stanford University
[15] Ricardo Panciroli (2016): Assessment of Structural Health Monitoring Technique through Synthetic Data
Generation , Faculty of Mechanical Engineering, Belgrade, University of Niccole Cusano from Italy.
833
Paper ID- 162
Numerical Simulation of Combined Pile-Raft Foundation under Horizontal

Loading
Mahalakshmi M✉, Hashini C D

Department of Civil Engineering, Bannari Amman Institute of Technology, Sathyamangalam, India
mahalakshmice@gmail.com
Abstract
Raft foundations generally have the ability to reduce differential settlement, contrarily causing excessive
settlement. In order to overcome this, piles are used along with the raft termed as combined pile-raft foundation
system. Due to the lack of availability of simplified tools and intricacy of work, the combined pile raft
foundation remains as an untouched area in research. This paper explores the performance of 2x2 numerically
simulated combined pile-raft foundation embedded in sand subjected to pure horizontal load using ABAQUS
3D. The effect of horizontal loading in terms of displacement is studied by varying the raft thickness, length to
diameter ratio and spacing to diameter ratio for the combined pile-raft foundation having a pile diameter of
500mm. The model is validated using the existing results. From this study, it is concluded that the effect of raft
thickness remains inconsequential whereas the length to diameter ratio and spacing to diameter ratio has an
impact on the static response of the combined pile-raft foundation system.
Key Words: Combined pile-raft foundation · Sand · Horizontal loading · ABAQUS 3D
1 Introduction
Population migration and explosion goes hand in hand leading to rapid urbanisation. This has resulted in the rise
of colossal multi-story buildings and skyscrapers, where the foundation of the structure and soil plays a major
role during the transfer of load. Deep foundations are generally used to withstand enormous loads and reduce
settlements on soil deposits of relatively weak to moderate nature. Thus, in order to overcome the settlement and
load transfer mechanism in such soils, a new technique called combined pile- raft system is used which provides
an effective way to restrict the settlement to permissible limit and safe transfer of loads to the subsoil.
Poulos [1] figured that the idea of combining the shallow and deep foundation was first proposed by Leonardo
Zeevaert for the construction of “Tower Latino Americana” in Mexico City. Katzenbach et al. [2,3] mentioned
the major design criterion to be the deformations and later marked that a safe and economic piled raft foundation
can be achieved by also considering the soil-structure interaction. Combined Pile raft foundation (CPRF) term
was coined by Katzenbach et al. [4] which consists of pile, raft and subsoil as the load bearing elements. After
idealizing the various soil profile for combined pile raft foundation, Poulos [5,6] concluded that a uniform soil
profile of relatively stiff clay or dense sand would be the favourable situations for CPRF. Analysis techniques
for pile group was first developed by Butterfield and Banerjee [7] as mentioned by Randolph [8], then the
analysis technique was developed for pile raft foundation by Davis and Poulos [9] and further simplified
analysis methods were also developed by Hooper [10], Sommer et al. [11], Franke at al. [12], Ai et al. [13].
Another form of design approach in which the piled raft foundation was modelled as plate loaded spring was put
forth by Griffith et al. [14], Clancy and Randolph [15], Kitiyodom P and Matsumoto [16-18].
834
Finite Element Method analysis was first used by Hooper [10] to understand the complex piled raft foundation
behaviour. The hybrid method was also used to analyse the complex behaviour of the piled raft which combined
one or two numerical analytical methods. Ta and Small [19,20], Chow and Small [21] compared Finite Element
Method (FEM) and Finite Layer Method (FLM) to investigate the pile raft behaviour. Burland and Kaira [22]
and Prakoso and Kulhawy [23] performed 2-D plane strain finite element analysis for elastic and elastic-plastic
models to compare the settlement and to design optimum pile raft respectively. Patil et al. [24] experimentally
studied the behavior of eccentrically loaded piled raft embedded in sand and proposed that addition of piles near
the edges of the raft increases the load carrying capacity and prevent the CPRF system from tilting. In addition
to the earlier works carried out on laterally loaded CPRF, a focus on the effect of static horizontal loading and
its behavior is studied using numerical simulation.
2 Validation of Model
CPRF was modelled and analysed using a finite element software ABAQUS 3D. Kumar et al. [25] has modelled
a 2x2 CPRF in PLAXIS 3D using the 1-g experimentation conducted by Matsumoto [26] for a series of vertical
and horizontal loading. The soil properties chosen for the experiment were based on the consolidated drained
triaxial test carried out by Matsumoto et al. [27]. The present model is validated by comparing the modelling
and analysis of CPRF under horizontal loading. The normalized lateral displacement of the present numerical
model obtained was found to be in accordance with Kumar et al. [25] and hence the current numerical model
can be used for assessing the performance of CPRF under different raft thickness, length to diameter ratio and
spacing to diameter ratio.
Fig. 1(a). Kumar et al. [25] Fig. 1(b). Present Study
835
3 Finite Element Modelling
The static response of 2x2 Combined Pile-Raft Foundation is studied by simulating the model in a finite element
software package ABAQUS 3D using the guidelines from Hibbit et al. [28]. The model consist of a soil
continuum, piled raft, interaction and loadings. A length to diameter ratios(L/d) of 20, 30 and 40 were adopted
terming L to be the length of the pile and d to be the diameter of the pile. The spacing to diameter ratio(s/d)
adopted were 3, 5 and 7 where s is the spacing of the piles. The raft thickness was varied from 500mm to
1000mm. The diameter of the pile was assumed to be 500 mm.
Fig. 2. ABAQUS Model
The dimension of the soil model is varied according to the size of the Combined Piled-Raft Foundation. The size
of the soil was modelled to take 1.5 times the length of the pile vertically as reported by Kumar et al. [29] and 5
times the width of the raft horizontally. The soil was assumed to take Mohr-Coulomb plasticity criterion and
was modelled to be elasto-plastic. The piled raft was modelled to be linear elastic. The properties of the soil and
piled raft were defined as in Table 1. The soil and piled raft were modelled using 3D deformable solid parts that
were extruded. The loading was given in two steps having an initial step and a static, general step.
An isotropic directional contact having a tangential mechanical behaviour with a penalty type friction
formulation having a coefficient of friction 0.3 was assigned. A standard surface to surface interaction with a
path contact tracking was used to provide a master-slave constraint. A finite sliding formulation having no slave
adjustment and surface smoothing is used to create the contact pair. The master and slave surface were the
surface area of the pile and the soil element in contact respectively. The bottom nodes of the soil continuum
were completely arrested and the side face nodes are restricted from moving horizontally. A standard element of
a quadratic geometric order having a 10-node quadratic tetrahedron (C3D10) element was used.
836
Table 1. Properties of Soil and Piled Raft
Property Soil Piled Raft

Young’s Modulus E (kN/m2) 40,000 4,00,00,000
Poisson’s ratio ν (no unit) 0.3 0.2
Mass Density γ (kN/m ) 3
16.3 24
Angle of internal friction ϕ 31º
Dilatancy Angle ψ 1º
2
Cohesion (kN/m ) 10
4 Parametric Study
The parametric studies were carried out on the effectiveness of thickness of raft, length to diameter ratio and
spacing to diameter ratio under horizontal loading using ABAQUS 3D.
4.1 Effect of Raft Thickness
To find the effect of raft thickness on a CPRF, the length to diameter ratio and spacing to diameter ratio were
taken as constant. The horizontal load applied on the raft acts eccentrically being the vertical loads absent. This
causes an excessive deflection of the combined piled raft system. Even though the increase in raft thickness
increases the load carrying capacity, it does not improve the settlement characteristics of the combined piled raft
foundation. The thickness of raft does not remarkably reduce the settlement. This phenomenon also shows that
raft is the load carrying element and hence the pile aid in reducing settlement. As shown in fig. 3(b), the change
in the thickness of the raft does not have a larger difference in response of combined piled raft system which is
in accordance with Oh et al. [30].
15
15
Displacement (mm)
Displacement (mm)
12
9 10
t=500mm
6
t=750mm 5
3
t=1000mm
0 0
2 4 6 8 10 500 750 1,000
Load (kN) Raft Thickness (mm)
Fig. 3(a). Effect of Raft Thickness Fig. 3(b). Effect of Raft Thickness
4.2 Effect of Length to Diameter Ratios
The length to diameter ratios that were adopted were 20, 30 and 40. For analysing the effect of length to
diameter ratios, an optimum raft thickness of 500mm is used. It is notable that slender piles reduce settlement
feebly on comparing to piles with smaller length to diameter ratios due to the eccentricity in loading.
837
15
Displacement (mm)
12
6
L/d - 20
3 L/d - 30
L/d - 40
0
2 4 6 8 10
Load (kN)
Fig. 4. Effect of length to diameter ratio

Since the loading is purely horizontal, a higher degree of eccentricity ensue while transferring load. This dictates
that piles with larger length to diameter ratios are not economical to serve as settlement reducers. In the case of
vertical loading, interfacial stress mobilises at higher depths which is not applicable in our case. The length to
diameter ratio has spurious claims as it has detrimental effects which are mainly due to the reduction in load
carrying capacity after a certain amount of settlement. The pile group in the sand is end bearing hence the
peripheral area plays only a very meagre role. The stress contour as shown in Fig. 5. shows that the stress
concentration is predominant near the piled raft which is mainly due to the impact of material property. In other
words, the Young's Modulus of the pile and soil affect the stress concentration.
Fig. 5. Stress Contour
4.3 Effect of spacing to diameter ratios

In mandate to understand the effect of spacing to diameter ratio of piles, three different spacing of 3d, 4d and 5d
were adopted. It is apparent that the increase in the spacing of the piles consequently increases the lateral
dimensions of the raft. Hence, under pure horizontal loading, the raft performance is notably good when
compared to that of lower spacing which may also be the consequence of the increase in load bearing volume of
the raft. Although the settlement reduces as the spacing increases the raft starts behaving like a simple mat
foundation as the load transferred to the soil beneath increases thereby losing the interaction between the piles
and they start acting like individual piles.
838
This shows that the raft proffers carrying load and piles the settlement reducers. The increase in spacing would
be uneconomical for higher ratios as the element does not act as a combined piled raft foundation. As the
spacing increases, settlement increases which are speciously due to a relatively lesser number of piles that aid in
reducing settlement. The stresses in the rear pile row are found to decrease as the spacing between the piles
increases which may be due to the shadow cast on the rear row which is an effect of the overlap of stresses
(shadowing effect).
15
Displacement (mm)
12
6 s/d - 3
s/d - 4
3
s/d - 5
0
2 4 6 8 10
Load (kN)
Fig. 6. Effect of spacing to diameter ratio

5 Conclusion
The static response of combined piled raft foundation under horizontal loading in the sand has been studied in
this paper. Using ABAQUS 3D, the combined piled raft foundation system was modelled having a soil
continuum with properties of sand using the Mohr-Coulomb plasticity model. The effect of raft thickness was
insignificant in terms of displacement when subjected to horizontal loads. The consequence of the length to
diameter ratios on the displacement was notable in piles having a greater length to diameter ratios as they
behaved differently when subjected to horizontal loading compared to short piles which were a result of
eccentricity. Likewise, the larger spacing to diameter proves to be inefficient for CPRF, since fails to validate
the inception. Thus, the displacement under larger L/D and S/D ratios needs cautious insight so as to evade the
failure due to higher stress concentration around the pile and to neglect the incompetency in the design of CPRF
in terms of horizontal loading.
6 References
1. Poulos HG (2005) Piled raft and compensated piled raft foundations for soft soil sites. Advances in Designing
and Testing in Deep Foundations Engineering, Geotechnical special publication 129, ASCE, Reston, VA,
214-234. doi: 10.1061/40772(170)2
2. Katzenbach R, and Reul O (1997) Design and performance of piled rafts. Proc. XIV ICSMFE, Balkema,
Rotterdam, Netherlands, 2253– 2256.
3. Katzenbach R, Arslan U, and Moormann C (2000) Piled raft foundation projects in Germany. Design
applications of raft foundations, Thomas Telford, London, 323–391. doi:10.1680/daorf.27657.0013
4. Katzenbach R, Schmitt A, Turek J (2005) Assessing Settlement of High-rise Structures by 3D Simulations.
Computer-Aided Civil and Infrastructure Engineering. vol.20, 221-229. doi: 10.1111/j.1467-8667.2005.00389
839
5. Poulos HG (2001) Piled Raft Foundations: Design and Applications. Géotechnique 51, (2): 95-113.
doi: 0.1680/geot.2001.51.2.95
6. Poulos HG (2001) Methods of Analysis of Piled Raft Foundations. TC18 Report, Int. Society of Soil
Mechanics and Geotechnical Engineering. doi: 10.4028/www.scientific.net/AMR.838-841.334
7. Butterfield R and Banerjee P K (1971) The Elastic Analysis of Compressible Piles and Pile Groups.
Géotechnique 21, No. 1,43-60. doi:10.1680/geot.1971.21.1.43
8. Randolph M F (1983) Design of piled raft foundations. Proc., Int. Symposium on Recent Developments in
Laboratory and Field Tests and Analysis of Geotechnical Problems, 525–537.
9. Davis E H and Poulos H G (1972) The Analysis of Piled Raft Systems. Aust. Geomechs. J., G2: 21-27.
10. Hooper J A (1973) Observations on the Behavior of a Piled- Foundation on London Clay. Proceedings of the
Institution of Civil Engineers ,E-ISSN 1753-7789 Volume 55 Issue 4, PART 2, pp. 855-877.
doi:10.1680/iicep.1973.4144
11. Sommer H, Wittmann P and Ripper P (1985) Piled Raft Foundation of a Tall Building in Frankfurt Clay.
Proc.11 ICSMFE, San Francisco, 4:2253-2257.
12. Franke E, Lutz B and El-Mossallamy Y (1994) Measurements and Numerical Modelling of High-Rise
Building Foundations on Frankfurt Clay. Geotechnical Special Publication 40, ASCE, 2: 1325-1336, New York:
American Society of Civil Engineers.
13. Ai Z Y, Han J and Yan Y (2005) Elastic analysis of single pile-rigid circular raft system in layered soils.
Advances in Deep Foundations, Geotechnical special publication 132, ASCE, Reston, VA, 1–14.
doi: 10.1061/40778(157)24
14. Griffiths D V, Clancy P and Randolph M F (1991) Piled raft foundation analysis by finite elements. Proc.
Int. Conf. of the International Association for Computer Methods and Advances in Geomechanics (IACMAG
91), Cairns, Queensland, Australia, (eds. G. Beer et al.), Pub. Balkema, pp.1153-1157.
15. Clancy P and Randolph M F (1993) An Approximate Analysis Procedure for Piled Raft Foundations.
International Journal of Numerical and Analytical Methods in Geomech., London, 17(12), 849-869.
doi: 10.1016/0148-9062(94)91196-7
16. Kitiyodom P and Matsumoto T (2002) A simplified analysis method for piled raft and pile group
foundations with batter piles. Int. J. Numerical and Analysis Methods in Geomech., 26(13), 1349–1369.
doi: 10.1002/nag.248
17. Kitiyodom Pile and Matsumoto T (2003) A simplified analysis method for piled raft foundation in non-
homogeneous soils. International Journal of Numerical and Analytical Methods in Geo-mechanics; 27:85-109.
doi: 10.1002/nag.264
18. Kitiyodom P and Matsumoto T (2005) A Simplified Analysis Method for Piled Raft Foundations Subjected
to Ground Movements Induced by Tunneling. Int. J. Numer. Anal. Meth. Geomech., 2005; 29:1485–1507.
doi: 10.1002/nag.469
19. Ta L D and Small J C (1997) An Approximation for Analysis of Raft and Piled Raft Foundations.
Computers and Geotechnics, vol. 20, no. – 2, 105-123. doi: 10.1016/S0266-352X(96)00012-2
20. Ta L D and Small J C (1996) Analysis of Piled Raft Systems in Layered Soils. International Journal of
Numerical and Analytical Methods in Geomech.,20, 57-72.
840
21. Chow H and Small J (2006) Settlement of a Piled Raft Foundation Considering Lateral Pile Resistance.
Computers and Geotechnics. Vol. 26, pp. 1 - 21.
22. Burland J B, Kaira J C (1986) Queen Elizabeth II Conference Centre, Geotechnical Aspects, Proc. ICE, part
I, No.- 80, 1479 - 1503.
23. Prakoso W A and Kulhawy F H (2001) Contribution to piled raft foundation design. Journal of Geotechnical
and Geoenvironmental Eng., 10.1061/(ASCE)1090-0241(2001)127:1(17), 17–24.
doi: 10.1061/(ASCE)1090-0241(2002)128:8(709)
24. Patil J, Vasanwala S A and Solanki C H (2016) An experimental study of eccentrically loaded piled raft.
International Journal of Geotechnical Engineering, 10:1, 40-45. doi:10.1179/1939787915Y.0000000006
25. Kumar A, Choudhury D and Katzenbach R (2015) Behaviour of Combined Pile-Raft Foundation (CPRF)
under Static and Pseudo-static Conditions using PLAXIS 3D. 6th International Conference on Earthquake
Geotechnical Engineering, Christchurch, New Zealand.
26. Matsumoto T (2014) Implication for Design of Piled Raft Foundations subjected to Lateral Loading.
Advances in Foundation Engineering; ISBN: 978-981-07-4623. doi: 10.3850/978-981-07-4623-0_KN-08.
27. Matsumoto T, Fujita M, Mikami H, Yaegashi K, Arai T and Kitiyodom P (2010) Load tests of piled raft
models with different pile head connection conditions and their analyses. Soils and Foundations, 50(1):63–81.
28. Hibbit H D, Karlsson B L and Sorrensen P (2007) Abaqus theory manual. SIMULIA, Providence, RI.
29. Kumar M A, Deepankar Choudhury M and Katzenbach R (1988). ASCE 04016013-2 International Journal
for Geomech.
30. Oh E Y N, Huang M, Surrak C, Adame R and Balasubramaniam A S (2008) Finite element modelling for
piled raft foundation in sand. Eleventh East Asia-Pacific Conference on Structural Engineering & Construction
(EASEC-11), Building Sustainable Environment, Taiwan, pp. 1-8.
841
Paper ID- 163
Fluid-Structure
Interaction Problem in
Elevated Water Tank
Sutanuka Nath1 and Atanu Kumar Dutta2

1P.GStudent, Department of Civil Engineering, 2Associate Professor, Department of Civil
Engineering, Jorhat Engineering College, Jorhat, Assam-785007, India
1
Email- sutanuka888@gmail.com
Abstract. This paper investigates the Fluid-structure interaction effect in the

elevated water tank due to seismic excitations. The intz type water tank design
details of which is available from Krishna Raju (2015), has been adopted for this
study and is modelled in ANSYS Workbench® using finite elements for solid as
well as fluid elements. The bases of tanks are considered to be fixed. Time
history analysis is performed considering multiple earthquakes. A comparison
has been drawn between the gravity load analysis and the time history analysis.
It is found from the study that the structure is safe under gravity load but it fails
under seismic load.
Keywords: Fluid-Structure Interaction, Elevated Water Tank, Time History

Analysis.
1 Introduction
Elevated water tanks are one of the most essential structures constructed for
distribution of water and storage purpose according to necessity. In past decades there
are many severe cases of failure of elevated water tank during earthquake hazards.
During ground motion movable fluid inside the tank produces hydrodynamic pressure
on the wall, which may cause permanent deformation of the tank. This interaction is
known as fluid- structure interaction. Failure of elevated water tank depend on various
reasons such as geometry of tank, material used, supporting system, soil failure, fluid
sloshing, characteristics of earthquake etc.
Fluid sloshing is a phenomenon of fluid-structure interaction. In liquid storage
tanks fluid sloshing gives maximum impact in case of failure mechanism of tank. To
check this sloshing effect or slosh-dynamics, computational fluid structure simulation
is carried out. There are many cases of elevated tank structures getting collapsed due
to sloshing effect during an earthquake.
In this study the model of elevated water tank which is considered from Krishna
Raju (2015) is checked under gravity loading and also seismic analysis is carried out
considering three representative earthquakes.
This study attempts to address the fluid-structure interaction in elevated water tank
considering different ground motions. It also compares between static load analysis
and seismic analysis.
842
2 Literature Review
The concept of fluid-structure interaction as a spring mass model is first introduced

in the 1950s. Single spring mass model takes the completely filled tank as a lumped
mass and the vertical shaft is considered as cantilever beam as shown in (Fig.1).
Vertical sloshing of water is neglected and hence the structure fails when there is
sloshing inside the tank.
Fig.1. Elevated tanks and the single lumped-mass model after Dogangün and Livaoglu(2004)
In 1963 Housner studied the behaviour of Fluid-structure interaction as a double

spring mass model for fixed base elevated tanks (Fig. 2). Fluid inside the tank
divided into two parts, upper region of water is considered as convective mass and
lower region is considered as impulsive mass. Sloshing of water takes place in
convective mass of water and impulsive mass moves along with the tank. In this
model Convective mass of water considered as lumped mass which is connected to
the walls of tank by springs of stiffness Kc and impulsive mass is rigidly connected
to the walls. Another model was proposed by FEMA guidelines. Convective mass of
water considered as lumped mass which is connected to the walls of tank by springs
of stiffness Kc and impulsive mass is rigidly connected to the walls.
Fig.2. Two-mass model for the elevated tanks after Housner (1963)
Another model was proposed by FEMA guidelines. Convective mass of water

considered as lumped mass which is connected to the walls of tank by springs of
stiffness Kc and impulsive mass is rigidly connected to the walls. Additional higher-
mode convective masses may also be included for the ground-supported tanks, as
shown in (Fig.3). But for most accurate analysis higher modes of masses are
neglected as higher modes of sloshing has negligible influence on the forces exerted
by the container wall.
843
Fig.3. Spring-mass analogy for ground supported cylindrical tank after Dogangün and
Livaoglu(2004)
In the above literature fluid is modelled as spring-mass model to check FSI, which is
quite difficult in case of three dimensional modelling. Visualization of fluid-
structure interaction is checked by modelling the water using different finite fluid
elements.
3 Modelling
The fixed base elevated water tank is modelled using ANSYS Workbench®. The water inside
the tank and the frame staged elevated water tank are modelled with proper finite elements.
The model of frame staged elevated water tank is adopted from Krishna Raju N.
(2015). The capacity of elevated water tank is 1 million liters supported on 16 meter
long frame staging system with 8 columns and the Diameter of cylindrical section of
tank is 12 meter. M20 concrete grade is considered for modelling the concrete
structure. The dimensions of various components are given in Table1. The water tank
is checked under both gravity loading and seismic loading considering three
representative earthquakes. Time history analysis is performed through “TRANSIENT
STRUCTURE” tool and gravity load analysis is performed by “STATIC
STRUCTURE” tool in ANSYS Workbench®.
Table 1. Sizes of various components of Intz type tank after Krishna Raju (2015)
Components Dimensions
Top Dome 120 mm thick

Top Ring Beam 300 mm×300 mm
Cylindrical Wall 200 mm thick
Bottom Ring Beam 300mm×300mm
Circular Ring Beam 1200mm×600mm
Bottom Dome 300 mm thick
Conical Dome 600 mm thick
Braces 500mm×500mm
Radius of cylindrical Section 6000mm
Physical and mechanical properties of the materials of the structure are listed in below
Table 2.
844
Table 2. Material properties after L. Kalani Sarokolayi et al. (2008)
Material Specific mass Modulus of Poisson’s

Kg/m3) Elasticity (N/m2) ratio
Concrete 2400 2.23× 1010 0.27

Water 1000 Bulk modulus -
= 2.2 × 109
2.3 Description of modelling in ANSYS Workbench®

Design modeller is used for building the geometry. Material and geometric properties
are assigned. Appropriate mesh is generated. Depending upon the meshing method,
sizing different solid elements are assigned by default. Boundary condition is assigned.
Contact pair between different types of solid material is program controlled.
SOLID187 is used in modeling the concrete elevated water tank and staging using
coarse mesh ascribing different material properties. SOLID187 is a tetrahedral
element. While modelling this elevated water tank TARGE170 is taken as target
element and CONTA174 is considered as contact element. These surface to surface
contact elements allow modelling of fluid pressure penetration loads. These interfaces
are capable of performing fluid-structure interaction. The fluid element is capable of
exchanging pressure and deformation properties with solid concrete elevated tank
structure. As visualization of fluid-structure interaction is effective with sloshing effect
of water during free mode vibration, water inside the tank is modelled with FLUID80
element.
3D frame-staged elevated water tank is shown in below (Fig.4) modelled using

ANSYS Workbench® in tank full condition.
Fig. 4. 3D model of frame staging elevated water tank considering fixed base
2.4 Time History Data

Time history analysis is computed for three representitive earthquakes. These are El-
Centro (18th May, 1940), Kobe (17th january,1995) and Kocaeli (18th May, 1999)
earthquake.
845
These ground motions are considered on the basis of their Maximum peak ground
acceleration.
Ground acceleration data for north-south component of El-Centro is obtained from
http://www.vibrationdata.com/elcentro.htm. Kobe and Kocaeli earthquake database
are selected from Pacific Earthquake Engineering Research Next Generation
Attenuation (PEER-NGA) strong motion database records available online at
http://peer.berkeley.edu/nga. Table 3 shows the ground motion records.
Table 3. Gound motion characteristics
Different ground Station Magnitude Time Total PGA(g)

motions steps time
(sec) Period
(sec)
El-Centro Imperial Mw=6.9 0.02 31.14 0.319
(1940) valley
Kobe Takarazuka, Mw = 7.2 0.02 40 0.8
(1995) Japan
Kocaeli Sakaria Mw= 7.6 0.01 20 0.628
(1999) station
Turkey
The input acceleration-time plot of El-centro earthquake, Kobe, Kocaeli earthquake

are shown in Figure 5 through Figure 7.
Fig. 5. Accelerogram plot of El-Centro(1940) earthquake
Fig. 6. Accelerogram plot of Kobe (1995) earthquake
846
Fig. 7. Accelerogram plot of Kocaeli (1999) earthquake

Static load analysis is done to check the safety of structure against gravity. Fig.8 shows
the equivalent stress on elevated water tank due to static load.
Fig.8. Equivalent stress on elevated water tank due to gravity load
Maximum stress in concrete structure is 2.57 N/mm2. Maximum limit of stress in

concrete under compression is fck/ 1.5 =13.33 N/mm2 (considering M20 grade of
concrete). The structure is safe under gravity loading.
Time history analysis is performed through “TRANSIENT STRUCTRE” tool in

ANSYS Workbench® considering three representative earthquakes. The acceleration is
applied in horizontal X. The maximum equivalent stresses in concrete due to time
history analysis are computed. The maximum equivalent stresses in both the concrete
structures due to El-Centro (1940), Kobe (1995), Kocaeli (1999) earthquakes are
shown in table 4.
Table 4. Maximum equivalent stress on concrete due to three representative

earthquakes
847
Name of the Maximum Equivalent stress in

earthquake frame staging (N/mm2)
El-Centro 62.0
earthquake(1940)
Kobe earthquake(1995) 33.90
Kocaeli 2.90
earthquake(1999)
(i) Against El-Centro and Kobe earthquake stress values exceeds maximum
limit of compression in concrete, i.e 13.33N/mm2
(ii) Agianst Kocaeli earthquake stress values on concrete are within permissible
limit.
It has been observed that the structure is not safe for seismic activity as per time
history analysis considered for two representative earthquakes. Considering
Kocaeli(1999) earthquake, the structure is safe. It is clear that structural configuration
as per Krishna Raju (2015) was arrived at using static analysis only and hence cannot
be used for earthquake prone region.
5 CONCLUSIONS
The paper concludes that frame-staged elevated water tank is not safe for seismicity as
per time history analysis with two representative earthquakes. But it is safe under
Kocaeli (1999) earthquake. It is clear that structural configuration as per Krishna Raju
(2015) was arrived at using gravity load analysis only and hence cannot be used for
severe earthquake prone region. The equivalent stress on concrete structure due to El-
Centro (1940) earthquake is more comparing to Kobe (1995) earthquake.
6 REFERENCE
1. ANSYS Theory Manuel, 1994. Edited by Peter Kohnke, Twelfth Edition. SAS IP,
Inc, pp.1266.
2. Ahmad S., Al-Hussaini, T. M. & Fishman, K. L. 1996. Investigation of active isolation
of Machine Foundations by Open Trench, Journal of Geotechnical Engineering, 117
(4), 616—636.
3. Dutta, S.C., Jain, S.K. & Murthy, C.V.R., 2000. Assessing the seismic torsional
vulnerability of elevated tanks with RC frame-type staging, Soil Dynamics and
Earthquake Engineering, vol.19, P.P. 183-197.
4. Housner G.W. 1963. Dynamic analysis of fluids in containers subjected to
acceleration, Nuclear Reactors and Earthquakes, Report No. TID 7024, U. S. Atomic
Energy Commission, Washington D.C.
848
5. Krishna Raju N., 2015. Advanced Reinforced Concrete Design (IS: 456-2000)
(English) 3rd Edition, CBS Publisher.
6. Kalani Sarokolayi, L., Navayineya, B., Hosainalibegi M., & Vaseghi Amiri. J., 2008.
Dynamic analysis of water tanks with interaction between fluid and structure, The 14th
World Conference on Earthquake Engineering.
7. Kumar D. & Patel B. 2016. Response of overhead water tank staging considering
Fluid-Structure-Soil Interaction, International Journal of Advanced Engineering and
Research Development, Vol. 3, Issue 4.
8. Livaoğlu, R. & Doğangün, A. 2005. Seismic evaluation of fluid-elevated tank-
foundation/soil systems in frequency domain, Structural Engineering and Mechanics
21: 1, 101-119.
849
Paper ID- 164
Buckling Analysis of braced frames under axial and lateral

loadings: The Effect of Bracing Location on Frame Configuration
Narayan*(0000-0003-0254-3490), Krishna Kant Pathak (0000-0001-9853-6054)
Department of Civil Engineering, IIT (BHU), Varanasi 221005 India

Emails: * narayan.rs.civ17@itbhu.ac.in, kkpathak.civ@itbhu.ac.in
Abstract: In this study, buckling analyses of 45 frames are carried out considering various combinations of 3
bays and 3 stories with and without braces under lateral and vertical loads separately. The main objective of
the study is to investigate the effect of particular bracing location on the overall buckling behavior of frames
incorporating a single diagonal brace or an X (cross) brace in each loop of the frame individually. Beams,
columns and braces are considered to have the same cross-section. The effects are compared for bare frames,
fully braced frames, one other bracing type and braces having same slenderness as that of beams and
columns. It has been found that for some cases braces may provide strength to the frame but it may cause pre-
mature buckling failure at buckling loads less than that for bare frames, so may not be beneficial in all the
cases as far as lateral loads are concerned. Linear perturbation buckling analysis has been conducted using
Abaqus software considering two node cubic beam element for each member. Such type of analysis is useful
for the safe design of framed structures against buckling.
Key-words: Frame buckling, Bracing level, Ordinary brace, Lateral loading, Vertical loading, Eigen modes,
Critical load, Stability, Slenderness, Diagonal brace, X brace, and Chevron brace.
1. Introduction
The braced frames are used to provide strength and stability to the structures. Researches have
been done to improve the buckling behavior of braced frames using various arrangements and
techniques. Buckling is an event which doesn’t depend on the strength of structure. Under
compressive loading the slender members are prone to buckling. As long as the load is small, axial
shortening of the member is observed and once the certain critical load is reached, a sudden out-
sideways bowing of the member occurs giving rise to large deformation, in turn causing the
collapse of the member [1]. This critical load is the limiting load under which axial compression in
an unbent configuration is possible. The transition from straight to bent configuration takes place
because at the buckling load straight configuration ceases to be stable. So the buckling load marks
the limit of stability. The essence of buckling is the disproportionate increase in displacement as a
result of small increase in load. This implies that buckling analysis is a subtopic of nonlinear
mechanics. Stress and strains are assumed to obey hooks law; therefore the nonlinearity is purely
geometrical [2]. So ascertaining the working of structure by preventing buckling of its members is
very complex and need to be taken up very seriously in analysis and design.
Egor P. Popov [3] suggested employing concentric braces when Moment resisting frames (MRFs)
cannot be designed sufficiently stiff economically for resisting wind forces. In the study the main
focus was on the importance of links used in eccentric braces compared the influence of link
length on the stiffness of braced frames. MRFs are found to have minimum stiffness, and
concentrically braced frames (CBFs) are found to have maximum stiffness but as the behavior of
850
CBFs is unreliable under cyclic loadings so eccentric braces with short links were appreciated.
Rafael Sabelli [4] gave an equation of frame behavior based on statical relations to estimate brace
deformation demand. A. Moein Amini et al [5] considered multi-storied buildings using concentric
X, V and chevron braces for two bracing patterns. One pattern is throughout bracing of 2 central
bays (adjacent) and other is of 2 end corner bays (non-adjacent). Modeling was done using RAM-
perform 3D software. For all braces non-adjacent bays were found to have lower stiffness but
higher strength in comparison to adjacent bays configuration. Hyung-Joon Kim et al [6] reviewed
the design requirements of beams and columns in braced frames, considering inverted V (chevron)
braced and suggested to use bigger sized column to prevent weak stories at lower stories in
ordinary CBFs (OCBFs).
The subject matter of this article is the effect of bracing location on the buckling behaviour of steel
braced frames. The behaviour is examined in terms of critical buckling load values obtained from
numerical analysis using linear perturbation buckling analysis method which is based on
eigenvalue problem. The analysis has been done using Abaqus CAE 6.14 [7] to model the frames
and to get the Eigenvalues referring to buckling load values. The frames are subjected to
compressive loads in vertical and horizontal (lateral loading) direction separately. As this study is
purely based on buckling behaviour of braces and the draft NBC of India [8] emphasises so much
on braces and demanded some suggestions and comments, so this study can be considered as a
suggestive report for better design and provisioning of braces.
2. Geometrical and material properties

2.1 Frame Configurations
2D frames having all the members being rigidly connected with each other including braces, are
considered. All frames have fixed end supports. All beams and columns are of length 3 metres.
The arrangement of braces is such that all the braces are concentric. All members have same cross-
section, including braces. For various combinations of 3 bays and 3 stories, frame cases with and
without braces have been considered for accessing the effect of each braces at each bracing
location in a frame by incorporating single diagonal brace or X brace in each loop of frame
individually. For comparison, full braces are provided for all above mentioned cases using single
diagonal and X braces separately and for comparing with other type of bracing chevron brace is
considered. For another comparison, single diagonal brace has been provided with same
slenderness as that of other structural members for combinations of single bay and three stories.
2.2 Material Properties
Steel is used as material for construction of all the members of frame. Elastic properties of steel
used are, Young’s modulus, E= 2e11 and Poisson’s ratio= 0.3, since effects of self-weight
of frame are neglected and are assumed to be incorporated in vertical loading as a part of it, so the
density of steel has no effect on buckling load value.
2.3 Profile and Section
Circular profile with radius of 0.05 m is used for all structural members including braces keeping
cross-section of each member same. For comparison a case for braces having slenderness equal to
that of beams and columns is taken, where radius of 0.07071 m is used. Homogeneous sections are
used with sectional properties as, E= 2e11 , Shear modulus, G= 7.7e10 .
851
2.4 Nomenclature and representation of model

For the study of effect of location of brace on the buckling behavior of braced frames, single brace
is placed individually at a particular location. For defining nomenclature to the bracing locations
they are numbered as shown in Fig. 1, where Br represents brace and Nbr represents no brace.
To denote number of bays and stories, abbreviations Bn, Sn are used (n is the number of respective
bays and stories) respectively. All other abbreviations are given in their context itself.
Br4 Br5 Br6

Story
S
Bay
Br1 Br2 Br3
B
Fig 1. Numbering of bracing loops in rigid frame with fixed supports (in Abaqus Model)
Diagonal brace oriented at 135º angle with x axis and cross (X) brace are used to study the
buckling behavior of frame including braces at various bracing locations and to compare the
results with fully braced and bare frames. These results are compared with results of chevron
braced frames. Bracing types are shown in Fig. 2.
Fig. 2. Types of braces used are, diagonal brace, cross (X) brace and chevron brace, from left to right
3. Modelling and Methodology

Parameter of concern here is critical buckling load under vertical and lateral loadings (from both
sides). Eigen mode values are obtained from simulation and the first positive Eigen value has been
chosen to be the critical buckling load value for each frame case. Abaqus CAE 6.14 software has
been used for analysing the frames considering two node cubic beam element using linear
perturbation buckling approach.
As it is known that strong column weak beam is ideal in the case for ductile behaviour under the
influence of seismic loading. It is also found that in the case of braced frames a strong beam is
good for handling the unbalanced forces that come in presence after the post-buckling condition
resulting in the loss of strength in compression member of conventional brace and yielding in
tension member of buckling resistant brace. After initial buckling the frame resists lateral load
similar to an eccentrically braced frame. Now the brace resists the load in conjunction with beam,
where the beam resists the vertical component of brace force also. So a stronger beam is required.
But inclusion of strong beam throughout the frame causes weak story problem in lower stories
resulting in a brittle failure. So to prevent other such complexities in present study same circular
cross-section is considered for simplicity and easy understanding of phenomenon. Concentric
(concentrated) vertical loads are provided at all the nodes connecting beams and columns whereas
concentric lateral loads are provided at open end corner nodes only. For unsymmetrical bracing
852
configuration the lateral loads are applied from both sides of the frames separately to access the
minimum value of buckling load on either side of the frame. The loadings are shown in Fig. 3.
V V H1 H2
V H1 H2
Fig. 3. Vertical force V, lateral force H1 and lateral force H2, from left to right, for B1S2 frame
To access the effect of location of braces on the stability of frame, 45 frames made from various
combinations of 3 bays and 3 stories with and without braces are considered for separately applied
lateral and vertical loads, using single brace at each bracing location individually. Lateral load is
applied from either side individually. The results of brace placed at 135º angle orientation can be
followed for corresponding bracing location having orientation 45º which is mirror image of brace
placed at 135º. As shown in Fig. 4.
H1 H2
H1 H2
Fig. 4. Lateral force H1 for 45º oriented brace corresponds to lateral force H2 for 135º oriented brace
Firstly, the 45 frames have been analysed using a single diagonal brace in each loop of the frame at
each bracing level resulting into 180 load cases and if considering 45 º angle orientation cases
separately, it results in 360 load cases . Then considering cross brace having all end conditions and
configurations same as above, 96 load cases are analysed as some cases are symmetric because of
symmetric bracing type and also for some case lateral loading application from both side is also
not required because of symmetry, resulting into 276 total load cases for both types of braces at
different bracing locations individually including bare frame cases. For comparison, full braces are
provided for all cases using single diagonal and X braces separately giving 18 more frame cases
resulting into 45 more load cases. For comparison of effectiveness of brace having same cross-
section as that of beams and columns with the brace having same slenderness as that of beams and
columns, frames with single bay and stories varying from one to three are considered giving 3
frame cases resulting into 9 load cases. For comparing the effect of single brace at particular
bracing location, a frame constituting single bay and three stories is used giving 3 frame cases of
single brace and 1 case of fully braced frame resulting into 12 load cases.
For comparison with other bracing type, chevron braced is used. Comparison of fully braced
frames is done considering single bay and stories varying from 1 to 7 contributing to 49 load cases
as chevron and X brace have symmetric construction. Analysis is done for frames made up of all
combinations of single story with bays varying from 1 to 4 to access the best location of braces
and to get the best combination of inserting braces at particular locations to get maximum stability
out of them best cases are used for comparison and for per bracing location effect, frames made up
of all combinations of single bay and stories varying from 1 to 4 are also analyzed both
contributing to 182 load cases for single brace at various locations and for fully braced and bare
frame cases and many more cases for finding best location of multiple braces in the frames having
single story with bays varying from 1 to 4. Linear perturbation procedure is used to do buckling
853
analysis, to get required number of eigenvalue. 9 to 18 buckling modes are obtained for various
frames depending on their complexity to understand the buckling behavior under various mode
shapes. Out of which first positive Eigen value is taken as critical buckling load value.

4.1 Effect of Location of Brace on Critical Buckling Load Value of Steel Frames
The effect of location of braces on behaviour of frame under buckling loads is ascertained here by
applying lateral and vertical loading separately on the frames made from various combinations of
3 bays and 3 stories. Single brace placed at a bracing location individually contribute to one frame.
The types of braces used for analysis are diagonal brace and X brace. The tables showing buckling
load values for each frame case are given below per story with varying number of bays. Braces,
columns and beams have same cross-section. Unit of load is Newton, N. Buckling loads H1 and
H2 are lateral (Horizontal) loads, and V is vertical load as shown in Fig. 3. H is the lower value
among H1 and H2.
Single story with varying number of bays
Effect of location of brace on singly braced single storied frame for varying bays is given in terms
of overall critical buckling load in Table 1.
Table 1. Bucking loads (N) for single storied frame with single brace in each closed loop individually
Diagonal Brace Cross Brace
Brace Load
Number of Bays Number of Bays
No. Direction
B1 B2 B3 B1 B2 B3
H 3340130 3250880 2961450 3340130 3250880 2961450
No Brace
V 804617 829638 854520 804617 829638 854520
H1 1323730 1329790 1333200 5989060 6172900 6177220
H2 2445050 2115050 1994900 5989060 2665640 2174980
Br1
H 1323730 1329790 1333200 5989060 2665640 2174980
V 2936390 2940360 2941470 4514540 3049600 2960880
H1 -over- 1343480 1349150 -over- 2665640 2668440
H2 2610510 2166730 6172900 2668440
Br2
H 1343480 1349150 2665640 2668440
V 2937500 2940260 3049600 3044950
H1 -over- 1350530 -over- 2174980
H2 2616270 6177220
Br3
H 1350530 2174980
V 2938410 2960880
Diagonal Brace
For these braced frames horizontal critical load is minimum for Br1, whereas vertical one is
maximum for Br1. With increase in number of bays, horizontal critical load decreases for no
bracing but increases for both horizontal and vertical loads in case of bracing. Considering bracing
condition alone, braces with minimum vertical critical load give maximum horizontal critical load
vice-e-versa. In a frame as the bracing location is raised upwards, horizontal critical load increases
and vertical critical load decreases and also, as the brace moves far from horizontal load
application direction, the horizontal critical buckling load increases and vertical critical buckling
load decreases with respect to other locations for same number of total bays. Single brace is
detrimental under horizontal load whereas for vertical load it is beneficial.
X Brace
The behaviour of cross brace is very much different from single brace. Critical buckling loads are
very much higher than single brace. Overall horizontal and vertical critical buckling loads decrease
854
with increase in number of bays, except central bay braced B3S1 configuration under lateral load,
which is higher than braced cases of B2S1. Bracing location with minimum critical horizontal load
also has minimum critical vertical load. Bracing location with maximum critical horizontal load
also has maximum critical vertical load.
Two stories with varying number of bays
Effect of location of brace on singly braced two storied frame for varying bays is given in terms of
overall critical buckling load in Table 2.
Table 2. Bucking load (N) for two storied frame with single brace in each closed loop individually
Brace Load
Number of Bay Number of Bay
No. Direction
B1 B2 B3 B1 B2 B3
No H 1386560 2469360 2614520 1386560 2469360 2614520
Brace V 378228 401593 415398 378228 401593 415398
H1 590560 607401 616721 1856080 3059270 2873010
H2 1128110 1270990 1295470 1855760 2165270 1757030
Br1
H 590560 607401 616721 1855760 2165270 1757030
V 671486 707964 733934 706937 722725 745707
H1 793410 613084 623578 1612630 2165290 2321350
H2 1199150 1641510 1638350 1612630 3059300 2321350
Br2
H 793410 613084 623578 1612630 2165290 2321350
V 445769 702163 729906 470687 722725 738889
H1 -over- 854428 620070 -over- 2057280 1757030
H2 1396420 1634430 1931500 2873010
Br3
H 854428 620070 1931500 1757030
V 448382 729789 463202 745707
H1 849064 869440 1931500 1978630
H2 1407010 1486180 2057280 1859540
Br4 H 849064 869440 1931500 1859540
V 454538 455513 463202 466821
H1 -over- 894403 -over- 2293060
H2 1599900 2293060
Br5
H 894403 2293060
V 455732 460111
H1 874188 1859540
H2 1335780 1978630
Br6
H 874188 1859540
V 459312 466821
Diagonal Brace
Horizontal critical load is minimum for Br1, whereas vertical one is maximum for Br1. With
increase in bays, critical load increases for both horizontal and vertical load cases. For buckling
due to vertical load, brace at bottom end loop is preferable. Considering bracing condition alone
mostly braces with minimum vertical critical load give maximum horizontal critical load vice-e-
versa. For a frame, as the bracing location is raised upwards, horizontal critical load increases and
vertical critical load decreases. Like frame cases of single story, condition of location of brace
moving far from loading point is not satisfied here. Single brace is detrimental under lateral load
whereas beneficial under vertical load.
X Brace
For all the considered frame cases of single and double story, as the location of braces rise the both
the horizontal and vertical buckling loads reduce. For the frame cases of three stories, only vertical
load reduces with rise in bracing level and in each frame for individual story the mid brace has
lower vertical critical load but higher horizontal buckling load. Side braces give better vertical
critical load.
855
Three Stories with varying number of bays

Effect of location of brace on singly braced three storied frame for varying bays is given in terms
of overall critical buckling load in Table 3.
Table 3. Bucking load (N) for 3 storied frame with single brace in each closed loop individually
Brace Load
Number of Bays Number of Bays
No. Direction
B1 B2 B3 B1 B2 B3
No H 657001 1256340 1755950 657001 1256340 1755950
Brace V 233663 252234 263112 233663 252234 263112
H1 375212 388554 396061 805894 1880280 1837530
H2 566004 719650 778346 805894 1229470 1309370
Br1
H 375212 388554 396061 805894 1229470 1309370
V 327279 351102 364405 340416 356969 369235
H1 397025 391151 399457 750600 1229470 1902580
H2 556975 1060780 1115200 750600 1880280 1902580
Br2
H 397025 391151 399457 750600 1229470 1902580
V 295722 348468 362627 313214 356969 366295
H1 649610 435625 398490 664898 1102230 1309370
H2 649993 677719 1039400 664898 925805 1837530
Br3
H 649610 435625 398490 664898 925805 1309370
V 241458 297813 362579 244268 308122 369235
H1 -over- 427760 445735 -over- 925805 1071510
H2 789960 738024 1102230 1014030
Br4
H 427760 445735 925805 1014030
V 301832 302682 308122 310531
H1 790840 455639 135546 1235900
H2 920392 901090 951571 1236050
Br5
H 790840 455639 951571 1235900
V 256781 302804 258133 305973
H1 779729 442009 951571 1014030
H2 1169020 754851 135546 1071510
Br6
H 779729 442009 951571 1014030
V 256955 305236 258133 310531
H1 -over- 814825 -over- 1422740
H2 1067530 1118830
Br7
H 814825 1118830
V 266587 267458
H1 839150 1610550
H2 1384620 1610550
Br8
H 839150 1610550
V 266652 267363
H1 815929 1118830
H2 116452 1422740
Br9
H 815929 1118830
V 266688 267458
Diagonal Brace
Horizontal critical load is minimum for Br1, whereas vertical one is maximum for Br1. With
increase in bays, critical load increases for both horizontal and vertical load cases. All mid braces
in each row have higher horizontal critical buckling load. For buckling due to vertical load, brace
at bottom end loop is preferable. Considering bracing condition alone braces with minimum
vertical critical load give maximum horizontal critical load vice-e-versa. As the bracing location is
raised upwards, horizontal critical load increases and vertical critical load decreases. For
horizontal load single brace is detrimental whereas for vertical load it is beneficial.
X Brace
Both horizontal and vertical critical buckling loads increase with increase in number of bays,
bracing level being same. As the bracing level rises, critical horizontal and vertical load decreases,
except top bracing level of B3S3 has higher horizontal buckling load than lower storey.
856
Remark for all Considered Cases of Frames Singly Braced by Diagonal Brace
With the increase in the number of bays for the same number of stories both vertical and horizontal
critical loads increase whereas for first story of no brace condition, horizontal critical load
decreases. With the increase in story for same number of bays both horizontal and vertical critical
loads get reduced (i.e. increase of stories is detrimental). H1 value is less than H2 value as there is
no tension brace available against H1 loading. Considering bracing condition alone, mostly the
braces with minimum vertical critical load give maximum horizontal critical load vice-e-versa. As
the bracing location raises upward horizontal critical load increases and vertical critical load
decreases. For buckling due to vertical load, brace at bottom end loops is preferable. Bracing at
top stories has more horizontal critical buckling load but it is not enough to resist buckling in
comparison to bare frame. For horizontal load single brace is detrimental whereas for vertical load
it is quite beneficial.
Remark for all Considered Cases of Frames Singly Braced by X Brace
The behaviour of cross brace is very much different from single brace. Here symmetry for some of
the frame cases can be seen. With the increase in number of bays for the same number of stories
both vertical and horizontal critical loads increase. With the increase in number of stories for the
same number of bays both horizontal and vertical critical loads get reduced. Cross braces have
high vertical load resistance to buckling but in comparison to diagonal brace. It is more beneficial
under horizontal load, provided the braces are enough to produce buckling resistance more than
that by bare frame. For buckling due to vertical load, brace at bottom stories are more beneficial.
Bracing at bottom stories has more horizontal critical buckling load than at above locations but
doesn’t resist buckling substantially.
Remark for all Considered Singly Braced Frames
For in-plane loadings, bracing in single loop of frame is not enough for restraining the buckling
due to lateral loading. What to speak of increasing the buckling resistance, under lateral load, in
some cases it is detrimental for stability of the structure as it degrades the buckling resistance of
the frame severely. If a single ordinary brace is introduced in any of the loops of the frames in any
of the configurations of bay and stories, it is for sure that the stability under lateral load is either
going to reduce or not going to increase substantially, except for B1S1 configuration (gets fully
braced), where X brace is found beneficial for considered profile, which has more buckling
resistance than that of bare frame. If talking about the strength gain against buckling due to axial
(vertical) load even a single brace provides substantially high restraint against buckling.
In further sections tables and charts are given for each fully braced frame cases. The tables and
charts regarding per bracing location for each frame case are large in number as for each particular
location of brace in a frame there is a table and a chart, so cannot be accommodated here. But the
explanations are given here for both study of buckling load and also for study of comparison ratios
regarding each bracing location in each considered braced frame case.
4.2 Comparison of Buckling Loads between Best of Singly Braced and Fully Braced Frames
Overall buckling load values of best of singly braced frame are compared with bare frame and
fully braced frame values using diagonal brace and using cross brace. In tables, simply S1, S2, S3
corresponds to number of stories and values under them are the buckling loads for the best case of
857
singly braced frame, ‘Nbr’ represents the bare frame cases and ‘Full’ represents the fully braced
configurations. Braces, columns and beams have same cross-section.
Story-wise variation for increasing number of bays [B1 (S1, S2, S3); B2 (S1, S2, S3); B3 (S1,
S2, S3)]
The overall buckling loads values using diagonal brace are shown in Table 4 and using X brace in
Table 5. Considering diagonal brace, for vertical loading, in all cases fully braced frame is more
stable than singly braced frame and singly braced is more stable than bare frame, but under lateral
loading, in case of single bay with increase in number of stories, fully braced frame shows less
stability in comparison to best of singly braced frame cases, both being less than bare frame
condition due to premature buckling. For two and three bays the fully braced frame has higher
stability than singly braced frame (except for B2S3 configuration) both having stability less than
bare frame condition, due to premature buckling. Similar to X brace, in all cases with the increase
in number of stories, the buckling load values for fully braced frame, singly braced frame and bare
frame tend to approximately converge to each other’s buckling load values corresponding to same
story whereas with increase in the number of bays, under lateral loading the difference increases
i.e. it reduces convergence of values but doesn’t provide more stability with provision of braces.
Table 4. Maximum critical load for single location braced and fully braced frames using diagonal brace
Lateral loading (N)
H S1 S2 S3 Nbr S1 Nbr S2 Nbr S3 Full S1 Full S2 Full S3
B1 1323730 793410 649610 3340130 1386560 657001 1323730 699189 610302
B2 1343480 854428 790840 3250880 2469360 1256340 2309540 1267030 774168
B3 1350530 894403 839150 2961450 2614520 1755950 2591670 2034270 1162320
Vertical Loading (N)
B1 2936390 671486 327279 804617 378228 233663 2936390 1618850 1037860
B2 2940360 707964 351102 829638 401593 252234 3003320 1664850 1069990
B3 2941470 733934 364405 854520 415398 263112 3010350 1670450 1076050
Considering X-brace, for B1 with stories varying from 1 to 3, it is found that under lateral loading,
the best case of singly braced frame gives buckling resistance similar to that of fully braced case,
that is more than bare frame case, but for two and three bays with stories varying from 1 to 3, fully
braced frame is more stable in comparison to singly braced and the bare frame but the frame singly
braced shows stability even less than the bare frame (except for B3S3 configuration) because of
premature buckling. For vertical loading, in all cases fully braced frame is more stable than the
singly braced frame and singly braced is more stable than bare frame. In all cases with the increase
in number of stories, the buckling load values for fully braced frame, singly braced frame and bare
frame tend to approximately converge to each other’s buckling load values corresponding to same
story whereas with increase in the number of bays, under lateral loading the difference increases
i.e. it reduces convergence of values so provide more stability with provision of braces.
Table 5. Maximum critical load for single location braced and fully braced frames using X brace
Lateral Loading (N)
H S1 S2 S3 Nbr S1 Nbr S2 Nbr S3 Full S1 Full S2 Full S3
B1 5989060 1855760 805894 3340130 1386560 657001 5989060 1968540 857207
B2 2665640 2165290 1229470 3250880 2469360 1256340 4953470 4731260 1990910
B3 2668440 2321350 1902580 2961450 2614520 1755950 4744430 4449750 3541930
Vertical Loading (N)
B1 4514540 706937 340416 804617 378228 233663 4514540 2393950 1552600
B2 3049600 722725 356969 829638 401593 252234 4468670 2541160 1662490
B3 3044950 745707 369235 854520 415398 263112 4396480 2540050 1699740
858
Critical buckling load value under vertical loading

For all configurations of bays with the increase in number of stories (on x-axis), buckling load (on
y-axis) pattern of curves for both X brace and single diagonal brace are decreasing convex
downward. Maximum critical buckling load value is obtained for fully braced frames, then for best
of the bracing at single location and lowest value is obtained for bare frame condition.
Critical buckling load value under lateral loading
With the increase in number of stories, buckling loads of bare frames and X braced frames (both,
best of singly braced and fully braced) have opposite pattern of curves for single bay (B1) cases
when compared with frame cases of B2 and B3, as B1 case gives decreasing convex downward
curve whereas other two give decreasing convex upward curves. For all the bay cases with
increasing number of stories, best of singly braced frames using diagonal brace have decreasing
convex downward pattern of buckling loads. When fully diagonal braced, pattern for B1 and B2
cases is similar that is decreasing convex downward but decreasing convex upward for three bays
cases. This shows that analysing behaviour of a brace considering single bay and extrapolating the
results for more number of bays can lead to misleading results.
Bay-wise variation for increasing number of stories [S1 (B1, B2, B3); S2 (B1, B2, B3); S3 (B1,
B2, B3)] (For this section, tables can be derived from Tables 1, 2, 3, 4, 5)
Critical buckling load under vertical loading
For second (S2) and third (S3) stories with increasing number of bays, either having full brace,
best case of single brace or without brace, both X brace and diagonal brace have nearly similar
pattern of rising curves for critical buckling load value, except the fully braced B3S2, which has
buckling load less than B2S2 configuration using X brace; in these frame cases maximum value is
obtained for fully braced, then for best of singly braced and least in case of bare frame. For single
story with varying number of bays, very different behaviour is shown by both the braces. For
diagonal brace, fully braced and the best of singly braced, both are rising with increase in bays and
also have nearly similar values that are much higher than bare frame condition but for X brace,
fully braced and the best of singly braced both have falling curves of buckling load values with
increase in number of bays. Single X brace condition has much lower values in comparison to
fully braced frame but both are higher than bare frame condition. So, it can be said that for critical
buckling load under vertical loading, single story doesn’t represent the behaviour of upper stories.
For single story, even single diagonal brace has buckling loads nearly equal to that of fully
diagonal braced cases.
Critical buckling load under lateral loading
For the cross-sections considered in this study, in comparison to frame braced using diagonal
brace, bare frame has more buckling resistance in comparison to both fully braced and the best of
singly braced frames because of premature buckling, although they provide better strength and
buckling resistance against vertical load. In bare frame condition for first story, the buckling
resistance decreases with increase in number of bays which is contradictory to the buckling
resistance of upper stories as there is ample rise in buckling resistance with increase in number of
bays. For fully braced with single diagonal, the pattern of variation of buckling load with
increasing number of bays for first story is convex upward but for upper two stories it is convex
859
downward whereas both are rising. Under lateral loading, frame fully braced using X brace
provide buckling resistance more than bare frame for all stories with increasing number of bays
but the pattern of variation with bays is different from that of first story the buckling load decrease
with increase in number of bays. For second story rise and fall curve is obtained as maximum
value is obtained for second bay. For third story rising convex downward pattern is obtained for
increasing number of bays. Bare frame has lower buckling load value than singly X braced frame
for first story but more in upper stories (except B3S3 configuration).
4.3 Diagonal braces having same slenderness as that of beam and column
The overall buckling loads of frames using brace having same cross-section as that of beams and
columns, (D) are compared with the buckling loads of frames using brace having same slenderness
as that of beams and columns, (D_SS), as given in Table 6.
Table 6. Comparison with diagonal having same slenderness as that of beams and columns (Full bracing)
D_SS No Brace D
Stories
H (N) V (N) H (N) V (N) H (N) V (N)
B1S1 2773330 3049350 3340130 804617 1323730 2936390
B1S2 1283320 1865070 1386560 378228 699189 1618850
B1S3 640445 1227780 657001 233663 610302 1037860
Considering fully braced frame, D_SS has more stability against both the vertical and lateral
loadings in comparison to D. The difference between lateral buckling load resistance of D_SS and
D abruptly decreases as the bracing level rises in the same frame and from third story for
configuration considered here, the buckling load values of both reaches near to that of bare frame
which is higher in all considered cases. Between D and D_SS, the difference in vertical buckling
load resistance increases slowly with the rise in bracing level, both have higher stability than bare
frame under vertical load. The behaviour is shown in Fig 4.
4.0E+06 3.5E+06
3.5E+06 3.0E+06
Critical load, N
D_SS
Critical load, N
3.0E+06 2.5E+06
D_SS
2.5E+06
2.0E+06 Nbr
2.0E+06 Nbr
1.5E+06
1.5E+06 D
D 1.0E+06
1.0E+06
5.0E+05
5.0E+05
0.0E+00
0.0E+00
B1S1 B1S2 B1S3
B1S1 B1S2 B1S3
Frame Configuration
Frame Configuration
Fig. 4. Comparison chart for both variants of braces under lateral and vertical loading from left to right
respectively.
Considering per bracing level, single braced D frame has nearly equal stability against vertical
loading in comparison to the frame with single braced D_SS frame. Whereas, against lateral
loading D_SS has higher value at lower stories but has decreasing pattern and D has rising curve
such that reaching third story for configuration used here it has more stability than D_SS. The
numerical analysis results are given in Table 7.
Table 7. For B1S3 per brace buckling load (N) for single bracing per story
Braces D_SS D
H V H V
Br1 609901 334765 375212 327279
Br2 530273 303784 397025 295722
Br3 547660 242873 649610 241458
860
This shows that single diagonal brace of D_SS type is detrimental for stability against lateral
loading at higher stories. Also except first story for all considered cases column buckles first in
case of D_SS whereas brace buckles in case of D. To access the effects of braces per bracing level
B1S3 configuration frame is used. The buckling load pattern is shown in Fig. 5.
7.E+05
6.E+05
Critical load, N
5.E+05 D_H
4.E+05 D_SS_H
3.E+05
D_V
2.E+05
D_SS_V
1.E+05
0.E+00
Br1 Br2 Br3
Brace Number (Numbering bottom to top)
Fig. 5. Combined chart for both variants of braces under lateral (H) and vertical (V) loading per bracing
4.4 Comparison of considered braced frames with braced chevron (Ch) brace
The results of buckling load values obtained using diagonal and X braced are compared with
chevron brace from single to seven stories, as given in Table 8.
Table 8. Comparison with fully Braced Chevron up to seven stories
Loading
Bracing B1S1 B1S2 B1S3 B1S4 B1S5 B1S6 B1S7
direction
H 3340130 1386560 657001 320815 231757 155549 110870
No Brace
V 804617 378228 233663 164549 125409 100642 83710
H 1323730 699189 610302 336254 223053 153578 112901
Diagonal
V 2936390 1618850 1037860 744676 541909 465473 389857
H 5989060 1968540 857207 475604 300986 206849 150473
X Brace
V 4514540 2393950 1552600 1128370 894775 713446 599160
H 2812400 1419690 941974 572641 346046 229557 162603
Chevron
V 3399200 1818780 1181290 861855 673011 549354 462696
Story-wise
For both the load types the critical buckling load value decreases with increase in number of
stories but the rate of decrement varies in both the conditions. The variation in buckling load value
on application of lateral load is very much erratic whereas for vertical load all braced frames have
nearly similar pattern of reduction in buckling load. In the case of lateral loading after 6th story, the
effect of bracing is negligible for the frames braced with the braces having profile as that of
columns and beams. Story-wise variation of fully braced frames under lateral loading is shown in
Fig 6.
7.E+06
6.E+06
5.E+06
Critical Load, N
Nbr
4.E+06
D
3.E+06
X
2.E+06 Ch
1.E+06
0.E+00
B1S1 B1S2 B1S3 B1S4 B1S5 B1S6 B1S7
Frame Configuration
Fig. 6. Buckling load values under lateral loading
861
For vertical loading the differences in buckling load capacities with different bracings get closer to
each other with increase in stories but are always more than bare frame condition. Considering X
Brace, for lateral loading there is abrupt reduction in buckling load capacity of X-braced frame
with respect to other bracing conditions but for vertical loading it is best with nearly similar
reduction pattern as in all other braced conditions. Story-wise variation of fully braced frames
under vertical loading is shown in Fig. 7.
5.E+06
5.E+06
4.E+06 Nbr
Critical load, N
4.E+06
3.E+06 D
3.E+06 X
2.E+06
2.E+06 Ch
1.E+06
5.E+05
0.E+00
B1S1 B1S2 B1S3 B1S4 B1S5 B1S6 B1S7
Frame Configuration
Fig. 7. Buckling load under vertical loading
Considering diagonal brace, in case of lateral load application, for the considered profile of brace,
buckling load capacity is even less than the bare frame but from 3 th story, it catches up near to bare
frame condition. For vertical load case it has buckling load less than other two frames but more
than bare frame. So it is effective in vertical load condition, but being unsymmetrical it has
different influence of loading from either sides and could buckle at lesser loads. Considering
Chevron brace, for both lateral and vertical loadings rate of decrement is small. For lateral loading
in the case of bottom stories the value of buckling load is less than that for X-brace and bare frame
but from upper stories the reduction in capacity of chevron brace is less and buckling load values
surpassed all other bracing types. So it can be said that considering either single story or multi-
story doesn’t give complete information about the stability characteristics of a particular brace.
Considering both lateral and vertical loading, chevron is optimal.
Bay-wise
Single story with varying number of bays (BnS1)
As it is known the bracing increases strength of structure, regarding stability nothing can be said as
stability is not directly dependent on strength of structure but depends on the geometry, material
properties and support conditions. The concern is more of lateral buckling load, because vertical
buckling load value has shown increment with increase in the strength of structure for all the
bracing types but it’s not the case with lateral buckling loading as it doesn’t show any clear
relation with strength increment. For same cross-section of all the members, the vertical buckling
load values are nearly equal for all the bracing type considering the values at various bracing
locations, for fully braced frame the values are bit different, even then in all cases the values are
much more than bare frame condition. The strength is increased by addition of all types of braces
used here but for considered configuration of frames and profile of members, the stability
characteristics doesn’t comply with that of strength. X-brace shows huge increment in both
vertical and lateral loading cases. Chevron also shows the same results except for B1S1
configuration. Whereas single diagonal brace shows increment in vertical loading case but a
862
drastic downfall in lateral loading case due to premature buckling and this decrement in buckling
load value is so much reduced that it is even lesser than bare frame condition. Considering effect
of bracing locations, the brace at central location is found to provide more stability. The
comparison of buckling loads (N) using various braces are given in Table 9.
Table 9. Buckling loads (N) for fully braced frames configuring single story and multiple bays
Bays Load X Ch D Nbr
B4S1 H 4705860 5809680 2582660 2670500
V 4357000 3450050 3010970 867350
B3S1 H 4744430 5551050 2591670 2961450
V 4396480 3449750 3010350 854520
B2S1 H 4953470 4756960 2309540 3250880
V 4468670 3445790 3003320 829638
B1S1 H 5989060 2812400 1323730 3340130
V 4514540 3399200 2936390 804617
The effect of bay-wise arrangement of braces in terms of buckling loads (N) is given in Table 10.
Table 10. Best arrangements giving higher buckling loads (location-wise bracing, 1 means brace and 0
means no brace)
B4S1 X brace Chevron Diagonal Diagonal
3 Brace 1101 1101 1101 1110
H 4817110 4820230 2593350 2154070
V 4418790 3442680 3005480 2955890
2 Brace 1001 1001 1001 0110
H 5285640 4101420 2152800 2033150
V 3416970 3357960 2954430 3003400
2 Brace 101 101 101 110
H 5117640 4356880 2147890 2158130
V 4479790 3438020 3003420 2954770
B4S1 and B3S1

For each bracing location buckling lateral load values for chevron are equal upto 2nd decimal point
then little lesser in comparison to X-brace. Buckling lateral load for diagonal brace is very low
even lower than bare frame condition. Buckling vertical load values for chevron and diagonal are
equal till 2nd decimal point then lesser in comparison to X-brace. But in fully braced frame the
values are quite different. For lateral load, in contrast to brace location results chevron shows
buckling load values much higher than X-brace condition. Same follows for B3S1 configuration.
When considering the best possible arrangement of bracing bay-wise, the best arrangement of each
brace is the one having braces at both side corners of the frame. For the arrangement having one
brace less than full bracing stability against vertical loading is approximately equal to that of fully
braced condition. For B4S1 configuration, including any of the considered bracing types, the best
arrangement is, when the braces are provided at both side corners of the frame and one extra brace
next to corner brace. For B3S1 frame, the arrangement includes single brace at each side corners
of the frame. X brace shows more stability for the arrangement having one brace less than fully
braced case for both lateral and vertical load in both B4S1 and B3S1 frames. For chevron, vertical
load value is similar to fully braced case with either 3 or 2 braces in B4S1 frame and 2 braces in
B3S1 frame but horizontal buckling load values is lesser in comparison to fully braced condition.
Diagonal brace provides stability against both lateral and vertical loading approximately equal to
fully braced case with the best arrangement of one brace less than fully braced case in B4S1 frame
863
and nearly equal stability against vertical loading for best 2 brace arrangement but lesser against
lateral loading in B3S1 frame.
B2S1
When considering the effect of bracing at each bracing location for both lateral and vertical load
case, the values of buckling load are very close in case of chevron and X brace. Among them
values of chevron are bit lesser. In case of diagonal brace for vertical buckling load, the values are
less than other braces but are close to them. In case of lateral loading the buckling load values for
diagonal brace are very less and even lesser than bare frame condition due to premature buckling.
When considering fully braced frame similar results are seen as for B4S1 and B3S1.
B1S1
Different behaviour in comparison to more number of bays is observed. For same cross-sectional
area, only X brace is found to be effective in vertical and lateral/ horizontal loading both. For
lateral loading chevron and diagonal brace are found to buckle prematurely at critical buckling
load less than that of bare frame condition and for vertical load both of them work better than bare
frame condition.
On the basis of single story results it can be concluded that the character of a brace cannot be
decided for higher stories and higher bay cases. Chevron buckles prematurely when used for single
story, but for higher stories it gives higher stability against lateral load, even more than X brace.
Chevron provides stability against vertical load having critical buckling load much higher than
bare frame condition. Also when considering from bracing location perspective both X and
chevron have nearly similar values of critical buckling load. For the diagonal brace vertical load
resistance against buckling is always less than that of other two bracing types.
4.5 H/V Ratio
In relation with different types of braces
The ratio of buckling load value under lateral loading to the corresponding vertical load value for
same braced frame configuration is denoted as H/V ratio. It is the ratio which indicates uniformity
in increase or decrease of either loading in comparison of other with increase in number of bays or
stories. When compared with bare frame case ratios, it indicates the change in behaviour of frame
under various loadings on introducing braces. In simple words, it’s a representative of optimised
increase in stability of frame against both lateral and vertical loading on introducing braces.
As it is seen in above context that each type of brace increases stability under vertical load but for
some profiles of brace it may show decrement in buckling load under lateral load. So when
comparing two types of braces providing good increment in stability under both vertical and lateral
direction, the brace having uniform H/V ratio throughout must be preferred. It’s not recommended
to use it for the braces having low increment in stability as H/V ratio can be more or uniform but
not representative of good bracing type as seen here for diagonal brace. Although stability of
diagonal braced frame under lateral loading is even less than bare frame but H/V is more than X
brace for higher stories. So this ratio is useful for comparing optimised increase in stability for the
best type of braces having very similar buckling load values provided both provide stability more
than bare frame case under both lateral and vertical loading in all cases. Then we can compare
their uniformity in increase of stability for various bays and stories. Some researchers use single
864
story to estimate the performance of brace which may not be followed for higher stories and more
number of bays. So this ratio can be used to check uniformity in increase in stability. X brace and
chevron are the two best types of braces among all braces considered here. Comparison of H/V
ratio of fully braced frames using various braces is given in Table 11.
Table 11. Comparison of H/V ratio for fully braced frames.
Brace B1S1 B1S2 B1S3 B1S4 B1S5 B1S6 B1S7
No Brace 4.15 3.67 2.81 1.95 1.85 1.55 1.32
Diagonal 0.45 0.43 0.59 0.45 0.41 0.33 0.29
X Brace 1.33 0.82 0.55 0.42 0.34 0.29 0.25
Chevron 0.83 0.78 0.80 0.66 0.51 0.42 0.35
For chevron, except for first story all values of buckling load under lateral and vertical loadings are
higher than bare frame. So if considering stories more than one, it can be seen from Table 11,
chevron has less deviation in values of H/V ratio so we can say that it provide optimised stability
with increasing stories.
As per the bracing location for three storied frames with bays varying from 1 to3
Since stability against vertical load always increases with bracing and there is so much uncertainty
about resistance against lateral loading, to ascertain the better arrangement amongst different
bracing types H/V ratio is being used here. Arrangement having more H/V ratio for same
configuration of frame will be suggested to work better in an optimized way under vertical and
lateral loadings, provided the buckling load resistance is higher than that of bare frame for both
lateral and vertical loadings separately. In all cases of singly braced frames, H/V ratio is less than
that of bare frame. Using single brace in B1S3, H/V ratio shows same pattern of increment with
increasing brace level for both diagonal and X brace i.e. H/V ratio is maximum for 3 rd story. In
B2S3 and B3S3 H/V ratio using single diagonal brace is similar, that is increasing with bracing
level. But using single X brace in B2S3 configuration and in B3S3 configuration H/V ratio is
lowest for bracing level at middle story. In B3S3 for each story, the central bay gives more
stability in comparison to corner bays.
Including one more kind of ratio that is ratio of buckling loads under lateral and vertical loadings
of braced frame to the respective loadings applied on bare frame. Let, Hr=H/Hnbr is the ratio of
overall buckling load value of braced frames to that of respective bare frame buckling load values
and Vr=V/Vnbr is the ratio of buckling load value of braced frames to that of respective bare
frame buckling values. Considering the value of buckling load upto five decimal points, it is found
that Hr/Vr ratio is maximum at central bay for all the cases of single brace and this ratio increases
with rise in bracing level of single brace for the frame. It indicates that central bay bracing
provides better stability under both vertical and lateral loading. In this study it is also observed
that, although bracing in central bay is better amongst all bays when single brace is provided or
when bracing is provided throughout for a single bay with multiple stories but when providing
bracing for multiple bays, both left and right end corner bays plays vital role in providing better
stability. Hr/Vr ratio is not equivalent to H/V ratio. Hr ratio and Vr ratio are independent of H/V
ratio. In all diagonal braced and X braced frame cases Hr and Vr ratios are approximately same at
same bracing level irrespective of bay location for each frame case. Although Vr decreases with
increase in bracing level for all singly braced frames, it is greater than one for all singly braced
865
frames using diagonal brace or X brace. Whereas using diagonal brace, Hr is less than one for each
bracing level. For single X brace except B1S3 and central brace at lowest story of B3S3, the Hr
ratio is less than one. For diagonal brace, the Hr ratio increases with rise in bracing level whereas
Vr decreases, both converging towards one with rise of bracing level whereas the gap of
convergence between Hr and Vr increases with increased number of bays. Gap of convergence to
reach one increases with the increase in number of bays for X brace also. In B1S3 the pattern of Hr
and Vr is same for single X braced frame that is decreasing with increase in bracing level reaching
more than one. For B2S3 middle story has minimum and lowest story has maximum Hr ratio. For
B3S3 central bay has more Hr ratio in comparison to corner bays for all bracing levels whereas in
both central and corner bays middle story level gives least Hr ratio.
Considering fully braced frames, with increase in number of bays, H/V ratio increase for all braced
and bare frames. H/V ratio of bare frame is higher than that of diagonal braced frames and X
braced frames for each bay. With increase in number of bays the rate of increase is also more for
bare frame. Amongst braced frames, the rise for X braced frame is higher in comparison to
diagonal braced frame. Both Hr and Vr ratios are higher for X brace. For diagonal brace Hr is
maximum for single bay. For diagonal brace Hr decreases for second bay and increases for third
bay having same number of stories. In X braced frame Hr increases with increase in number of
bays. In above all cases considered here, Vr is above 4 for diagonal and above 6 for X brace, both
decreases with increase in number of bays for same story.
The comparison of H/V ratio for fully braced frames and bare frame are given in Table 12. Table
has two parts first one showing bay-wise variation, with three stories and varying bays and the
second part shows story-wise variation, with three bays and varying stories.
Table 12. H/V ratio comparison for 3 stories and varying bays and for 3 bays and varying stories
Bays diagonal X brace No Brace Stories diagonal X brace No Brace
Full H/V H/V H/V Full H/V H/V H/V
B1S3 0.59 0.55 2.81 B3S1 0.86 1.08 3.47
B2S3 0.72 1.20 4.98 B3S2 1.22 1.75 6.29
B3S3 1.08 2.08 6.67 B3S3 1.08 2.08 6.67
As per the bracing location for three bays with stories varying from 1 to 3
In diagonal braced frame cases, Although for same frame at each bracing level with varying bay
locations H/V ratio values are very close, H/V ratio is more when the brace is introduced at top
bracing level and amongst bays, little higher at central bay. In X braced frame central bays gives
maximum H/V ratio for all cases and difference in values from side bay is more for frame having
more number of stories and also for higher bracing level in the same frame. In bare frame also H/V
ratio for frame having more number of stories for same number of bays is more. Top story braces
of each frame case gives maximum H/V ratio. For diagonal braced frame, H/V ratio and Hr and Vr
has same pattern that is for each frame with the change in location of brace among bays at same
bracing level/story have nearly same ratios. For both diagonal braced and X braced frames Hr
ratios are less than one except for a single X brace at central bay of B3S3 frame. For X braced
frame central bay has maximum Hr ratio for each bracing level. Even though Vr is above one in
each case but opposite to pattern of Hr ratio, Vr is minimum for central bays.
866
The comparison of Hr and Vr ratio for fully braced frames and bare frame are given in Table 13.
Table has two parts, one is for shows bay-wise variation, with three stories and varying bays and
the second part shows variation story-wise, with three bays and varying stories.
Table 13. Hr and Vr ratio comparison for 3 stories and varying bays and for 3 bays and varying stories
Bays Diagonal X Brace Stories Diagonal X Brace No Brace
Full Hr Vr Hr Vr Full Hr Vr Hr Vr
B1S3 0.93 4.44 1.30 6.64 B3S1 0.88 3.52 1.60 5.14 1
B2S3 0.62 4.24 1.58 6.59 B3S2 0.78 4.02 1.70 6.11 1
B3S3 0.66 4.09 2.02 6.46 B3S3 0.66 4.09 2.02 6.46 1
Considering fully braced frames, pattern of H/V ratio for three bay with stories varying from 1 to 3
is very similar to that of three stories with bays varying from 1 to 3 that is rising curve for bare
frame with increase in stories and having ratio more than both diagonal and x braced frames for all
considered cases. X brace has increasing pattern with increase in number of stories. Only
difference here is that for diagonal frame, H/V ratio decreased for third story. A bit different from
the case of three stories with bays varying from 1 to 3, Vr for both diagonal and X braced frame
rises with increase in the number of stories with X brace having higher ratio. Hr also rises for X
braced frames in similar fashion as Vr but Hr ratio of diagonal braced frame decreases with
increase in number of stories even though for a particular frame Hr increases with rise in bracing
level of single diagonal frame having profile similar to that of considered here i.e. having same
cross-section as that of beams and columns.
5 Conclusion
It is observed that the lateral buckling resistance is generally influenced by very few braces at few
lower stories; except for the diagonal brace (having same cross-section as that of beams and
columns) for which bracing at few top bracing levels plays vital role and other braces are idle
regarding lateral buckling control. Against lateral buckling, inclusion of tension brace along with
compression brace provides stability to the structure. For vertical load, in general the bottom
braces are more significant in resisting buckling but for substantial buckling resistance increase,
every brace in each story is significant. As the stories increase the effect of bracing reduces
drastically for X-braced frames, but in case of chevron the reduction is minor. Even though under
vertical loading, reducing brace size at higher stories is not expected to influence the overall
stability of structure much but to prevent soft story and weak story problems, stiffness and strength
of few consecutive stories need to be nearly equal satisfying the limits of standard codes of that
country in which the structure is to be constructed. For this reason we need to provide bracing at
each story level so that there is no significant change in stiffness or strength of stories. In most of
the cases, symmetric bracing type with symmetric arrangement of bracing is expected to work well
against both vertical and lateral load and avoids premature buckling to some extent. For a
symmetrically framed structure, under lateral load applied from either side of the frame, one side
of frame acts under the action of tensile force and other under compressive force in a similar
fashion. For the frames made from the various combinations of 3 bays and 3 stories considering
the effect of particular bracing location under applied in-plane loadings, bracing in single loop of
frame is not enough for restraining the buckling due to lateral loading. Under lateral loading, for
B1S1 configuration (which is a fully braced configuration), using single X brace substantial
867
increase in buckling load value is observed, otherwise single brace is detrimental for strength of
the structure as it degrades the buckling strength severely. The structure can resist more lateral
load without any brace rather than the introduction of an ordinary diagonal brace in a single closed
loop of frame of any configuration (except B1S1configuration, for braces having buckling load
resistance more than that of bare frame). But if talking about the stability improvement against
buckling due to axial load (loading in vertical direction) even a single brace increases so much
stability against buckling due to vertical loading. Diagonal brace provides strength to structure and
even provides good stability under vertical loading but it is found to be detrimental for stability of
structure under lateral loading. The braces have cross-section same as that of beams and columns
and the slenderness ratio is nearly half of the maximum allowable for steel members but the
slenderness ratio is more than that of beam and columns, which makes it more prone to buckling.
For diagonal braced frame it is found that when the brace is made with providing slenderness ratio
equal to other main members of structure, even then the stability against lateral loading is lower
than that of bare frame either for the singly braced frame having maximum value at a particular
bracing level or for fully braced frame. For diagonal braced frame, what to speak of minimum
critical buckling load value amongst both lateral load directions, when the load is applied in the
lateral direction for which the diagonal brace is in tension then also in many cases the overall
stability of structure has been found to be less than that of bare frame due to buckling of beam or
column.
Many researchers suggest bracing central bays portion for both strength and stability against lateral
and transverse loading. In our study also for many cases central arrangement is found to be nearly
equivalent to the best possible cases. In general, bracing central bay throughout all stories, gives
better results in comparison to other for structure having configuration of single bay having multi-
stories but when it comes to more than one bay to be braced central is not always better. In this
study it is found that for more than one bay bracing, bracing at least both side end corner bays
gives higher buckling load capacity and are found to be the critical elements in deciding the
amount of increment in the buckling load resistance of frame in comparison to bare frame. It is
also found that in the arrangements not having corners braces, the beams of un-braced end corner
bay tends to buckle at a low buckling load. So for multiple bays our study suggest that corner
braces are very crucial to ensure stability against lateral loading and vertical loading and as a
wholesome conclusion are found to be the best choice regarding stability. Regarding stability, it’s
not necessary that more braces will resist more buckling load for every bracing type and
configuration. Story-wise we are supposed to brace at every bracing level to prevent the condition
of soft story or weak story but for bay-wise bracing we are not restricted to brace throughout bay-
wise also. So we must go for the best bracing arrangement. Here we have found that for stability
bracing corner bays is crucial. For bay-wise bracing it’s not necessary that full bracing would
provide more buckling resistance as in some cases it’s found that even bracing half the bays
provide nearly equal or some times more buckling resistance in comparison to full bracing.
As most of the researchers use single story frame to ascertain the buckling resistance of brace type
and based on that they conclude the competent behaviour of that brace. But it’s not necessary that
the brace providing better resistance against buckling for low-story structures would show same
868
pattern of resistance for high rise frames. Also even in an high rise structures itself, it’s not
necessary that a bracing type providing better resistance at lower stories would be showing same
kind of resistance pattern at upper stories. Also individually each brace could be more effective at
particular location in the structure but as a combined effect of whole braced frame its effect might
be significantly changed. But it’s found that the braces working well at particular location are
generally found to work well under the combined effect of whole bracing system too, if not
completely same but in near proximity of that effect. It is observed that the behaviour of braced
frame is not continuous for all the bracing types both story-wise and bay-wise. In a frame, the
effect of brace in restraining buckling in lower stories may be lesser or higher in comparison to
upper stories but it may respectively again rise or fall with the change in bracing levels. So
concluding based on arbitrarily considered number of stories or bays may not show the actual
behaviour of braced frame for either low-rise or high-rise building. It might give misleading results
(similar results are shown between fully braced chevron and X brace). According to our study we
suggest to go for at least 3 variations of bays and 3 variations of story levels as low, medium and
higher stories to approximately access the characteristic behaviour of a bracing type.
References
1. Chajes, A.: Principles of structural stability theory. Prentice-Hall Civil Engineering and Engineering
Mechanics Series, New Jersey (1974)
2. Brush, Don O and Almroth, Bo O.: Buckling of bars, plates, and shells. Vol. 6. McGraw-Hill, New York
(1975)
3. Popov, Egor P and Kasai, Kazuhiko and Engelhardt, Michael D.: Advances in design of eccentrically
braced frames. Bulletin of the New Zealand National Society for Earthquake Engineering. Vol. 20, No. 1
(1987)
4 Sabelli, R.: Research on improving the design and analysis of earthquake-resistant steel-braced frames.
EERI, NEHRP Professional Fellowship in Earthquake Hazard Reduction, Final Report, Oakland (2001)
5. Amini, A Moein and Majd, M and Hosseini, M.: A Study on the Effect of Bracing Arrangement in the
Seismic Behavior Buildings with Various Concentric Bracings by Nonlinear Static and Dynamic Analyses.
15th WCEE, Lisbon, Portugal (2012)
6. Choi, Kyung-Suk and Park, Jin-Young and Kim, Hyung-Joon: Numerical Investigation on Design
Requirements for Steel Ordinary Braced Frames, Engineering Structures Vol. 137, pp. 296-309, (2017). doi:
10.1016/j.engstruct.2017.01.066
7. Abaqus, Ver.: Abaqus 6.14 Documentation. Abaqus User’s Manual (2014)
8. BIS: Draft National Building Code of India: Part 6 Structural Design, Section 6 Steel, Third Revision of
SP7 (Part 6/Section 6), CED 46 (8041) WC, New Delhi (2015)
869
Paper ID- 165
Structural Evaluation of Existing RCC Bridges for Durability

Durgesh Hingnekar a, Abhay Tawalare b
a- PhD Scholar, Department of Applied Mechanics, Visvesvaraya National Institute of Technology, Nagpur-
440010; Email: durgesh.hingnekar@gmail.com
b- Assistant Professor, Department of Civil Engineering, Visvesvaraya National Institute of Technology,

Nagpur-440010; Email: abhaytawalare@civ.vnit.ac.in
Abstract
The paper discusses the structural evaluation of some major highway RCC bridges with precast girders constructed
about six years ago as per new codal provisions of IRC 6:2016. The existing bridges were designed as per IRC 6: 2010.
Recently some cracks were observed on the deck slab and longitudinal girders of these bridges due to heavy traffic
leading to nearby port. The objective of structural evaluation was to assess the durability of the existing bridge in
future. For this purpose, it is necessary to verify the existing bridges for structural safety, fatigue safety and
serviceability based on the requirements of current IRC 6:2016. Therefore, major focus of the evaluation was to check
the adequacy of existing deck slab and longitudinal girders to carry the new loads as per IRC 6:2016, specially the IRC
Class SV loading with 20 axles and a cumulative load of 385 tonnes. A moving load analysis was carried out for SV
loading as per the specification of IRC 6:2016 by using the finite element software package. The stresses in concrete
and reinforcement were calculated for new load to check whether it is crossing the limit considered in old designs. In
addition, fatigue safety was also carried out with 3 axles loading with cumulative load of 40 tonnes as per IRC 6:2016.
Based on the analysis suggested in IRC 58, life to failure was calculated for individual structural elements. Finally,
remedial strengthening measures for the new loadings are suggested based on the structural evaluation results.
1. Introduction
During the last three decades after globalisation, India has seen tremendous amount of growth in commerce which has
led to large infrastructure development activities especially with regards the road connectivity. A lot of construction
activities with regards to bridges have happened since then. But now due to tremendous speed of growth the traffic
volumes and the running speed of vehicles are increasing day by day. Not only has that with advancement in
automobile heavy axle trucks with heavier road freights are becoming a common picture. So, some of the earlier
bridges are now subjected to demands which they are not designed for. This greater demand has resulted into
reassessment and requirement for strengthening of many existing bridges.
Extensive research on the complex phenomenon of deterioration of concrete in service and premature durability issues
has been reported around the globe. Tay et al. [1] reported about the In situ investigations related to strength of
deteriorated concrete. Deterioration of concrete structures has been studied by Basheer et al. [2] using predictive
models. Hover [3] studied some specific problems in safety evaluation of concrete bridges. Broomfield [4] discussed
different options to repair and rehabilitate the existing damaged RCC structures. Gjorv et al. [5] carried out the study on
the advances in durability design and performance based specifications.
Existing bridges require regular maintenance to ensure sufficient structural safety and serviceability. Maintenance
decisions are supported by various monitoring and inspection methods. The performance of existing bridges may be
characterized by the durability, probability of failure and the associated risks. This paper gives an overview of the
structural evaluation of some existing reinforced concrete bridges with precast girders.
2. Study Approach
2.1 Assessment Methodology
The assessment of the bridges is generally carried out in line with some standard process, one of the example of such
process is shown in the flowchart in Figure 1.
870
Figure 1: Flowchart for the general assessment of existing structures (from ISO 13822:2012, Annexure E)[6]
2.2 Structural Evaluation

During the service life of any bridge, it is subjected to fatigue and wear-tear due to moving vehicle loads. Overloading
due to increase in wheel loads and continuous exposure to aggressive environment may aggravate the deterioration.
Hence, to evaluate the existing structure generally the in-situ material properties of the structure are used. The actual
loads which the bridge structure is currently likely to be subjected to are also evaluated. From durability point of view
generally analysis also includes moving load analysis and fatigue safety evaluation.
2.3 Moving Load

Structural evaluation of bridge for moving load is generally carried out as per IRC 6:2016[7]. In some cases where the
bridges are constructed on highways frequented by the heaviest multi-axle loads, the special vehicle loading proposed
in IRC 6: 2016 should be considered. The IRC Class (Special vehicle) SV loading, multi axle hydraulic vehicle with 20
axles and a cumulative load of 385 tonnes is the heaviest loading specified in the code. The details of the loading are as
follows:
Figure 2: Typical 20 axle load for IRC class SV loading (385 tonnes) (from IRC 6:2016)
871
2.4 Fatigue Load

In order to check the fatigue induced in the deck slab due to frequent movement of the traffic, fatigue load analysis is
generally carried out during the durability assessment process of existing bridges. According to the definition of the
fatigue life, the approaches for fatigue analysis can be classified into the Crack propagation method or Fracture
mechanics approach, the Strain-life method and the Stress-life method or S-N curve approach. The Stress Life or S-N
method is one of the most popular method to calculate fatigue life, in fact it is one of the oldest method which is still
used in design specifications. To obtain S-N curve, several test specimens are subjected to cyclic load with uniform
amplitude. The number of cycles required to failure of specimens is observed. The logarithm of the number of cycles is
plotted against the maximum stress level for a wide range of stress levels gives S_N curve.
IRC 58[10] gives the S-N curve values using the following formulae,
[ ]
The fatigue load considered as per IRC 6:2016, clause 204.6 states that “ The stress range resulting from the single
passage of the fatigue load along the longitudinal direction of the bridge, shall be used for fatigue assessment with the
fatigue load so positioned as to have worst effect on the detail or element of the bridge under consideration.”
Figure 3: Typical Fatigue load (40T)
3. Case Study Description

The durability assessment of some major bridges using the approach described in the previous section is discussed in
this section.
3.1 Problem
These major highway RCC bridges with precast girders were constructed about six years ago as per codal provisions of
IRC 6:2010[8]. These bridges were located in western region of India and cater to a major port with heavy traffic.
Recently some cracks were observed on the deck slab and longitudinal girders of these bridges due to heavy traffic
leading to nearby port.
872
The bridges were initially designed for the moving load combinations as per IRC 6:2010[8]. But as the bridges were
constructed on highways frequented by the heaviest multi-axle loads, the bridges were needed to be checked for the
special vehicle loading proposed in IRC 6: 2016[7]. The adequacy of existing deck slab and longitudinal girders to
carry the new loads as per IRC 6:2016, specially the IRC Class (Special vehicle) SV loading with 20 axles and a
cumulative load of 385 tonnes, was a concern. A moving load analysis was carried out for SV loading as per the
specification of IRC 6:2016 by using the finite element software package STAAD Pro V8i [9].
Visual inspection of the bridges revealed cracks on the deck slab and precast girders Some cracks which were observed
in the deck slab as shown in the pictures in Figure 4.
Figure 4: Cracks in deck slab

3.2 Structure description
The RCC T girder bridges were designed as Major Bridge on normal highway crossing with deck slab arrangement
supported on four longitudinal girders. The carriage way width was 11.25m and overall width was 14m. The length of
span varied from 16.5m to 21m. The cross sectional details of the bridges are as given in Figure 5.
873
Figure 5: Typical section of Bridges and girder details
3.3 Structure Analysis-Moving load

The superstructure was modelled in STAAD Pro. V8i using beam and plate elements, a plan view of typical STAAD
model is given in Figure 6.
Figure 6: Typical details of STAAD model geometry (Plan)
The load cases considered in the analysis as well as the load factors are as per the clause 204.5.4 of IRC 6:2016. The
load for the Special vehicle is considered to be acting as a pressure load on deck modelled as plate elements. The
magnitude and area of application pressure loading is calculated tyre area given in the IRC 6 code dispersed as per
Pigeaud’s method. The load is acting at the centre of the carriageway as specified in given in IRC 6:2016. Figure 6
shows the load as applied in STAAD model.
Figure 7: Typical IRC class SV loading as applied in STAAD model - 3D view
3.4 Structure Analysis- Fatigue Load
874
IRC 6:2016, clause 204.6, states that “The bridges close to areas such as ports, heavy industries and mines and other
areas, where generally heavy vehicles ply shall be designed for the stress induced due to 10 x 106 cycles.” So number
of cycles as per IRC 6:2016 is taken as 10 x 106.Normally, concrete suffers fatigue failure prior to steel. Hence fatigue
in concrete is calculated and Applied in the model in the form of three different cases as shown in Figure 8, such that
maximum effect of the loading is observed.
Figure 8: Typical Fatigue load (40T)

4. Results
The results of the loadings are compared with the permissible stresses as per IRC 21:2000[11]. The results for SV
loading and Fatigue loading are presented in the following sections.
4.1 SV loading
Figure 9: Moments in decks slab of bridges
875
Table 1: Moments and stresses in decks slab of bridges
S. No Span Moment Comp stress Concrete Tensile stress in Steel

kNm/m MPa MPa
1 Mid Span 20m 43.8 Sagging 10.94 OK 302.00 Unsafe
2 Mid Span 16.5m 35.5 Sagging 8.86 OK 244.77 Unsafe
3 Mid Span18.8m 39.82 Sagging 9.94 OK 274.56 Unsafe
4 Mid Span 21m 47.25 Sagging 11.80 OK 325.79 Unsafe
Figure 10: Maximum tensile stresses in reinforcement of decks slab of bridges (permissible stress 240MPa)
From Table 1 and Figure 10, it is clear that the stresses in concrete of deck slab for special vehicle load is within
permissible limits but the same is not true in case of the slab reinforcement, where the reinforcement is stress beyond its
permissible limits. Hence, the deck slab defiantly needs strengthening, especially in the area highlighted in grey in
Figure 9.
Figure 11: Maximum compressive stresses in concrete for external and internal girder of bridges (permissible stress
13.33MPa)
876
Figure 12: Maximum tensile stresses in reinforcement of for external and internal girder of bridges (permissible stress
240MPa)
The maximum compressive stresses in concrete for external and internal girder of bridges are well within the
permissible limits for special vehicle loading as shown in Figure 11. But, the stresses in reinforcement of internal girder
of one of the bridges are increasing beyond the permissible limit of 240MPa as shown in Figure 12, hence this girder
needs to be strengthened.
4.2 Fatigue loading

Case one of Fatigue loading with three trucks standing nearly at the mid-span of the bridge deck gives most critical
results. The results of span moment of the deck slab is shown in Figure 13 below.
Figure 13: Moments in decks slab of bridges due to fatigue loading
877
Table 2: Fatigue safety check for decks slab of bridges
Fatigue Modulus
Depth Flexural stress Stress Design
Moment of rupture N Ratio
mm in concrete MPa ratio, SR Cycles
kNm/m MPa
17 225 2.0 4.43 0.455 2.72E+07 1.00E+07 2.72 OK

14.8 225 1.8 4.43 0.396 unlimited 1.00E+07 Safe OK
From Figure 13 and Table 2, it is clear that the bridge deck is safe for fatigue loads.
5. Conclusion and Recommendations

The cracks were observed in several panels both at the top and bottom of the deck slab. Cracks in some cases were
appeared to be due to shrinkage of concrete due to poor curing and their depths were limited to the cover portion only.
But the same cannot be said about all the cracks and therefore structural evaluation of the deck slab of these bridges
was needed. After structural evaluation it was found out that the tensile stresses in deck slab reinforcement were
exceeding the permissible limits. Hence the design of deck slab for SV loading was inadequate and hence strengthening
of the bridge deck was necessary. In line with the conclusion some possible strengthening measures are suggested in
this section. Several structural repair techniques for reinforced concrete bridges are employed in such cases based on
the type of damage.
I. For small cracks which do not penetrate beyond the clear cover of the deck slab a low viscosity epoxy can be
injected in the cracks. Otherwise non shrink grout of approved material can also be used.
II. For more serious cracks which led to reduction in the load carrying capacity of the deck slab and wherever the
deck slab is found out to be insufficient strengthening techniques can be used. Other strengthening methods
that can be used which are as follows:
i. Steel plate bonding: In this method steel plates of calculated thickness are bonded to the members to increase
their strength or stiffness. The bonding between steel plates and members is made by using adhesives or
anchors. This method requires a lot of drilling and hacking in the section to be retrofitted and the steel plates
are also prone to corrosion.
ii. FRP Strengthening: A Fibre Reinforced Polymer (FRP) is typically made by glass or carbon fibres which are
high tensile in nature; they are oriented in a desired direction where the tensile strength is required. FRP can be
really helpful in increasing the shear, flexural strength and enhancing the ductility of the member.
Strengthening using FRP is fast, less labour intensive and aesthetically pleasing unlike steel plates. FRP is also
light weight and is not prone to corrosion. Only issue that needs to be addressed in FRP is regarding the
duarability as it needs suitably designed protective coatings to protect it from exposure to adverse
environmental conditions like acidic environment, ultraviolet rays, impact, temperature variation, fire etc.
6. References
1. D.C.K. Tay and C.T. Tam, In situ investigation of the strength of deteriorated concrete, Construction and
Building materials, Vol. 10, No. 1, 1996.
2. P.A.M. Basheer, S.E. Chidiac and A.E. Long, Predictive models for deterioration of concrete structures,
Construction and Building materials, Vol. 10, No. 1, 1996.
3. Kenneth C. Hover, Special problems in evaluating the safety of concrete bridges and concrete bridge
components, Construction and Building materials, Vol. 10, No. 1, 1996.
4. Odd E. Gjorv, Advances in durability design and performance based specifications, Chapter 13, Concrete
Durability
5. John Broomfield, Repair methods, Chapter 11, Concrete Durability
878
6. ISO, C. (2005). 13822 Bases for design of structures-Assessment of existing structures. CEN Brussels
7. Indian Road Congress (IRC) 6:2016, Standard Specifications and Code of Practice for Road Bridges
8. Indian Road Congress (IRC) 6:2010, Standard Specifications and Code of Practice for Road Bridges
9. STAAD Pro v8i, Bentley System, Inc.
10. Indian Road Congress (IRC) 58:2002, Guidelines for design of rigid pavements, Indian Road
Congress, New Delhi, India
11. Indian Road Congress (IRC) 21:2000, Standard Specifications and Code of Practice for Road
Bridges
879
Paper ID- 166
Experimental study of concrete prepared as a partial replacement of

cement by various admixtures
Shamsuz Zaman1, Dr. Nayanmoni Chetia2
1
P.G Student, Jorhat Engineering College, Jorhat, Assam
2
Assistant Professor, Civil Engineering Department, Jorhat Engineering College, Jorhat, Assam
Email: shamsuzzamanaei62@gmail.com Email: nayanmoni.chetia@gmail.com
Abstract
This study investigates the behaviour and strength of concrete by replacing cement with various admixtures such
as Alccofine, Bentonite, Rice husk ash at different proportions for a mix of M25 grade of concrete. Many trials
are adopted at different proportion such as for Alccofine 5% to 15%, Bentonite 10% to 35%, RHA 5% to 15%
etc. Slump test, compaction factor test, test for bulk density, compression test and split tensile tests are carried
out to check the quality and strength of the specimen. Tests for Compressive strength are performed for 7days,
28 days and 56 days of curing and split tensile tests are carried out for 28 days of curing. After analysing the
various test results, it is observed that replacement of cement by Alccofine at 20%; Bentonite at 15% and RHA
at 5% by weight shows the maximum improvement factor and also the combination of the mentioned
admixtures in different proportions shows satisfactory results. Regression analysis namely SPSS technique has
been used for regression analysis and modelling and an equation has been proposed. Cost analysis has been also
provided for different proportions.
Keywords: Compressive strength; Split tensile strength; Improvement factors, Regression analysis.
1. Introduction and literature alccofine. Suresh et al. (2017) performed tests with
reviews different water-binder ratio of 0.45 and 0.50
designed for M25 with 10%, 20%, 30% replacement
Concrete is the most widely used construction by bentonite. The compressive strength and tensile
material; being used for different purposes to make it strength are determined at 7, 28 days. The result
suitable in various conditions. Sometimes ordinary shown at 10% replacement of cement by bentonite
concrete may fail to exhibit the required quality gives maximum compressive strength as well as
performance or durability. There has been worldwide tensile strength, while increasing in percentage of
attempt to find suitable and effective alternative bentonite strength will decreases.
materials which can considerably minimize the use
of cements, improve the quality of concrete and also 2. Methodology and Materials
suitable for environments. In this paper some 2.1 Methodology
admixtures have been used as partial replacement The materials used in the project work are OPC – 43
OPC cement are Alccofine , Bentonite, Rice husk grade of cement (Dalmia), coarse sand from
ash at various proportion. Magdum et al. (2016) and Kanaighat (Kalioni River), coarse aggregates
Pawar et al. (2013) prepared high performance (Bihubor). Locally available Rich husk ash, wood
concrete by replacing OPC with Alccofine -1203 in ash and wood powder are collected. CCB treated
different proportion. In their research work they Bamboo fibres and Bamboo powder collected from
found the fresh and harden properties of SCCs with Rain Forest Research Institute, Jorhat (ICFRE) and
adding Alccofine is superior that of the other another admixture Fly ash is taken from Rajasthan
variations. Marthong (2015) performed a series of NTPC with the help of TOPCEM INDIA Plant’s
tests to investigate the effects of concrete properties Quality Control Department. Others admixtures are
with OPC of varying grades, partially replacement collected from nearby market. The Mix design has
with RHA in the range of 0, 10, 20, 30 and 40% been performed as per IS 10262:2009. The code that
respectively. The compressive strength, water has been used for physical properties of cement, FA,
absorption, shrinkage and durability of concrete were CA and admixtures are IS 383-1970, IS 8112: 2013,
mainly studied. The study suggests that up to 20% IS 4031:1996 (P-1), IS 269-2015, IS 5513-1959, IS
replacement of OPC with RHA has the potential to 2720 (P-3), IS 1199-1959, IS 516-1959, IS 456:2000
be used as partial cement replacement, having good and IS 5816-1999.
compressive strength performance and durability. 2.2 Materials
M.S Pawar et al. (2013) from their experimental
investigations studied the filling ability, passing 2.2.1 Alccofine: Alccofine is mineral admixture
ability and resistance to segregation are the self- produce by Ambuja Cement Ltd. It has unique
compatibility characteristic are increased by adding characteristics to enhance “performance of concrete‟
alccofine in SCC mixes and they found the fresh and in fresh and hardened stages due to its optimized
harden properties of SCCs with adding of 10% particle size distribution. It is easy to use and can be
alccofine is superior that of 5% and 15% of added directly with cement, ultrafine particle of
880
alccofine provide better and smooth surface finish. It 2.2.3 Rice husk ash: Rice milling industry generates
also lowers the water/binder ratio. If the percentage a lot of rice husk during milling of paddy which
level of alccofine is increased beyond that level it comes from the fields. It is estimated that about 70
acts as a filler material and yields good workability million tons of RHA is produced annually
to the concrete. There are two types of Alccofine, worldwide. India being the second largest rice
Alccofine1203 is essential in terms of reducing heat producer in the world, systematic approach to this
of hydration and strength at all stages whereas material can give birth to a new industrial sector of
Alccofine 1101 can be used as a grouting purpose. rice husk ash in India. Every year approximately 120
million tons of rice is produced in India. This gives
around 24 million tons of rice husk. This rice husk is
mostly used as a fuel in the boilers for processing of
paddy. Rice husk ash (RHA) is about 25% by weight
of rice husk when burnt in boilers. Therefore about 5
million tons of Rice Husk Ash is produced every
year in India
Fig. 1: Alccofine – 1203

Alccofine 1203: It is an alccofine with low calcium
silicate. Alccofine 1200 series is of 1201, 1202, 1203
which represents fine, micro fine, ultrafine particle
size respectively. Alccofine 1203 is a slag based
SCM having ultra-fineness with optimized particle
size distribution. Alccofine 1203 provides reduced
water demand for a given workability, even up to
70% replacement level as per requirement of
concrete performance Fig. 3: Rice Husk Ash
2.3 Physical properties of materials and Concrete
2.2.2 BENTONITE: Bentonite is a fine clay
material mined from the earth, formed by the The various physical properties of cement are given
decomposition of volcanic ash. It is an absorbent in Table 1
aluminium phyllosilicate, impure clay consisting Table - 1: Physical properties of OPC – 43 Grade
mostly of montmorillonite. It presents strong Cement
colloidal properties and its volume increases several Tests Value
times when coming into contact with water, creating Consistency of cement 31.50%
a gelatinous and viscous fluid. Bentonite acts as Initial Setting Time 2.05 hours
natural pozzolan in ordinary Portland cement. A
Final Setting Time 9.10 hours
pozzolan is siliceous or aluminous material which
itself possesses equivalent to zero percent cementing Fineness 92%
properties, but in the presence of moisture it Avg. Specific gravity 3.14
chemically reacts with calcium hydroxide at ordinary Colour Grey
temperature to form compounds possessing
cementitious properties. When water is added in a The various physical properties of FA and CA are
mixture of OPC and pozzolan, its silica component given in Table 2
reacts with liberated calcium hydroxide in hydrated Table -2: Physical properties of FA and CA
cement paste. It is widely used as a drilling mud for Sample Avg. Water Avg. Specific Size in mm
oil and natural gas wells. absorption gravity
FA 1.181 2.77 4.75 down
CA 0.523 2.525 20 down
The various physical properties of Admixtures are

given in Table 3
Table -3: Physical properties of various Admixtures
Sample Fineness Avg. Specific Colour
gravity
Alccofine 96.5% 2.875 White
Bentonite 74% 2.195 Brown
RHA 19.19% 2.25 Black
Fig. 2: Bentonite
881
Table - 8: Compressive strength of specimen with

The various physical properties of Concrete are Rice husk ash as an admixture
given in Table 4 % of Average Compressive strength Improvem
Admixture in N/mm2 ent factor
Table - 4: Slump and Compaction factor used 7 days 28 days 56 days at 28 days
Sample Slump Compaction factor NC 0% 28.29 39.26 42.92 1.0
NC 48 0.931 R1 5% 20.33 35.23 41.89 0.897
A4 60 0.953 R2 10% 13.44 27.88 31.67 0.710
B2 41 0.933 R3 15% 8.780 18.56 20.99 0.473
R1 38 0.906
AB4 44 0.937 The comparison on compressive Strength at 7, 28
days and 56 days of curing of different admixtures in
3. Result and discussion various proportions with 0% admixture are given in
the fig. 4, 5, 6 and 7
The cubes with various proportion, with different
amounts admixtures are tested for Compressive
50
strength at 7, 28, 56 days of curing are given in the
Compressive strength in
40 7 days
Table 5, 6, 7 and 8 having w/c 0.45
30
28 days
N/mm2
Table - 5: Compressive strength of specimen with 20
Alccofine as an admixture 10
% of Average Compressive Improve 0 56 days
Admixture strength in N/mm2 ment NC A1 A2 A3 A4 A5
used 7 days 28 days 56 days factor at % of Admixture used
28 days
Fig. 4: Alccofine used as admixture
NC 0% 28.29 39.26 42.92 1.0
A1 5% 24.88 34.28 41.00 0.873
A2 7.5% 24.44 36.78 42.33 0.937 50
Compressive strength
A3 10% 24.44 40.78 43.45 1.039 40 7 days

A4 12.5% 25.34 42.67 44.11 1.087 30
in N/mm2
28 days
A5 15% 24.11 35.56 40.33 0.906 20
10 56 days
Table - 6: Compressive strength of specimen with 0
Bentonite as an admixture NC B1 B2 B3 B4 B5 B6
% of Average Compressive Improve % of Admixture Used
Admixture strength in N/mm2 ment
used factor at
Fig. 5: Bentonite used as admixture
7 days 28 days 56 days
28 days
60
NC 0% 28.29 39.26 42.92 1.0
B1 10% 22.35 32.00 34.22 0.815 7 days

40
B2 15% 24.82 34.00 35.77 0.866
28 days
N/mm2
B3 20% 19.33 24.44 27.33 0.623

20
B4 25% 17.21 21.78 25.11 0.555
B5 30% 12.31 18.22 22.88 0.464 56 days
0
B6 35% 9.070 12.89 17.56 0.328 NC R1 R2 R3
% of Admixture used
Table - 7: Compressive strength of specimen with
Alccofine and Bemtonite as an admixture Fig. 6: Rice Husk Ash used as admixture
% of Average Compressive Improvem
50
Admixture strength in N/mm2 ent factor
Compressive Strength in
used at 28 days 7 days

NC 28.29 39.26 42.92 1.0
AB1 22.67 37.56 39.56 0.957 28 days
N/mm2
AB2 20.67 3.890 36.67 0.889

AB3 19.56 32.00 34.44 0.815 0 56 days
AB4 26.67 39.33 43.11 1.002 NC AB1 AB2 AB3 AB4
% of Admixture used
Fig. 7: Alccofine and Bentonite used as
Admixture
882
Some failure specimen at 28 days of compressive Some failure specimen at 28 days of split tensile
strength is give in fig. 8 and 9 strength is given in fig. 11 and 12
Fig. 8: Alccofine (10%) Fig. 11: Alccofine (12.5%)
Fig. 9: Rice husk ash (5%) Fig. 12: Normal concrete

The Split Tensile Strength and Bulk density of
various admixtures with 0% admixture are given in 4. Regression Analysis
the Table 9
Regression analysis is a set of statistical processes
Table – 9: Split tensile strength and Bulk Density
for estimating the relationships among variables. It
Admixt Split tensile Improveme Avg. Bulk includes many techniques for modelling and
ure used strength at 28 nt Factor Density analysing several variables, when the focus is on the
days relationship between a dependent variable and one or
NC 2.900 1.000 2320.000 more independent variables (or 'predictors').
A4 3.254 2.971 2213.333
B2 2.759 2.051 2234.667 4.1 Regression by Minitab
R1 2.122 2.405 2205.333
AB4 2.688 2.829 2205.333 Minitab is a software product that helps us to analyze
the data; which is designed essentially for the Six
Sigma professionals. It is a popular & powerful
Comparison on Split Tensile Strength at 28 days of
statistical software package that provides a wide
curing of different admixtures in various proportions
range of data analysis capabilities. It provides a
with 0% admixture are given in the fig. 10 simple, effective way to input the statistical data,
manipulate that data, identify trends and patterns,
1.200 and then extrapolate answers to the current issues.
Improvement Factor
1.000 NC Minitab provides a quick, effective solution for the

A4 level of analysis required in most of the Six Sigma
0.800
projects.
0.600 B2 The F-ratio in the ANOVA Table tests whether the
0.400 R1 overall regression model is a good fit for the data.
0.200 AB4
The Table - 10 shows that the independent variables
statistically significantly predict the dependent
0.000
.
variable, F (6, 12) =18.25 (i.e., the regression model
Types of concrete used
is a good fit of the data).
Fig. 10: Split Tesnsile Strength
883
Table - 10: ANOVA Table 4.3 Regression Equation
Source DF Adj SS Adj MS F- P- From the regression co-efficient obtained from

Value Value Minitab, the predict compressive strength at 28 days
Regression 6 1172.42 195.403 18.25 0.000 expressed in general form of the equation i.e
CC 1 0.20 0.199 0.02 0.894
A 1 0.18 0.180 0.02 0.899 Predicted Average compressive strength at 28 days
B 1 0.09 0.088 0.01 0.929 in N/mm2 = -176 + 0.52 × Cement content + 0.50 ×
R 1 0.03 0.029 0.00 0.959 A + 0.34 × B + 0.20 × R + 0.52 × AB11 + 0.46 ×
AB11 1 0.19 0.189 0.02 0.897 AB22 …………Eq.(a)
AB22 1 0.16 0.157 0.01 0.906
Error 12 128.52 10.710 - - 5. Cost Analysis
Total 18 1300.94 - - - The rate of different materials and labours are taken
as per market price and cost analysis of concrete
The Table - 11 provides the R, R2, adjusted R2, and work are carried out and given in the Table –14 and
the standard error of the estimate, which can be used Table - 15
to determine how well a regression model fits the
data: Table - 14: Rate of different Items and labours as per
Table - 11: Model Summary Market price
S R-sq R-sq(adj) R-sq(pred) Materials Quantity Rate
3.27258 90.12% 85.18% 53.79% CA 7.630 cu.m Rs. 2100/cu.m
FA 4.687 cu.m Rs. 1000/cu.m
Cement 2.883 cu.m Rs. 440/bag
The co-efficient obtain from Minitab for calculating
Admixture Alccofine Rs. 27/kg
predicted compressive strength at 28 days are given
Bentonite Rs. 5/kg
in Table – 12
RHA Rs. 0
Table – 12: Regression Co-efficient
Labour:
Term Coef SE T- P- VIF
Coef value value Head mason 1/3 Rs. 450 Rs. 116.67
Constant -176 1585 -0.11 0.913 Mason 2 Rs. 400 Rs. 660
CC 0.52 3.79 0.14 0.894 34162.65 Labour (male) 12 Rs. 300 Rs. 3600
A 0.50 3.82 0.13 0.899 9628.71 (Female) 20 Rs. 250 Rs. 5000
B 0.34 3.81 0.09 0.929 57094.22 Water boy 6 Rs. 250 Rs. 1500
R 0.2. 3.92 0.05 0.959 6765.67 Total Labours Cost per 10cu.m of concrete work
AB11 0.52 3.89 0.13 0.897 5671.39 Rs. 10876.67
AB22 0.46 3.83 0.12 0.906 11682.22
Table - 15: Cost Analysis of per cu.m of concrete
The Measured and Predicted Compressive Strength work as per Market price
at 28 days of testing are given at Table 13 Mixed Rate per % of decreased in
Total
Proportion cu.m cost per cum work
Table – 13: Measured and Predicted Compressive NC 77813 7781.34 0.00
Strength at 28 days A4 88800 8879.98 -14.12
Average B2 75061 7506.05 3.54
Predicted
Compressive R1 75686 7568.59 2.73
Compressive
Notation strength in N/mm2 AB4 76020 7602.04 2.30
strength in N/mm2
at 28 days of
at 28 days of curing
curing
6. Conclusion
NC 39.26 38.93
A 34.28 38.52 The experimental observation had shown different of
B 32 31.49 improvement in compressive strength for various
R 35.23 32.32 admixtures at different proportion. The partial
AB11 37.56 38.93 replacement of cement was carried out for M25
AB22 32 34.59 grade of concrete.
▪ At 12.5% partial replacement of cement by
4.2 Fixing Dependent variable and predictors Alccofine, observed improvement factor
1.087 and it increased 14.12% cost per
a. Dependent Variable: Average Compressive cu.m of concrete work.
strength in N/mm2 at 28 days of curing ▪ At 15% partial replacement of cement by
b. Predictors: CC, A, B, R, AB11 and AB22 Bentonite, observed improvement factor
884
0.866 and reduce 3.54% cost per cu.m of those who have given their undivided support will
concrete work. not be forgotten.
▪ At 5% partial replacement of Rice Husk
Ash, observed improvement factor 0.897 Reference
and reduce 2.73% cost per cu.m of
concrete work. 1. Marthong, C (August – 2012) “Effect of Rice
▪ When the partial replacement of Alccofine Husk Ash (RHA) as partial replacement of
+ Bentonite is considered at a proportion of cement on concrete properties” International
10% of Alccofine + 10% of Bentonite Journal of Engineering Research & Technology
observed improvement factor 1.002 and (IJERT) Vol. 1 Issue 6
reduce 2.30% cost per cu.m of concrete 2. Magdum, Manisha M et al. (2016) “Influence of
work. Mineral Admixture (Alccofine-1203) On the
In this study some alternatives are tried to improve Properties of Hybrid Fiber Reinforced Concrete”
the quality of concrete, reduce in the Carbon dioxide American Journal of Engineering Research
emission and make an eco-friendly environment. (AJER) p-ISSN: 2320 - 0936 Vol 5, Issue-10, pp
72-75
3. Pawar, M.S; Saoji, A.C (2013) “Effect of
7. Acknowledgements Alccofine on Self Compacting Concrete” The
I would like to express my special thanks of gratitude International Journal of Engineering and Science
to Dr. Nayanmoni Chetia, Assistant Professor, Civil (IJES) Vol-2, Issue-6, Pages (05-09) | ISSN(e):
Engineering Department, Jorhat Engineering 2319 – 1813 ISSN(p): 2319 – 1805
College, Jorhat, Assam for her excellent guidance 4. Suresh, R; Soundarya, M K; Anjan M (2017)
and encouragement and support during the course of “Incorporating Bentonite Partially in place of
my work that I came to know about so many new Cement-an assessment of Strength and
things. I am very thankful to all the Scientists of Durability properties” International Journal of
RFRI, Jorhat for their constant support. I would also Civil Engineering and Technology (IJCIET) Vol
like to thank all the members of TOPCEM INDIA - 8, Issue 11, pp. 294 – 305 ISSN Print: 0976 –
Plant’s Quality Control Department for helping in 6308
the collection of sample. Finally, thanks again to
Notations
CA Coarse Aggregates B3 Bentonite 20%
FA Fine Aggregates B4 Bentonite 25%
OPC Ordinary Portland Cement B5 Bentonite 30%
CCB Copper Chrome Boron B6 Bentonite 35%
CC Cement Content R1 Rice Husk Ash 5%
ANOVA Analysis of Variance R2 Rice Husk Ash 10%
NC Normal concrete R3 Rice Husk Ash 15%
A1 Alccofine 5% AB1 Alccofine 7.5% + Bentonite 15%
A2 Alccofine 7.5% AB2 Alccofine 10% + Bentonite 15%
A3 Alccofine 10% AB3 Alccofine 7.5% + Bentonite 10%
A4 Alccofine 12.5% AB4 A Alccofine 10% + Bentonite 10%
A5 Alccofine 15% AB3 Alccofine 10% + Bentonite 0%
B1 Bentonite 10% AB4 Alccofine 0% + Bentonite 10%
B2 Bentonite 15%
885
Paper ID- 168
“Dampness: A silent killer in structural buildings- A Case study”

Leishangthem Yaiphaba Meitei1, Thokchom Kiranbala Devi2
1
M.Tech Student Civil Engineering, Manipur Institute of Technology, Imphal, India
2
Asst. Professor Civil Engineering, Manipur Institute of Technology, Imphal, India
Email id. yaiphapple@gmail.com, kiranbala_th@yahoo.com
Abstract:Endless damage in structural building by dampness has become present issue in

Manipur. In this paper, two different spots of structural buildings constructed at urban and rural
area was taken for comparing, proper field survey regarding the dampness mishap was inspected
and research for finding better solution for low change of dampness.This paper examines causes,
effects and remedies of dampness in Imphal, Manipur. Dampness affects both the exterior and
interior face of the structures, risk factor for health and demolishes the affected part, repairs this
will make extra cost and become less economy. This havoc can be minimized by using
waterproofing compound, easily available in market namely Dr Fixit 101 and Aqutitewater
proofing. It can be used to reduce the pore on the concrete and reduce the rising moisture inside
the structures. A sorptivity test was conducted in 63 cubes of 100x100x100mm as M15, M20 and
M25 nominal mix, by using different quantity of water proofer and from the experimental results
using Dr Fixit (2ml, 3ml, 4ml and 5ml) in M25 sorptivity value is approximately equal to using
Aqutite (20ml, 17ml, 14ml, 12ml and 10ml) in M20 sorptivity value respectively and referred to
used Dr Fixit and Aqutite in concrete grade not less than M20. Higher the amount of Dr. Fixit;
lower the sorptivity value and higher the amount of Aqutite; higher the sorptivity value.
Keywords: Dampness building, Causes, Effect, Remedies,Waterproofing.
1.0 Introduction
Manipur is a state in Northeast India, with the city of Imphal 24.8170°N, 93.9368°E as its capital.
The State covers an area of 22,327square kilometers and has a population of 23, 88,634. Concrete
is the most widely used material all over the world and difficult to find out alternative materials
for construction which is suitable as that of such material from durability and economic point of
view [1]. Permeability through water passage also major concerned about dampness in Manipur,
common havoc were due to un-maintained thickness of Damp Proof Course (DPC) and low
concrete grade while constructing structural buildings. Dampness refers slightly moisture in the
wall due to capillary action. It is the common phenomena in older buildings and occurs when basic
precaution was not taken from the beginning of the constructions. Dampness in buildings is
believed to some extend associated with climate factors [2].
886
Manipur has receiving annual rainfall of 933mm to 2593mm according to Indian Council of
Agricultural Research (ICAR) Imphal centre. This becomes a source of water close to a building
will also be one of the problems related to dampness [3].Field survey has been performed with
proper inspection of damp residential building in several parts of the states. The reason for this
case study is to analyses in moisture movement at Damp Proof Course (DPC) using water proofer
(Dr. Fixit & Aqutite).With the addition of water proofer, the dampness issue can be reduced at a
certain limit, in order to make a structural building free from moisture content. This problem was
very common in old structural buildings which was commonly experienced the physical
impairment such as cracks, paint peeling, spalling plaster, brick damage, skirting failure, stained
wall, efflorescence and if maintained was neglected, it will become more serious defects by
shortening the life span of the construction materials from the threat of climate and environment
factors [4].
Dampness in buildings which includes history of water damage, leakage, wet spots, visible mould,
and mould odor was a common problem. In a cold climate the prevalence of water damage and
dampness exposures has been estimated to be 10% to 60% [1]. It will generally occur in buildings
during its life use, if some basic precautions are not taking during the construction of such
Buildings from the beginning [5].A high percentage of dampness in building is due to rising
moisture from beneath, condensation from within the structure, rain penetration, high temperature,
vegetation growth near the buildings and so on.
In Manipur, Imphal the major problem of dampness is due to less idea of the sources, non-
engineering building construction, drainages problem and less aware of home dampness signs is
associated with health risk factor. It is found that dampness not only can favor house dust mites
(HDM) and microbial growth, like fungi, bacteria and mould, but also can generate by itself
chemical or biological degradation of buildings materials and furnishes [6]. In order to make
stable buildings, it is an essential requirement to supervise and inspection during construction of
structural buildings and used acceptable quantity of water proofer in order to prevent from above
havoc.
887
2.0 Area of study and building details

Before starting the survey a proper location were selected for better convenience in transportation
to make proper data collection.The selected sites were at Singjamei1 and Takyelpat2, Imphal-West
District in Manipur. The location is 4.1km south and 4km west respectively from the state capital,
Imphal respectively. After taking permission, a proper site survey and investigation was
conducted.
The damp building which is located at Singjamei, which latitude and longitude are 24°46´46.6´Ń,
93°56´20.2´É, near the National Highway (NH)-102 and Asian Highway (A)-1, had constructed
in year 2000. This building type is a ground floor building, shallow foundation (isolated
foundation), 0.6m plinth level, 0.05m DPC provided at floor, tiles not provided, roof are made of
galvanized iron sheets, place in urban area and surrounding is well building constructed. The site
is 100m away from Imphal River. This area is very much developed in building construction.
Besides the building, a small drainage is passes through which breadth and heights are (0.5x0.5)
m. The wall of the building is constructed with brick (local 2nd class) and inner face side of the
wall is plastered with mortar (1:3). The survey was conducted on August 28th 2018.
Fig. 1.Site visit exterior wall, interior wall and (GPS) location Singjamei, Imphal West, Manipur
Other building which is at Takyelpat, latitude and longitude are24°48´06´Ń, 93°54´14.6´É

respectively, had constructed in 2014 according to the report from construction centre. This
building is also ground floor building, shallow foundation (isolated foundation), 1m plinth level,
0.10m DPC provided at floor, tiles not provided, roof are made of galvanized iron sheets which
was exposed to sunlight and plywood’s ceiling at internal room, place in rural area and situated at
marshy land. The buildings were quite exposed to wind. This said areas were having few buildings
structure at surrounding. There was no drainage condition at nearby building, the wall of the
building is constructed with brick (local 2nd class) and plastering is provided 0.2m from flooring.
The survey was conducted on August 28th 2018.
888
Fig. 2.Site visit exterior wall, interior wall and (GPS) locationTakyelpat, Imphal West, Manipur
3.0 Causes, effect and remedies details

3.1Causes and effect for dampness of wall in buildings
Dampness is the most common problem and burning issues in Manipur. During the inspection at
said location i.e. Singjamei and Takyelpat, Imphal west, Manipur, some causes and effect of
dampness can identified as follows-
Table 1.Location Singjamei, Imphal-West, Manipur, latitude and longitude are 24°46´46.6´Ń,
93°56´20.2´É respectively.
Causes Effect
Poor DPC, no water proofer Damage brickwork
moisture rises Decay mortar in the joint
Water tank near the building Stained wall surface, unpleasant smell
Drainage is too closed Visible wetting, spalling of plaster
Vegetation growth Tension to the owner, efflorescence
Non Engineer construction Formation of mold, growth of mosses
Table 2.Location Takyelpat, Imphal-West, Manipur, latitude and longitude are are24°48´06´Ń,
93°54´14.6´É respectively.
Causes Effect
Poor DPC, no water proofer Floor crack
Marshy Land White colour deposited
Water logging Plaster spalling
Poor construction practices Decay mortar in joint
Vegetation growth Growth of Fungus, mosses
Non Engineer construction Poor aesthetic view
The above observations were found from the site visit Singjamei and Takyelpat, Imphal West,
Manipur. From the table, most failure in the buildings is due to poor Damp Proof Course (DPC)
and Non-Engineering type of construction. It is extremely important to take action against the
causes of rising moisture in the structure in order to control above effect and durable to the
structural buildings. Most building material is porous in their nature which make a passage way
for water can flow. Water is sharper than tiny needle; it can easy pass through micron holes very
easily. Capillarity is the main force behind dampness, moisture introduced into the face of wall by
deposited salts. The salts are left behind the wall of the structure when the moisture present inside
889
evaporates and this salt disturbed the moisture present in the air. The height of rising moisture
increases with the increasing of wall thickness of the structure.
3.2 Remedies for dampness of wall in buildings
Both the locations have most common phenomena about causes and effect. This impact may be
corrected mainly by using waterproofing compound during constructions, removal of vegetation,
provide proper ventilation, etc.
In this paper, as a remedial from rising moisture, it is recommended to usewater proofer (Dr. Fixit
and Aqutite) at certain amounts.The Water proofer acts as healable, curable, algae resistant
waterproofing compound. For better solution, experimental test were performed to identify how
much quantity of water proofer were acceptable to use in structural buildings. The details are
discussed in this paper.
4.0Materials and method
4.1 Based on local available materials
i. Cement. Portland Pozzolana Cement (PPC) of grade 33 was used. As per IS: 2720- Part 3, the
cement satisfies its requirement.
ii.Fine aggregate. Local available river sand (Serou sand) was used and performed sieve analysis
by using 500gm of sand sample.
Table3.Sieve analysis of the local available fine aggregate
Sieve size Mass (Mass/500)X Cumulative Cumulative -100

obtained 100 (gm) mass (gm) (gm)
(gm)
4.75mm 8.30 1.66 01.66 98.34
2.36mm 5.70 1.14 02.80 97.20
1.18mm 14.10 2.82 05.62 94.38
600micron 343.5 69.7 74.32 25.68
300micron 98.50 19.7 94.02 05.98
150micron 24.10 4.82 98.84 01.16
LP 5.80 1.16 100 00.00
From IS: 383-1970 , Table 4, (Clause 4.3), the fine aggregate is Grading Zone I, so it is convenient
to used in structural buildings design.
iii.Aggregates. Local available stone crush chips were used. As per IS: 383-1970, Table 2,
(Clause 4.1 and 4.2) the sizes of the aggregate were combinations of 12.5mm and 16mm w.r.t
percentage of passing as per code.
iv. Water.Clean portable water was used in this study. After testing by litmus paper, the pH value
of water is 6.9. Water cement ration used is 0.6 for M15, 0.52 for M20 and 0.46 for M25
concretes.
v. Water proofer. In this paper, Dr. Fixit 101LW+ and Aqutite were used as a chemical. As per
instruction, 4ml per kg of cement for Dr. Fixit while mixing of concrete and 12.5 ml per kg of
cement for Aqutite in slurry cement and apply to the surface for waterproofed.
890
vi.Water repellent paint. Water proof paint was used to prevent absorb of water from four side of
the cube, in order to performed unidirectional flow of water from beneath.
4.2Method
Sorptivity test
Sorptivity is materials property which transport of liquid in porous solid by capillary action [7]. It
is a unidirectional absorb of water inside the specimens. This test is simple and rapid to determine
the tendency of concrete to absorb water by capillary suction [8].
Specimen preparationand procedure and InstructionNominal mix of concrete of 1:2:4, 1:1.5:3 and
1:1:2 were making for M15, M20 and M25 grade concrete respectively. The amounts of water
proofer used per kg of cement are- Dr. Fixit (0.0ml, 4.0ml, 5.0ml, 6.0ml, 7.0ml, and 8ml) in M15
grade concretes, Dr. Fixit( 0.0ml, 3.0ml, 4.0ml, 5.0ml and 6.0ml) in M20 grade concretes, Dr.
Fixit ( 0.0ml, 2.0ml, 3.0ml, 4.0ml and 5.0ml) in M25 grade concretes, used during mixing of
concrete and Aqutite (20.0ml, 17.0ml, 14.0ml, 12.0ml and 10ml) in M20 and M25 grade
concretes, it should apply after curing at the surface to be waterproofed.
Table 4.Instruction for applying water proofer i.e. Dr. Fixit and Aqutite
Dr Fixit Aqutite
Portable water to be used in concrete mix Mixed with water just before use
Added during concrete mix Prepare slurry cement with minimum water
Maintained the quantity Surface should be clean, free from dust
Mixed 2 to 3 minutes thoroughly Mix well, applies at the surface
No extra water should added during casting wet the surface before applying
Casting should be done within 5-20minutes After applying need 2 to 3days curing
The test specimens were 100x100x100mm cubes and prepared 3 specimens for each test, and take
average value as per IS: 456-2000. After casting, the samples were placed for steam curing 2days
at 95°C and drying in oven at temperature of 100°C. The sides of the specimens were sealed in
order to achieve unidirectional flow. A water repellent paint (enamel) was used as sealant. The
weights of the specimens were taken as initial dry weight (W1). A clean water, tray, stop watch,
stick, digital weight machines, towels were used to performed this test. Samples were immersed at
the water in tray, depth of 25mm. At 60minutes the samples were removed from the water, the
excess water at the surface of the cube was blotted off and takes the weight of the sample (W2).
891
Sorptivity can be determined by, S= I/ t0.5, mm/min0.5
Where; S = Sorptivity in mm,
t = elapsed time in min., I = (W2 - W1)/ A.d,
W1 = dry weight of cube in grams,
W2 = weight of cube after 60min capillary suction in water in grams,
A = surface area which water penetrated in mm2,d= density of water in gm/cm3
5.0 Experimental results

After getting the value of initial dry weight (W1) and weight after 60min. in water of the sample
(W2), calculate the sorptivity value using the above formula. The values obtained are quite small
but this small amount of rising of moisture will make a structure safe for a long period of
times.Table 5 and 6 gives the Sorptivity test result of using water proofer Dr. Fixit 101 (ml) and
Water Aqutite (ml) in concrete for 2days steam curing at 95°C. The Water proofer (Dr. Fixit 101)
in ml verses Sorptivity in 10-4, in mm/min0.5 and water proffer (Aqutite) in ml verses Sorptivity in
10-4, in mm/min0.5 are graphically shown in figure A and B respectively.
Table 5.SorptivityValue at 60min for nominal mix M15, M20&M25grade concretes using Dr.
Fixit water proofer during mix
Nominal Water Dry weight of Weight of cube Sorptivity Value

mix proofer Dr. cube, W1(gm) after 60min in 10-4,
proportion Fixit(ml) capillary suction mm/min0.5
grade in water, W2
concrete (gm)
M15 0.0 2225.0 2250.2 3.253
1:2:4 with 4.0 2215.0 2240.1 3.240
water: 5.0 2235.5 2255.4 2.569
cement is 6.0 2280.0 2296.4 2.117
0.6 7.0 2280.2 2293.5 1.717
8.0 2350.5 2363.5 1.678
M20 0.0 2345.9 2352.6 0.865
1:1.5:3 with 3.0 2333.0 2338.1 0.658
water: 4.0 2308.0 2312.1 0.529
cement is 5.0 2349.2 2352.1 0.374
0.52 6.0 2236.5 2238.7 0.284
M25 0.0 2290.5 2294.7 0.542
1:1:2 with 2.0 2304.0 2307.1 0.400
water: 3.0 2303.9 2306.9 0.384
cement is 4.0 2260.0 2262.2 0.258
0.46 5.0 2285.0 2286.3 0.168
892
3.5
3 M15
Sorptivity in 10-4, mm/min0.5

M20
2.5
M25
2
1.5
1
0.5
0
0 2 4 6 8 10
Water proofer in ml
Fig 3: Graph plot for Water Proofer (Dr Fixit) for M15, M20 and M25 in mlVersus Sorptivity
value in 10-4, mm/min0.5, using the respective values from table 5.
Fig. 4.Mixing and casting, steam curing, set for capillary suction and set for weight for M15, M20
and M25 grade concrete respectively.
Table 6.SorptivityValue at 60min for nominal mix M15, M20&M25grade concretes using Aqutite
water proofer, applied at the surface to be water proofed
Nominal mix Water Dry weight of Weight of cube Sorptivity

proportion proofer cube, W1(gm) after 60min, Value in 10-4,
grade Aqutite (ml) applied water mm/min0.5
concrete proofer, W2
(gm)
M20 20.0 2311.5 2315.5 0.516
1:1.5:3 with 17.0 2462.1 2465.8 0.478
water: 14.0 2309.2 2312.5 0.426
cement is 12.0 2304.0 2306.5 0.323
0.52 10.0 2372.5 2374.8 0.297
M25 20.0 2497.9 2500.0 0.271
1:1:2 with 17.0 2353.1 2355.1 0.258
water: 14.0 2464.2 2466.1 0.245
cement is 12.0 2362.8 2364.5 0.219
0.46 10.0 2348.6 2350.2 0.206
893
0.6
0.5
Sorptivity in 10-4,
mm/min0.5
0.4
0.3
M20
0.2
M25
0.1
0
0 5 10 15 20 25
Water proofer in ml
Fig.4: Graph plot for Water Proofer (Aqutite) for M20 and M25 in mlVersus Sorptivity value in
10-4, mm/min0.5, using the respective values from table 6.
6.0 Conclusion
Based on theexperimentalinvestigation, sorptivity of grade concretesmade from local available
materials, by using different amount of water proofer in the nominal mix M15, M20 and M25
grade concrete.The salient observations are made from the respective graph plot on Fig. 3, Fig. 4
and Table 5, 6 some reliable solutions were obtained as
(a) From table 5, observed thatsorptivity values werelow at high grade concrete M25> M20>
M15 i.e. 1:1:2 > 1:1.5:3 > 1:2:4.
(b) From the table 5, it can conclude that M15 is not applicable at construction practices
because high permeability of rising moisture. So it can damage the structure very fast
beyond expectation.
(c) From table 5, it can also said that higher the amount of water proofer (Dr.Fixit) in per kg
of cement; lower the sorptivity value in M15, M20 and M25.
(d) In construction practices for concrete grade M20 &M25; acceptable limit of Dr.Fixitmay
be 6ml and 5 ml per kg of cementwith water cement ratio were 0.52 and 0.46 respectively
as per table 5.
(e) From table 6, higher the amount of water proofer (Aqutite), higher the Sorptivity value.
Using 10ml might slightly reduce the sorptivity value of water cement ratio 0.46 at M20
grade concrete.
(f) From table 6, M25 grade concrete, sorptivity values for 20ml, 17ml, 14ml, 12ml and 10
ml were slightly equal. So 10ml of Aqutite used in M25 is little bit advantages than
others.
(g) Comparing table 5 and 6, a classic solution obtained asquantity of water proofer used in
M25 (Dr. Fixit 101) is approximately equal to M20 (Aqutite water proofer) and may be
the most economicaland effective for controlling dampness.
(h) Used of water proofer might increase, the time taken by water to risedamp andmake
durable than normal structural buildings.
On these results, it can recommend that the benefits of using water proofer at D.P.C during
construction are quite safer and reliable to the structural buildings. The approximate minimum
thickness of the D.P.C should be 100mm and further studies were conducted to control already
occurs dampness in the structural buildings in Manipur, India.
894
References
1. Preeti Tiwari, Rajiv Chandak, R.K Yadav (2014). Effects of salts on compressive
strength of concrete ISSN: 2248-9622 Vol.4, Issue 4, pp.38-42.
2. Pei Zam H, Emilia ZA, Karmegan K and Mohad Sapuan (2017). Dampness and Mould
exposure in buildings as a risk factor for health effects Vol. (1): 28-40.
3. Halim, A.A., Harun, S.N. & Hamid Y. (2012). Diagnosis of dampness in conservation of
historic buildings Journal Design+Built, 5, ISSN: 1985-6881.
4. Al-Hafzan Abdullah Halim, Dr. Siti Norlizaiha Harun, Dr. Md Yusof Hamid (2012).
“Diagnosis of dampness in conservation of historic building” ISSN: 1985-6881 Vol 5.
5. Dr. Gana A. J (2015). An investigation in the dampness in structures (A case study of
high rise buildings). ISSN: 2349-4506 page 17-21.
6. Zhijing Lin, Zhuohui Zhao, Huihui Xu, Tingtinng Wang, Haidong Kan, Dan Norback
2015 Home dampness sign in association with asthma and allergic disease in 4618
preschool children in Urumqi, China-The influence of ventilation/cleaning habits PLOS
ONE│DOI: 10.1371/journal.pone.0134359 page 1-13.
7. Jayeshkumar Pitroda, Dr F S Umrigar (2013). “Evaluation of sorptivity and water
absorption of concrete with partial replacement of cement by thermal industry waste (fly
ash)” ISSN: 2277-3754, Vol.2, Issue 7 page 245 to249.
8. Salmabanu Luhar, Urvashi Khandelwal (2015) A study on water Absorption and
Sorptivity of Geopolymer concrete. SSRG-IJCE-Vol2 Issue 8 page 1-9.
9. Gladstone Sena Kportufe (2015). “The Devastating Effects of Rising Damp in the
Construction Industry” ISSN: 2224-6096 (Paper) ISSN: 2225-0581 (Online), Vol.5,
No.10, 2015.
10. Kofi Agyekum, Joshua Ayarkwa, Christian Koranteng and Emmanuel Adinyira (2013).
“An Overview of the Symptoms Associated with Dampness in walls of Residential
Buildings in Major Towns in Ghana” ISSN: 2224-5790 (Paper) ISSN: 2225-0514
(Online), Vol.3, No.9, 2013.
11. IS: 2720- Part-3.
12. IS: 2386- Part-3.
13. IS: 383-1970.
14. IS: 456-2000.
10
895
Paper ID- 170
PREDICTION OF COMPRESSIVE STRENGTH OF HIGH STRENGTH FIBER
REINFORCED CONCRETE USING ARTIFICIAL NEURAL NETWORKS
Mohan Lal Chowdary1
1
Dept. of Civil Engineering, Vasavi College of Engineering, Hyderabad, Telangana, INDIA,
Mobile: 9989673220
cmohanlal69@gmail.com
ABSTRACT
The structural and architectural design of tall and large size structures subjected to various types of
loads demand the use of high strength reinforced concrete in construction. The compressive strength of
concrete cubes is the main guiding factor in deciding the quality of concrete. For a concrete mix
proportion, compressive strength can only be determined by testing cubes after 28 days of casting. The
experimental finding of compressive strength may further delay the decision-making process due to
lack of uniform code procedures for mix design of high strength fiber reinforced concrete with mineral
admixtures. The present study uses a back propagation artificial neural network model to predict the
compressive strength of concrete for a given mix proportion. A data base of 54 mix proportions with
different binder content and steel fiber content are prepared and tested under standard conditions in the
laboratory. The input variables for the neural network model are the cement content, silica fume, water
content, fine aggregate, coarse aggregate, fiber content and the age of concrete. The high strength
concretes were prepared by replacing the cement with silica fume (0 to 15%) and by addition of steel
fiber content (0 to 1%). Neural network tool box of MATLAB R2017b software is used to develop the
models with 10-50 hidden neurons at step size of 10. Levenberg Marquardt algorithm, Bayesian
Regularization algorithm and Scaled Conjugate Gradient algorithm of the MATLAB software are
made use of for training, testing and validation of the data sets. The present study demonstrates that the
developed neural network models are effective in predicting the compressive strength of high strength
fiber reinforced concrete for a given mix proportion. Application of this neural network model will
also help to avoid delay in decision making process.
Keywords: High strength concrete, Neural network model, Binder, Steel fiber, Mix proportion.
1. Introduction:
In recent years there is rapid increase in construction of tall and large size structures in many cities of
India. Most of these structures are of reinforced cement concrete. As the structures grow taller, the
demand for higher compressive strength of concrete increases. From the literature it is observed that
high strength reinforced concrete can be developed with the addition of supplementary cementitious
materials such as GGBS, silica fume, fly ash etc. Nowadays concrete which is tested after 28 days of
896
casting and having a compressive strength of 60 MPa and above is treated as high strength concrete.
There are no uniform code procedures for concrete mix design of high strength fiber reinforced
concrete. The experimental finding of compressive strength by testing cubes after28 days of casting
may delay the decision-making process. If the concrete has not reached the design strength on 28 th day
of testing it may be too late to take a decision. In this circumstance, it is necessary to have a tool that
predicts the strength of concrete.
Cheng Yeh I (1999) has developed a method of optimizing high performance concrete mix
proportioning for a given workability and compressive strength using Artificial Neural Networks.
Cheng Yeh I (2006) has described a method of modeling slump of fly ash and slag concrete using
Artificial Neural Networks. Jui-Sheng Chou P.E., Chien-Kuo Chiu P.E., Mahmoud Farfoura and
Ismail Al-Taharwa (2011) have attempted to predict the accuracy of the compressive strength of high-
performance concrete by two machine learning models, one statistical model and two metaclassifier
models. Gokmen Tayfur, Tahir Kemal Erdem and Onder Kirca (2013) have used Fuzzy Logic and
Artificial Neural Networks model for strength prediction of high strength concrete with five different
binder contents of silica fume. Anh-Duc Pham, Nhat-Duc Hoang and Quang-Trung Nguyen (2016)
have developed a model for predicting high performance concrete compressive strength by hybridizing
firefly algorithm and the least square support vector regression. Behzad Abounia Omran, Qian Chen
and Ruoyu Jin (2016) have used nine data mining models in predicting the compressive strength of a
new type of concrete containing three alternative materials as fly ash, Haydite light weight aggregate
and Portland limestone cement. Mohan Lal Chowdary (2017) has applied back propagation Artificial
Neural Networks to predict the design of reinforced concrete rectangular short columns subjected to
biaxial bending. Chengyao Liang, Chunxiang Qian, Huaicheng Chen and Wence Kang (2018) have
developed a neural network model for prediction of compressive strength of concrete in wet-dry
environment. Mohan Lal Chowdary (2018) has presented a model for design of rectangular combined
footings with Artificial Neural Networks.
In the present study, a back propagation artificial neural network (ANN) approach is used to predict
the compressive strength of high strength fiber reinforced concrete for a given mix proportion.
Application of this neural network model will also help to avoid delay in decision making process.
2. Neural Network Model:
The architecture of the neural network model of the present work is shown in Fig 1. The model
consists of 7 neurons in input layer and one neuron in output layer. The input variables for the neural
network model are the cement content, silica fume, water content, fine aggregate, coarse aggregate,
fiber content and the age of concrete. The cement is replaced with silica fume between 0 to 15% at a
897
step size of 5%. The steel fiber having length of 50 mm with a constant aspect ratio of 50 is used in
preparing the concrete. The quantity of steel fiber added in the concrete varies from 0 to 1% at a step
size of 0.5%. In addition, the quantity of water content, coarse aggregate and fine aggregate are also
varied, and the mix proportions are created as shown in Table 1. A data base of 54 mix proportions
with different binder content and steel fiber content are prepared and tested for compressive strength
under standard conditions in the laboratory. For each mix 3 cubes were prepared and a total of 162
cubes were casted and tested to obtain 7 days and 28 days of compressive strength. Table 1 shows the
input training data sets and output target data sets for 7 days and 28 days of testing. Figure 3, figure 4
and figure 5 show the graphical representation of test results of compressive strength of concrete with
varying percentages of silica fume and fiber content. Neural network tool box of MATLAB R2017b
software is used to develop the models with hidden neurons between 10 to 50 at step size of 10. The
neural network model created in MATLAB software is shown in Fig 2.
CC – Cement content SF – Silica fume

WC – Water content FA – Fine aggregate CS – Compressive strength
CA – Coarse aggregate FC – Fiber content
Fig 1: Neural network architecture
Fig. 2: Neural Network model created in MATLAB
898
Table 1: Data base of mix proportions

Input training data sets Output target data sets
Mix S.F Fiber Cement Water F. A C.A FIBER S. F C.S.7 days C.S.28days
(%) (%) (Kgs) (Lts) (Kgs) (Kgs) (Kgs) (Kgs) (MPa) (MPa)
A 5 0 475 140 744 1116 0 25 52.23 66.08

B 10 0 450 140 744 1116 0 50 61.07 75.42
C 15 0 425 140 744 1116 0 75 65.69 79.11
D 5 0.5 475 140 744 1116 39.25 25 56.18 69.08
E 10 0.5 450 140 744 1116 39.25 50 64.67 76.41
F 15 0.5 425 140 744 1116 39.25 75 67.28 79.32
G 5 1 475 140 744 1116 78.50 25 57.47 76.5
H 10 1 450 140 744 1116 78.50 50 69.42 83.95
I 15 1 425 140 744 1116 78.50 75 72.03 85.60
J 5 0 522.5 154 718.4 1077.6 0 27.5 57.45 70.12
K 10 0 495 154 718.4 1077.6 0 55 68.9 86.43
L 15 0 467.5 154 718.4 1077.6 0 82.5 78.23 87.21
M 5 0.5 522.5 154 718.4 1077.6 39.25 27.5 59.7 73.43
N 10 0.5 495 154 718.4 1077.6 39.25 55 69.15 85.26
O 15 0.5 467.5 154 718.4 1077.6 39.25 82.5 73.1 86.73
P 5 1 522.5 154 718.4 1077.6 78.50 27.5 62.08 79.42
Q 10 1 495 154 718.4 1077.6 78.50 55 72.6 86.58
R 15 1 467.5 154 718.4 1077.6 78.50 82.5 75.55 87.67
S 5 0 570 168 692.8 1039.2 0 30 58.48 69.63
T 10 0 540 168 692.8 1039.2 0 60 67.33 84.21
U 15 0 510 168 692.8 1039.2 0 90 69.74 91.34
V 5 0.5 570 168 692.8 1039.2 39.25 30 60.13 75.32
W 10 0.5 540 168 692.8 1039.2 39.25 60 75.06 90.21
X 15 0.5 510 168 692.8 1039.2 39.25 90 73.63 89.34
Y 5 1 570 168 692.8 1039.2 78.50 30 61.65 85.15
Z 10 1 540 168 692.8 1039.2 78.50 60 77.90 95.91
$ 15 1 510 168 692.8 1039.2 78.50 90 70.67 96.71
899
Fig. 3 Compressive Strength with varying percentages of silica fume, binder & 0% fiber content
Fig. 4 Compressive Strength with varying percentages of silica fume, binder & 0.5% fiber c ontent
900
Fig. 5 Compressive Strength with varying percentages of silica fume, binder & 1.0% fiber content
3. Training of neural network

Levenberg Marquardt algorithm, Bayesian Regularization algorithm and Scaled Conjugate Gradient
algorithm of the MATLAB software are made use of for training, testing and validation of the data
sets. From the 54 data sets, 70% data sets are used for training, 15% for validation and the remaining
15% for testing. The training of the neural network model has been carried out with the number of
hidden neurons between 10 to 50 with a step size of 10. The regression values obtained from the
training process are listed in the Table 2. A regression value close to 1.0 indicates that output values
are in close agreement with the target values.
Table 2 Regression values (R) from the training of Neural Networks

Regression values (R)
Number of hidden
neurons Levenberg- Bayesian Scaled Conjugate
Marquardt Regularization Gradient
10 0.96751 0.97921 0.9747
20 0.9614 0.99306 0.8929
30 0.98131 0.94525 0.93751
40 0.93297 0.9446 0.84757
901
The proposed artificial neural network model has been trained on the neural network tool box of
MATLAB R2017b software using Levenberg Marquardt algorithm, Bayesian Regularization algorithm
and Scaled Conjugate Gradient algorithm. The training, validation and testing graphs having
corresponding regression coefficients obtained by Levenberg Marquardt algorithm with hidden
neurons from 10-50 are shown in Figure 6 to Figure 10. The variation of regression values with
varying number of hidden layers are also listed in the Table 2 and are represented in the Figure 11. It is
observed that the Levenberg Marquardt algorithm gives the best regression coefficient of 0.98131
when the model is trained with 30 hidden neurons. But, the Bayesian Regularization algorithm and
Scaled Conjugate Gradient algorithms give the best regression coefficients of 0.99306 and 0.9747
when the model is trained with number of hidden neurons of 20 and 10 respectively. Among all the
combinations, regression value of 0.99306 with 20 hidden neurons by the Bayesian Regularization
algorithm is the best result. This regression coefficient value using Bayesian Regularization algorithm
indicates that the output values are in good agreement with the target values. Hence, Bayesian
Regularization algorithm with 20 hidden neurons was used to predict the compressive strength of the
concrete. Figure 12 shows the predicted value of compressive strength of concrete as 90.7597 MPa.
The corresponding input values for testing neural network model are: cement content-600 kg, silica
fume-50 kg, water content-168 liters, fine aggregate-692.8 kg, coarse aggregate-1039.2 kg, fiber
content-78.5kg and age of cube testing-28 days.
Fig. 6 Regression plot for Levenberg-Marquardt with 10 hidden neurons
902
Fig.7 Regression plot for Levenberg-Marquardt with 20 hidden neurons
903
904
Fig. 11 Variation of regression value (R) with number of hidden neurons
Fig. 12 Prediction of compressive strength of concrete using ANN tool box of MATLAB
905
5. Conclusions
In the present study an effort has been made to predict compressive strength of concrete using artificial
neural networks. Effect of adding different percentages of fiber content and silica fume on compressive
strength of concrete is also presented. Among the various neural network training algorithms, Bayesian
Regularization algorithm is found to be efficient in predicting compressive strength of concrete as the
regression coefficient obtained from this algorithm is close to 1.0.
Acknowledgement
I would like to thank the Principal and the management of the Vasavi College of Engineering,
Hyderabad for giving an opportunity to conduct the research work in the college by providing
MATLAB R2017b software. I would also like to thank for sponsoring registration and for presentation
of the research work at national conference.
References
1. Cheng Yeh I : Design of high-performance concrete mixture using neural networks and
nonlinear programming, vol.13, No.1, Journal of Computing in Civil Engineering, Jan 1999,
ASCE, pp. 36-42.
2. Cheng Yeh I : Exploring Concrete Slump Model Using Artificial Neural Networks, vol.20,
No.3, Journal of Computing in Civil Engineering, May 2006, ASCE, pp. 217-221.
3. Jui-Sheng Chou P.E., Chien-Kuo Chiu P.E., Mahmoud Farfoura and Ismail Al-Taharwa:
Optimizing the Prediction Accuracy of Concrete Compressive Strength Based on a Comparison
of Data-Mining Techniques, vol.25, No.3, Journal of Computing in Civil Engineering, May
2011, ASCE, pp. 242-253.
4. Gokmen Tayfur, Tahir Kemal Erdem and Onder Kirca : Strength Prediction of High-Strength
Concrete by Fuzzy Logic and Artificial Neural Networks, vol.26, No.11, Journal of Materials
in Civil Engineering, Oct 2014, ASCE, pp. 1-7.
5. Anh-Duc Pham, Nhat-Duc Hoang and Quang-Trung Nguye : Predicting Compressive Strength
of High-Performance Concrete Using Metaheuristic-Optimized Least Squares Support Vector
Regression, vol.30, No.3, Journal of Computing in Civil Engineering, Feb. 2016, ASCE, pp. 1-
4.
6. Behzad Abounia Omran, Qian Chen and Ruoyu Jin,: Comparison of Data Mining Techniques
for Predicting Compressive Strength of Environmentally Friendly Concrete, vol.30, No.6,
Journal of Computing in Civil Engineering, Aug 2016, ASCE, pp. 1-13.
7. Mohan Lal Chowdary : Application of Artificial Neural Networks for Design of RCC
Rectangular Short Columns Subjected to Biaxial Bending, vol.7, International Journal of
Engineering and Advanced Technology, Dec 2017, pp. 185-188.
8. Mohan Lal Chowdary : Modeling Design of Rectangular Combined Footings with Artificial
Neural Networks, vol.5, issue 2, Journal of Emerging Technologies and Innovative Research,
Feb.2018, pp.99-103.
9. Chengyao Liang, Chunxiang Qian, Huaicheng Chen and Wence Kang : Prediction of
Compressive Strength of Concrete in Wet-Dry Environment by BP Artificial Neural Networks,
vol. 2018, Advances in Materials Science and Engineering, April 2018, pp. 1-11.
906
Paper ID- 171
Effective Public Private Partnership Model for Redevelopment of Government

Buildings
Pritesh D. Aher1 (0000-0001-6554-7187), Rajanikant V. Devalkar2 (0000-0001-9522-7141) and Vaishnavi Naik3
Abstract
Post-independence, India witnessed tremendous economic instability. For economic development, India
adopted mixed economy and people started moving to cities. Urbanization is people migrating from rural areas to
urban areas for better lifestyle. With modernization and industrialization there is ever growing number of nuclear
families, in which each family splitting further, engage a greater number of new units of houses in urban areas.
Increasing urbanization is leading to scarcity of land, housing units and commercial centres. In every new
development plan, exterior of prevailing areas is brought under development which requires huge funds for
infrastructure construction and its implementation. One of the principal challenge in front of developer is high
cost land. As the serious problem is the absence of a clear title property due to such problem financial institutions
and real estate developers hesitate to invest in projects of real estate. As there is no redevelopment and
concentration on available lands, land remains underutilized and thus contributing to shortage of land and hike in
land prices. But there are many existing plots in the core of the city which can be redeveloped using PPP models
for efficient use of land. So Public-Private Partnership (PPP) is appropriate solution to overcome the problem in
housing sector by developing different PPP models. CSF in housing sector are identified and analyzed using SPSS
software and criticality of factor is determined. Two case studies are undertaken in Nashik city for redevelopment
of government building using effective PPP model.
Keywords:
Public Private Partnership, SPSS, Cost analysis, Critical Success Factor.
1 Introduction
In the context of rapid urbanization, the lack of public space development is one of the major issues of
commercial development in the core part of cities.[1] Undeveloped infrastructure is also the reason of low-quality
life of the population, the high illness rate and death rate. However all the infrastructure projects are so expensive,
and the matters of their financing will be the sharpest.[2]. One of the principal challenge in front of developer is
high cost land. Due to high cost land it becomes difficult for developer to deliver the affordable housing
project.[3]. With modernization and industrialization there is ever growing number of nuclear families, in which
each family splitting further, engage more number of new units of houses in urban areas [4]. Despite of efforts
taken by private and government the shortage is increasing day by day. This problem requires huge amount of
funds which cannot be handled either by private sector or government autonomously, so Public-Private
Partnership (PPP) is appropriate solution to overcome the problem. PPP is a contractual agreement between
private firm and Government aimed towards implementing infrastructure services and facilities that were
traditionally managed and provided by public sector. The concept is to allow the public agency to satisfy a specific
public need while minimizing the use of limited public funds.[5]. It is possible to utilize the strengths of private
actors i.e. innovation, technical knowledge and skills, managerial efficiency and entrepreneurial spirit, and the
role of public actors, including social responsibility, social justice, public accountability and local knowledge, to
render the high quality infrastructure and services[6]. There is requirement for extreme participation of the private
907
sector in encouraging the capacity of the public sector to deliver more and better services through PPP; this has
allowed for a more formal transfer of responsibilities from the state to private sector companies. In PPP public
and private sectors work together with hand in hand that results in to blend and utilize their special skills to provide
services to public [7]. Most importantly, PPP emphasizes value for money- focussing on reduced costs, better risk
allocation, faster implementation, improved services and possible generation of additional revenue. PPP offers
monetary and non-monetary advantages for the public sector. [8]. Many public infrastructure projects across a
wide range of industries have been privatized worldwide for improved quality, efficiency and effectiveness.[9].
India witnesses many constraints for both sectors, as in public sector witnesses lack of expertise and
deficiency of quality whereas private sector falls short of development control guidelines, and non-coordination
with local authorities who are accountable for providing off-site infrastructure, tax concessions, land acquisition,
establishing special laws, delay in approvals etc. In India due to rapid urbanization, the increasing shortage for
housing units and the shortage of land is to be satisfied by constructing a greater number of units rapidly and
economically. According to housing to all scheme the Government would require two crore housing units by
2022.[10] To fullfill this huge requirement various PPP models are desirable.
2 Literature Review
The literature review intends to study the vital observations and recordings of present knowledge including
rational findings along with theoretical and methodological inputs to this topic.
A. Review of Technical Papers
A0.
a) Olabode Emmanuel Ogunsanmi Results suggest that CSSFs for successful implementation of PPP project are
integrity, complexity of project, project management expertise, realistic cost/benefit assessment, government
guarantee private sector financial capability and adequate financial market that must be considered by all
stakeholders. Implications of this study for policy makers in government and stakeholders is for government to
provide adequate guarantee for implementation of PPP project that would enable equitable participation of the
Private sector and for practice concessionaires with good integrity, financial capability and project management
expertise must be considered for future PPP projects. To implement Public Private Partnership projects efficiently
and effectively it may be necessary for all stakeholders to be fully aware of the various success factors that can
aid its successful implementation. Critical success factors (CSF) is defined as key area of activity in which
favourable results is absolutely necessary for a particular manager to reach his goals. Study of critical success
factors for PPP utilized for private contractors in competitive tendering and negotiation in contracts, stable
political and economic solutions. The study recommends that government at all levels must provide good
guarantee for PPP projects implementation[11].
b) Robert Osei-Kyei, Albert P.C. Chan As identified CSFs are risk allocation and sharing, strong private
consortium, political support, community/public support and transparent procurement. The research approaches
adopted are case study, questionnaire survey and mixed methods. The findings revealed provide an overview of
CSFs for PPPs to enhance future implementations. Finally a checklist of CSFs for PPPs has been developed,
which could be adopted for further empirical studies[12].
908
c) Urmi Sengupta This paper studies the housing in Kolkata using public private partnership. The aspects of
implementation of PPP housing schemes in Kolkata are the roles of and relationships between partners, land
supply, financial arrangements; target groups and affordability. Literature specific to PPPs in housing in
developing countries, however, is relatively sparse and the concept is yet to have a separate niche in either the
theoretical or the practical landscapes. In this paper the affordability is also accounted. Affordability and
accessibility have been central concerns of partnership schemes in Kolkata. It is observed that purchasing a house
under a PPP has generally become more affordable owing to government stipulations regarding minimum size
and sale price for LIG units. The performance of PPP in Kolkata is apparent from the improvements made in the
condition, quality and stock of housing as the joint approach brings together the efficiency in production and
technical and marketing expertise of the private sector with the accountability and righteousness of the public
sector. The state government’s business will be seen in its ability to deal with the risks associated with fluctuating
macro-economic conditions by maintaining policy principles while adjusting details without constraining
competition in the production or supply of housing. The challenge for the private sector partners will be to
minimize risks resulting from inclusion of affordable elements in housing provision, uncertainty of returns at
market rates, and dealing with scores of other regulatory constraints such as protracted bidding processes or
decisions on when to sell properties imposed by the project[13].
d) Xiaodan Li, Hao Yang, Wenjing Li, Zhiting Chen This paper studies the residential Brownfield redevelopment
(RBR) which is rapidly growing in US. The paper also throws light on the opportunities and challenges in
restructuring RBR’s ecological environment and economic benefits. To meet the requirement a collaborative
model of public private partnership is adopted. It creates an institutional framework in which the public sectors
provide strategic profits to the private sectors, while the private entities implement and develop the public sector’s
plan. Two case studies are selected for RBR that are Summerset at Frick Park and Hazelwood, both in Pittsburgh,
Pennsylvania, for analysis purpose. Through the comparison of successful and unsuccessful cases, our statement
is future concluded: consensus, public owned properties and involvement of academic institutions contributed
notably in PPP and RBR due of the potential benefits to residential brownfield redevelopment (RBR) in
economical, ecological and community development, it is important to study public private partnership in RBR
as well as its crucial factors. The successful case in restructuring ecological environment and promoting
sustainable development also provides a paradigm for other cities that are withstanding similar RB or RBR
issues.[14]
e) Yongjian Ke, Shou Qing Wang, and Albert P. C. Chan, The paper studies several methods to pinpoint the threat
allotment in PPP assignments of Hong Kong Special Administrative Region and China by questionnaire survey
by equating the threats in Greece and U.K. The analyses showed that risk allocation among four countries depict
that the public sector in the U.K. was supreme to allocate the PPP threats to the private entity, trailed by Greece,
Hong Kong, and China. Respondents from Greece revealed that strong backing for the public segment is to
preserve the macro level threats. Risk is inevitable and hard to deal with and involves a proper management
structure in cooperation of theoretically and practically. The comparative analyses would be adopted to identify
the level of common dis- position toward risks and provide an indication of the potential to collaborate. For future
research events the analyses will be beneficial for risk allocation and potential teamwork. The results illustrate
that the public partner desired to retain social risks and political. The comparative study concludes that in every
909
PPP market, private parties that invest should participate more actively in managing the risks. Also the paper
emphasizes risk transferring for productive delivery of PPP[15].
f) Xueqing Zhang, Shu Chen, A systematic framework for infrastructure development through public private
partnerships Public private partnerships (PPPs) play an essential part in fetching private partys competition to
public controls in infrastructure expansion and in amalgamation of resources of both parties. A great number of
infrastructure ventures have been fruitfully developed through PPPs with considerable cost savings. On the part
of the public client, it moves the attention concerning consortium selection away from the lowest price to other
important issues, one of which is value for money. On the part of the private consortium, norms besides price are
used to recommend the client, and to reduce competitors innovative solutions are used. It is conveyed that U.S.
state and regional governments have regularly witnessed 1040% enhancements in service quality and cost savings
and asset management using PPPs. Structure of organized methodology of PPP involve of Four-phase framework,
Public procurement principles, Public private win win solution, Rationale of the framework. The framework
integrates four broadly distributed phases in delivery process such as (1) design of a workable concession, (2)
competitive concessionaire selection, (3) financial regulation of the selected concessionaire during the concession
period, and (4) periodic concession rebidding to allow new entry for the concession for successful delivery of
projects[16].
g) Tharun Shastry L, A Study on Public-Private Partnerships with Reference to Indian Infrastructural Projects In
this paper it is witnessed that considering the current requirement for the PPP is not an alternative but its necessity.
The advantages of PPP are discussed such as large investment in public (both urban and rural) infrastructure,
efficient service delivery, cost-effectiveness, contracts that are performance-based, sharing of risks, effective use
of assets and opportunities of long-term investment, time efficiency, greater convenience, increased reliability and
saving costs along with easy availability of information. Considering the governments constraints to increase
funds on its own, it has resorted to pioneering exercise identified as Public-Private Partnership (PPP) in several
subdivisions. PPP refers to a form of a bond amongst the public and private segment for exact time in order to
enable projects that require enormous capital amount. The paper shows main infrastructure expansion projects in
Maharashtra which accounts to more than 50% which are established on the P3 model. In the 2000s, states of
Karnataka, Madhya Pradesh, Gujarat, and Tamil Nadu too implemented this model. In order to make PPP projects
successful in India, risk should only be transferred to those parties who are best in managing it, because
transferring risk will result in a huge impact. Over-specification and miss-specification of project result can cause
huge damage. Hence, the government will have to define its goals, desires, principles and requirements as clearly
as possible. The whole sum of bidders should be enough to have effective competition but should be limited in
order to have control over transaction cost, quality of bids and probability of success. It is always better to have
standardized contracts in order to avoid transaction cost, delays and uncertainties. In terms of agreement reward
method, the global competitive bidding resulted 35% of investment in India followed by national competitive
bidding of 26%[17].
h) LiYaning Tang, Qiping Shen, Martin Skitmore; and Eddie W. L. Cheng, Ranked Critical Factors in PPP
Briefings Paper shows Public-private partnerships (PPPs) are rapidly growing to obtain Australian infrastructure
projects. As the initial preparation stages are very pivotal in governing a triumphant outcome in construction
projects. During this period it is seen that there is insufficiency of organized examination on the type and nature
of the critical factors affecting the effectiveness and productivity of PPP. Analysis of literature is done of PPP in
910
Australia, out of which four main classifying factors (risk, procurement, finance and stakeholder) are recognized
along with many sub-factors. A questionnaire survey describing participation of state governments stakeholders
is seen and a mathematical framework that ranks the factors d is generated. To upgrade the PPP briefing stage
inspection of the possible of the factors for both public entity and private entity is done[18].
i) Ashwin Mahalingam, PPP experiences in Indian States: Bottlenecks, Enablers and Key Issues given India’s
infrastructure needs, PPPs are a necessity and not just an option. However, in order to enable a better insight on
developing infrastructure efficiently through PPPs there are a mass issues that are required to resolved. Recent
studies indicate that India must invest close to USD 400 Billion for coming next five years in infrastructure
development, at least 40% of which must be developed through Private Public Partnerships (PPPs). The main
reasons for the failure of PPP projects are original rise in tariffs, macroeconomic swings in currency or purchasing
power, non-regulatory and institutional domains, societal dissatisfaction against the private entity and political
etc. A casual scan at the statistics on the India Infrastructure website shows that only 144 PPP projects have been
completed or being completed where a majority of which are in the transportation sector. This is a relatively small
fragment of the total number of infrastructure investments in India. Several PPP proposals for projects at the state
level face roadblocks due to the inefficiency of enabling PPP legislation. In some cases, the existing law enactment
mandate that only the public entity be authorized to particular infrastructure services. Various bottlenecks
observed are Bottlenecks at the Institutional Level, Organizational Bottlenecks, and Project level Bottlenecks etc.
To reply to the research questions, we require inputs and insights from engineering, management economics,
sociology, political science along other aspects of social science. The paper emphasizes on requirement of
passionate and dedicated minds to solve the issues identified[19].
j) Ashwin Mahalingam, PPP Experiences in Indian Cities: Barriers, Enablers, and the Way Forward. The paper
studies policy gaps and its identification in the prevailing environment of urban Indian PPP and recommends
strategic approaches that can overcome barriers which are existing. This paper also studies the urban infrastructure
job in India using PPP to sustain the growth to fulfill the requirement to invest close to $500 billion in
infrastructure enhancement and maintenance over the period ranging from 2006 to 2011. Such large requirement
money which is involved and the vast amount of infrastructure that is to be constructed, it makes very clear that
the engagement of the private division will be necessary, for financing as well as in terms of implementation of
infrastructure projects. Thus Public-private partnerships are sole option has to be reviewed to be inevitable in the
prevailing Indian infrastructure. The implied observation points out the information that in India only 178 PPP
projects relating to transportation, housing, water and sanitation, and other basic urban services have been or are
being adopted in urban centers in whole country a majority of which are in the transportation sector. But this is a
very small fragment of total infrastructure investments. Also the non-rural Indian PPP domain stresses hurdles
that contributes to project failure and which could threaten the success of the PPP model in India. These hurdles
are then analyzed, studied, crafted and rectified to PPP enablers for successful delivery of projects. Also the paper
identifies various policy fissures in the non-rural current Indian PPP domain and proposes additional strategic
ways that can be engaged to combat existing hurdles and to safeguard sustainable PPP projects[20].
911
3 Problem Statement and Methodology
3.1 Problem Statement
Increasing urbanization is leading to scarcity of land, housing units and commercial centers in Nashik. In every
new 5 year development plan, exterior of prevailing areas is brought under development which requires huge
funds for infrastructure construction and its implementation. One of the principal challenge in front of developer
is high cost land. As the serious problem is the absence of a clear title property due to such problem financial
institutions and real estate developers hesitate to invest in projects of real estate. As there is no redevelopment and
concentration on available lands, land remains underutilized and thus contributing to shortage of land and hike in
land prices. But there are many existing plots in the core of the city which can be redeveloped using PPP models
for efficient use of land.
3.2 Methodology
a) The problem statement is defined with the help of literature.

b) Through expert interviews and literature, factors are found out and a questionnaire is prepared using importance
index.
c) A questionnaire survey is carried out which includes respondents like Government officials, contractors,
builders etc.
d) Collected responses are analysed using SPSS software to find the mean of all responses.
e) Ranking of these analysed factors is calculated to find critical factors.
f) Rating is given according to the mean.
g) Cost analysis models using Public Private Partnership (PPP) are prepared.
4 Data Collection and Data Analysis
4.1 Data Collection
The study of literature and discussion with the experts shows a number of factors are studied and identified in the
past as critical success factors. The respondents are requested to evaluate the factors on important index scale
starting with 1 as not important, 2 as slightly important, 3 as moderate important, 4 as very important, and 5 as
extremely important. Critical factors in housing sector using Public Private Partnership are studied by collecting
the data in the form of questionnaire survey from Government officials, contractors, builders, top managerial
authorities.
4.2 Data Analysis
For questionnaire survey 43 factors are collected which are arranged under heads of General factors, Technical
factors, Managerial factors, governing factors, Contractual factors, Financial factors, Operational factors and
Other factors are analyzed using SPSS software to calculate the mean.
912
Table 1 Descriptive Statics
N MEAN
Pre-project assessment 87 3.8276
Availability of finance & its provision 87 3.8046
Audience acceptance 87 3.3103
Differences in working 87 3.0000
Poor decision making 87 2.9655
Planning & designs with approvals 87 3.4713
Applicability 87 3.3908
Public guarantee for loan 87 3.8506
Project duration 87 4.5862
Geotechnical conditions 87 3.7701
Construction technology & method 87 3.9425
Design deficiency 87 3.7126
Poor quality of workmanship 87 3.4253
Resource availability 87 3.6092
Transparent procurement process 87 3.3448
Latest technology 87 3.0460
Coordination in system within consortium 87 4.1264
Scheduling and controlling 87 4.5402
Procedures for transferring project to client 87 3.5977
Dispute resolution system 87 3.3793
Risk resolution 87 3.0690
Governing body 87 3.0115
Unstable government 87 2.1839
Legal framework 87 3.9310
Political environment 87 2.9540
Investment schedule & guarantee revenue system 87 3.8161
Guarantee 87 3.7816
Government acts ex rera 87 2.6322
Prequalification of contractor 87 4.0805
Partnership formation 87 3.8966
Delay in approvals permits 87 3.8966
Formation of clauses 87 3.9425
Inflation 87 3.7471
Price changes& tariff changes 87 3.8621
Public credit 87 3.7356
Concessionaire changes 87 4.1034
Escalation 87 3.2874
913
Operational cost overrun 87 4.0460

Safety consideration 87 3.8851
Assistance in ppp 87 3.2874
Force majeure 87 3.8736
Political & social support in drafting phase 87 3.5517
Good governance 87 3.8966
Valid n (listwise) 87
From these mean values ranks are given to the critical success factors. Following table shows the ranks of critical
success factors in descending order of mean value.
Table 2 Ranking of Critical Success Factors
Factors Ranks Mean value

Project duration R1 4.5862
Scheduling and controlling R2 4.5402
Coordination in system within consortium R3 4.1264
Concessionaire changes R4 4.1034
Prequalification of contractor R5 4.0805
Operational cost overrun R6 4.0460
Formation of clauses R7 3.9425
Construction technology and method R8 3.9424
Legal framework R9 3.9310
Partnership formation R10 3.8966
Table 3 Rating of Critical Success Factors According to Their Criticality
Mean value Impact

0-1 Not critical
1-2 Fairly critical
2-3 Average critical
3-4 Very critical
4-5 Extremely critical
Table 4 Rating of Critical Success Factor and Their Numbers

CSF Total Number
Not critical 0
Fairly critical 0
Average critical 4
Very critical 33
Extremely critical 6
From above analysis of data collected the land cost and the availability of plots in the heart of the city is serious
problem but there are many government plots which can be redeveloped by public and private sector working
together.
914
5 Case study for Nashik district, Maharashtra, India
5.1 Gangotri Vasahat site, Nashik
Effective PPP model for redevelopment of government of buildings in golf club ground area on Trimbakeshwar
road. This area lies in the heart of the city. There is no availability of plots in this area. This area is densely
populated. Also there are government old offices and residential quarters on 45 m and 24 m road respectively.
Considering these offices and quarters redevelopment is appropriate option. Redevelopment can be suggested
using PPP for this location like commercial complexes and offices on 45m wide road and residential units on 24m
road. Area detail of the plot to be redeveloped is 8805 m2
Figure 1 Map of Gangotri Vasahat site

Proposed plan for Class 3 Staff quarters, Lokmanya Tilak Vasahat and Class 3 & 4 quarters Gangotri Vasahat,
Trimbak road, Nashik-422002 consist of residential and commercial units.
Site details
Plot Area 8805 Sqm
Basic Zonal FSI 1.10
Premium FSI permissible on Payment of Premium to
Authority 0.50
14088.00
Total FSI =1.6*8805 Sqm
Plot Rate = Rs. 6,967.67

Ground Floor Commercial Unit Rate = Rs. 14,000.00
First Floor Commercial Unit Rate = Rs. 11,500.00
Second Floor Commercial Unit Rate = Rs. 11,500.00
Third Floor Commercial Unit Rate = Rs. 9,500.00
Fourth Floor Commercial Unit Rate = Rs. 6,500.00
Constructions Cost = Rs. 2,500.00
Residential Unit Rate = Rs. 7,500.00
Constructions Cost = Rs. 1,500.00
Rate of interest = 12% pa
915
Commercial 3 units = 20593.16 Sqft.

Ground Floor 3 units = 20593.16 Sqft.
First Floor 3 units =20593.16 Sqft.
Second Floor 3 units =20593.16 Sqft.
Third Floor 3 units =20593.16 Sqft.
Fourth Floor 1 unit =4563.68 Sqft
Total commercial =86936.32 Sqft.
Residential 128 units = 127345 Sqft
Total Residential area = 214281.85 Sqft
5.1.1 Alternative I- Cost analysis for various models

Model 1 – for 60% -40%
60% for government
Total 60% saleable area = 128569.11 Sqft
Residential area = 87382.79 Sqft
Commercial area = 41186.32 Sqft
Cost of Construction
Cost of construction for residential = 1500*87382.79
Cost in figures = Rs 13,10,74,185 /-
Cost of construction for commercial = 2500*41186.32
Cost in figures = Rs 10,29,65,800/-
Total cost of construction = Rs 23,40,39,985/-
40% for Private entity
Commercial area = 45750 Sqft
Cost of construction for commercial = 2500*45750
Total cost of construction of project (R+C) before interest = Rs 40,83,59,095/-
Interest is applied on 50% amount for 2 years = Rs 4,90,03,091.4/-
Total cost of project after interest =Rs 45,73,62,186.4/-
Market selling price
Selling price for residential = 39962.74*7500
916
Selling price for commercial = (20593.16*14000)+(20593.16*9500)+(4563.68*6500)

= Rs 51,36,03,180/-
Total selling price (R+C) = Rs 81,33,23,730/-
Profit Margin of contractor
Profit = Market cost – construction cost (Govt. + Private)
= 81,33,23,730-45,73,62,186.4
= Rs 35,59,61,543.6/-
Model II – 50% 50%
50% for government
• 50% for Private entity
Selling price for residential = 61390.93 *7500
= Rs 51,36,03,180/-
=97,40,35,155 -45,73,62,186.4
= Rs 51,66,72,968.6/-
917
Model III – 55% 45%

55% for government
= Rs 51,36,03,180/-
= 89,36,79,405 -45,73,62,186.4
= Rs 43,63,17,218.6/-
5.1.2 Alternative II- Cost analysis for various models
Public private partnership projects require large funds and investment for its completion. Arrangement of these
large funds is difficult and involves risk. To solve this problem two or more private entities can form partnerships
and raise this huge amount of funds and share risk together. Following calculations are done for 3 private entities
including cost of construction for government.
918
Model I- 60% 40
60% for government
Cost of construction = Rs 23,40,39,985/3
= Rs 7,80,13,328.33/-
Residential area = 39962.74/3 Sqft
= 13320.9133 Sqft
Commercial area = 45750/3 Sqft
= 15250 Sqft
Cost in figures = Rs 1,99,81,369.95/-
Total cost of construction = Rs 5,81,06,369.95/-
Total cost of construction of project (R+C) before interest
= Rs 13,61,19,698.3/-
Interest is applied on 50% amount for 2 years
= Rs 4,90,03,091.4/3
= Rs 1,63,34,363.8/-
Selling price for residential = 13320.9133*7500
Selling price for commercial = Rs 17,12,01,060/-
Total selling price (R+C) = Rs 27,11,07,909.8/-
=27,11,07,909.8 -15,24,54,062.1
= Rs 11,86,53,847.7/-
Model II – 50% 50%

50% for government
= Rs 6,72,99,233.33/-
=20463.6433 Sqft
919

= 15250 Sqft
= Rs 13,61,19,698.3/-
= Rs 4,90,03,091.4/3
= Rs 1,63,34,363.8/-
=32,46,78,384.8-15,24,54,062.1
= Rs 17,22,24,322.9/-
Model III – 55% 45%

55% for government
= Rs 7,26,56,283.33/-
= 16892.2766 Sqft
Commercial area = 45750/3 Sqft
= 15250 Sqft
920

= Rs 13,61,19,698.3/-
= Rs 4,90,03,091.4/3
= Rs 1,63,34,363.8/-
= 29,78,93,134.5-15,24,54,062.1
= Rs 14,54,39,072.4/-
6 Cost Comparison Models
Table 5 Cost Comparison Models for Alternative I (Gangotri Vasahat)

Model 60%-40% 50%-50% 55%-45%
Sector Government Private Government Private Government Private
Total land
in area in 8804.26
Sqft
Built up 11944.46 7962.97 9953.72 9953.72 10949.09 8958.34
area in Sqft
Cost of
26,34,41,839 19,39,20,346 22,63,99,245 23,09,62,940 24,49,20,550 21,24,41,363
Constructio
.8 .6 .7 .7 .3 .1
n in Rs
Selling 81,33,23,730
----- -------- 97,40,35,155 -------- 89,36,79,405
price in Rs
Profit
35,59,61,543.6 51,66,72,968.6 43,63,17,218.6
margin Rs
921
Table 6 Cost Comparison Models for Alternative II (Gangotri Vasahat)
Model 60%-40% 50%-50% 55%-45%

Sectors Government Private Government Private Government Private
Total land
in area in 94768.215
Sqft
Built up 42856.37 28570.913 35713.641 35713.641 39285.005 32142.277
area in Sqft
Cost of
8,78,13,946. 6,46,40,115. 7,54,66,415. 7,69,87,646. 8,16,40,183. 7,08,13,878.
Constructio
61 49 23 85 42 61
n in Rs
Selling 27,11,07,909 32,46,78,384 29,78,93,134
--------- -------- --------
price in Rs .8 .8 .5
Profit
11,86,53,847.7 17,22,24,322.9 14,54,39,072.4
margin Rs
7 Conclusion
The above study concludes that conventional practices of planning, designing, execution, special planning and
management of resources such as men, material and money etc. are required for economically and timely
accomplishment of the project. Also, the daily increasing requirement for housing due to urbanization and
modernization has led to shortage of both land and housing units in the core parts of the city and adds pressure on
government. To meet this shortage there is requirement of large amount of funds along with time constraints. PPP
is seen as a financing instrument for bridging the demand supply difference in relation of private capital and
experience. Various concepts, framework, the critical success factors, risks allocation, criteria for selection of
private partners of PPP are studied. Redevelopment of government structures meet the goal of inner cities in-
filling them and lead to smart growth of cities. The site selection for the study in center of the city is done for
Nashik city. A questionnaire is developed and the factors affecting in housing, regenerated and redevelopment are
identified through questionnaire survey. The critical success factors are determined using IBM SPSS 23.0 which
are beneficial for redevelopment of government buildings in housing sector. The Critical success factors helps in
developing PPP models for the site selected. From the data collected a plan is proposed for redevelopment of
government quarters and cost analysis for various models like 60%-40%, 50%-50%, 55%-45% wherein greater
share is given to Government and lesser to private entity. Two alternatives are calculated wherein alternative one
is for single private entity along with Government forming a PPP and second alternative is where three private
entities come together form a partnership amongst themselves and the Government. It is found out that 50%-50%
model is the best possible model for redevelopment of government quarters for selected site using PPP which
results in win-win situation for both parties
922
8 References
1. Abhilash, R.S., Nirmal, A., Chandy, M.: ScienceDirect Ensuring public realm through private- private
partnership , in the context of rapid urbanization ( Case of CBD area Kochi ). Procedia Technol. 24, 1652–
1659 (2016). doi:10.1016/j.protcy.2016.05.181
2. Polyakova, I., Vasilyeva, E.: Benefits of public-and-private partnership for the creation of the
infrastructure of the urbanized territories in Russia. Procedia Eng. 165, 1380–1387 (2016).
doi:10.1016/j.proeng.2016.11.868
3. Alleviation, U.P., Bhawan, N., Delhi, N.: Model Public Private Partnership Policy for Affordable Housing
in India. (2017)
4. Zhang, X., Asce, M.: Paving the Way for Public – Private Partnerships in Infrastructure Development.
131, 71–80 (2005)
5. Gurgun, A.P., Ph, D., Touran, A., Ph, D., Asce, F.: Public-Private Partnership Experience in the
International Arena : Case of Turkey. 30, 1–11 (2014). doi:10.1061/(ASCE)ME.1943-5479.0000213.
6. Roehrich, J.K., Lewis, M.A., George, G.: Social Science & Medicine Are public e private partnerships a
healthy option ? A systematic literature review. Soc. Sci. Med. 113, 110–119 (2014).
doi:10.1016/j.socscimed.2014.03.037
7. Rwelamila, P.D., Fewings, P., Henjewele, C., Asce, A.M.: Addressing the Missing Link in PPP Projects :
What Constitutes the Public ? 31, 1–10 (2015). doi:10.1061/(ASCE)ME.1943-5479.0000330.
8. Projects, D.: Developing Public-Private Partnerships in Local Infrastructure and Devleopment Projects.
A PPP Manual for LGUs. Volume 1: Understanding PPP Concepts & Frameworks. (2012)
9. Zhang, X., Asce, M.: Financial Viability Analysis and Capital Structure Optimization in Privatized Public
Infrastructure Projects. 131, 656–668 (2005)
10. Alleviation, U.P.: Government of India Ministry of Housing & Urban Poverty Alleviation (RAY Mission
Directorate). 1–6 (2017)
11. Ogunsanmi, O.E.: Stakeholders ’ Perception of Critical Success Sub- Factors ( CSSFs ) for
Implementation of Public – Private Partnership Projects. 4, 89–102 (2014)
12. Osei-Kyei, R., Chan, A.P.C.: Implementing public-private partnership (PPP) policy for public
construction projects in Ghana: critical success factors and policy implications. Int. J. Constr. Manag.
3599, 1–11 (2016). doi:10.1080/15623599.2016.1207865
13. Sengupta, U.: Government intervention and public – private partnerships in housing delivery in Kolkata.
30, 448–461 (2006). doi:10.1016/j.habitatint.2004.12.002
14. Li, X., Yang, H., Li, W., Chen, Z.: Public-private partnership in residential brownfield redevelopment :
case studies of Pittsburgh. Procedia Eng. 145, 1534–1540 (2016). doi:10.1016/j.proeng.2016.04.193
15. Ke, Y., Wang, S., Chan, A.P.C.: Risk Allocation in Public-Private Partnership Infrastructure Projects :
Comparative Study. 16, 343–351 (2010)
16. Zhang, X., Chen, S.: A systematic framework for infrastructure development through public private
partnerships. IATSSR. 36, 88–97 (2013). doi:10.1016/j.iatssr.2012.11.001
17. Invention, M.: A Study on Public-Private Partnerships with Reference to Indian Infrastructural Projects
Tharun Shastry L. 3, 56–62 (2014)
923
18. Tang, L., Ph, D., Shen, Q., Asce, M., Skitmore, M., Cheng, E.W.L.: Ranked Critical Factors in PPP
Briefings. 29, 164–171 (2013). doi:10.1061/(ASCE)ME.1943-5479.0000131.
19. PPP EXPERIENCES IN INDIAN STATES: BOTTLENECKS, ENABLERS AND KEY ISSUES
Ashwin Mahalingam 1.
20. Mahalingam, A., Asce, M.: PPP Experiences in Indian Cities : Barriers , Enablers , and the Way Forward.
136, 419–429 (2010)
924
Paper ID- 172
Key Performance Indicators for Public Private Partnership Projects in India-

a case study
Rajanikant Devalkara*(0000-0001-9522-7141),Krupesh Chauhanb (0000-0002-6854-7956),
Sanjay Yadavb(0000 0001 8480 2227)
a,b
Sardar Vallabhbhai National Institute of Technology, Surat,395007,India
Abstract
Public Private Partnership Projects (PPP) have become an important option in India. Though there are many
problems faced, PPP’s are still considered as an interesting option. To achieve maximum Value for Money, Key
Performance Indicators can be viewed as an effective method; KPI is the performance measurement tool in the
construction industry. This article describes in detail the 7 Performance Packages and 44 Performance Indicators.
This data collection is carried out from comprehensive literature survey, discussion with expert and various case
studies from all over the world.Questionnaire survey was conducted in all over the world which are related to
PPP````` projects to get the ratings for each indicators separately. Likert Scale technique was used to rank the
indicators. The responses so obtained were then put into SPSS software, where the mean, standard deviation and
rank were obtained for each performance indicator using mean score ranking technique. A case study of Nashik
smart road pilot project has been taken. From the obtained list of KPI’s 5 Project performance metrics are suggested
which relate to Time management, Cost management, Quality standards, Risk management and Management of
resources for the smart road development scheme.
Keywords
Public Private Partnership (PPP); Key Performance Indicators; performance measurement
1 Introduction
Public-Private Partnerships (PPPs) have become an integral part of infrastructure procurement strategy in many
governments across the world [1] so they are attaining the popularity in developing countries. Major challenge is
financial crisis for implementing of PPP [2]. The rationale of PPP is nothing but to focus on taking up of his concept
and making it useful for the public by providing with the services in a good quality [3] However, with respect to
their value of money PPP is a good option for public projects. In performance measurement the Key Performance
Indicators (KPI) are the most important factors, viewed as efficient way to deliver monetary value[4].
Performance measurement is an essential part of management and thus may have been practiced since many years.
Industrial performance measurement in construction has normally depends on efficiency, returns, and profitability
[5],in terms of progress i.e. work completion within time[6]. Various projects which come under infrastructural
development are considered to be undertaken by Public Private Partnerships [7].PPP needs clear specifications as
the transaction which under PPP are complex and the services to be provided should be clear and to understand how
the risks are allocated between public and private sector[8]. There has been an increase in PPPs because of many
reasons such as performance improvement in the public sector, cost stabilization to provide services, innovations in
925
operations, maintenance of methods used. As there are many constraints in the budget of government they should be
reduced by the private stakeholders investments. For the public services the private entities have always been
involving for delivering these services. PPP is a very old concept which is being used for many years. It was first
initiated in 1970 in the USA. First of all, the importance was given for economic infrastructure and then importance
was given for social infrastructure and other services [9].
2 Key Performance Indicators
Key Performance Indicators are one of the factors which help in making the PPP project become successful
and help work the project efficiently. KPI is the performance measurement tool in the construction industry. The
construction industry KPI was first published in 1999, and is updated manually by the UK working group. This
improvement target formed the basis for the national headline KPI’s in the UK, which were designed to show how
improvement would be demonstrated [10]. These KPI’s are now used in the construction industry to measure the
performance of the project. For a successful PPP project, it is important to check on the performance of the project.
If a PPP project is not able to fulfill its main objectives then the project is to undergo a failure so to avoid this, KPI
can guide us to find out which are the critical factors in the project, to concentrate on those factors to work
efficiently. The improved KPI model suggests 41 PI, which indicate that performance enhancement in PPP projects
is strongly depend on procurement, design and planning in the public sector projects, efficient process control in the
private sector and the ultimately the satisfaction of both the public and private sectors. These 41 PI calculates the
strengths and weaknesses of all types of PPP construction projects and improve performance of PPP projects [4].
The KPIs can also be use for empirical study, the number of KPI should be kept minimum and a consensus needs to
be obtained from interviews ,future work surveys and discussions with professionals for effective application in
project [11].
3 Methodology
3.1 Literature Review
In literature review different papers are studied for studying different success factors. Primary data from
literature search, academic books, research papers and Journals are main sources used in this study. After all the
survey and research work has done 44 performance indicators has found out and they are clubbed under 7 different
Key Performance Indicators. This stage will be identifying research objectives and research needs, and will include
a comprehensive review and background discussion of the topic to help define the methodology and final results.
3.2 Data Collection (Part A)
The second stage consists of data collection (Part A) of various indicators which affects the implementation of PPP
projects. This data collection is carried out from comprehensive literature survey, discussion with expert and various
case studies from all over the world.
926
3.3 Preparation of Questionnaire Survey
The third stage consists of preparation of questionnaire survey form by compiling the factors and sending
to respondents. The random sampling technique is use to select number of respondents. The questionnaire form is
send to the respondents by both the method online as well as offline. For online method the questionnaire form is
created in Google forms and sent it to the various respondents all over the world which are related to PPP projects.
For offline method i.e. for direct respondents face to face interview is carried out and questionnaire form is filled.
3.4 Data Collection (Part B)
The fourth stage consists of data collection (Part B) of questionnaire form from the respondents who
responded to online forms.
3.5 Data Analysis
The fifth stage consists of analysis of data collected from various respondents. For data analysis the SPSS
software were used in which descriptive statistical analysis is carried out. Then mean, rank, standard deviation and
frequency for all factors were found out.
4 Performance packages
The performance packages were collected from various international journals, books, literature, internet
and self- learning. The performance packages may differ from country to country. The performance packages in this
article are relatable to the Indian economy. Following were the performance packages that were selected for the
study.
4.1 KPI1 - Financial
The PI’s in this package include those factors which are related to the PPP project financially. This package
is further divided into 11 performance indicators.
4.2 KPI2- Physical Parameters
In this package the physical conditions related to the project are included. Here 7 indicators are included.
4.3 KPI3 - Innovation
This package includes all the new techniques that can be used in a PPP project. 7 performance indicators
come under this category.
4.4 KPI4 - Legislative
This package includes those factors which are related to the governance, political issues, laws etc. 3
performance indicators come under this group.
927
4.5 KPI5 - Labour
The factors which are related to the labours are included in this package. 5 indicators come under this
category.
4.6 KPI6 - Management
All the managing activities or factors come under this package. 10 indicators come under this category.
4.7 KPI7 - Health and Safety
This package includes the indicators related to the safety and health of labours working on site. 3
performance indicators come under this category.
5 Likert Scale Technique
A Likert Scale Technique was used to take the responses from the respondents to evaluate by giving it a
quantitative value, with level of agreement or disagreement as dimensions. Often five ordered response levels are
used. In this article also 5 order response levels are used. The format of a typical five – level Likert scale or use for
survey analysis is as follows:
 Not Important
 Less Important
 Moderate
 Important
 Very Important
6 Mean Score Ranking Technique
The questionnaire survey included a table ranking questions which asked the respondents to rank a list of
choices according to its importance. In order to analyse these specific type of questions, the mean score ranking
technique was used. Options were given rating scale numbers from 1 to 5 for use in calculations and statistical
analysis. For example, for the level of importance ranking questions, not important was given a score of 1, and very
important was given a score of 5. The mean score of each option in the question is then calculated by using the
following formula: MS = Σ (f×s) / N where 1≤MS≤5 Where s = score of each factor by respondents, ranking from 1
to 5, f = Frequency of each rating ,N = Total number of responses.
7 Questionnaire Survey
An online questionnaire was prepared using Google Forms. Experts in the field of PPP were sent the
questionnaire online through google forms. The respondents were from national as well as international levels. The
questionnaire was sent to about 145 experts in the field of PPP in different countries all around the world, of which
70 people respondent.
928
Table 1 List of KPI and PI

KPI Package Code Performance Indicator
F1 Global financial crises
F2 Favourable economic conditions
F3 Demonetization
F4 Inflation
F5 Investing parties
KPI1 - Financial
F6 Sound financial analysis
F7 Safe investments
F8 Insurance coverage
F9 GST (India)
F10 Stock and market conditions
Pp1 Site conditions
Pp2 Design complexity
Pp3 Type of construction
KPI2 - Physical parameters Pp4 Purpose of construction
Pp5 Governments knowledge of PPP
Pp6 Flexibility of project
Pp7 Quality standards
I1 Use of new technology
I2 Creative financial packages
I3 Use of Hi-tech instruments/machineries
KPI3 - Innovation I4 Technology transfer
I5 Employee training
I6 Digitalising work
I7 CCTV surveillance
L1 Stable legal environment
KPI4 - Legislative L2 Stable laws
L3 Changing rules and regulations
La1 Wages of labour
La2 Provident funds for labour
KPI5 - Labour La3 Skilled labours
La4 Optimum salaries for
La5 Good relationship amongst the workers
M1 Stakeholders management
M2 Management of activities on site
M3 Cost management
KPI6 - Management
M4 Time management
M5 Risk management
M6 Management of resources
929
M7 Safety management
M8 Administrative management
M9 Conflicts management
M10 Transfer management
H1 Safety of labours on site
H2 Predetermining risks
KPI7 - Health and safety
H3 Undertaking free health check-ups for on-site
workers
8 Use of SPSS Software
SPSS software was used to get the mean and standard deviation of the indicators by putting the scale given
by the respondents to the indicators. SPSS Statistics is a software package used for logical statistical analysis. Using
Likert scale the number of responses for a particular indicator was calculated. Mean and standard deviation were
calculated. After getting the values of means and std. deviation ranks were given to the indicators based on the mean
value. The indicator having maximum mean value was given 1st rank, while the indicator having minimum value
was given the last rank. For example, the mean value of the indicator F1 (Global financial crises) was calculated to
be 3.8429 and std. deviation 0.84503, mean value of Pp4 (Purpose of construction) was 3.9149 and std. deviation
0.84687, similarly mean value of L3 (Changing rules and regulations) was 4.2143 and std. deviation 0.74001, for
M6 (Management of resources) 4.3714 and std. deviation 0.68464, for I1 (Use of new technology) 3.9000 and std.
deviation 0.80127, etc. in this way the mean and standard deviation for all the performance indicators were
calculated, which thereby were given ranks.
For the selection of factors which are best for the implementation of PPP, the factors are ranked by using
SPSS descriptive analysis. The result obtained from survey and the factors are arranged in their decreasing score.
According to the mean value, ranking is given to each performance indicator. In the KPI list we see “Time
management” indicator having mean value 4.6087 which is highest and the “Demonetization” having mean value
3.3696 which is the lowest in the Performance Indicators list.
10 Grouping of Survey Results
After calculating mean and ranking of the performance indicators, the selection of some indicators which
are having mean value >= 4.000 is done. 19 performance indicators were considered to be most important for the
performance of a PPP project which had their mean value >=4. Following is the table showing the important
indicators having the mean value greater than 4.
Table 2 List of Pi having mean value >=4
Performance indicators Mean Std. deviation Rank
Time management 4.6087 0.57651 1
930
Risk management 4.5870 0.61738 2

Quality standards 4.5111 0.62603 3
Cost management 4.4565 0.58525 4
Stable laws 4.4444 0.69267 5
Management of resources 4.4222 0.69048 6
Stable legal environment 4.3778 0.80591 7
Safety of labours on site 4.3478 0.67387 8
Favourable economic conditions 4.3261 0.59831 9
Skilled labours 4.2826 0.71997 10
Good relationship amongst the workers 4.2826 0.80727 11
Stakeholders management 4.2667 0.71985 12
Changing rules and regulations 4.2222 0.76541 13
Safety management 4.1739 0.70881 14
Investing parties 4.1522 0.75916 15
Management of activities 4.1333 0.72614 16

Predetermining risks 4.1087 0.73721 17
Sound financial analysis 4.0870 1.15135 18
Administrative management 4.0000 0.78881 19
11 Discussion on Frequency Table
Using SPSS descriptive statistics mode we analyze individual factor. We can calculate frequency,
cumulative percent and by using charts we can interpret the data on the bar chart. Following table shows the
frequency, cumulative percent for the Purpose Of Construction factor. In that we see 4 respondents selected Purpose
of construction is Less important . Then 16 respondents said that Purpose of Construction factor for the successful
implementation of PPP is Moderately important. Likewise 32 respondents selected it as important and 18 selected it
as very important.Y-axis represents percentage and X-axis ratings scheme. Likewise for all the factors individually
bar charts and percentages are calculated.
Table 3 Purpose of Construction
Frequency Percentage Valid Cumulative
percent percent
Not important 0 0 0 0
Less important 4 5.7 5.7 5.7
Moderate 16 22.9 22.9 28.6
Important 32 45.7 45.7 74.3
Very important 18 25.7 25.7 100
Total 70 100.0 100.0
931
Figure 1 Bar Chart Showing the

he Frequency of Responses for a Single Indicator
12 Project Performance Metrics for Nashik, INDIA Smart Road Pilot Project a Case Study
12.1 Introduction to Performance Metrics
A project needs quantitative data that can be used to track the project’s health, performance. Project
management metrics are useful to implementing practical processes in any organization. The owners want to be
show how the project is running.. This process will help us to target what and how can be measured and ultimately
display the best metrics to show the success and health of the projects.
12.2 About Nashik Smart Road Development Scheme – A Pilot Project
A tender for the smart road pilot project work from Ashok Stambh to Trimbak Naka by NMC (Nashik
Municipal Corporation) is proposed. A draft plan was prepared for the 1.1km road. The survey of the road was
completed in three phases. This road will be model for other proposed roads.
12.3 Inclusions of the Plan
There were various amenities giv

given in the scheme which includes kiosks providing information on
important places of the city ,pedestrian
pedestrian walkways on both sides of the road, underground electric cables ,CCTV
cameras, seating arrangement for senior citizens on the walkways, one metre wide cycle tracks
racks on both sides of the
road and deck to prevent digging of roads.
CBS to Trimbak Naka and Meher to Red Cross signal will be one-ways,
ways, another road development includes
inclu from
Ambedkar statue to CBS.
The final plan of the project scheme is not yet prepared. It will be done after discussion with the police
commissioner, traffic
affic police, MSRTC and MSEDCL.
12.4 Basic Details of the Plan

Estimated Cost
Cost: Rs 16,15,04,998 /-
932
Period of Contract
Contract: Construction : 6 months
(rainy season also included)
Figure 2 Map Showing the Road Selected For Smart Road Pilot Project from Trimbak Naka to Ashok
Stambh
13 Suggested Project Management Metrics for the Smart Road Pilot Project
13.1 Metrics 1: Schedule and Effort Variance Metrics
In this metrics the key is to measure the progress and performance of the project against the baseline. This
metric is very important and is used for profitability of project.
Planned Value (PV): The amount spent as much as is planned. Another name for PV is Budget at Completion
(BAC).
PV at any stage = (Planned % Complete) X (BAC)
Earned value (EV): The value of work that is completed and which is expressed in terms of the budget and which is
assigned to that work or activity.
Actual cost (AC): The total cost of the eactual work completed to date and recorded in accomplishing work
performed for an activity.
Schedule Variance (SV): schedule performance of the project.
SV = EV – PV
Cost Variance (CV):: cost performance on the project.
CV = EV – AC
933
Figure 3. Graph Showing Planned Value, Earned Value, Actual Cost

13.2 Metrics 2: Project Time Metrics
This metric measures time spent on maintaining status quo vs forward-looking projects.also measures
capital hours vs. expense hours.The projects need to be properly categorized as new development, maintenance,
support, and also deciding misc. work activities and misc. non-work activities, projects as expense. The actual hours
are recorded using timesheets in the project management system.
These metrics are communicated to manager team in every month. The data so collected is to be put in place which
may include appropriate management, timely checking of timesheet submitting the report of time prepared and
categorizing the project.
13.3 Metrics 3: Project Performance Metrics
In this metric the comparison of performance of the project is done by the budget which is approved and
the date that is committed.
The important attributes of the project are identifying the true reason for the changes and appropriate time. The
PMO captures the performance baseline work effort and the current date, and the actual hours are captured using
timesheets and are used to report the actual working hours in the project management system.
This metric is communicated to engineers and project 0managers.
13.4 Metrics 4: Project Scorecard Metrics
In this metric a scorecard is maintained to check the project performance against defined criteria that need
to be achieving for the project to be a success. It includes Cost of the project, performance of Quality and Schedule,
and Customer Satisfaction.
User Satisfaction is measured by conducting a stakeholder survey. There are various questions asked to the
customers and the scores are given to the project.
The project managers or coordinators prepares the scorecard for each project and make changes accordingly.
934
13.5 Metrics 5: Quality & Customer Satisfaction Metrics
Quality Assurance is the main priority of project manager during execution of project. It proves the health
of the product to the end user and drives the Customer Satisfaction. Good quality should be reported throughout the
life of the project or work and little defects are also shows impact on the work or project
Defect age: Number of days since the defect is seen and not repaired. It can also be stated as the customer has been
waiting for their issues to get resolved under quality metrics, following are the key ones to track:
𝐓𝐨𝐭𝐚𝐥 𝐧𝐮𝐦𝐛𝐞𝐫 𝐨𝐟 𝐝𝐞𝐟𝐞𝐜𝐭𝐬 𝐟𝐨𝐮𝐧𝐝
𝐃𝐞𝐟𝐞𝐜𝐭 𝐝𝐞𝐧𝐬𝐢𝐭𝐲 =
𝐌𝐞𝐚𝐬𝐮𝐫𝐞 𝐨𝐟 𝐬𝐢𝐳𝐞
𝐓𝐨𝐭𝐚𝐥 𝐧𝐮𝐦𝐛𝐞𝐫 𝐨𝐟 𝐝𝐞𝐟𝐞𝐜𝐭𝐬 𝐫𝐞𝐬𝐨𝐥𝐯𝐞𝐝

𝐃𝐞𝐟𝐞𝐜𝐭 𝐫𝐞𝐬𝐨𝐥𝐮𝐭𝐢𝐨𝐧 𝐫𝐚𝐭𝐞 =
𝐓𝐨𝐭𝐚𝐥 𝐞𝐟𝐟𝐨𝐫𝐭 𝐬𝐩𝐞𝐧𝐭
Figure 4 Graph Showing Defects Assigned Vs Defects Resolved
14 Conclusion
For the infrastructural development in the country various measures are being undertaken by the
government. PPP is one such case. It is concluded that PPP is an effective way to deal with shortage of funding
issues only by one party. Based on factors by using SPSS software Key Performance Indicators were analyzed. The
present study examined the Key Performance Indicators for PPP implementation. The results indicate that all the
performance indicators were rated properly and based on the results from the total respondents factors in the top
rankings include the indicators given in Table II of the paper. Among these 19 PI’s, 5 performance indicators are
chosen i.e. Time management, Cost management, Quality standards, Risk management and Management of
Resources and project performance metrics are suggested to the Nashik smart road development project from
Trimbak road to Ashok stambh. These metrics when applied while executing the project can contribute to make this
PPP project successful.
935
15 References
1. Liu, J., Love, P.E.D., Smith, J., Sing, C.P.: Testing a PPP Performance Evaluation Framework. 681–686
(2016). doi:10.1680/tfitsi.61279.681
2. Akhmetshina, E.R., Mustafin, A.N.: Public-private Partnership as a Tool for Development of Innovative
Economy. Procedia Econ. Financ. 24, 35–40 (2015). doi:10.1016/S2212-5671(15)00609-7
3. Toor, S. u R., Ogunlana, S.O.: Beyond the “iron triangle”: Stakeholder perception of key performance
indicators (KPIs) for large-scale public sector development projects. Int. J. Proj. Manag. 28, 228–236
(2010). doi:10.1016/j.ijproman.2009.05.005
4. Yuan, J., Ph, D., Wang, C., Skibniewski, J., Ph, D., Asce, M., Li, Q., Ph, D.: Developing Key Performance
Indicators for Public-Private Partnership Projects : Questionnaire Survey and Analysis. 28, 252–264 (2012).
doi:10.1061/(ASCE)ME.1943-5479
5. Bassioni, H.A., Asce, S.M., Price, A.D.F., Hassan, T.M., Asce, M.: Performance Measurement in
Construction. 20, 42–50 (2004)
6. Ahmad, S.B.S., Svalestuen, F., Andersen, B., Torp, O.: A review of performance measurement for
successful concurrent construction. Procedia - Soc. Behav. Sci. 226, 447–454 (2016).
doi:10.1016/j.sbspro.2016.06.210
7. Chou, J.S., Pramudawardhani, D.: Cross-country comparisons of key drivers, critical success factors and
risk allocation for public-private partnership projects. Int. J. Proj. Manag. 33, 1136–1150 (2015).
doi:10.1016/j.ijproman.2014.12.003
8. Ke, A.P.C.C.J.F.Y.Y.C.C.P.Y.S.Q.W. and Y.: Empirical Study of Risk Assessment and Allocation of
Public-Private Partnership Projects in China. J. Manag. Eng. 27, 81–90 (2013).
doi:10.1061/(ASCE)ME.1943-5479.0000049.
9. Kaganski, S., Majak, J., Karjust, K., Toompalu, S.: Implementation of Key Performance Indicators Selection
Model as Part of the Enterprise Analysis Model. Procedia CIRP. 63, 283–288 (2017).
doi:10.1016/j.procir.2017.03.143
10. Sibiya, M., Aigbavboa, C., Thwala, W.: Construction Projects’ Key Performance Indicators: A Case of the
South African Construction Industry. Iccrem 2015. 2018, 954–960 (2015). doi:10.1061/9780784479377.111
11. Yuan, J., Zeng, A.Y., Skibniewski, M.J., Li, Q.: Selection of performance objectives and key performance
indicators in public-private partnership projects to achieve value for money. Constr. Manag. Econ. 27, 253–
270 (2009). doi:10.1080/01446190902748705
936
Paper ID- 173
Performance of Reinforced concrete Beams with M-Sand

as Alternate Fine Aggregates
Praveen Kumar K1* (0000-0001-6641-3404), Lokesh Kumar G2 (0000-0002-9060-
8866), Radhakrishna 3 (0000-0003-2151-3393) and S M Basutkar4 (0000-0002-0616-
2364)
1 Rashtreeya Vidyalaya College of Engineering, Bengaluru, India
praveenkumark@rvce.edu.in
Abstract. Natural sand is used as fine aggregate in manufacturing of cement

concrete, whose availability is decreasing day by day. There is an acute need to
address this issue; by finding suitable alternate material, which can replace
natural sand, without compromising the properties. This study explores the
behaviour of reinforced concrete beams with alternative fine aggregate, M-Sand.
M 25 grade concrete was designed with natural sand as fine aggregates. Natural
sand was substituted by M-sand at 0, 50 and 100%. Constant workability of 90-
110 mm slump was maintained at each replacement level by varying dosage of
superplasticizer. Reinforced concrete beams of size 150mm x 200mm x 1500mm
was designed as per Indian standards. Nine Beams were cast with concrete
having natural sand and at 50,100% replacement of M-sand as fine aggregates.
Beams were tested over 1300mm simply supported span under two point loading
system till failure. First crack load, Load –Deflection characteristics, Deflection
at first crack, variation of compressive and tensile stresses with strain, Energy
absorption capacity, Ductility factor, failure load, Deflection at failure, Ultimate
moment carrying capacity and crack width were studied. Microstructure of
concrete was analysed using SEM and EDS. Results indicate that workability of
concrete reduces whereas compressive strength increases with increase in M-
sand percentage. Substitution of M-sand results in dense and compact structure
with enhanced bonding between fine aggregates and cement paste. Enhanced
energy absorption capacity, failure load, ductility factor, moment carrying
capacity and ultimate deflection were observed in M-sand concrete beams as
compared to beams with natural sand. First crack was seen at a lesser load and
lesser deflection was observed, which shows slightly brittle nature of M-sand
concrete. Good correlation was observed between experimental and analytical
results with respect to Ultimate moment carrying capacity. Hence use of M-sand
can be recommended advantageously in concrete for production of reinforced
concrete beams.
Keywords: Concrete, SEM, EDS, Beams, Ultimate load, Moment, Crack width,
Ductility factor
1 Introduction
Presently world is staring at a possibility that, in the near future, raw materials required
by the construction industry are fully consumed and are no longer available for further
937
2
construction works. One such material is natural sand. The availability of this
conventional raw material, which is used as fine aggregate in manufacturing of cement
concrete, is decreasing day by day [1,2]. Fine aggregate fills about 35% of volume of
concrete and plays an important role in imparting the properties to concrete in its fresh
and hardened state [3]. Normally, natural sand from river beds or banks is used in
conventional concrete as fine aggregates. There is a huge demand for natural sand in
developing countries like India to cater rapid infrastructural growth. Excessive
extraction of natural fine aggregates from riverbed is causing threat to environment and
in many places; already toppled the ecological balance as well as lowered the ground
water table. With a concern to protect the environment, its recreational value and
biodiversity, government agencies have laid down restrictions on sand quarrying
process across the world and this resulted in scarcity as well as significant increase in
the cost of natural sand. To satisfy the increased demand for fine aggregate from
construction industry, research is oriented towards alternative materials, which can
replace natural sand in concrete without effecting the properties. Review of literature
shows active research on use of wide range of alternative materials such as Pond ash,
M-sand, Processed slag sand, recycled aggregates etc in preparation of cement mortar
and Concrete [2 One of them is M- sand, which is obtained by crushing rocks using
machines such as Vertical Shaft Impact (VSI) crushers, washing and grading. It can be
an alternative for natural sand in concrete, however its suitability to be investigated.
Grading limits of M-sand generally falls within grading limit II as specified by IS: 383-
2016 [4]. Shape and properties of M-sand are comparable with those of river sand.
Grading of M-sand with stretched ‘S’ shaped curve yields better workability when
compared to M-sand with reverse S shape or parabola.[5]. Mortars with M-sand shows
better workability, higher water retentivity, better compressive strength and similar
stress-strain characteristics as compared to mortar with natural sand [6]. On contrary,
Beixing et.al [7] reported that manufactured sand particles are more angular and have
rougher surface as compared to river sand. Concrete containing M-sand exhibits higher
compressive, flexural strength and improved abrasion resistance. Cabrera et. al. [8]
concluded that admixture dosage as per manufacturers guidelines are not much
effective in M- sand mortars, as in case of mortars prepared with natural sand. Water
requirement for achieving certain workability increases by around 25-45 kg/m3. Tao Ji
et al. [2] developed a design method for proportioning M-sand concrete; based on
minimum paste theory and concluded that presence of microfines in moderate amount
in M-sand improves the workability and mechanical properties. Cepuritis et al. [9]
reported that characteristics of particle sizes of 0.125-2 mm and < 0.125 mm are
predominant factors affecting the fresh properties of concrete, when there is no change
in total grading of sand. By varying the properties of crushed filler (<.125mm), huge
enhancements in the rheology and fresh properties of concrete can be accomplished.
Long term compressive strength development in M-sand concrete is similar to river
sand concrete, and presence of stone powder within 13% shows beneficial effect [10].
Presence of stone powder in M-sand upto 13% is beneficial for long term tensile
strength development [11]. Shen et al. [12] reported that ultra-high strength concrete
prepared with sandstone M-sand exhibits higher strength than that prepared from river
sand. The microstructure of ultra-high strength concrete prepared with M-sand is very
938
3
dense and the corresponding hydration product sizes such as that of ettringite, Calcium
Sulphoaluminate hydrates, calcium hydroxide, monosulfate, and C-S-H gel becomes
much finer than those in the river sand concrete. M-sand based concrete has narrower
ITZ, presence of small amount of AFt crystals in the ITZ, than that of River Sand
concrete, while the AFm , stick AFt and portlandite crystals are enriched at the ITZ of
River sand concrete. The microstructure of M-Sand concrete is denser than that of River
Sand [12]. Abrasion resistance of roller compacted concrete improves and cantabro loss
reduces with increase in M-sand percentage [13]. The presence of more fines in M-
sand enhances the water demand, but it contribute to an increase in paste volume in
SCC. Low and medium strength SCCs of 25-60MPa can be developed by using M-
sand [14]. Substitution of river sand by M-sand in SCC enhances fresh properties and
results in improved mechanical properties in presence of 30% fly ash as replacement
for cement [15]. Mukherjee and Rai [16] experimentally investigated the flexural
behaviour of Reinforced concrete (RC) beams after reaching ultimate capacity and then
retrofitted with pre stressed carbon fibre reinforced composite (CFRC) laminates.
Rehabilitated beams have shown improved performance in terms of higher failure load
as well as lower deflection. CFRC beams continued in the elastic zone for considerably
higher loads, which resulted in increased area under load-deflection curve. High
strength concrete beams reinforced with steel fibers, when loaded under two point
loading exhibits linear variation of deflection and moment curvature relation upto first
crack load and thereafter non-linearly till ultimate load [17]. Theoretical investigations
on crack width evaluation of various types of RC members subjected to two point
loading considering three parameters namely reinforcement ratio, arrangement and
grade was reported by Allam et. al. [18].
Preliminary investigations reveal favorable substitution of natural sand with M-sand in
mortar and concrete at varying percentages. Numerous investigations were reported on
characteristics of mortar as well as concrete with M-sand. The influence of fines on
fresh and hardened properties of M-sand concrete is well reported; and varying
percentage of fines is the reason on conflicting results. M-sand shows better
performance in Roller compacted concrete and SCC also; when used as a replacement
for natural sand. However; the properties of RC beams with M-sand concrete are not
reported. Hence present investigation is oriented to evaluate the behaviour of RC beams
having concrete with natural sand and at different replacement percentages of M-sand.
2 Objectives of the Research

The objective of the present study is to determine the fresh, hardened properties and
microstructure of concrete at different replacement levels of M- sand and hence to
evaluate the structural behaviour of reinforced concrete beams under two point loading.
Fresh properties as well as compressive strength of concrete with M-sand and
behaviour of RC beams with M-sand are compared with that of natural sand concrete
and RC beams with natural sand to understand the effects of replacement of M-sand.
939
4
3 Materials and Mix Design
In preparation of concrete following materials were used

(i) Ordinary Portland Cement(OPC)
(ii) Natural sand and M-sand
(iii) Coarse aggregates
(iv) Tap water
(v) Superplasticizer
Ordinary Portland cement (OPC) of 53 Grade conforming to IS 12269:2013 [19],

having specific gravity of 3.1 was used as binder. Natural sand and M- sand were used
as fine aggregates (F.A) whose properties are as shown in Table 1.
Table 1. Physical properties of fine aggregates
Property Natural Sand M- Sand

Water Absorption (%) 1.2 1.1
Moisture content (%) 0.4 0.3
3
Loose Bulk Density(kg/m ) 1351.5 1691.3
3
Compacted Bulk Density(kg/m ) 1504.6 1771.7
Fines content (%) 1.1 7.4
Zone as per IS 383:2016 [4] II II
Particle size distribution of Natural sand and Slag sand are shown in Figure 1. The
gradation curves of Natural sand (NS) and M-Sand (MS) are almost similar and both
are well graded. Coarse aggregate (CA) derived from granite source having specific
gravity of 2.60 were used in the present study. Water absorption and Moisture content
in coarse aggregate were 0.72 and 0.42% respectively. Tap water satisfying
requirements of IS 456-2000 [20] and commercially available Poly Carboxylic Ether
based superplasticizer with a specific gravity 1.08 conforming to BS: 5075-3:1985 [21]
and IS: 9103:1999[22] were used in preparation of concrete.
940
5
100
80
Percentage Finer
60
0%
40
100% MS
20
0
0.01 0.1 1 10
Sieve size in mm
Figure 1. Particle size distribution curves of Natural sand and M-sand

M25 grade concrete was designed as per IS: 10262-2009 [23] and the Final mix was 1:
1.99: 2.97 with w/c ratio of 0.4. Mix details are given in Table 2. High Yield Strength
Deformed (HYSD) bars having a yield strength of 500 N/mm2 confirming to IS: 1786:
2008 [24] were used in preparation of RC beams.
Table 2. Mix Proportion for M25 concrete
Mix F.A (kg/m3) Water Admixture
Cement C.A Content dosage
(kg/m3) River M- (kg/m3) (Liters/m3) (kg/m3)
sand sand
NS 394 787.2 0 1172.02 157.6 0.394
MS50 394 393.6 393.6 1172.02 157.6 0.591
MS100 394 0 787.2 1172.02 157.6 0.749
Under reinforced RC beam having size of 1500mm x 150mm x 200mm supported over
simply supported span of 1300mm were used in the investigation. Bottom Tension
reinforcement of 2-Y12, and hanger bars of 2-Y10 at top were used. 2Legged-
Y8@125mm c/c stirrups were used as shear reinforcement.
4 Methodogy
Concrete mix of M25 grade was prepared with natural sand as fine aggregates to obtain
90-110mm slump. Two more types of concrete, one with 50% M-sand and other with
100% M-sand were also prepared. Required slump was obtained by changing the
superplasticizer dosage. 3 number of RC beam were cast for each of 0, 50 and 100%
M-sand replacement concrete. Needle vibrator was used to ensure proper compaction.
After 24 hours beams were demoulded and cured for 28 Days. Compressive strength of
941
6
concrete was determined by casting and testing 150mm cubes in Compression Testing
Machine as per IS: 516-1959 [25]. RC beams were tested under two point loading
system in a loading frame with a load cell having capacity of 500kN. Test set up is
shown in Figure 2. Three dial gauges were attached, two under the loading points and
one under central point for noting the deflections. Two demec gauges, one each for
measurement of compressive strain and tensile strain were installed. Stress- strain
properties, Deflection at midspan, First crack load and Ultimate load were noted.
Ductility factor is calculated by taking the ratio of ultimate load deflection to the
yielding load deflection. Experimentally obtained moment carrying capacity were
compared with theoretical moment carrying capacity as per IS: 456-2000 [20].
Figure 2. RC beam testing arrangement

Workability of concrete decreases with increase in M-sand percentage. Dosage of
superplasticizer required were 0.10, 0.15 and 0.19% respectively for NS, MS50 and
MS100. Rougher surface texture and lesser sphericity of M-sand compared to natural
sand generates more friction, which necessitates higher dosage of admixture.
Compressive strength of concrete at 28 days for NS, MS50 and MS 100 are 32.33, 37.11
and 36.17 N/mm2 respectively. The SEM and EDS images of NS, MS50 and MS100
are shown in Figure 3a, 3b, 3c, 3d and 3e and 3f respectively. It is evident that MS50
and MS100 concretes show slightly denser microstructure than natural sand concrete.
These contain more paste volume due to presence of fines. Ca/Si ratio was found to be
1.80 for control concrete. Ca/Si ratio for MS 50 and MS100 concrete are 2.5 and 2.3
respectively, which are in line with the strength of corresponding concretes. Denser
structure, better particle packing and better bonding of M-sand particle with cement
paste is responsible for enhanced strength properties.
942
7
(a) SEM for NS (b) EDS image of NS
(c) SEM for MS50 (d) EDS Image of MS50
(e) SEM for MS100 (f) EDS Image of MS100

Figure 3: Sem and EDS image of concrete
943
8
The variation of normalized compressive strain and normalized tensile strain for RC
beams with 0, 50 and 100% replacement of M-sand (NS, MS50 and MS100
respectively) are given in Figures 4 and 5 respectively. Load deflection curve for all the
beams are shown in Figure 6. Maximum compressive stress were observed as 17608,
17457, 19923 kN/m2 and corresponding strains were 0.0017, 0.0018 and 0.0018
respectively for NS, MS50 and MS100. Maximum tensile stress observed were 18216,
18210, and 19923 kN/m2 and corresponding strains were 0.0192, 0.0258 and 0.0208
respectively for three types of concretes. Stress and strain values are in good agreement
with values specified by IS: 456-2000[20].
25000
20000
Stress (kN/m2)
15000
NS
10000 MS50
5000 MS100
0
0 0.001 0.002
Strain
Figure 4. Variation of Compressive stress with Compressive strain
25000
20000
Stress(kN/m2)
15000 NS
10000 MS50
5000 MS100
0
0 0.01 0.02 0.03
Strain
Figure 5. Variation of Tensile stress with tensile strain
944
9
120
100
80
Load(kN)
60 NS
MS50
40
MS100
20
0
0 5 10 15 20 25
Deflection(mm)
Figure 6. Load deflection curve at midspan

Maximum midspan deflection recorded were 15.99, 18.76 and 19.089mm respectively
in NS, MS50 and MS100 beams at 92, 98 and 103kN loads. It is evident that load
carrying capacity and ductility of beams with M-sand is more as compared to natural
sand beams. First crack load, deflection at first crack, Ultimate load, Ultimate deflection,
Energy absorption capacity, Ductility factor and moment carrying capacity are as
indicated in Table 3 for various beams.
Table 3. Test results for RC beams
RC beam type NS MS50 MS100

2
Compressive strength of concrete (N/mm ) 32.33 37.11 36.17
First crack load(kN) 44.5 41.4 43.5
Deflection at first crack(mm) 3.0 2.6 2.7
Ultimate load (kN) 95.1 98.1 103.2
Ultimate Deflection (mm) 16.1 18.7 19.1
Energy Absorption Capacity(kN/mm) 1159.8 1463.1 1542.4
Ductility Factor 2.2 2.8 2.9
Ultimate Moment carrying Capacity (kN-m)- 20.7 21.4 22.5
Experimental
Ultimate Moment carrying Capacity (kN-m)- 16.5 16.6 16.6
Theoretical as per IS:456-2000
945
10
The compressive strength of concrete increases with increase in percentage of M-sand

replacement. Compressive strength of MS50 and MS100 concrete are 14.78 and 11.87
% more respectively as compared to NS concrete. This may be due to more angular
nature of M-sand particles and dense microstructure of M-sand concrete. First crack in
M-sand beams were noted at a lower loads than natural sand concrete because of
slightly brittle nature of M-sand concrete; but ultimate load carrying capacity is higher.
Deflection at first crack are slightly lower in M-sand concrete beams due to lesser first
crack load. 3.2 and 8.5% enhancement can be seen in ultimate load carrying capacity
of MS50 and MS100 beams. Corresponding deflections are 16.3 and 18.3% more as
compared to NS beams. Energy absorption capacity of MS50 and MS 100 beams are
26.15 and 32.98% more as compared to NS beams. Higher moment carrying capacity
by 3.38 and 8.70% can be seen in MS50 and MS100 beams. Enhanced ductility factors
of 2.8 and 2.9 are obtained in MS50 and MS100 as compared to 2.2 in NS beams.
Experimentally obtained moment carrying capacity are cross checked by computing
Ultimate moment of resistance of doubly reinforced beams as per IS: 456-2000 [Ref].
Good correlation between experimental and analytical results with respect to Ultimate
moment carrying capacity can be observed. Due to superior properties of M-sand
concrete and similar behaviour to natural sand concrete, M-sand concrete can be
recommended in reinforced concrete beams.
6 Conclusions
 Physical Properties of M-sand grains obtained from VSI crushing is similar to
that of natural sand; but they are slightly angular or cubical as compared to
spherical grains of natural sand. M-sand is well graded and belongs to Zone II
as per IS:383-2016 [4]
 M-sand concrete is less workable compared to natural sand concrete and
increased dosage of admixtures is essential for obtaining required workability.
 Compressive strength of concrete is more at 50 and 100% replacement of M-
sand. Partial replacement of natural sand by 50% M-sand gives optimum
values.
 M-sand beams exhibit higher Ultimate load, energy absorption capacity,
ductility factor, and moment carrying capacity as compared to natural sand
beams. Hence use of M-sand turns out to be an economical and safe alternate
material for natural sand in construction of RC members.
Acknowledgment. Authors wish to acknowledge authorities of Rashtreeya Vidyalaya

College of Engineering, Bengaluru, India for providing facilities to carry out the
experimental works.
946
11
References
1. V. Syam Prakash, Ready Mixed Concrete Using Manufactured Sand as Fine Aggregate,
Proceedings of 32nd Conference on Our World in Concrete & Structures: 28 – 29 August
2007, Singapore.
2. Tao Ji, Cai-Yi Chen., et al, A Mix proportion design method of manufactured sand concrete
based on minimum paste theory, Construction and Building Materials, Vol 44, (2013), 422-
426.
3. Sanjay Mundra, P.R. Sindhi., et al, Crushed Rock Sand - An Economical and Ecological
Alternative to Natural Sand to Optimize Concrete Mix. Perspectives in Science, Vol 8,
(2016), 345-347.
4. IS: 383-2016, Specification for coarse and fine aggregates, Bureau of Indian Standards, New
Delhi, India.
5. Fuhai Li, Jinliang Gao，Guhua Li ,Chuan Yu, Influence of manufactured sand
grading to workability of concrete, Applied Mechanics and Materials, Volume 174,
2012, pp 545-550.
6. Venkatarama Reddy B V, Suitability of Manufactured sand (M-sand) as fine
aggregate in Mortar and concrete”, CSIC Project report, January 2012, IISc,
Bangalore.
7. Li Beixing ,Ke Guoju, Zhou Mingkai , Influence of manufactured sand
characteristics on strength and abrasion resistance of pavement cement concrete ,
Construction and Building Materials, Volume 25,2011, pp 3849–3853.
8. O. A. Cabrera , L. P. Traversa , N. F. Ortega , Effect of crushed sand on mortar and
concrete rheology, Materiales de Construccion, Vol 61,303,2011, pp 401-416.
9. Rolands Cepuritis , Stefan Jacobsen , Bard Pedersen , Ernst Mortsell, Crushed sand
in concrete e Effect of particle shape in different fractions and filler properties on
rheology, Cement and Concrete Composites, Volume 71, 2016, pp 26-41.
10. Xinxin Ding , Changyong Li , Yangyang Xu , Fenglan Li , Shunbo Zhao,
Experimental study on long-term compressive strength of concrete with
manufactured sand , Construction and Building Materials 108,2016, pp 67–73.
11. Shunbo Zhao, Xinxin Ding , Mingshuang Zhao, Changyong Li , Songwei Pei,
“Experimental study on tensile strength development of concrete with
manufactured sand”, Construction and Building Materials 138, 2017,pp 247–253.
12. Weiguo Shen, Yi Liu , Zhongwen Wang , Lianghong Cao , Dinglue Wu, Yujie
Wang , Xiaoli Ji , Influence of manufactured sand’s characteristics on its concrete
performance, Construction and Building Materials 172, 2018, pp 574–583.
13. S. Krishna Rao, P. Sravana, T. Chandrasekhar Rao, “Investigating the effect of M-
sand on abrasion resistance of Roller Compacted Concrete containing GGBS”,
Construction and Building Materials 122, 2016, pp 191–201.
14. Prakash Nanthagopalan, Manu Santhanam, Fresh and hardened properties of self-
compacting concrete produced with manufactured sand, Journal of Cement and
Concrete Composites, Vol. 33, 2011, pp 353–338.
15. Praveen Kumar K, Radhakrishna, Characteristics of SCC with Fly Ash and
Manufactured Sand, IOP Conf. Series: Materials Science and Engineering 149
(2016) 012111 doi:10.1088/1757-899X/149/1/012111
947
12
16. Abhijit Mukherjee, Gopal L. Rai, Performance of reinforced concrete beams

externally prestressed with fiber composites, Construction and Building Materials,
23, 2009, pp 822–828.
17. I. H. Yang, C. Joh, and B. S. Kim, Flexural strength of ultra-high strength concrete
beams reinforced with steel fibers , Procedia Engineering 14, 2011, pp793–796.
18. Said M. Allam, Mohie S. Shoukry, Gehad E. Rashad, Amal S. Hassan, Crack width
evaluation for flexural RC members, Alexandria Engineering Journal 51, 2012, pp
211–220.
19. IS: 12269-2013, Ordinary Portland cement 53 grade-specification, Bureau of
Indian standards, New Delhi, India.
20. IS: 456-2000, Code of practice for plain and reinforced cement concrete, Bureau
of Indian standards, New Delhi, India.
21. BS: 5075- 3:1985, Concrete admixtures. Specification for super plasticizing
admixtures, British Standards.
22. IS 9103-1999, Concrete admixtures- specification, Bureau of Indian standards,
New Delhi, India.
23. IS: 10262-2009, Concrete Mix Proportioning- Guidelines, Bureau of Indian
standards, New Delhi, India.
24. IS:1786:-2008, High strength deformed steel bars and wires for concrete
reinforcement, Bureau of Indian standards, New Delhi, India.
25. IS: 516-1959, Method of tests for strength of concrete, Bureau of Indian standards,
New Delhi, India.
948
Paper ID- 177
LOAD-MOMENT INTERACTION BEHAVIOUR OF

AXIALLY LOADED SHORT COMPRESSION
MEMBERS
Biswajit Samal 1, Shashwati Soumya Pradhan2
1
Ph.D Research Scholar, National Institute of Technology, Silchar, Assam, India, Tel: +91-
9778104234, e-mail: biswajitsamal777@gmail.com
2
Faculty, Centre for Advanced Post Graduate Studies, BPUT, Rourkela, Odisha, India, Tel:
+91-9439351426, e-mail: saswatipradhan09@gmail.com
ABSTRACT:-Every column is subjected to bending in addition to axial forces and they need
to be proportioned to resist both. Columns will bend under the action of moments and those
moments will give Compression on one side of the columns and tension on another side.
Depending upon the relative Magnitudes of the axial loads & moments, section failure may
occur. The capacity of a short column is the same as the capacity of its section under primary
stresses, irrespective of its length.The design & behaviour of short compression members
subject to axial compression members combined with uniaxial bending i.e. bending with
respect to either major or minor axis, the loading condition is statically equivalent to a condition
of uniaxial eccentric compression where in the factored axial load Pu is applied at an
eccentricity e = Mu/Pu with respect to the centroidal axis. In this study the design strength of
different column section are analysed as well as load-moment interaction curve is carried out
for the column section corresponding to the neutral axis. Non- Dimensional interaction
diagrams are also carried out as per Indian standard codes. By using SCI-LAB programming
various non- dimensional interaction curves are plotted with all grades of steel i.e. Fe 250, Fe
415 & Fe 500 with different d’/D values.
Keywords: Load-Moment Interaction curve, Short column, Neutral-Axis, Compression

Members, Uniaxial Bending.
1 INTRODUCTION
Short R.C.C. columns are under axial load & uniaxial bending acts in a different manner that
when it is subjected to axial load, however columns subjected to axial load can also carry some
moment that may appear during construction. The behavior of these columns and the three
modes of failure are explained. Hence, the needed interaction diagram of columns, plot of P &
M, taking different positions of neutral axis, either outside or within the cross-section of the
column. Depending on the position of the neutral axis, the column might or might not have
tensile stress to be taken by longitudinal steel. In the compression region however, longitudinal
steel will carry the compression load along with the concrete as in the case of axially loaded
short column.
949
For rectangular and square columns having reinforcement placed on the end faces only,
the reinforcement is assumed to consist of two equal thin strips parallel to the compression face
of the section. For rectangular and square columns having steel reinforcement equally
distributed along all four faces of the section, the reinforcement was considered to consist of a
thin rectangular or square For square and circular sections having steel bars arranged in a circle,
the reinforcement was considered to be a thin circular tube.
Various columns are subjected to bending in one direction, though subjected to

relatively small bending moment in orthogonal direction. These columns are designed by using
the interaction diagrams and checked for the adequacy of capacity in the orthogonal direction.
However, some columns, in the case of corner columns, are subjected to equally significant
bending moments in two orthogonal directions.
2 MODE OF FAILURES OF COLUMN
Generally, there are two types of failure modes, (i)Failure due to compression or Compression
failure, when N.A lies outside, as a result of which creates compression throughout the section,
and (ii) Failure due to tension or Tension failure, here N.A lies within the section developing
tensile strain on the left side of the N.A. With compression failure & tension failure, another
mode of failure is discussed below i.e. balanced failure.
2.1 BALANCED FAILURE
Yielding generally occurs at left side of outermost longitudinal row in this type of failure. Here,
maximum compressive strain of concrete reaches 0.0035. As per IS:456:2000 the yield strains
of different type of reinforcement are given below:-
0.87𝑓𝑦
 For Mild Steel (Fe 250) 𝜀𝑦 = 𝐸𝑠
0.87𝑓𝑦
 For High yield strength deformed bar (Fe 450 & Fe 500) 𝜀𝑦 = + 0.002
𝐸𝑠
2.2 COMPRESSION FAILURE
When eccentricity of loading results a small value corresponding to balance failure, yielding is
not occur in longitudinal reinforcement & failure at compression edge of column occurs due to
crushing of concrete. Depending upon the grades of steel & its proximity the compression
reinforcement may or may not be yield.
950
2.3 TENSION FAILURE
When eccentricity of loading results a higher value than that of corresponding balanced failure
condition, failure will occur by yielding of tension reinforcement. Initially yielding occur at
the outermost longitudinal reinforcement in the tension side of the N.A & successive inner
rows on the tension side of the N.A, may leads to increase in strain. Ultimately, collapse occurs
at the compression edge due to crushing of concrete.
3 INTERACTION DIAGRAM
( Fig 1 : A typical load-moment interaction diagram)
A R.C.C column with specific number of longitudinal reinforcement has different carrying
capacities of of Pu and Mu before its failure depend on the eccentricity of the load. Figure 1
represents interaction diagram which gives carrying capacities ranging from P0 with zero
eccentricity on the vertical axis to M0 in the horizontal axis. The vertical axis relates to load
with zero eccentricity while the horizontal axis represents infinite value of eccentricity. A radial
line joining the point O of Fig.1 to point 2 imparts the load with the minimum eccentricity. The
radial line represents a particular eccentricity of the loading.Each point on the interaction
diagram imparts a unique pair of Pu and Mu causing the state of incipient failure. It has three
zones of failure: (a) from point 1 to point 5 is compression failure zone, (b) point 5 is balanced
failure and (c) from point 5 to point 6 is tension failure zone. Large eccentricity causes failure
in tension reinforcement by yielding while small eccentricity causes failure in compression.
Within point 5, the critical point at which compression & yielding occurs concrete & steel
simultaneously.
951
4 PREPARATION OF DESIGN CHARTS
Design charts are prepared using the equations by taking N.A within & outside of the section.
The actual grade of concrete is to be taken into account, however ,it may not be workable to
follow this properly. The procedure of preparation these design charts are explained below. It
is significant that the values of (Pu/fckbD) and (Mu/fckbD2) are determined by taking different
locations of the N.A for the cases given below.
(P /f bD)=0.4+(p/100f ) (0.67f 0.4f )………………………………………………..(1)

u ck ck y ck
The Pu & Muobtained from Eq.10.51 c as per cl.39.3 of IS 456:2000. From the condition
of cl. 39.3 of IS 456 and considering minimum eccentricity as 0.05D, we have
2 2
M = (P ) (0.05)D = 0.02 f bD + (0.05 pbD /100) (0.67 f - 0.4 f ) ………….(2)
u u ck y ck
2
Divide both sides of the above equation by fckbD , we have
2
(M /f bD ) = 0.02 + (0.05p/100 f ) (0.67 f - 0.4 f )…………………………...(3)
u ck ck y ck
4.1 COMPUTER PROGRAMMING BY USING SCI-LAB
The different computer programs by using SCI-LAB are analyzed by taking various numerical
examples of short column design. The rectangular sections with reinforcement distributed
equally on two sides as per SP:16 (Charts 27-38) are plotted. The two sides refer to the sides
parallel to axis of bending which means there are no inner rows of bars & each outer row has
an area of 0.5As. Corresponding to this, there are as many as 12 no. of charts available, covering
3 grades of steel (Fe 250, Fe 415, Fe 500) with 4 values of dˈ/D ratio for each grade (i.e. d'/D=
0.05, 0.10, 0.15, 0.20) are also plotted. Each of the 12 charts covers a non-dimensional design
interaction curves with p/fck values ranging from 0 to 0.26 are given below.
(Graph 1: Column non-dimensional interaction curve for Fe 250, d'/D=0.05 & 0.10)
952
(Graph 3: column non-dimensional interaction curve for Fe 415, d'/D=0.05 & 0.10)
953
4.2 COMPARISION WITH SP-16 CODE BY USING PLOT DIGITIZATION

SOFTWARE
As per SCI-LAB programming the various non- dimensional interaction curves are plotted with
all grades of steel i.e. Fe 250, Fe 415 & Fe 500 with different d’/D values. But the charts
available in SP-16 code i.e. very close enough with the numerical values obtained from the
short column design by using PLOT DIGITIZATION software. There are other situations,
encountered in practice, which are not amenable for the use of SP-16 charts. This includes case
of unsymmetrically arranged reinforcement in rectangular sections & non-rectangular sections
such as L-shaped, T-shaped, H-shaped etc. In such cases it becomes necessary to construct
proper interaction diagrams in order to obtain accurate & reliable solutions. The final data
points obtained with respect to SP-16 code are given below.
954
(Graph 7: Data plots as per SP-16 compared with design values,

Fe 250,d’/D=0.05)

Fe 415,dˈ/D=0.05)
955

Fe 500,d'/D=0.05)
The non-dimensional interaction curves are plotted by using SCI-LAB programming & PLOT
DIGITIZATION software. The above points are plotted i.e. very close enough to design charts
of SP-16 code (Chart 27-38, reinforcement distributed equally on two sides, dˈ/D=0.05, 0.10,
0.15, 0.20).
5 CONCLUSION
From the present investigation it can be concluded that the non- dimensional interaction curves
plotted by using various computer programming & software & design charts obtained as per
SP-16 code are approximately same. Design of a member subject to combined axial load &
uniaxial bending will involve extensive calculations, the non-dimensional interaction curves
can used to overcome that difficulty. There are some other situations, encountered in practice,
which are not amenable for the use of interaction curves. It includes when reinforcement
arranged symmetrically in rectangular sections, non- rectangular & circular section.
956
6 REFERENCES
[1] Kent and Park.R, (1971), ―Flexural members with confined concrete, Journal of
Structural division, ASCE, V.97, ST-7, July 1971.
[2] Mander et al. (1982). "Stress-strain behaviour of concrete confined by overlapping

hoops at low and high strain rates." J. American Concrete Institute, 79.
[3] Hong, K. N., and Han (1985), High strength concrete–material properties and structural
behaviour, HSC. ACI SP-87, American Concrete Institute, Detroit.
[4] J. Kent Hsiao, Priestley M.J.N., Park R. (1987), ―Theoretical Stress- Strain Model for
Confined Concrete‖, Journal of Structural Engineering, publisher: ASCE,
http://www.asce.org/.
[5] Jaswant N. Arlekar and C. V. R. Murty (1998), Stress-Strain Model of High- Strength
Concrete Confined by Rectangular Ties‘, Journal of Structural Engineering, KSCE, V.
9, No. 3, 2005, pp. 225-232.
[6] Jiang and Teng (1999), Effects of Confinement on Interaction Diagrams of Square, vol
1, 2009.
[7] T. Jiang & J. G. Teng (2001), Reinforced Concrete Columns‘, Civil Engineering
Dimension, Vol. 11, No. 2, September 2009, 78-88.
[8] Krishnan, Swaminathan “Seismic Behaviour of Beam Column Joints in Moment

Resisting Reinforced Concrete Frame Structures,” submitted to Indian Concrete
Journal, October 2004.
[9] H. Akbarzadeh Bengar, A. A. Maghsoudi (2002) “pushover analysis of a 19 story

concrete shear wall building” submitted to 13th world conference on Earthquake
Engineering (2004).
[10] V.M. Shinde & J. P. Bhusari (1988), ―Theoretical Stress- Strain Model for Confined
Concrete‖, Journal of Structural Engineering, publisher: ASCE, http://www.asce.org/.
[11] Cheol-Ho Lee, Thomas H.-K. Kang (2007), Effects of Confinement on Interaction
Diagrams of Square Reinforced Concrete Columns‘, Civil Engineering Dimension,
Vol. 11, No. 2, September 2009, 78-88.
957
Paper ID- 178
“A study on Moment– Curvature relationships for REINFORCED CONCRETE

BEAMS with varying fire loading conditions”
Borgohain, Ankit 1, Bhattacharyya, Sriman Kumar2
1
U.G.Student, Department of Civil Engineering, NIT Silchar , Silchar,788010, India.
2
Professor, Department of Civil Engineering, IIT Kharagpur, Kharagpur , WestBengal, 721302,India.
ABSTRACT
In this paper a numerical study by three-dimensional modeling is done for reinforced concrete beam exposed to fire
conditions with various boundary conditions. Both geometric and material properties are taken into account in this
formulation because of the changes in material properties and the large variations experienced in fire,. The three stages
associated with the numerical procedure for evaluating fire resistance of RC beams; namely fire temperature calculation,
thermal analysis and strength analysis are modeled using a finite element (FE) model. The FE model is prepared in
ABAQUS software to study the response of an RCC beam under fire during loading conditions. The RCC beams are
analysed for different end conditions and fire loading surfaces. Initially it is kept at 25°C .Then the uniform loading of
one–third of the compressive strength of concrete is applied as a pressure loading in the beam (3D solid element).The other
faces of RCC beam have temperature boundary conditions. The temperature boundary condition is governed by
temperature –time curve as per ASTM E119. The moment-curvature (M–k) relationships are developed for the beam at
few critical positions of the RCC beam for various end conditions as well as temperature boundary conditions. These M-k
relationships may be used to carry out the strength analysis of the beam member and draw significant conclusions.
Keywords or phrases: M–k relationships, fire temperature calculation, thermal analysis
1. INTRODUCTION
2. LITERATURE REVIEW
Fire safety of structures has been of tremendous interest
in last few decades particularly after the September 11 Kodur V. et al.(1). Prepared batches of concrete with
attacks on Twin Towers, New York in 2001. This has led NSC , HSC(siliceous aggregate) , HSC ( carbonaceous
to varied research in the field of evaluation of response of aggregate),HSC(steel reinforcement),
building and other structural systems under fire. Fire is HSC(polypropylene) and found that deformation in beams
one of the most severe risks to buildings and structures. A are due to 3 factors namely load, thermal expansion and
limited range of relevant research has been reported creep. NSC has a higher deformation rate than HSC due to
previously, especially for high performance concrete. In the fact that HSC has higher young’s modulus. Carbonate
this paper a numerical study by three-dimensional aggregate has less spalling than siliceous aggregates due
modeling is developed for reinforced concrete beam to its higher specific heat value.
exposed to fire conditions. Fire is one of the most severe A. Lau (2) et. Al. studied concrete after being subjected to
design situations as it not only affects the strength of different elevated heating temperatures, ranging between
concrete, but also the structural stiffness and stability. A 105 °C and 1200 °C. The compressive strength, flexural
reinforced concrete beam, compared to other structural strength, elastic modulus and porosity of concrete
members, has most often to cope with vertical forces and reinforced with 1% steel fiber (SFRC) and changes of
bending moments. In addition, the use of high strength color to the heated concrete etc were studied . For
concrete (HSC) is becoming more popular due to the maximum exposure temperatures below 400 °C, the loss
improvements in structural performance such as higher in compressive strength was relatively small. Significant
strength and durability that it can provide compared to reductions in compressive strength are observed when
conventional Normal Strength Concrete (NSC). This paper temperature exceeds 400 °C. When steel fibers are
presents the development of a computer model for incorporated at 1%, an improvement of fire resistance and
predicting the fire behavior of RC beams under realistic crack development was observed.
fire loading scenarios. The model is based on a Jae H. Chung (3) developed a finite difference model that
macroscopic finite element approach and a series of simulates coupled heat and mass transport phenomena in
moment–curvature relationships for tracing the response reinforced concrete structures exposed to rapid heating
of the beam. conditions such as fires. A mathematical and
computational model for simulating the multidimensional,
958
thermo-hydrological response of reinforced concrete The Thermal conductivity Vs temperature curve is

structural elements is developed and used to study the presented in Fig 1.
effects of steel reinforcement on thermodynamic state
variables. The effects of steel reinforcement on heat and 2
mass transfer in the surface region of an R/C beam are
Thermal conductivity
found.
1.5
(W/m-K)
1
3. EXPERIMENTAL PROGRAM
A reinforced concrete beam of 3-meter length and cross- 0.5
sectional area 250mm×350 mm used for the study. An
attempt is made to compute the moment vs. curvature 0
relationship at various locations of the beam when the fire 0 200 400 600 800 1000 1200
Temperature (0C)
loading is applied in various faces of beam. In the present
study a roller–pinned support condition is considered and
is subjected to a constant pressure load of 1.333E7 N/m2. Fig. 1:Thermal conductivity as a function of
temperature (ENV 1992-1-2, 1995)
The beam is exposed to ASTM E119 standard fire for
different exposure times at different faces. M40 grade of 3.3 Specific heat
concrete and Fe 415 grade of steel have been used and The specific heat is a function of temperature
geometric parameters of considered beam are described in and according to Euro code, it is valid to both siliceous
given in Table-1. and calcareous concrete (ENV 1992-1-2, 1995). The
Table 1. Geometric parameter of RCC beam Specific heat Vs temperature curve is presented in Fig 2
Beam length (mm) 3000

Concrete cross-section (mm2) 350×250
Reinforcement diameter 20
(mm)
Percentage reinforcement 1.795 %
Transverse reinforcement 12-8 dia. @250 mm c/c
(mm)
Concrete cover (mm) 40
Support conditions roller-pinned
Fire exposure ASTM E119 Standard
method fire
Exposure time in hours 0.5,1, 2, 3, 3.5 Fig.2. Specific heat as a function of temperature for ordinary
siliceous concrete (ENV 1992-1-2, 1995)
3.1 Reinforced concrete beam model
3.4 Thermal properties of steel
For the numerical analysis, on three faces of the referred Steel is considered an isotropic material in
RCC beam, model temperature boundary conditions are temperature calculations and its thermal properties
applied. Investigations done for the temperature, stress, could be described by three different material
strain, curvature distribution pattern as field output and at properties: thermal conductivity, specific heat and
specific points as history output to characterize the beam thermal expansion co-eff and respective graphs are
exposed to fire. The number of element created were a) presented in Fig 3,4 and 5.
Concrete beam block 264 elements; b) Longitudinal
reinforcement bar 33 elements . 60
3.2 Thermal conductivity 50
Thermal conductivity
The thermal conductivity for siliceous 40

concrete is a function of temperature and is expressed 30
(W/m-K)
by equation (ENV 1992-1-2, 1995). Upper limit of 20

thermal conductivity,
10
λc=2-0.2451(θ/100) +0.0107(θ/100)2 W/mK
0
for 20 C ≤ θ ≤ 1200 C0 0
lower limit of thermal conductivity 0 500 1000 1500
λc=1.36-0.136(θ/100) +0.0057(θ/100)2 W/mK Temperature (C)
for 200C ≤ θ ≤ 12000C Fig.3.Thermal conductivity as a function of temperature for steel
(ENV 1993-1-2, 1995)
where λc is thermal conductivity in [W/m-K]
θ is the concrete temperature in [0C]
959
Specific heat of steel is a function of 4. RESULTS AND DISCUSSION

temperature and expressed by the graph below(ENV
1993-1-2, 1995). 4.1 Overview of the analysis.
It is observed that initially the reinforced
3000
concrete beam is experienced a simultaneous
contraction (at top fibres) and expansion (at bottom
2500
Specific heat (J/kg-K)
fibres) due to constant pressure load at the topmost

2000 surface. But when it is exposed to fire it experienced an
1500 expansion in elements and strain is developed. Due to
1000 the heating , curvature of the entire beam configurations
500 changes and varies accordingly with time. Due to the
0
cost of computation only data for some selected
element of the beam is calculated such as 16(mid span),
0 200 400 600 800 1000 1200 8,25(quarter span) etc. Fig 7 shows A reinforced
Temperature (0C)
concrete beam of 3-meter length subjected to a constant
Fig.4. Specific heat as a function of temperature for steel (ENV
pressure load of 13333.33 kN/m2 on the X-Z surface
1993-1-2, 1995) and exposed to fire according to the fire curve ASTM
E119 Standard fire (Transient ) with varying exposure
Density of steel and ordinary Siliceous condition to fire. The three other surfaces of the beam
concrete are considered a constant value 7850 kg/m3 are kept at a constant temperature at 25°C.
and 2400 kg/m3 respectively.
0.00002
thermal expansion
0.000015
Coefficient of
0.00001
(/0C)
0.000005
0
0 500 1000 1500
Temperature (0C)
Fig.7 Fire load applied on three faces
Fig.5 Thermal expansion co-efficient Vs temperature
3.5 Fire exposure methods FIRE LOADING ON ONE FACE (SINGLE ROOM
In this analysis ASTM E119 Standard fire UNDER FIRE)
method is considered.
1500 I. ELEMENT 8 (ONE-FOURTH OF LENGTH)
Temperature (0C)
1000
0.00E+00
2 hours
500 -0.3 -0.2 -0.1 0
Section moments(Nm)
-5.00E+05
2.5
0 hour
-1.00E+06
0 50 Time (minute)
100 150 200 3 hours
-1.50E+06
1 hour
Fig.6: Temperature-time curve for the standard fire (ASTM
E119, 2007) -2.00E+06
30
The temperature-time relationship on the boundary minutes
-2.50E+06
member is defined by equation. Curvature (1/m)
Tf = T0 + 750(1- exp(-3.79553th1/2)) + 170.41th1/2 Fig 8: Moment vs curvature relations for different durations of
th, time (h); fire loading for element 8.
T0, initial temperature (° C); and
Tf, fire temperature (° C)
960
II. ELEMENT 16 (MID POINT) Fig11: Deflections in Y direction under influence of pressure
loading and fire loading for 30 minutes.
0.00E+00 2
-0.3 -0.2 -0.1 0 hours
-5.00E+05
2.5
-1.00E+06 hour
3
-1.50E+06 hours
Fig12 : Curvature vs Temperature for element 8 for a fire
-2.00E+06 1 hour loading duration of 30 minutes.
-2.50E+06
Fig 9: Moment vs curvature relations for different durations of
fire loading for element 16.
III. ELEMENT 25 (THREE- FOURTH OF

LENGTH)
0.00E+00 Fig 13: Curvature vs Temperature for element 16 for a fire

-0.08 -0.06 -0.04 -0.02 0 loading duration of 30 minutes.
-5.00E+04
-1.00E+05 30 min
1 hour
-1.50E+05
2 hour
Fig 14 : Curvature versus Temperature for element 25 for a fire
loading duration of 30 minutes.
-2.00E+05 3 hour Results after 180 minutes of fire loading:
2.5
-2.50E+05 hour
-3.00E+05
Fig 10: Moment vs curvature relations for different durations

of fire loading for element 25.
Fig 15: Deflections in Y direction under influence of pressure
FIRE LOADING ON TWO OPPOSITE FACES loading and fire loading for 180 minutes.
(OPPOSITE ROOMS UNDER FIRE)
Results after 30 minutes of fire loading:
Fig 16: Curvature vs Temperature for element 8 for a fire

loading duration of 180 minutes
961
Fig 17: Curvature vs Temperature for element 16 for a fire Fig 20: Curvature vs temperature plot for the central element of
loading duration of 180 minutes the RCC beam (wire-beam element 16)
4.3 Results for different durations (1, 2, 3, 3.5 hours) of

fire loading:
Analysis has been done for 5 different conditions of fire

loading and deflected shape for all 5 cases have been
extracted. Also curvature Vs temperature curves for all
5 cases for different elements have been extracted.
Fig 18: Curvature vs Temperature for element 25 for a fire

loading duration of 180 minutes
FIRE LOADING ON THREE FAC ES (OVERHEAD

BEAM IN A ROOM UNDER FIRE)
Results after 30 minutes of fire loading:

Fig 21: Displacement U2 (along global Y) contour for entire
beam at the end of 2 hour the heating step.
The Curvature Vs Temperature for different exposure

durations have been presented in fig 11,12 and 13.
From the curvature data, Moment curvature relationship
have been obtained and presented in Fig
Fig 19: Displacement U2 (along global Y) contour
Initially the curvature is constant during the analysis of the

loading step as there is no temperature change during the
loading step. Temperature changes monotonically only in Fig 22: Curvature vs Temperature plot for 120 minute
the fire load step and both time and curvature magnitude
keeps on increasing until heating stops. This is a condition
which is observed at all the plots for curvature vs
temperature as the static loading step is same in all the
cases. Different durations for fire loading exposure for
0.5hr,1hr,2hr,3hr and 3.3 hr have been tried for. Maximum
absolute value for U2 is 2.425e-1 m at element 1 solid
beam part. The Curvature Vs temperature plot for wire
beam element for 30 minute fire exposure has been
presented in fig 9
962
Fig 23: Curvature vs Temperature plot for element no. 16 (mid- From the moment Curvature relationships obtained,
point element) of the central embedded wire-beam element for a the values of respective moments have been
duration of 3 hours.
presented in table 2.
Table 2. Moment and curvatures for salient points

for different fire exposures
Location 1 Location 2 Location 3
(Element 8) (Element 16) (Element 25)
Sl Mom Mom Mom
Expos Curva Curva Curva
. ent ent ent
ure ture ture ture
N (kNm (kNm (kNm
time (/ km) (/ km) (/ km)
o. ) ) )
30 1664. 99.267 1936. 115.53 148.2 8.8446
1 minut 293 51 96 08 866 18
es
Fig 24: Curvature vs Temperature plot for element no. 8 (one 60 1664. 99.265 1903. 113.50 149.7 8.9290
fourth length element) of the central embedded wire-beam 2 minut 266 9 005 56 022 52
element for a duration of 3 hours. es
120 1695. 101.14 1903. 113.56 151.2 9.0221
3 minut 789 61 984 4 628 37
es
180 1664. 99.267 1902. 113.50 151.0 9.0067
4 minut 3 94 958 28 055 94
es
210 1.66E 99.265 1850. 110.35 8.9322
149.7
5 minut +03 99 176 46 52
558
es
Fig 25: Displacement U2 (along global Y) contour for entire

beam at the end of the heating step. TABLE 3: Moment versus curvature values for
element 8 for different exposure conditions and fire
loading durations.
Element 8 Face 1 Face 3

Sl. Mome Curvatu Mome Curvatu
Exposu
No nt re nt re
re time
. (km) (/ m) (kNm) (/ m)
1664.2 .099.267
30 1669.2 0.15545
1 93 51
minutes 64
Fig 26: Curvature vs Temperature plot for element no. 8 (one 1664.2 .099265
fourth element) of the central embedded wire-beam element for a 60 1664.2
2 0.1555 66 9
duration of 3 hours 30 minutes. minutes 54
1695.7
The moment curvature relationship for all durations 120 1681.7 .1011461
3 0.10065 89
were extracted. Fig 16 shows moment Vs curvature minutes 89
for 2 hr duration
150 1692.4
4 0.15814 NIL NIL
2500 minutes 04
2000 -
1664.3 .099267
180 1679.6
1500 5 0.10053 00 94
minutes 42
Moment(kNm)
1000
500
TABLE 4: Moment versus curvature values for
0 element 16 for different exposure conditions and fire
0 100 200 300 400 loading durations.
Curvature(1/km)
Element 16 Face 1 Face 3
Fig 27: Moment vs curvature for element 16 (2hr duration) Sl. Exposu Mome Curvatu Mome Curvatu
N re time nt re nt re
963
o. (kNm) (/ m) (kNm) (/ m) gradual .Hence the effect in the moment vs curvature

curve is very less making the difference between all
1890.4
1936.9 0.11553
30 69 0.11314 the curves seemingly smaller.
1 6 08
minutes
5. The curve for lower time of exposure lies above the
1903.0 0.11350 curve for higher time of exposure.

60 1903.4 0.11353
2 05 56 6. After exposure to fire for 120 minutes the peak
minutes 81
1903.9 0.11356 Moment values and corresponding curvatures are
120 1930.1 0.11513
3 84 4 reached for all the three locations elements 8, 16, 25
minutes 63
respectively in Table 2 .
-
150 1939.2 0.11567
4 NIL NIL
minutes 87
- 6. ACKNOWLEDGMENT
1902.9 0.11350
180 1927.7 0.11498
5 58 28
minutes 47
I would like to express my special thanks to Prof.
Sriman Kumar Bhattacharyya, IIT Kharagpur for his
excellent guidance and encouragement.
I would also like to thank Prof. Arghya Deb, IIT
The moment–curvature curve at any position in a beam
Kharagpur for his valuable suggestions. I express my
has coincidental curves (at same locations for various
sincere gratitude to Indian Academy of Sciences for
times of exposures to fire). It is so because the loading
step is same for all the different conditions and time of offering me this opportunity.
exposure to fire.
7. REFERENCES
5. CONCLUSION 1. Kodur VKR, Cheng F-P, Wang T-C, Sultan MA(2003). Effect
of Strength and Fiber Reinforcement on Fire Resistance of High-
Strength Concrete Beams. J Struct Eng [Internet].
1. The moment-curvature values for one face exposure 2003;129(2):253–9.
is more than the moment–curvature values for three 2. Lau A, Anson M.(2006) Effect of high temperatures on high
performance steel fibre reinforced concrete. Cem Concr Res.
faces of exposure to fire. 2006;36(9):1698–707.
3. Chung JH, Consolazio GR.(2005) Numerical modeling of
2. The maximum moment in moment curvature curve is transport phenomena in reinforced concrete exposed to elevated
at location 2( element 16) at time 150 minutes and is temperatures. Cem Concr Res. 2005;35(3):597–608.
of magnitude 1939.257 kNm at curvature of 0.1156 m

-1.
for one sided exposure. The next highest moment
in moment curvature curve is at location 1( element
16) at time 120 minutes and is of magnitude 1936.96
kNm at curvature of 0.11553 m -1. (TABLE 4 ).
3. The moments at location 3 are very small in
comparison. It is due to the non uniform boundary
conditions applied at the cross-sections: hinge at one
cross-section and a roller with constraints in any one
direction in another direction.(TABLE 2)
4. The difference in the curves for Moment vs curvature
for different times of exposure is very small due to the
fact that after 30 minutes of fire exposure in fire
loading step the increase in temperature is very
964
Paper ID- 179
Effect of Slab Thickness on Period of

Vibration of Reinforced Concrete Building
Prabhat Kumar Soni 1*(0000-0003-2419-2601),

S. K. Dubey 1(0000-0003-3421-6945),
and Prakash Sangamnerkar 2(0000-0003-3059-3727)
1*
Maulana Azad National Institute of Technology (MANIT), Bhopal, India
Prabhatsoni15@gmail.com
1
Maulana Azad National Institute of Technology (MANIT), Bhopal, India
dubeysk2000@yahoo.com
2
M.P.Housing Board, Bhopal (M.P.), India, Bhopal, India
nerkar17@rediffmail.com
Abstract. Fundamental period of vibration of reinforced concrete (RC)

buildings can be evaluated using empirical expressions given in Indian codes of
seismic design to derive design base shear. Indian seismic code IS1893 (Part-1)
– 2016 provides empirical expression to estimate the approximate fundamental
natural period of vibration which widely depend on basic parameters such as
number of stories or height of the building. The Consideration of effects of
some other parameters of the structures is also seems to be required. There is
scope of further improvisation in these equations. It can be observed that there
is wide use of fundamental period obtained using these expressions. It is
therefore, in the seismic design of the structures, it is very important to use
realistic values of time period. This study deals with the evaluation of
fundamental period of vibration of symmetrical RC buildings which
incorporates the parameters like slab thickness, stiffness of the structure, for a
particular number of bays in either direction, base width / plan area of the
building, in addition to the height of the structure. In the paper, different values
of the time period have been obtained by performing dynamic analysis on
building / structural configurations for different parameters of the building and
earthquake zone III, as per the provisions given in the Indian seismic code and
results have been presented with comparative analysis. Computer software
STAAD has been used to analyze the building model.
Keywords: Nonlinear regression analysis · Fundamental period · Number of

bay · Dynamic analysis· Stiffness
1 Introduction
The distribution of mass and stiffness are considered along the height of the building
for analysis of fundamental period of vibration. An estimation of seismic base shear
requires the fundamental natural period of vibration T of the building. The magnitude
of seismic induced dynamic forces and its directions vary with time cause
considerable inertia forces on the buildings. The approximate fundamental natural
period of vibration (T a in seconds) of a moment resisting frames building without
brick infill panels can be estimated by empirical expression as given in Indian seismic
code IS1893 (Part-1) -2016 [5].
965
Ta = 0.075h0.75 for R.C. frame building (1)
Ta = 0.085h0.75 for steel frame building (2)
Ta = 0.09 h / √d for all other buildings (3)
The fundamental period can be evaluated using simplified expressions (1) to (3)
found in codes, time periods predicted by these expressions are widely used in
practice, although it has been pointed out by Khan and Hoqueb [6], Verderame et.al.
[7], and has been discussed about scope for further improvement in these equations
since the height alone is inadequate to explain period variability. It is also known that
the period of a reinforced concrete (RC) frame structure differs depending on the
direction of the structure considered; the longitudinal or transverse direction.
Currently period of vibration in seismic analysis can be estimated by any of the

following methods:
1. Experimental observations on similar buildings (which almost never happens
in practice)
2. Any rational method of analysis (referring to dynamic analysis), or
3. Using the empirical expressions prescribed in the design code
The fundamental natural period can be evaluated using simplified expressions

found in codes, which are based on earthquake recordings in existing buildings,
laboratory tests and numerical or analytical computations. It can be observed from the
earlier research work that very few buildings are equipped with the vibration
measuring instruments and hence actual measured data were also very few in
numbers. These data sets were used by various researchers to perform regression
analysis. Simple empirical expressions to evaluate fundamental period were proposed
in terms of height, because of the following reasons:
1. Height of the structure is the most influencing parameter which affects the
time period.
2. Proposed equation allows the designer to evaluate fundamental period and
hence base shear and other response quantities without going much into the
detailed for the sizes of structural elements.
These technical codes provide expressions which depend primarily on basic

parameters such as building height or number of stories.
2 Literature study
Evaluation of the fundamental period of vibration required for the simplified design
of RC structures has been done for many years using a simplified formula relating the
period to the height of the building.
Verderame et. al [7] pointed out that height alone seems inadequate to explain
period variability and suggested that global parameter (e.g., plan area) should be
added in simplified relationships for rapid period evaluation. Therefore, they included
the plan area in an expression of the following equation
T = αHβ Sγ (4)
Where, S is the product of two principal plan dimensions of the building; Lx and Ly.
966
Shakir and Mandal [11] pointed out that the current seismic design code does not
consider the effect of beam-column joint stiffness. In the literature, far too little
attention has been paid to incorporate this effect in approximate formulas to calculate
by hand the natural periods of semi-rigid steel frames. New formulas have been
developed for rigid-jointed plane steel frames and validation of approximate formulas
was verified by a finite element analysis using ABAQUS software. A parametric
study was conducted to find the effect of semi-rigid connections on the natural
periods of vibration of plane steel frames. Further, it has been observed that the
flexibility of the connection has a considerable effect on the natural frequencies (or
natural periods) of a steel frame.
Nyrko et. al [8] proposed seven different equations in their study. In order to
determine more accurate expressions for the elastic period they considered seven
basic expressions which take different parameters into consideration such as the ratio
between the number of bays in the longitudinal and transversal directions, number of
bays parallel to the considered direction, the product between the number of bays in
the longitudinal and transversal directions in addition to the number of floors.
Following are the expressions proposed to evaluate period of vibration:
T = C N C2 (5)
1
T = C N C2 BC3 (6)
1
T = C N C2 + C BC4 (7)
1 3
KC3
 
C2  B x
 (8)
T=C N
1 B 
 y
(
T = C N C2 Bx .B y
1
)
C3 (9)
T = C NC2 + C Bx .B y
1 3
( )
C4 (10)
C4
B 
+C  x  (11)
T=C N C2
1 3 B 
 y 
Where, N is the number of stories, B is the number of bays of the building

parallel to the considered direction, Bx is the number of bays in longitudinal direction,
By is the number of bays in transversal direction, C1, C2, C3 and C4 are (unknown)
parameters.
Goel and Chopra [10] collected data measured from eight Californian earthquakes,
from 1971 (San Fernando earthquake) until 1994 (Northridge earthquake) and showed
that code formula generally underestimates the periods of vibration measured from 27
RC frames, especially those above sixteen stories. Therefore, different formulas were
proposed resulting from semi empirical analysis, with Figure 1, the best-fit plus 1
standard deviation recommended for displacement-based assessment, whilst the best-
fit minus 1 standard deviation recommended for conservative force-based design.
967
Figure 1: Results of RC MRF using eigen value analysis [Figure from Goel and Chopra, 1997
(10)].
3 Analysis
Different values of the time period have been obtained by performing dynamic
analysis on building / structural configurations for different parameters of the building
and earthquake zone III, as per the provisions given in the Indian seismic code.
Computer software STAAD [18] has been used to analyze the building model.
General arrangement of beams and columns are depicted as figure below:
Figure 2: Plan for Two - Bay square shaped building for dynamic analysis
968
Table 1: Details of building configuration, material used & structure
Building Configuration Material Specification Structural Details

Multi-storey Concrete M-25, 0.5×0.5 m
Type of Material Size of columns
rigid jointed Reinforcement 0.4×0.4 m
structure used
plane frames Fe-415. 0.3×0.3 m
No. of Type of Type -II, Medium
GF to G+5. Slab thickness 150 mm
storey soil soil as per IS-1893
Imposed load 4.00 kN/m2
5000√fck
Floor height 3.6 m Ec
N/mm2
Floor finish 1.00 kN/m2
100 mm, Water proofing 2.5 kN/m2

Slab 0.7√fck
150 mm, Fcr
thickness N/mm2 Specific weight
200 mm 25 kN/m3
of R.C.C.
Table 2: Period values (sec) for column size 500×500 mm – 2 bays
Slab Thickness
Building Height (m)
100.00 mm 150.00 mm 200.00 mm
3.60 0.75 0.83 0.89
7.20 1.35 1.47 1.58
10.80 1.96 2.13 2.28
14.40 2.55 2.75 2.95
18.00 3.15 3.39 3.62
21.60 3.85 4.12 4.38
Slab Thickness
Building Height (m)
100.00 mm 150.00 mm 200.00 mm
3.60 0.93 1.02 1.01
7.20 1.61 1.76 1.89
10.80 2.30 2.49 2.66
14.40 2.95 3.18 3.40
18.00 3.60 3.88 4.14
21.60 4.38 4.69 4.99

Slab Thickness
Building Height (m)
100.00 mm 150.00 mm 200.00 mm
3.60 1.29 1.42 1.54
7.20 2.16 2.35 2.53
10.80 3.01 3.26 3.49
14.40 3.84 4.14 4.42
18.00 4.68 5.04 5.37
21.60 5.68 6.08 6.46
969
Figure 3: Variation of period values (sec) for column size 0.4×0.4 m – 2 bays. Height of
building is in metre
building is in metre.
building is in metre.
Tables (2) – (4) and Figures (3) – (5) show that time period is also increasing with
increase in thickness of the slab, and hence the effect of the same cannot be ignored.
Figure No. (3) represents the increment in time period as 18.7% & 13.8% for slab
thickness (100mm & 200mm), with column size 500mm×500mm and for building
height 3.6m & 21.6m respectively. It can be observed from Figure No. (4) that the
970
variation in time period comes out as increase of 8.6% & 13.9% with column size
400mm×400mm. similarly, time period increase to 19.4% & 13.7% with column size
300mm×300mm(Figure No.5) for beam thickness 100mm & 200mm for building
height 3.6m and 21.6 m.
4 Conclusion
It can be observed from Figure No. (3) to (5), with variation of slab thickness, time
period is also showing variation in its values and hence the effect of the same cannot
be ignored. Therefore, height alone seems inadequate to evaluate period of vibration
and the results of this study suggest that thickness of slab along with stiffness of the
structure, also affect the time period of vibration. Hence these factors should also be
incorporated in simplified relationships given in IS 1893,for evaluation of time period
of vibration in seismic analysis.
References
1. Panzera, F., Lombardo G., Muzzetta I.: Evaluation of building dynamic properties through
in situ experimental techniques and 1D modeling: The example of Catania, Itly, Physics
and Chemistry of the Earth, Elsevier, 2013, 63, pp. 136 - 146.
2. Sha, J., Shi, W., Wang, J. : Regional study on structural dynamic property of buildings in
China, Earthquake Engineering and Structural Dynamics, John wiley publ., 2013, 42, pp.
1013-1029.
3. Oliveria, C. S., Navarro, M.: Fundamental period of vibration of RC buildings in Portugal
in-situ experimental and numerical techniques, Bull Earthquake Engineering, Springer,
2010, 8, pp. 609 – 642.
4. Anil, O., Altin, S.: An experimental study on reinforced concrete partially infilled frames.
Eng Struct, 2007, 29(3), pp.449-60.
5. Indian standard Criteria for Earthquake resistant design of structures-Part-1: General
Provisions and Buildings, IS 1893(Part 1):2016, Bureau of Indian Standards, New Delhi.
6. Khan, M. A., Hoqueb, E.: A rational for determining the natural period of RC building
frames having infill, Engineering Structures, Elsevier,2006, 28, pp 495-502.
7. Verderame, G. M., Iervolino, I., Manfredi, G.: Elastic period of existing RC-MRF
buildings, Eurocode 8 Perespectives from Indian standpoint Workshop, Napoli, Italy,2009,
pp 79-94.
8. Nyrko, M. H., Morie, D., Draganic, H., Nyarko, E.K.: New direction based (Fundamental)
periods of RC frames using Genetic Algorithms, 15th World Conference on earthquake
engineering, Lisboa, 2012.
9. Kwon, O.S., Kim, E.S.: Evaluation of building period formulas for seismic design.
Earthquake Engineering and Structural Dynamics, 2010, 39, pp.1569-1583.
10. Goel, R. K., Chopra, A. K.: Period formulas for moment-resisting frame buildings,
Structure Eng Div ASCE, 1997, 123 pp 1454-1461.
11. Al-Aasam, H.S., Mandal, P.: Simplified procedure to calculate by hand the natural periods
of semirigid steel frames”, Journal of Structural Engineering, 2013, 139,1082-1087.
12. Gong, M., Sun, J., & Xue, L.: Emperical Formula of Fundamental Period for Steel
Structure Based on Seismic Response Record. International Conference on Civil
Engineering (ICETCE), Lushan, China, 2011, pp. 283-286.
13. Guler, K., Yuksel, E., Kocak, A.: Estimation of the fundamental vibration period of
existing RC buildings in Turkey utilizing ambient vibration records, Journal of Earthquake
Engineering 12(SUPPL. 2), 2008, pp. 140–150.
14. Hong, L.L., Hwang, W.L.: Empirical formula for fundamental vibration periods of
reinforced concrete buildings in Taiwan, Earthquake Engineering and Structural
Dynamics, 2000, 9(3), pp. 327–337.
15. Crowley, H., Pinho, R.: Simplified equations for estimating the period of vibration of
existing buildings. First European Conference on Earthquake Engineering and
Seismology, Geneva, Switzerland, 2006.
971
16. Gallipoli, M.R., et al.: Empirical estimates of dynamic parameters on a large set of
European buildings, Bulletin of Earthquake Engineering, 2010, 8(3), pp. 593–607.
17. Ingle, R.K.: Time Period of elevated water tower, Indian Concrete Journal, 1997, 71, pp.
497.
18. STAAD Pro-V8i –Structural analyses and designing software by Bentely.
19. Indian standard code of practice for structural safety loadings standards: IS-875-1987
Part-1, 2 Bureau of Indian Standards, New Delhi.
20. Sangamnerkar, P., Dubey, S. K.: Effect of Base Width and Stiffness of the Structure on
Period of Vibration of RC Framed Buildings in Seismic Analysis, Open Journal of
Earthquake Research, 2015, 4, pp. 65-73.
21. Sangamnerkar, P., Dubey, S. K.: Parameters Affecting Period of Vibration of Symmetrical
Reinforced Concrete Buildings in Seismic Analysis International Journal of Scientific &
Engineering Research, March-2016, Volume 7, Issue 3, PP. 763-775.
22. Sangamnerkar, P., Dubey, S. K.: Estimation of Period of Vibration of Symmetrical
Reinforced Concrete Buildings in Seismic Analysis, The IUP Journal of Structural
Engineering,
23. Sangamnerkar, P., Dubey, S. K.: Equations to evaluate fundamental period of vibration of
buildings in seismic analysis, Structural Monitoring and Maintenance, 2017, Vol. 4, No. 4
PP. 351-364.
24. Santhi, M.H., Knight, G.M.S., Muthumani, K.: Evaluation of seismic response of soft-
storey infilled frames”. Computers and Concrete, 2005, 2(6), 423–437.
25. Al-Aasam, H.S., and Mandal, P.: Simplified procedure to calculate by hand the natural
periods of semirigid steel frames”, Journal of Structural Engineering, 2013, 139, 1082-
1087.
26. Arulselvan, S., Subramanian, K.: Experimental investigation on three dimensional RC
infilled frame – RC plane frame interactions with slab for seismic resistance, American
Journal of Applied Sciences, 2008, 5(4), 328–333.
27. Varadharajan, S., Sehgal, V.K., Saini, B.: Determination of Inelastic seismic demands of
RC moment resisting setback frames , Archives of Civil and Mechanical Engineering
,2013, 13, 370-393.
972
Paper ID- 180
Coir Fibre Reinforced Concrete mixed with GGBS
Rahul B M, Nauman Ahmed, Rishikesh Hinda and Bharti Prasad
Nitte Meenakshi Institute of Technology, P.B. No. 6429, Govindapura, Gollahalli, Yelhanka,
Bangalore, 560064
rahulbm96@gmail.com,naumanahmed736@gmail.com,rishikeshhinda@gmai
l.com and bharti.prasad@nmit.ac.in
Abstract. Concrete ever since discovered has proven itself as a versatile building
material. Further its association with steel has given it unmatched strength to
build structures beyond one’s imagination. But introduction of steel turns out to
be costly, especially in small non-load bearing structural element. Present study
is made to reinforce the plain cement concrete by introducing 0.5 %, 1 %, 2%
and 3% of the raw coir fibre by weight of cementitious material having same
aspect ratio. The ideal estimation of fibre content in concrete has been studied.
The cement is also partially replaced by GGBS to give a more sustainable and
green concrete mix capable of taking tensile strength. The present study reveals
that coir fibre can be an excellent replacement of steel reinforcement for small
structural element like lintels, sunshade etc. for taking tensile stresses. The failure
and crack pattern of specimen is also been studied which gives an idea that intro-
duction of coir fibre gives potential reduction in crack development.
Keywords: Coir Fibre, Ground Granulated Blast-furnace Slag, Compressive

Strength, Split-Tensile Strength and Flexure Strength.
1 Introduction
Rapid development in science and technology has revolutionized our society and its
demand, leading to drastic expansion of urban area in each and every dimension of
society. In between all these the economically weak section of society, who are also
backbone of our development is still struggling to get basic shelter and infrastructure
facilities. Thus, it is time to use our science and technology to find a way to provide a
much sustainable and economical solution for the low-cost housing facilities.
Concrete is a versatile composite material used in construction of most of our mod-
ern infrastructure facilities. But in spite of its proven capability in many folds it is weak
in taking tensile stresses and can only effectively take compressive stresses. To enhance
its tensile properties, it is conventionally reinforced with steel which in turn increases
the overall cost of the structure making it unaffordable for many. Also, the natural
sources of getting steel are depleted at high rate making it unsustainable material.
973
Natural fibre are filament, good in tension, that can be obtained from plants or ani-
mals like silk fibre, jute fibre, coir fibre etc. These fibres are part of construction since
age long like bird uses grass fiber for making their nest, also ancient Egyptian uses
straw to reinforce brick made out of mud. The present study is to find the feasibility of
natural fibre as a composite material in concrete.
The natural fibre used in the present study is coir fibre. Shreesail [1] studied the
deformation properties of concrete with fibers under static loading condition and the
behavior of structural components in terms of compressive strength stating the ad-
vantages of coir fibre. Also, as the area of study is South India where coir fibre are
locally available in abundant, coir fibre was found to be preferred natural fibre for the
present study. The workability of concrete is also increased to some extent as the coir
is introduced [2]. According to Chandel,[3] strength of the concrete can be increased
by introducing concrete composites and to cast the concrete so as to minimize the pol-
lution effect. Among natural fibers, coir fibers give maximum tensile strength and are
durable in nature. Also, Sahaya [4] investigated the behavioral study of natural fiber in
concrete structure. The coir fiber is treated using natural latex before using in concrete,
to prevent moisture content. In this experiment compressive strength and split tensile
strength tests are carried out using coir fiber of length 20mm, 25mm and 30mm. Mary
[5] claimed that coconut shell and coir fibers are used in concrete as these are agricul-
tural wastes and also it is economical to a great extent. It will help to produce light
weight and economically profitable materials in construction fields. Here coconut shell
and coir fibers are partially replaced as coarse aggregates. Its compressive strength,
tensile strength, temperature resistance, water absorption, electrical resistance, chemi-
cal resistance and pH tests of the sample are performed. The addition of fly ash is done
to increase the strength and workability of the concrete. Shriram [6] elaborated that the
coir fibers can be used as reinforcement in place of traditional steel reinforcement for
the building components like lintels, over door, window openings and almirah shelves.
Steel reinforcement is expensive to many people in most developing countries. In re-
gions where, natural fibers are abundantly available which when utilized will reduce
the cost of the construction. Coconut fiber used are 40mm in length. Compressive
strength and modulus of rupture of CFRC specimens were determined following stand-
ard procedures at different curing ages and also crack-patterns was monitored by Ko-
lawole [7]. Soni [8] studied the concrete behavior of plastic and hardness variation de-
pending on the concrete materials, mixed proportion, fiber types and length, quantity
of fibers added. Bending stress was increased when the sample was cured in NaCl and
MgSO4 mix water. According Santra [9] the experimental study was carried out on the
use of coconut fiber reinforced concrete which can reduce conventional building mate-
rials’ cost for the rural and urban development. Othuman[10]studied that foamed con-
crete is good in compression but weak in tension and tends to be brittle. By adding
fibers, the weakness could be overcome. The use of fibers also alters the behavior of
the fiber matrix composites after it has cracked, thereby improving its toughness.
974
2 Material and Methodology
The present experimental study is a preliminary work carried out to found the capability
and compatibility of coir fibre mixed with concrete. To obtain a more economical and
sustainable composite material cement is also being partially replaced by ground gran-
ulated blast-furnace- slag (GGBS). GGBS is obtained by quenching of molten ash from
furnace with help of water and has an excellent cementitious property.
The basic constituents of concrete are cement, fine aggregate, coarse aggregate, and
water. For the present study Portland cement by Zuari cement, Ennore sand and 20mm
down coarse aggregate has been used. Portable tap water has been used for mixing the
various constituents. GGBS is collected from a Ready-Mix Plant located at Bangalore.
All the basic testing on material has been carried out at laboratory facility available
at the institute except for GGBS and Coir, for which test are conducted at external la-
boratories.
Mix Design of M30 conforming to IS 10262-2009 has been adopted for the casting
of specimens of concrete mixed with GGBS. The previous research has shown that
GGBS can replace cement effectively even up to 70% but optimum results are obtained
at 25%. Thus 25% of cement has been replaced by GGBS. Coir fibre has been intro-
duced by percentage by weight of total cementitious material. The different model spec-
imen has been casted by varying the percentage of coir fibre as 0%, 0.5 %, 1%, 2 %
and 3%. For the present study here, the cementitious material include cement and
GGBS both. Table 1 gives the detail of quantity of various material required for test
specimen.
Table 1. Quantity of material required by test specimen (M 30 Mix Design).
Material Mix 1 Mix 2 Mix 3 Mix 4 Mix 5

Cement Cement Cement Cement Cement
+GGBS+ +GGBS+ +GGBS+ +GGBS+ +GGBS+
0.0% Coir 0.5% Coir 1.0% Coir 2.0% Coir 3.0%Coir
Cement (kg) 22.08 22.08 22.08 22.08 22.08
GGBS (kg) 07.34 07.34 07.34 07.34 07.34
Water (litres) 11.76 11.76 11.76 11.76 11.76
Fine Aggregate (kg) 32.04 32.04 32.04 32.04 32.04
Coarse Aggregate (kg) 56.99 56.99 56.99 56.99 56.99
Coir Fiber (kg) 00.15 00.29 00.59 00.88
3 Test Specimen and Testing
For studying the different strength parameter of concrete with and without coir various
combination of specimen has been casted. The length of the coir as shown in Figure 1is
kept in a varying range from 10mm to 50mm for all the specimens. The strength
975
parameter for all the test specimen has been carried out after 7 days of curing and 28
days of curing conforming to IS 456. The test specimen for compressive strength and
flexural strength is made as per IS 516:1959. The cubes of standard dimension 150 x
150 x 150 mm had been casted for compressive strength test while beam of standard
dimension of 100 x 100 x 500 mm were casted for flexural strength test. The split tensile
strength test is conducted conforming to IS 5816:1999. Table 2 gives the detail of test
specimen to be casted for each test.
Fig. 1. Weighing of Coir Fibre (10mm to 50mm of length)
Table 2. Test Specimen for different Test
Composite Material Number of sample casted for Number of sample casted for
each test (after 7 days curing) test each (after 28 days cur-
ing)
Cement +GGBS 3 3
Cement +GGBS+ 0.5% Coir 3 3
4.1 Effect on fresh concrete
Workability. The workability of the fresh concrete kept on reducing with increase in
coir fibre content in the mix. Up to 1% of coir content there was fairly not much effect
on the workability but after 2% addition of coir the workability started reducing
976
drastically. To maintain the slump value without altering w/c ratio superplasticizer ad-
mixture was used. The optimum dosage of the admixture was fixed by trial and error
method for attaining the required slump and was found to be 0.5% and 0.6% respec-
tively for test specimen with 2% and 3% of coir fibre respectively. The workability test
is conducted as per IS :1199-1959.
Mixing. Mixing of various constituent of concrete was done using concrete mixer up
to the test specimen having 1% of coir, further with addition of coir formation of lumps
were observed in the mixer. Thus, for test specimen having 2% and 3% of coir dry mix
was done in pan concrete mixer while mixing of water was done manually. Vibration
table machine was used for compaction.
4.2 Effect on Harden Concrete
Effect on strength. The test result of compressive strength, flexural strength and split
tensile strength for 7 days and 28 days is given in Table 3 and 4. There is a considerable
increase in compressive strength and flexural strength of the fibre reinforced concrete
as compared with plain concrete up to 1% of the coir but further addition of coir has
shown a declined trend for compressive and flexural strength as shown in Figure 2 and
3. The possible reason as observed could be the low specific weight of coir, due to
which it occupies more volume than any other constituent in the concrete and replace
the parent constituent of the concrete considerably . Thus, as the volume of coir in-
creases, the volume of concrete decreases leading to decrease in compressive strength
beyond addition of 1 % of fibre. While the split tensile has an increasing trend with the
increase in % of coir added as the fibre is capable of taking tension, the tension carrying
capability of the specimen also kept on increasing with increase of coir fibre content as
shown in Figure 4. From the various test conducted on different test specimen it can be
concluded that the optimum percentage of coir by weight of cementitious material can
be 1% which gives considerable increase in compression, flexural and tension carrying
capacity of the given concrete mix.
Table 3. Test Result after 7 days of Curing (in N/mm2)
Specimen Compressive Flexural Split-Tensile

strength strength strength
Cement +GGBS 23.10 5.98 1.65

Cement +GGBS+ 0.5% Coir 27.56 7.50 2.07
Cement +GGBS+ 1.0% Coir 32.22 8.25 2.18
Cement +GGBS+ 2.0% Coir 21.50 6.64 1.75
Cement +GGBS+ 3.0% Coir 20.50 6.12 1.67
977
Table 4. Test Result after 28 days of Curing (in N/mm2)
Specimen Compressive Flexural Split-Tensile

strength strength strength
Cement +GGBS 34.22 11.42 1.64

Cement +GGBS+ 0.5% Coir 36.63 13.35 2.61
Cement +GGBS+ 1.0% Coir 40.22 15.00 2.78
Cement +GGBS+ 2.0% Coir 34.55 12.20 2.84
Cement +GGBS+ 3.0% Coir 27.11 12.00 3.76
60
40 36.63 40.22
34.22 32.22 34.55
(N/mm2)
23.1 27.56 27.11

20 21.5 20.5
0
0% coir 0.5% coir 1% coir 2% coir 3% Coir
Percentage Coir added
7 days compressive strength
28 days compressive strength
Fig. 2. Graph showing the variation in Compressive Strength with % increase of Coir Fibre
Flexural Strength
Flexural strength (N/mm2)
20
15 15
13.35 12.2 12
10 11.42
7.5 8.25 6.64
5 5.98 6.12
0
0% coir 0.5% coir 1% coir 12.2 3% Coir
7 days Flexural Strength 28 days Flexural Strength
Fig. 3. Graph showing the variation in Flexure Strength with % increase of Coir Fibre
978

4 3.76
2.78 2.84
(N/mm2)
2.61 2.18
2 1.64
1.65 2.07 1.75 1.67
0
0% coir 0.5% coir 1% coir 12.2 3% Coir
7 days Split Tensile Strength

28 days Spilt Tensile Strength
Fig. 4. Graph showing the variation in Split Tensile Strength with % increase of Coir Fibre
Effect on Crack. The crack pattern during the various test has also been studied re-
vealing that apart from enhancing the strength parameter of concrete coir is also capable
of reducing the crack effectively. Figure 5 shows the crack pattern of specimen after
failure and it is evident that coir not only increase the strength carrying capability but
also hold the concrete together even after failure avoiding complete crushing of harden
concrete.
Fig. 5. Picture showing test specimen of cube and cylinder with coir after test
979
5 Conclusion
The results of the present experimental study show that an economical composite con-
crete can be proposed using coir fibre which is not only capable of taking addition
compressive strength and tensile stress without use of conventional steel reinforcement.
But as a matter of fact, it is also evident that coir cannot replace the steel for major
structures as addition of coir after certain percentage lead to reduction of the compres-
sive and flexural strength of the mix. Thus, addition of coir fibre to conventional con-
crete can be used effectively for low cost housing structure or structural element like
lintel, sunshade etc where structure is subject to minimal amount of tension. The further
partial replacement of cement with GGBS makes the mix proposed more economical.
Also, from the observed crack pattern after failure it can be suggested to adopt the pro-
posed mix for structure at seismic zone to take care of minor seismic forces and avoid
tension crack.
Reference
1. B.H., S., Chougale, J.: Effects of Coconut fibers on the properties of concrete. International
Journal of Research in Engineering and Technology 03(12), 5-11(2014).
2. Parbhane, D.M., Shinde, S.B.: Strength properties of coir fiber concrete. International Jour-
nal of Civil Engineering and Technology 5(2), 20-24(2014).
3. Chandel, A., Shah, T.: A Comparative Strength Study of Coir Fibre Reinforced Concrete
(CFRC) Over Plain Cement Concrete (PCC). IOSR journal of Mechanical and Civil Engi-
neering 13(2), 101-103(2016).
4. Sahaya Ruben, J., Baskar, G.: Experimental study of coir fiber as concrete reinforcement
material in cement-based composites. International Journal of Engineering Research and
Applications 4(1), 128-131(2014).
5. Mary Ealias, A., A.P, R.: Improvement of strength of concrete with partial replacement of
course aggregate with coconut shell and coir fibers. IOSR Journal of Mechanical and Civil
Engineering 11(3), 16-24(2014).
6. Shriram Golde, M., Malviya, L.N.: Use of coir fibers for sustainable development. Interna-
tional Journal for Scientific Research and Development 4(7), 788-789(2016).
7. Kolawole Adisa, O., Adewale Doyinsola, A., Abiola Gabriel, M.: Strength development and
crack pattern of coconut fibre reinforced concrete (CFRC). Civil and Environmental Re-
search 4, 46-53(2013).
980
8. Soni, P., Parolkar, R.: Study & analysis of durability of fiber reinforced concrete. Interna-
tional Journal of Engineering Sciences and Research Technology 4(6), 743-750(2015).
9. Santra, S., Chowdhury, J.: A comparative study on strength of conventional concrete and
coconut fibre reinforced concrete. International Journal of Scientific and Engineering Re-
search 7(4), 32-35(2016).
10. Othuman Mydin, M.A., Rozlan, N.A. Ganesan, S.: Experimental study on the mechanical
properties of coconut fibre reinforced lightweight foamed concrete. J. Mater. Environ. Sci.
6 (2), 407-411(2015)
981
Paper ID- 181
Axial stress-strain model for plain concrete and concrete filled steel tube columns
Palash Dey1,* R. K. Gupta2 and A. I. Laskar3
1,2,3
Department of Civil Engineering, National Institute of Technology Silchar, Assam-788010, India
*Email: deypalash@gmail.com
ABSTRACT
In this work, the axial stress - strain model for plain concrete and concrete filled steel tube with
different cross-sectional configuration of short columns have been developed based upon finite
element (FE) and experimental analysis. The FE numerical model for the simulation of columns
subjected to axial load has been developed using the commercial software package
ABAQUS. The results estimated using FE analyses are found in good agreement with the
measured results. It has been found that the ductility of the concrete filled steel tube (CFST)
column is significantly high compared to the plain concrete column. The outcome of the study
also shows that the circular column profiles are good geometry to be considered since they have
outstanding performance concerning of ductility in axial loading condition.
Keywords: Concrete filled steel tube; ductility; confinement; failure mode
1. Introduction
The use of concrete filled steel tube (CFST) columns compare to plain concrete columns in
modern structures has been the field of attractiveness for many researchers in recent years. In
these CFST columns, the load is only executed on the concrete core. Thus, the function of steel
tube in concrete confinement is more effective. Moreover, the steel tube in CFST columns is
utilized as permanent formworks, leading to no demand for concrete shuttering and curing and
accordingly reduction in the construction duration and cost.
The application and development of CFST columns in high rise buildings in China were
demonstrated by Zhong and Zhang [1]. Varma et al. [2] studied experimentally the behavior of
high strength square concrete-filled steel tube beam-columns. The synergistic interaction
between steel tube and filled concrete was studied experimentally by Sakino et al. [3]. Han et al.
[4] demonstrated experimentally that the steel tube confined concrete columns show high levels
982
of ductility and energy dissipation. Ghannam et al. [5] performed tests on steel tubular columns
filled with normal and lightweight concrete to investigate the behavior of such columns. Ding et
al. [6] presented an experimental comparison between concrete-filled square steel tubular stub
columns confined by internal loop or spiral stirrups, traditional square concrete-filled steel
tubular stub columns, and inner stiffened square concrete-filled steel tubular stub columns.
Schnabl et al. [7] showed the analytical buckling of slender circular concrete filled steel tubular
columns with compliant interfaces. Ding et al. [8] presented a combined numerical and
theoretical study on the composite action between steel and concrete of circular steel tube
confined concrete stub columns. Oyawa et al. [9] demonstrated the structural response of
composite concrete filled plastic tubes in compression. The compressive behavior of glass-fibre
reinforced and unreinforced cementitious material-filled square steel tubular columns as
investigated by Kharoob and Taman [10]. The study of concrete filled cold-formed steel tubular
stub columns was carried out experimentally by Zhu et al. [11].
The prime objective of this study is to investigate numerically and experimentally the
performance of plain and CFST column subjected to axial compression. A number of tests were
arranged on axially loaded columns with circular, square and rectangular cross-section; stress
versus strain relation and the failure mode of the columns were presented.
2. Finite Element Model

The modeling of CFST columns is performed in three main segments containing confined
concrete, confining steel tube and interaction among steel tube and concrete core. Moreover, in
finite element analysis of these columns, the type of element selection and size of the mesh
should be proper for precise simulation of the behavior of the columns in an admissible
computational time. The behavior developed by numerical analysis of the concrete column has
been carried out considering the Drucker Prager model given in the ABAQUS library. The four
node shell element (S4R) elements have been chosen to model a hollow steel tube, which is
shown in Fig. 1(a). For infill concrete (see Fig. 1(b)), eight node linear brick elements (C3D8)
are found to be most suitable in the viewpoints of accuracy and time necessary to run the
programme. A typical assembly of circular CFST column is shown in Fig. 1(c).
983
Figure 1. FE model: (a) hollow steel tube, (b) infill concrete core, (c) assembly of CFST column
column.
3. Experimental programme
The application of axial compression on the column has been executed using Universal Testing
Machine of 2000 kN capacity. After completion of 28 days of curing, gradual loading has been
applied on the columns at a rate of 0.6 kN/s and sustained until mater
material
ial has failed. This type of
compression test was designed to produce the load bearing capacities and failure configuration
configurations
of CFST columns. The displacement and the applied force readings were recorded automatically
at each load increment using the data acquisition system. The experimental set
setup
up is depicted in
Fig. 2
2.
Figure 2. Experimental setup.
984
4. Results and Discussions

The numerically and experimentally obtained results are presented in this section. The steel
density, Young’s modulus of steel (E) and its Poisson ratio (µ) are 7.85×10-6 kg/mm3, 200GPa
and 0.3 respectively. The density of concrete has been obtained experimentally as 2.45×10-6
kg/mm3. The young modulus of concrete has been calculated based upon the expression
5000 f (given in IS 456-2000) for M25 and M40 grade of concrete and concrete Poisson
ratio (µ = 0.15) used for the analysis. The geometric properties are presented in the Table 1.
Table 1 Geometric properties of different configuration cross-sectional columns
Sl. Column Dimensions Wall Height Area of Area of

No. Configuration (mm) thickness (mm) Concrete Steel
(mm) core(Ac) (As)
in mm2 in mm2
1 Circular (C) 150 2 600 16741.55 929.9
2 Square (S) 132.934 x 132.934 2 600 16741.55 929.9
3 Rectangular (R) 165 x 107.1 2 600 16741.55 929.9
Table 2 Cross-sectional shape of the different short columns and their symbols
Section Symbol Meaning of the symbol
C1 M25 grade of plain concrete column
C2 M40 grade of plain concrete column
Circular (C) C3 M25 grade of CFST column
C4 M40 grade of CFST column
S1 M25 grade of plain concrete column
S2 M40 grade of plain concrete column
Square (S) S3 M25 grade of CFST column
S4 M40 grade of CFST column
R1 M25 grade of plain concrete column
R2 M40 grade of plain concrete column
Rectangular (R) R3 M25 grade of CFST column
R4 M40 grade of CFST column
985
The average compressive strength for M25 and M40 grade of concrete cubes were found
experimentally as 32.67 MPa and 50.40 MPa after 28 days of curing respectively. Combination
C contains C1, C2, C3, C4, C5 and C6. The combination S and R include S1, S2, S3, S4, S5, S6
and R1, R2, R3, R4, R5, R6 respectively. Table 2 illustrates the symbols and their meanings.
4.1 Axial stress-strain plot
Numerically and experimentally obtained axial stress-strain plots for different configuration of
columns for concrete grade M25 have been presented in Fig. 3 and Fig. 4 respectively. Similarly,
for M40 grade of concrete, these are illustrated in Fig. 5 and Fig. 6.
From these figures, it can be found that there are three classes of curves based on their shape:
(a) Class I: It represents the high level of confinement. Here, the axial stress keeps on increasing
with respect to axial strain, in the early stage at a constant rate, but later at a slower rate.
Furthermore, it is mentioned that the axial stress never decreases and there is no peak on the
axial stress-strain plot. (b) Class II: It illustrates the moderate level of confinement, the axial
stress increases with respect to the axial strain till a certain peak axial stress is gained. After
gaining the peak, the axial stress reduces moderately as the axial strain further increases. (c)
Class III: It describes the low level of confinement, the axial stress increases with respect to the
axial strain till a certain peak axial stress is achieved. After reaching the peak, the axial stress
reduces significantly as the axial strain further increases.
50
40
Axial Stress (MPa)
30
Circular-M25
20 Square- M25
Rectangular-M25
10 Circular-Plain-M25
Square-plain-M25
Rectangular-Plain-M25
0
0 0.01 0.02 0.03 0.04 0.05
Axial Strain
Figure 3. Numerical axial stress-strain plot for different configuration of columns (concrete grade M25).
986
50
40
Axial Stress (MPa)
30
Circular-M25
20 Square-M25
Rectangular-M25
10 Circular-Plain-M25
Square-Plain-M25
0
0 0.01 0.02 0.03 0.04 0.05
Axial Strain
Figure 4. Measured axial stress-strain plot for different configuration of columns (concrete grade M25).
60
50
Axial Stress (MPa)
40
30 Circular-M40
Square-M40
20 Rectangular-M40
Circular-Plain-M40
10 Square-Plain-M40
0
0 0.01 0.02 0.03 0.04 0.05
Axial Strain
Figure 5. Numerical axial stress-strain plot for different configuration of columns (concrete grade M40).
60
50
Axial Stress (MPa)
40
30 Circular-M40
Square-M40
20 Rectangular-M40
Circular-Plain-M40
10 Square-Plain-M40
0
0 0.01 0.02 0.03 0.04 0.05
Axial Strain
Figure 6. Measured axial stress-strain plot for different configuration of columns (concrete grade M40).
987
From Fig. 3 and Fig. 4 it can be found that the circular CFST column C3 (circular
(circular-M25
M25
M25))
follows the class I curve, which indicates the high level of confinement. On the other hand,
square, S3 (square-M25
(square M25)
M25) and rectangular
rectangular,, R3 (rectangular
rectangular
rectangular-M25
M25
M25)) columns follow the class II
curve, which describes the moderate level of confinement. The low level of confinement can be
noticed in case of plain concrete columns, namely C1, S1, and R1. Hence, it can be conclude
concludedd
that the ductility of the CFST is significantly increas
increased
ed when compared to the plain concrete
concrete
column alone. Moreover, it has been notic
noticed
ed that the post yield behaviour are more in case of
circular CFST columns as compare to square and rectangular because of it
itss uniform confinement
effects. Similarly, same thing can be observed in case of M
M--40
40 grade of concrete columns.
4.2 Failure modes
The failure modes of three types of specimens were presented in Fig 7 to Fig 9. It can be found
that
hat the shear failure is generat
generated
ed for plain concrete columns, viz., C1, C2, S1, S2, R1 and R2. In
case
se of empty steel tubes, failure was
was by local buckling at the ends. Thus, iitt is clear that the
existence of infill concrete core resist
resistss or delays local buckling of the steel tubes, hence
converting the steel tube to a restraining medium against lateral ex
expansion
pansion of the
the infill concrete
material. This is the essence of composite interaction where each material enhances tthe
he other
e.g., the steel tube confines and restrains the infill concrete core from bursting while the infi
infill
ll
concrete resists
resists the buckling of the steel tubes
tubes (C3, C4
C4, S3, S4
S4,, R3
R3, and R4)
R4).
Figure 77. Failure modes of different circular cross-sectional

cross sectional specimens after testing
988
Figure 88. Failure modes of different square cross-sectional

cross sectional specimens after testing
Figure 9. Failure modes of different rectangular cross

cross-sectional
sectional specimens after testing
5. Conclusions
This paper demonstrates the behaviour of axial stress

stress-strain
strain model for plain concrete and
concrete filled steel tube with different cross
cross--sectional
sectional configuration of short columns based
upon finite element (FE) and experimental analysis. The FE numerical model for the simulation
of columns subjected to axial lload
oad has been establish
established
ed using ABAQUS software. The results
estimated using FE analyses are found in good agreement with the experimental results. It has
been found that the ductility of the concrete filled steel tube (CFST) column is significantly high
compared
ompared to the plain concrete column. Moreover, the study also shows that the circular column
profiles are good geometry to be considered since they have outstanding performance
concerning of ductility in axial loading condition. At the end of this paper tthe
he failure modes are
demonstrated.
989
References
[1] Shan-tong Zhong and Sumei Zhang, Application and development of concrete-filled
steel tubes (CFST) in high rise buildings, Advances in Structural Engineering, 2(2)
(1998) 149-159.
[2] A. H. Varma, J. M. Ricles, R. Sause and L. W. Lu, Experimental behavior of high
strength square concrete-filled steel tube beam-columns, Journal of Structural
Engineering, 128(3) (2002) 309-318.
[3] K. Sakino, H. Nakahara, S. Morino and I. Nishiyama, Behavior of centrally loaded
concrete-filled steel tube short columns. J. Struct. Eng., 130(2) (2004) 180-188.
[4] Han, L.H., Yao, G.H., Chen, Z.B. and Yu, Q. (2005). Experimental behaviour of steel
tube confined concrete (STCC) columns. Steel and Composite Structures 5(6):459–484.
[5] S. Ghannam, O. Al-Rawi and M. El-Khatieb, Experimental study on light weight
concrete-filled steel tubes. Jordan Journal of Civil Engineering, 5(4) (2011) 521-529.
[6] F. Ding, C. Fang, Y. Bai and Y. Gong, Mechanical performance of stirrup-confined
concrete-filled steel tubular stub columns under axial loading. Journal of Constructional
Steel Research, 98 (2014) 146–157.
[7] S. Schnabl, G. Jelenić and I. Planinc, Analytical buckling of slender circular concrete-
filled steel tubular columns with compliant interfaces. Journal of Constructional Steel
Research 115 (2015) 252 –262.
[8] F. Ding, L. Tan, X. Liu and L. Wang, Behavior of circular thin-walled steel tube
confined concrete stub columns. Steel and Composite Structures, 23(2) (2017) 229-238.
[9] W. O. Oyawa, N. K. Gathimba and G. N. Mang’uriu, Structural response of composite
concrete filled plastic tubes in compression. Steel and Composite Structures, 21(3)
(2016) 589-604.
[10] O. F. Kharoob and M. H. Taman, Behavior of fibre reinforced cementitious material-
filled steel tubular columns. Steel and Composite Structures, 23(4) (2017) 465-472.
[11] A. Zhu, X. Zhang, H. Zhu, J. Zhu and Y. Lu, Experimental study of concrete filled
cold-formed steel tubular stub columns. Journal of Constructional Steel Research, 134
(2017) 17-27.
990
Paper ID- 186
BAMBOO REINFORCED CONCRETE BEAM –

DESIGN APPROACH FOR LOW COST HOUSING
Vijay R Wairagade1, Ishwar P Sonar2
1. Indira College of Engineering and Management, Pune, India

2. College of Engineering, Pune, India
ABSTRACT
Most of the Indians (60%) live below the poverty line and since the population is on the rise the demand
of basic needs increases. Engineer’s role is to provide the best facilities at optimum cost. And as such
Concrete is the most consumed construction material in the entire world because of its various advantages
such as low cost, availability, easy to mould, good compressive strength, fire resistance etc. But it cannot
be used alone because of its low tensile strength. And therefore, it is usually reinforced with steel which
is very strong in tension. Due to the increasing cost, unavailability and other drawbacks of the steel, it
become necessary to use an alternative material as reinforcement for low cost housing.
Bamboo which is abundantly available renewable material is used as a construction material from the
earlier times due to its eco-friendly, advantageous, economical and versatile properties. As it is good in
tension and bending properties it has drawn the attention of researchers to use it as reinforcement in
cement concrete for low cost constructions. The feasibility for usage of bamboo as reinforcement in
concrete is evaluated through a series of experimental investigations in the present study. In this paper
we have tried to explore the structural behaviour and the future applications of bamboo reinforced beam
for a low cast housing preferably in rural areas where bamboo is available in abundant as a step towards
sustainable development.
Keywords
Bamboo, Sustainability, mechanical and physical properties, concrete, reinforcement.
1 Introduction
Concrete is the mainly used material in building construction. It is used largely because it is economical,
readily available and has suitable building properties, such as high compressive strength. It has low
tensile strength due to this, it is often reinforced with steel bars. These bars provide the high tensile
strength to concrete. But use of steel has some disadvantages, like higher cost and non-renewability of
steels. Production of steel is responsible for a major source of greenhouse gas emission. Hence, many
attempts are carried out by researchers to provide a low-cost sustainable alternative of steel by using
locally available material. Many researchers investigated the possibilities of using vegetables, fibres
materials as reinforcement in concrete. Vegetables fibres materials which have been studied are 1) jute,
2) coconut, 3) coir, 4) sisal, 5) babadua, 6) date palm, 7) raffia palm, 8) bamboo and bamboo fibers, etc.
although, most of these studied yielded good results still bamboo has a clear advantage over other natural
reinforcing material.
There are several good reasons why bamboo is considered as reinforcement for concrete:
• It is of low cost compared with steel
• It is readily available
• Its strength to weight ratio compares favorably with steel.
991
Bamboo is one material which reaches its full growth in just a few months and reaches its maximum
mechanical strength in just few years. It exits abundance in tropical and sub-tropical regions and makes
it an economically advantageous a lightweight design, better flexibility, and toughness due to thin walls
with discretely disturbed nodes and its great strength make it a good construction material. It is used as
structural material for scaffolding at construction sites, building roads, furniture, rugs, diapers, clothes,
promote fertility in cows, medical purposes and many more accessories.
1.1 Basic Characteristics of Bamboo
Bamboos are giant grass-like plants and not trees. Bamboo is seismically resisting material and for
sustainable environment development without harming our global environment. Bamboo exhibits certain
characteristics which limit its effectiveness as concrete reinforcement. But this can be overcome by
adopting proper methods.
1.2 Durability of Bamboo as an Engineering Material
The durability of bamboo depends strongly on the preservative treatment methods in accordance with
basic requirements. Its chemical composition should not have any effect on the bamboo fiber and once
injected it must not be washed out by rain or humidity.
1.3 Bond
If seasoned (dry) bamboo is used as reinforcement, when the concrete is poured it will absorb water and
swell. Later, as the concrete dries, the bamboo will shrink and the bond will be broken. If unseasoned
(green) bamboo is used, it will lose water and shrink as the concrete dries and again the bond will be
broken. The bond can be improved in a number of ways.
1.4 Low modulus of elasticity
The relatively low modulus of elasticity can cause problems in respect of the following:
Cracking and deflection: A bamboo reinforced element will crack and deflect perhaps 50% more than a
steel reinforced element of equivalent section.
2 Laboratory Program
2.1 Selection and preparation of bamboo strips
To check the feasibility and reliability as reinforcing material and to select appropriate kind of bamboo
specimen, series of tests are conducted on the selected bamboo samples to find out physical and
mechanical properties. Bamboo samples were collected from local market of Talegaon, Pune and
Ahmednagar. The species of bamboo are Dendrocalmus Strictus which is predominantly found in India.
The age of bamboo used is 3 to 4 years having brownish appearance and samples which are cut in winter.
Diameters of bamboos are ranging from 30 mm to 50mm and thickness 6mm to 20mm. The bamboo
splints of desired size are cut from these bamboo culms were prepared for testing. To improve the bond
strength, bamboo splints are coated with oil paint (coating materials) & sprinkled with sand being
economical.
For the reinforcement, the bamboo samples are treated in three stages. Firstly bamboo specimen are
polished by sand paper. Then a coat of black color oil paint is applied on the specimen. Finally, a coat of
fine aggregates (dry sand) is applied on the specimen to improve the bonding between the bamboo and
concrete.
2.2 Testing of bamboo Samples
The various tests conducted on the selected Bamboo samples, where the outcome found to be very much
competitive and satisfactory to use the material in concrete, we have decided to cast and test the bamboo
reinforced concrete beams.
992
Figure 1 Collection of bamboo samples from local market and testing of bamboo splints
2.3 Casting and Testing of BRC Beams
The various tests conducted on the selected Bamboo samples, where the outcome found to be very much
competitive and satisfactory to use the material in concrete, we have decided to cast and test the bamboo
reinforced concrete beams. Thus BRC beams of 2.4 mtrs in length are casted with different number of
bamboo splints are used in cement concrete M30 grade and cured for 28 days. The strain gauges are also
installed under the expert’s supervision.
The BRC beams are then tested in UTM of 400KN capacity with the proper arrangement to support 2.4
meter length beam by fabricating a built up section. The BRC beam installed with strain gauges are also
tested under the expert’s supervision. Two point loading test is done on bamboo reinforced beams to
study their failure load and failure pattern.
Figure 2 Casting and testing of BRC beams
993
Further from these observations we tried to find out some conclusion. In the first trial of RC & BRC
beams. It is observed that the BRC beams are following the similar trend like RC beams. This build
hopes and brings confidence to continue the work. The strength observed was 60% of RC beams with
the average of 2.70 % of bamboo as a reinforcement. The stresses developed in bamboo at failure load
observed were very less.
Figure 3 Comparison of RC & BRC Beams
3 Design Constants – Bamboo and Concrete

The outcome of the earlier test results of BRC beams leads to develop some constants so as to further
design and test the beams and to verify the assumed standards. So with some assumptions and the values
obtained from the test results of the bamboo specimens, I tried to calculate constants. Since we have
decided to use M30 grade of concrete for casting of final beams, I have plotted the variation of Mr with
bamboo percentage only for M30. From the graph it is clearly reflected that it is better to have maximum
4.36 percentage of bamboo as a reinforcement.
Figure 4 Variation of Mr with Pb
Figure 4 Variation of Mr with Pb
994

So accordingly I have casted few more beams with 1.96%, 2.93% and maximum 3.91 percentage of
bamboo splints as reinforcement. Out of these beams strain gauges are attached to bamboo splints and
the concrete surface at different positions in two beams. Total 33 beams were casted out of which the
shortlisted test results of 12 nos. of beams are considered. Two point loading test is done on bamboo
reinforced beams of dimensions 1200 mm X 150 mm X 250 mm to study their failure load and failure
pattern. For all the beams (RC & BRC) ultimate experimental load was more than calculated load. Also
except 1.96% reinforced BRC beams, every other beams shows better load carrying capacity at allowable
deflection of 3mm (span/350). If we could extend the allowable deflection for BRC upto 4.2mm
(span/250), all beams exhibits much safer load compared to designed load.
Figure 5 Casting and testing of BRC beams
Also, very important outcome of these tests are the highly satisfactorily results by 3.91% of BRC beams
compared to 0.48% RC beams. The results at 3.91% of bamboo reinforced beams are higher than the
results of 0.48% RC beams. If we could consider the span/250 as allowable factor for deflection in case
of BRC members, the results are much better than regular RC members.
995
References
[1] ATUL AGARWAL, BHARADWAJ NANDA, DAMODAR MAITY, “Experimental investigation on
chemically treated bamboo reinforced concrete beams and columns”, Construction and Building Materials,
ELSEVIER, 71, 2014, pp. 610–617.
[2] DINESH BHONDE, P. B. NAGARNAIK, D. K. PARBAT AND U. P. WAGHE, “Experimental Analysis of

Bending Stresses in Bamboo Reinforced Concrete Beam”, ELSEVIER - Proceedings of 3rd International Conference
on Recent Trends in Engineering & Technology 2014.
[3] DIRK E. HEBEL, FELIX HEISEL AND ALIREZA JAVADIAN, “Engineering Bamboo, The new
composite reinforcement”, Chair of Architecture and Construction, Future Cities Laboratory, Singapore ETH
Centre, Singapore.
[4] FRANCIS E. BRINK AND PAUL J. RUSH, “Bamboo Reinforced Concrete Construction”, U. S. NAVAL
CIVIL ENGINEERING LABAORATORY, Port Hueneme, California, February 1966.
[5] GHAVAMI K., 2005, “Application of Bamboo as a Low-cost Construction Material”, Proceedings of the
bamboo workshop, Nov. 14-18, 1988, pp. 270-279.
[6] JULES J.A. JANSSEN, “Designing and Building with Bamboo”, Technical report no.20, International Network
for Bamboo and Rattan 2000.
[7] KENT A. HARRIES, BHAVNA SHARMA, MICHAEL RICHARD, “Structural Use of Full Culm
Bamboo: The Path to Standardization”, International Journal of Architecture, Engineering and Construction Vol.1,
No. 2, June 2012.
[8] LEENA KHARE, “Performance Evaluation of Bamboo Reinforced Concrete Beams”, The University of Texas
at Arlington, December 2005, pp. 1-100.
[9] LIMAYE D., 1952, “Strength of Bamboo (Dendrocalmus Strictus)”, Indian Forester, Timber Mechanics Branch,
Forest Research Institute, Nov. 1952, pp. 558-575.
[10] TIMOTHY CLANCY FERGUSSON-CALWELL, “Viability of Bamboo reinforced concrete for residential
housing in Indonesia”, School of Engineering and Information technology, Charles Darwin University, 2015.
[11] WAKCHAURE M. R., KUTE S. Y., MEHETRE P. R., “Bamboo Reinforced Concrete For Sustainable
Development”, UKIERI Concrete Congress - Concrete Research Driving Profit and Sustainability, 2016, p.p. 1576-
1585.
[12] DR. I. P. SONAR, “Performance of Bamboo reinforced Concrete Composite Structural Elements”,
International Conference on Sustainable Technologies for Concrete Constructions, ACI, Mumbai, Sep. 3-4, 2010.
[13] VIJAY WAIRAGADE, “Study of Bamboo as Reinforcing Material in Concrete”, National Conference on
Innovations in Engineering and Management, NCIEM-2013, Pune, Jan.31–Feb. 01, 2013.
[14] VIJAY WAIRAGADE, DR. ISHWAR SONAR, “Review of Bamboo as Potential Reinforcement in
Concrete”, International Conference on Advancement in Engineering and Technology, ICAET-2014, Pune, Sep. 1-
3, 2014.
[15] VIJAY WAIRAGADE, DR. I. P. SONAR, “Bamboo as Potential Reinforcement in Concrete Panels”, Ancient
Science and Technology: Retrospection and Aspirations, ASTRA-2015, Pune, Jan. 10-11, 2015.
[16] VIJAY WAIRAGADE, “Experimental Study of Bamboo Reinforced Concrete Beam”, Regional Research
Conference, INNOVATION-2017, Pune, Nov. 15, 2017.
[17] PRAYUSH PAREKH, SAIPRASAD PATIL, PRANJALI SALUNKHE, VIJAY WAIRAGADE, “Study
of Structural Behaviour of Bamboo Reinforced Lintel Beams”, International Civil Engineering Symposium, IIT
Bombay , ICES, March-2018, p.p. 68-71.
[18] PRAYUSH PAREKH, SAIPRASAD PATIL, VIJAY WAIRAGADE, “Bamboo Reinforced for Sustainable
Construction for Rural Areas”, International Journal of Innovative Research & Studies, Volume 8, Issue IV, April-
2018, p.p. 284-287.
996
Paper ID- 188
Sampling based techniques for finite element model updating in

Bayesian framework using commercial software
Ayan Das1 and Nirmalendu Debnath1*
1
Department of Civil Engineering, National Institute of Technology, Silchar, Assam, India
e-mail: das.ayan53@gmail.com, *nirmalendu.debnath@gmail.com
Abstract
Finite element (FE) model updating in Bayesian framework, using sampling based techniques like Markov
Chain Monte Carlo (MCMC), is observed to be investigated by many researchers. The present work is focussed
on FE model updating using MCMC techniques where modelling is performed using commercial FE software to
avoid the difficulties with writing computer-program for FE modelling. In this present work, two prominent
MCMC techniques based on Metropolis–Hastings (MH) algorithm viz. enhanced-MCMC and transitional
MCMC are primarily used, while FE modelling is performed using a well-known FE software viz. SAP2000. A
reasonably complex structure in the form of a cantilever plate is considered in this study and modelled using
shell elements. Besides, damage is simulated in this plate structure by decreasing the Young’s modulus of few
of the elements of the discretized plate structure. Modal data in the form of frequencies and incomplete mode
shapes, evaluated from the damaged structure, are taken as the measured modal data. The technique of error
localisation and an improved parameter selection method are adopted for limiting the number of updating
parameters to facilitate better performance. Moreover Gibbs sampling which is an effective algorithm of MCMC
technique is also demonstrated using SAP2000. All the MCMC techniques for FE model updating are
performed using a computational framework based on interactions between MATLAB and SAP2000 with the
help of SAP2000 open application programming interface (OAPI). It is observed that level of performances in
FE model updating are most satisfactory while using enhanced-MCMC in comparison with others.
Keywords: Finite element method; Bayesian approach; model updating; Markov Chain Monte Carlo;
commercial FE software.
1. Introduction
Finite element (FE) model updating has become the most widely used technique as a means to bridge the gap
between the experimental modal properties of a structure and the predictions of a numerical model. Due to
various types of errors such as modelling errors, the results obtained by any FE model are different from those
actually measured. Hence FE model updating is a very significant application in this regard to reduce such
errors. The fundamental idea of FE model updating is to update / refine the initial numerical model so that the
modal properties of the updated model is highly correlated with that of the experimental / real structure. Thus
the updated / improved numerical model becomes capable of simulating the corresponding real structural-
system more accurately. The literature-review regarding various updating techniques is presented below in a
very concise manner.
Friswell and Mottershead [1] presented an extensive presentation on overall FE model updating where three
basic classes of model updating viz., (i) direct updating techniques (ii) iterative techniques and (iii) frequency
response function (FRF) based techniques, have been elaborately explained. Ewins [2] explained different types
997
of errors in creating an FE model of a structure. Moreover, Mottershead and Friswell presented on basic
physical understanding of FE model updating [3], an overview of different FE model updating techniques [4].
Among various updating techniques, the class of direct updating techniques were developed by the works of
various researchers e.g. Baruch [5-7], Baruch and Bar-Itzhack [8], Berman and Nagy [9], Caesar [10-11], Wei
[12], Friswell et al. [13].
Within the class of iterative techniques, the sensitivity technique [14] is usually observed to be considered as
quite significant. Yang and Chen [15] presented a comparative study on direct versus iterative techniques for FE
model updating. On the other hand, FRF based FE model updating was also investigated by various researchers
e.g. Lin and Ewins [16] introduced the concept of model updating by using incomplete set of FRF data, Imregun
et al. [17] investigated the FRF based updating using noisy and incomplete experimental data.
One of the most effective techniques of model updating in recent times is the technique of model updating in
Bayesian framework. One such class of model updating is the Bayesian probabilistic updating based on
maximum a posteriori (MAP). In this regard, one important work of Beck [18] may be mentioned, where a
unified Bayesian statistical framework was presented for system identification based on modal identification.
Also, Beck and Katafygiotis [19] presented a Bayesian FE model updating methodology (utilizing dynamic
response data) that can handle the inherent ill-conditioning and possible non-uniqueness in model updating
applications and Vanik et al. [20] presented a Bayesian probabilistic methodology using a sequence of identified
modal parameter for updating stiffness-parameters. Yuen [21, 22] developed a technique of Bayesian model
updating based on MAP which has the significant advantage of non-requirement of matching of mode shapes
between the measured and analytical models. Besides, a book by Yuen [23] elaborately describes Bayesian FE
model updating techniques, model class selection and various topics related to it. In a relatively recent work,
Mustafa et al. [24] presented FE model updating of an existing large bridge-structure in Bayesian probabilistic
framework using incomplete modal data and without requirement of mode matching. Lastly, Das and Debnath
[25] proposed a technique of updating in Bayesian framework based on MAP where the strictly positive nature
of the updating parameters is taken into consideration by assuming the structural parameters to follow lognormal
distribution.
The method adopted in this paper belongs to the class of sampling based FE model updating in Bayesian
probabilistic framework, where MCMC technique is primarily used in the process of maximising the posterior
PDF. Many researchers have successfully adopted modified versions of MCMC techniques namely adaptive
MCMC [26], transitional MCMC (TMCMC) [27], hybrid MCMC [28], enhanced MCMC [29-32] in FE model
updating. In the present work, two prominent MCMC techniques viz., enhanced MCMC and TMCMC which
adopt Metropolis–Hastings (MH) algorithm [33, 34] have been used for FE model updating employing a well-
known FE software viz. SAP2000 [35]. The need for commercial FE software arises due to the fact that it is not
always easy to write computer program for FE modelling of complex structures. The example considered in this
paper is a reasonably complex structure in the form of a cantilever plate andits numerical model is created in
SAP2000 using shell elements. Also, damage is simulated in the structure by decreasing the Young’s modulus
of few of the elements of the discretized plate structure. The eigenvalues and incomplete mode shapes
corresponding to selected degrees of freedom (DOFs) for first few modes of the plate structure are evaluated and
these serve as the measured modal data. Apart from these, 1 % noise is added to the actual modal data to form
998
noisy modal data used in model updating.In order to limit the number of updating parameters in the updating
procedure, the technique of error localisation [36] is employed. Besides, an improved updating parameter
selection method [37] is also adopted in this paper for improved performance in updating. It may be mentioned
that Gibbs algorithm [38], which is also an efficient algorithm of MCMC technique has been demonstrated in
this example in spite of difficulties in sub-structuring of structural matrices while using commercial FE
software.Various researchers [39-41] have successfully employed Gibbs sampling in successful updating of a
structure which do not require the procedure of mode-matching. The model updating procedure using all these
sampling techniques is performed using a computational framework where interactions between MATLAB
[42]and SAP2000 is performed using SAP2000 open programming interface (OAPI). A comparison has been
performed among all the three MCMC techniques in updating the plate structure employing SAP2000 both in
terms of performance and computational efficiency and it has been observed that level of performance in
updating and damage detection are most satisfactory for enhanced-MCMC followed by TMCMC and Gibbs
sampling while employing commercial FE software.
2. Theoretical background of Bayesian model updating
2.1 Formulation of posterior PDF
In Bayesian approach, the posterior probability is obtained from the prior probability incorporating likelihood
function of the observed data. Such Bayesian probability associated with continuous valued parameters in the
context of FE model updating may be expressed using Bayes’ theorem [43] as in Eq. (1).
p ( D | θ) p (θ)
p ( θ | D) = (1)
p ( D)
where, θ and D, represent the structural parameter-vector and observed data-vector respectively. Besides,
p (θ | D) is the posterior probability density function (PDF) of the parameter-vectorgiven the observed data-
vector, p ( θ ) is the prior PDF of the parameter-vector, p ( D | θ) is the likelihood function and p ( D ) is the
 
normalizing constant. The observed data in this paper consists of measured frequencies f and modeshapes Ψ .
Assuming the prior PDF to follow uniform distribution, Eq. (1) can be reformulated as shown in Eq.(2).
     
p  θ | f , Ψ  = k1 p  f , Ψ | θ  (2)
   
where, k1 is a normalizing constant.
Assuming the modal parameters from different modes to be independent of each other, the likelihood function
  
p  f , Ψ | θ  can be expressed as shown in Eq. (3).
 
       
pf,Ψ | θ = pf | θ p Ψ | θ
     
  m  
N N
= p  f r | θ  p  Ψ r | θ 
m
(3)
r =1   r =1  
999
The PDF of the measured eigenvalues for the rth mode p  f r | θ  can be expressed as shown in Eq. (4).

 
  
2

 
p f r | θ =
1  1
exp  − 2  f r − fr (θ)  
(4)
   
  2 k 
2k fr 
   

where, k is the standard deviation of the fractional error between the measured frequency f r and the calculated
one f r ( θ ) which is dependent on the uncertain structural parameter vector θ . On the other hand, the PDF of
the measured mode shape for the rth mode can be expressed as in Eq. (5).
   1 
exp − 2 (1 − MAC ( r ) )
1
p  Ψr | θ  = (5)
  2 k  2k 

where, k is the measure of uncertainty of the fractional error between measured mode shapes Ψ r and calculated
mode shapes Ψr ( θ ) for the rth mode and MAC ( r ) is the modal assurance criteria between the measured and
calculated mode shapes for the rth mode and is given in Eq. (6).
 2
Ψr Ψr (θ )
MAC ( r ) = (6)
 T   T
 Ψr Ψr  ( Ψr (θ) Ψr (θ))
 
After normalizing all mode shapes such that their Euclidean norms are equal to unity, Eq. (5) can be rewritten as
in Eq. (7).
  1  1   2

p  Ψr | θ  = exp  − 2 1 − Ψ r Ψ r ( θ )   (7)
  2 k  2k   
Hence, using Eqs. (4) and (7), the likelihood function shown in Eq. (3) can be written as in Eq. (8).
    1  m 1
N
p  f , Ψ | θ  = exp  − 2 J ( θ )  (8)
   2k  r =1 2 k
2
where, J ( θ ) is given as in Eq. (9).
   
2

Nm
 r r
f − f ( θ ) 
 2

J (θ) =   + 1 − Ψ r Ψ r ( θ )   (9)
  

r =1  
 fr 
 
Hence, the posterior PDF shown in Eq. (2) can be expressed as in Eq. (10).
    1 
p  θ | f , Ψ  = k2 exp  − 2 J ( θ ) (10)
   2k 
where, k2 is a normalizing constant.
1000
2.2 Sampling technique for model updating
In this paper, for structural model updating in Bayesian framework, MCMC technique is used to sample from
the posterior PDF. Conventionally, Metropolis-Hastings (MH) algorithm [33,34] is used to sample from the
posterior PDF but as described in [26], owing to narrow concentration of parameters space of interest
corresponding to the important region of the posterior PDF, it is not trivial to simply apply MH algorithm to
draw samples from posterior PDF in FE model updating. Also, further problem arises in case of multimodal
(multiple peaks) posterior PDF leading to non-ergodic Markov chain [27]. In such cases, modified versions of
MH algorithm viz., enhanced MCMC [29-32] and transitional MCMC or TMCMC [27] are employed in this
paper for successful model updating of a structure. Detailed explanation of enhanced-MCMC and TMCMC are
explained in details in [29-32] and [27] respectively.It may be mentioned that both the MCMC techniques
employ MH algorithm and the technique of mode-matching is required as can be evident from the expression of
likelihood shown in Eq. (10). Hence, another important algorithm of MCMC known as Gibbs sampling [38] has
been adopted where no mode-matching procedure is required. In case of the Gibbs sampling technique, the
conditional distribution of unknown parameters, given value of other parameters, are used in a sequential
manner. Generally, all the samples are accepted in Gibbs sampling technique, unlike the usual MH algorithm.In
model updating using Gibbs sampling [39-41], sub-structuring of the structural matrices corresponding to each
of the updating parameters is required. Sub-structuring of matrices is difficult to perform in case of model
updating using commercial software especially when the structural matrices are non-linear functions of the
updating parameters. Hence in this paper, updating parameter is selected in such a way that the structural
matrices are linear functions of the updating parameter so that the sub-structuring structural matrices are
constant and less complicated to obtain.
3. Numerical validation using commercial FE software
3.1. Description of the structure
In the present work, a complex structure in the form a cantilever plate consisting of 12 shell elements is
considered. Computer programming for the FE modelling of such structures is not easy to generate. Thus use of
commercial FE software can be very much helpful in FE model updating of any complex structure. In this
paper, FE model of the plate structure is created using SAP2000 version 15, which is one of the most popular
and commercially used FE software. The plate considered in this example is a cantilever rectangular plate
consisting of 96 DOFs. As seen from Fig. 1., the plate consists of 12 shell elements each having a thickness of 4
mm, Young’s modulus of 2.1 × 1011 N/m2 and density 7800 kg/m3. To simulate the experimental data, damage
is simulated in the structure and it is performed by discretizing one of the shell elements into 100 shell elements
and reducing the Young’s modulus of 9 elements to 1.2 × 1011 N/m2 as shown in Fig. 2. It is assumed that first
five modes of the damaged structure are considered as target / measured modes for model updating.Out of all
DOFs of the damaged model, 16 DOFs are taken as measured DOFs. These measured DOFs are considered only
along z-direction or out of plane directions associated with all joints corresponding to all the unrestrained joints
of the analytical model. Noise has been added to the frequencies and mode shapes of the damaged model to
have a more realistic simulation of experimental frequencies and mode shapes. In this regard, zero-mean
Gaussian noise has been generated using 1% coefficient of variation (COV) of the measured modal parameters.
1001
These simulated measured modal parameters (frequencies and mode shapes) are used as target in model
updating. Comparison of modal properties of the analytical and damaged structure in terms of frequencies and
modal assurance criterion (MAC) values are shown in Table 1.The process of model updating discussed in
Section 2 in Bayesian framework using both enhanced-MCMC and TMCMC employing MH algorithm and also
Gibbs sampling are performed using a computational framework based on interactions between MATLAB and
SAP2000 using SAP OAPI, where the structural matrices are extracted from SAP2000 and imported into
MATLAB for its further usage in the updating procedure.
Fig. 1. Analytical model of the plate structure
Fig. 2. Simulation of damage in the plate structure
3.2. Selection of updating parameter
The updating parameters considered in this example initially is the Young’s modulus (E) of the shell elements,
making a total of 12 updating parameters. The updating procedure starts by detection of location of modelling
errors by performing error localisation technique using a force balance method [36]. In this technique, an error
vector is obtained for each mode as shown in Eq. (11).
  (r ) 
 a
K −  M a  Φer = er  (11)
 
 (r )
where, M a , K a are the analytical mass and stiffness matrices,  is the rth experimental eigenvalue, Φ er is the
rth experimental eigenvector and e r is the error vector for rth mode. It may be mentioned that the measured mode
shape is limited to fewer DOFs due to limitation in number of sensors in ideal situations. Hence the
experimental mode shape shown in Eq. (11) is obtained by expanding the actual measured mode shape
1002
corresponding to all DOFs of the analytical model using System Equivalent Reduction Expansion Process
(SEREP) [44]. The average of the error vector for all the measured modes and corresponding to only the
measured DOFs is calculated and shown schematically in Fig. 3. From careful evaluation of the figure of error
localisation shown in Fig. 3, it shows that dominant errors in the analytical model are located around the
simulated damaged plate element location. It is a known fact that the performance of updating procedure is
satisfactory if number of updating parameters is small. Considering this issue, a sensitivity study as described in
[37] is performed on the updating parameters, where sensitivity of an objective function with respect to each
updating parameter is calculated. Two neighbouring parameters are merged into one if the sign of their
sensitivities are same. The objective function used for this example is the modal distance measure shown in Eq.
(9). The schematic diagram of the sensitivity study is shown in Fig. 4 where elements showing positive and
negative sensitivities are presented.
Fig. 3. Error localisation using force balance method
Fig. 4. Schematic diagram of the sensitivity study (blue colour indicates positive value of sensitivity value and
white colour indicates negative value).
After examining the results shown in Fig. 3 and 4, the updating parameter considered in this example is Young’s
modulus of 4th plate element of the analytical model as shown in Fig. 1.It may be mentioned that in case of
Gibbs sampling technique of model updating, the sub-structuring of stiffness matrix is necessary for its
parameterization procedure. Typical parameterization of a stiffness matrix is shown as in Eq. (12).
1003
N
K (θ) = K 0 + Σθ  K
l =1
l l
(12)
where, K l is the subsystem matrix for lth stiffness-parameter θ l and K 0 is the non-parameterized component of
stiffness matrix.However, in ourexample, Eq. (12) can be written as shown in Eq. (13).
K ( θ ) = K 0 + E4 K S (13)
where, E4 is the updating parameter i.e. Young’s modulus (E) of the 4 th plate element and K S is the sub-matrix
K
of the stiffness matrix associated with E4 . Usually, K S is obtained as K S = but this procedure is not
E4
likely to be possible while employing commercial software as in SAP2000. In this example, K S is obtained
using a technique as shown in Eq. (14).
K S = K ( ES ) (14)
where, ES = ( E1 ,E2 ,...,E12 ) and Ei is the Young’s modulus of the ith plate element such that El = 0 for all
values of l except l=4 and E4 = 1 . On the other hand, the non-parameterized part of the stiffness matrix K 0 is
obtained as in Eq. (15).
K0 = K ( E0 ) (15)
where, E0 = ( E1 ,E2 ,...,E12 ) such that El = 2 1011 N/ m2 for all values of l except l=4 and E4 = 0 . The
technique of obtaining the sub-matrix K S is possible due to the fact that the stiffness matrix is linear function of
the updating parameter E4 . It is not likely to obtain the sub-matrix using commercial software like SAP2000 if
the stiffness matrix is non-linear function of the updating parameters because in such situations the sub-matrix
K
K l associated with the updating parameter θ l can be obtained only using K l = unlike the procedure
θl
shown in Eq. (14).
The model updating of theplate structure described in Section 3 is performed using the sampling techniquesof
MCMC incorporating MH algorithm viz. enhanced-MCMC and TMCMC and another incorporating Gibbs
sampling algorithmas presented in Section 2. Apart from this basic difference of employing sampling algorithm,
another difference in requirement of mode matching is noteworthy. In this paper, out of the three sampling
techniques employed in updating the structure, enhanced-MCMC and TMCMC require mode matching
procedure while Gibbs sampling method does not require mode matching between the experimental and
analytical modes.
Performance in updating of natural frequencies for the damaged plate structure using enhanced-MCMC,
TMCMC and Gibbs sampling is shown in Fig. 5. As may be observed from Fig. 5, performance in matching of
1004
frequencies is quite similar in case of enhanced MCMC and TMCMC and slightly better in comparison to Gibbs
sampling for first two modes while Gibbs sampling perform better for the last three modes.On the other hand,
diagonal modal assurance criteria (MAC) values between initial and target mode shapes are also significantly
improved after updating (as shown in Table 2).It may be mentioned that the MAC values are closer to unity in
case of enhanced MCMC and TMCMC in comparison to Gibbs sampling. Also, the updated values of the
stiffness parameter or Young’s modulus (E) of the damaged plate element and also their updated COV values
are shown in Table 3 for all the three sampling techniques, where the COV values are smallest for enhanced-
MCMC followed by TMCMC and Gibbs sampling.It may be mentioned that one important algorithmic
parameter (A) [29-32] is used in enhanced-MCMC and (  ) in case of TMCMC [27] for controlling (usually
reducing) the level of dispersion of the posterior PDF from one level to another. The value of ‘A’ controls the
rate at which variance k2 shown in Eq. (10) of the likelihood function reduces from one level to another with its
value equal to unity at first level. On the other hand, the value of '  ' controls the constant rate at which the
power of the likelihood function increases from zero in first level to unity in final level.The updating results are
obtained using the values of ( A = 1.1) and (  = 0.0344) in case of enhanced-MCMC and TMCMC
respectively for model updating of the plate structure. On the other hand, number of samples considered for
Gibbs sampling after disregarding the built-in period is 54,803.Both the MCMC techniques employing multi-
level MH algorithm viz. enhanced-MCMC and TMCMC generate samples multiple times (i.e. multiple levels)
instead of generating samples once, where PDF of proposal distribution of a particular level is estimated based
on the samples of previous level. Number of samples is usually (as in the case of this present work) kept same
throughout the levels using both the methods andin this example number of samples considered in each
sampling level is taken to be 400. Computation time depends on number of sampling-levels in case of enhanced-
MCMC and TMCMC and on number of samples (i.e. number of sampling-iterations) in case of Gibbs sampling.
Moreover in case of enhanced-MCMC, number of sampling-levels depends on the considered value of
algorithmic parameter (A) – number of sampling-levels is increased with a decreasing value of this parameter
(A). On the other hand, in case of TMCMC, number of sampling levels depends on the value of the parameter
( ) – number of sampling-levels is increased with a decreasing value of this parameter ( ) . Six different
values of algorithmic parameter (A) for enhanced-MCMC are considered in this exercise as: 2.5, 2, 1.8, 1.5, 1.2
and 1.1 and six different values of algorithmic parameter (  ) for TMCMC are considered in this exercise as:
0.25, 0.111, 0.0714, 0.0526, 0.0416 and 0.0344. In such quantification of computational efficiency, a single
computer (make: Dell Inc., model: OptiPlex 9020) is used having key features like: (a) processor: Intel(R)
Core(TM) i7-4790 CPU @ 3.60GHz (8 CPUs) (b) RAM: 4096MB (c) hard disk drive: 500GB and (d) operating
system: Windows 8.1 Pro 64-bit. At first, evolution of modal-distance shown in Eq. (12) against the
computation time is shown in Fig. 6 and it has been observed that the enhanced-MCMC and TMCMC perform
better in comparison to Gibbs sampling. Additionally number of sampling-levels (for enhanced-MCMC and
TMCMC) and number of samples (for Gibbs sampling) are also plotted (as shown in Fig. 7) against
computation time for better understanding of computational efficiency. It may be mentioned that markers
(associated with enhanced-MCMC and TMCMC as shown in Figs. 6(a), 7(a) and Figs. 6(b), 7(b) respectively),
in the increasing order of computation time, indeed correspond to the above mentioned values of algorithmic
1005
parameter (A) and (  ) in the same order. It is interesting to study the performance of these MCMC techniques
based on the observation of final samples. In this regard, histogram of the final level samples for enhanced
MCMC and TMCMC and all the samples after disregarding the burn-in samples for Gibbs sampling are shown
in Fig. 8. From studying the histogram of the final samples in Fig. 8, samples show close to bell shaped normal
distribution in case of enhanced-MCMC, while the samples are slightly more dispersed in case of TMCMC and
samples are skewed like that of lognormal distribution or gamma distribution in case of Gibbs sampling.
Table 1. Comparison of modal properties of damaged plate and undamaged (analytical) plate
Frequencies (Hz)
Mode No. Experimental Analytical MAC values
Error (%)
(damaged) (undamaged)
1 3.14 3.32 5.56 0.9999
2 17.02 17.41 2.23 0.9979
3 18.35 19.45 5.97 0.9948
4 47.65 51.65 8.41 0.6199
5 51.04 52.14 2.15 0.7821
Table 2. MAC values after updating using enhanced MCMC, TMCMC and Gibbs sampling
Mode No. Enhanced MCMC TMCMC Gibbs sampling
1 0.9999 0.9999 0.9998
2 0.9979 0.9982 0.9894
3 0.9974 0.9980 0.9846
4 0.9792 0.9786 0.9755
5 0.9841 0.9837 0.9822
Table 3. Updated values of stiffness parameter and its COV after updating using enhanced MCMC, TMCMC
and Gibbs sampling
Enhanced MCMC TMCMC Gibbs sampling

Young’s modulus (E) (×
0.7826 0.7722 0.3978
1011 N/m2)
Updated COV 0.092 0.1777 1.1648
1006
Fig. 5. Comparison of frequencies of the target and updated models using enhanced-MCMC, TMCMC and
Gibbs sampling.
(a) (b) (c)
Fig. 6. Computational efficiency (modal distance against computation time) of (a) enhanced-MCMC (b)
TMCMC and (c) Gibbs sampling using SAP2000.
(a) (b) (c)
Fig. 7. Computational efficiency (repetition-measures against computation time) of (a) enhanced-MCMC (b)
TMCMC and (c) Gibbs sampling.
(a) (b) (c)
Fig. 8. Histogram of final samples of stiffness-parameters using values (a) enhanced-MCMC (b) TMCMC and
(c) Gibbs sampling.
5. Conclusion
1007
The FE model updating of a reasonably complex structure in the form of a cantilever plate using sampling
techniques popularly known as MCMC is performed with the help of a commercially recognised FE software
SAP2000since it is not easy to write computer program for FE modelling of any complex structure. In this
paper, two MCMC techniques which employ MH algorithm in multiple levels viz. enhanced MCMC and
TMCMC are employed in updating the structure which require mode-matching procedure. Besides, Gibbs
sampling is also adopted in updating of the structure which do not require mode-matching procedure but the
procedure of sub-structuring of the structural matrices is required. The procedure of sub-structuring of the
stiffness matrix in this paper with respect to the updating parameter is accomplished by employing SAP2000
since this method is required in successfully performing the Gibbs sampling. It may be mentioned that the
computational framework of the updating procedure is accomplished by interaction of SAP2000 and MATLAB
using SAP OAPI, which proves to be an effective application programming interface. Performances of the
MCMC techniques for model updating of the structure are studied for a damage scenario which is incorporated
by damaging one of the plate elements of the structure. Besides, computational efficiency of all the MCMC
techniques are also studied. It has been observed that the technique of enhanced-MCMC performs the best out
of the three chosen techniques followed by TMCMC and Gibbs sampling in successfully updating the structure.
References
[1] Friswell MI, MottersheadJE (1995) Finite element model updating in structural dynamics, Kluwer
Academic Publishers, Boston.
[2] Ewins DJ (2000) Adjustment or updating of models, Sadhana 25 pp 235-245.
[3] Mottershead JE, Friswell MI (2001) Physical Understanding of Structures by Model Updating,
Proceedings of the International Conference on Structural Identification, Kassel,Germany, pp81-96.
[4] Mottershead JE, Friswell MI (1993) Model updating in structural dynamics: A survey, Journal of Sound
and Vibration 167 pp 347-375.
[5] Baruch M (1978) Optimization procedure to correct stiffness and flexibility matrices using vibration tests,
AIAA Journal 16 pp 1208-1210.
[6] Baruch M (1982) Methods of Reference Basis for Identification of Linear Dynamic Structures.
Proceedings of the 23rd Structures, Structural Dynamics and Material Conference, New Orleans,
Louisiana, May pp 557-563.
[7] Baruch M (1984) Methods of Reference Basis for Identification of Linear Dynamic Structures, AIAA
Journal 22 (4) pp 561-564.
[8] Baruch M, Bar-Itzhack IY (1978)Optimal weighted orthogonalization of measured modes, AIAA Journal
16 (4) pp 346-351.
[9] Berman A, Nagy EG (1983) Improvement of large analytical model using test data, AIAA Journal 21 (8)
pp 1168-1173.
[10] Caesar B (1986) Update and identification of dynamic mathematical models, Proceedings of the Fourth
International Modal Analysis Conference, Los Angeles, pp 394-401.
[11] Caesar B (1987) Updating system matrices using modal test data, Proceedings of the Fifth International
Modal Analysis Conference, London, England, April, pp 453-459.
[12] Wei FS (1990) Analytical dynamic model improvement using vibration test data, AIAA Journal 28 (1) pp
175-177.
1008
[13] Friswell MI, Inman DJ, Pilkey DF (1998)The direct updating of damping and stiffness matrices, AIAA
Journal 36 (3), pp 491-493.
[14] Mottershead JE, Link M, Friswell MI (2011)The sensitivity method in finite element model updating: A
tutorial, Mechanical Systems and Signal Processing 25, pp 2275-2296.
[15] Yang YB, Chen YJ (2010) Direct versus iterative model updating methods for mass and stiffness matrices,
International Journal of Structural Stability and Dynamics 10, pp 165-86.
[16] Lin RM, Ewins DJ (1994) Analytical model improvement using Frequency Response Functions,
Mechanical Systems and Signal Processing 8 (4),pp 437-458.
[17] Imregun M, Visser WJ, Ewins DJ (1995) Finite Element model updating using frequency response
function data ˗ I. Theory and initial investigation, Mechanical Systems and Signal Processing 9 (2), pp
187-202.
[18] Beck JL (1996) System Identification methods applied to measured seismic response, Proceedings of the
Eleventh World Conference on Earthquake Engineering.
[19] Beck JL,Katafygiotis LS (1998) Updating models and their uncertainties. I: Bayesian statistical
framework, Journal of Engineering Mechanics, 124 (4), pp 455-461.
[20] Vanik MW, Beck JL, Au SK (2000) Bayesian probabilistic approach to structural health monitoring,
Journal of Engineering Mechanics, 126 (7), pp 738-745.
[21] Yuen KV (2006) An Extremely Efficient Finite-Element Model Updating Methodology with Applications
to Damage Detection, Proceedings of Enhancement and Promotion of Computational Methods in
Engineering and Science X, Sanya, Hainan, China, August 21-23, pp 166-179.
[22] Yuen KV, Beck JL, Katafygiotis LS (2006) Efficient model updating and health monitoring methodology
using incomplete modal data without mode matching, Structural Control and Health Monitoring, 13, pp.
91-107.
[23] Yuen KV (2010) Bayesian methods for structural dynamics and civil engineering, John Wiley &Sons
(Asia) Private Limited.
[24] Mustafa S, Debnath N, Dutta A (2015) Bayesian Probabilistic Approach for Model Updating and Damage
Detection for a Large Truss Bridge, International Journal of Steel Structures.
[25] Das A, Debnath N (2018) A Bayesian finite element model updating with combined normal and lognormal
probability distributions using modal measurements, Applied Mathematical Modelling, 61, pp. 457-483.
[26] Beck JL, Au SK (2002) Bayesian Updating of Structural Models and Reliability using Markov Chain
Monte Carlo Simulation, Journal of Engineering Mechanics, 128 (4), pp 380-391.
[27] Ching J, Chen YC (2007) Transitional Markov Chain Monte Carlo Method for Bayesian Model Updating,
Model Class Selection, and Model Averaging, Journal of Engineering Mechanics, 133 (7) pp 816-832.
[28] Cheung SH, Beck JL (2009) Bayesian Model Updating Using Hybrid Monte Carlo Simulation with
Application to Structural Dynamic Models with Many Uncertain Parameters, Journal of Engineering
Mechanics, 135 (4), pp 243-255.
[29] Lam HF, Yang J, Au SK (2015) Bayesian model updating of a coupled-slab system using field test data
utilizing an enhanced Markov chain Monte Carlo simulation algorithm, Engineering Structures, 102, pp
144-155.
1009
[30] Lam HF, Hu J, Yang JH (2017) Bayesian operational modal analysis and Markov chain Monte Carlo-
based model updating of a factory building, Engineering Structures, 132, pp 314-336.
[31] Lam HF,Alabi SA, Yang JH (2017) Identification of rail-sleeper-ballast system through time-domain
Markov chain Monte Carlo-based Bayesian approach, Engineering Structures, 140, pp 421-436.
[32] Lam HF, Yang JH, Hu Q, Ng CT (2018) Railway ballast damage detection by Markov chain Monte Carlo-
based Bayesian method, Structural Health Monitoring 17(3), pp 706-724.
[33] Metropolis N, Rosenbluth AW, Rosenbluth MN, Teller AH, Teller E (1953) Equation of State Calculations
by Fast Computing Machines, The Journal of Chemical Physics, 21 (6), pp 1087-1092.
[34] Hastings WK (1970) Monte Carlo sampling methods using Markov Chains and their applications,
Biometrika, 57 (1), pp 97-109.
[35] CSI Computer & Structures Inc. (2009). Three dimentional static and dynamic finite element analysis and
design of structures. SAP2000, Computer & Structures, Inc., Berkeley, CA.
[36] Fissette E, Stavrinidis C, Ibrahim S (1988) Error location and updating of analytical dynamic models using
a force balance method, Proceedings of the 6th International Modal Analysis Conference, Kimssimmee,
FL, pp. 1183–1190.
[37] Kim GH, Park YS (2008)An automated parameter selection procedure for finite-element model updating
and its applications, Journal of sound and vibration, 309, pp 778-793.
[38] Geman S,Geman D (1984) Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of
Images, IEEE Transactions on Pattern Analysis and Machine Intelligence 6 (6), pp 721-741.
[39] Ching J, Muto M, Beck JL (2006) Structural Model Updating and Health Monitoring with Incomplete
Modal Data Using Gibbs Sampler, Computer-Aided Civil and Infrastructure Engineering, 21, pp 242-257.
[40] Bansal S (2015) A new Gibbs sampling based Bayesian model updating approach using modal data from
multiple setups, International Journal for Uncertainty Quantification, 5 (4), pp 361-374.
[41] Cheung SH, Bansal S (2017) A new Gibbs sampling based algorithm for Bayesian model updating with
incomplete complex modal data, Mechanical Systems and Signal Processing, 92, pp 156-172.
[42] MATLAB 2015a, TheMathWorks, Inc.: 2015.
[43] Bernardo JM, Smith AFM (2000) Bayesian theory, John Wiley & Sons, England.
[44] O’Callahan J, Avitabile P, Riemer R (1989) System Equivalent Reduction Expansion Process (SEREP),
IMAC VII, pp. 29–37.
1010
Paper ID- 192
Stochastic structural optimization of multiple tuned mass damper

(MTMD) system
Kamalesh Bhowmik1* and Nirmalendu Debnath1
1
Department of Civil Engineering, National Institute of Technology, Silchar, Assam, India
e-mail: kamaleshbhowmik93@gmail.com*, nirmalendu.debnath@gmail.com
Abstract
This study deals with the optimization of parameters of multiple tuned mass damper (MTMD) system for
seismic vibration control considering uncertain bounded structural parameters. An approach based on Taylor-
expansion of objective function along with subsequent interval-extension is taken into account in this study. The
objective function involving the uncertain bounded parameters is transformed into two independents
deterministic sub-problems leading to the lower and upper bound solutions with the help of interval-extension.
The optimization strategy investigated in this optimization framework is the stochastic structural optimization
(SSO) where the root mean square displacement of the primary-secondary coupled system is minimized. A
numerical study is performed to observe the effect of uncertainties on the optimization of MTMD parameters
where the primary system is coupled with MTMD system with two, three, five and ten mass units. Performance
of these MTMD systems are also compared under various levels of uncertainties. Fundamental mode of the
system is considered for vibration-control, where natural frequency and damping ratio associated with the
fundamental mode are found to be nearly 10 rad/sec and 0.03 respectively. In this study, natural frequency and
damping ratio of the primary system, natural frequency and damping ratio of elastic filter, intensity of the power
spectral density of white noise excitation at the bed rock are taken as the uncertain parameters. It is observed
that SSO is suitable for higher levels of uncertainties associated with structural parameters. It may be mentioned
that such observations are found for various numbers of mass-units of MTMD system.
Keywords: Multiple tuned mass damper system, Stochastic structural optimization, Uncertainties.
1 Introduction
The tuned mass damper (TMD) is used as an important passive devices in the area of passive vibration control
of structural systems. TMD has been used as dynamic vibration absorber in the investigation of many
researchers [1, 2]. It is an interesting area of research at present to examine how tuned mass damper performs
under the base-excitation. Multiple-tuned mass damper systems are taken in the designing exercise because it
performs very well in vibration control [3]. TMD can be designed individually targeting any mode for single
degree of freedom (SDOF) system. For multi-degree of freedom (MDOF) system it can be designed in the same
way [4].
In various engineering field, the standard method to design the vibrating structures is to consider deterministic
models and parameters [5]. The uncertainties of different parameters (mechanical characteristics, geometric
parameters, external loading) are presented in the designing process for approving the hypotheses. The mean
values, extreme values of the parameters can be used for the hypotheses [5]. The analysis of uncertainties is also
considered to get optimal design. Tuned Mass Damper (TMD) has been used [6, 7] in passive vibration control
problems to reduce vibrations. Two different processes can be used to define uncertainties: the probabilistic
method and the non-probabilistic method [8]. The Monte Carlo simulations [9], the stochastic finite element
method, the orthogonal series expansion [8] are under the category of probabilistic method. These methods are
used in vibration control problems for various uncertainties. But these methods are not useful, if statistical
parameters are not sufficient. Sometimes, parameters are defined in a non-probabilistic process with the help of
mean and extreme values. These values are known as uncertain but bounded (UBB) parameters [10]. Interval
1011
analysis was introduced in presence of these kind of parameters for finding the extremum of static response for
structures [11, 12]. Nowadays, the extremum of dynamics response is solved by using this process [8, 13]. This
interval analysis is also used for the vibration control problems like stochastic structural optimization (SSO)
problems [14] and reliability based optimization (RBO) problems [15]. For these strategies, we have to find the
optimal parameters of tuned mass damper for UBB parameters. In stochastic structural optimization problem,
the standard deviation of the response of the structure is reduced and for reliability based optimization problem,
a failure probability associated to the response of the structure is reduced. The aim of this investigation is to
study and discuss the effectiveness of this strategy for various uncertainty levels associated with the stochastic
structural optimization strategy. In this strategy, the optimal values of the MTMD parameters are achieved by
reducing the root mean square displacement (rmsd) of the primary-secondary coupled system [R13]. It is
observed that SSO is suitable higher levels of uncertainties associated with structural parameters. It may be
mentioned that such observations are found for various numbers of mass-units of MTMD system.
2 Structural model and the equations

Fig.1 shows the primary system supported by n number of tuned mass dampers which have various dynamic
properties. The stiffness, damping constant and mass of the primary system are ks, cs, and ms respectively. The
damping ratio in the primary system is denoted by ξs. Similar to the primary system, the stiffness, the damping
constant and the mass of the jth TMD are kj, cj and mj respectively.
The tuning frequency ratio of the MTMD system is expressed by
T
f  (1)
S
where S = natural frequency of the primary system.
The natural frequency, j of the jth TMD is expressed by
  n 1  
 j  T 1   j  
2  n  1
(2)
 
n
j
and T   n
(3)
j 1
n  1
 (4)
T
where ωT = average frequency of all MTMD units; and β = non-dimensional frequency band-width of MTMD
system.
The damping constant of the jth TMD is denoted by
c j  2m jT  j (5)
Where T is the damping ratio and it is kept constant for all MTMD units).
The mass ratio (  ) of the MTMD system is writtten as

n
j 1
mj
 (6)
ms
The equation of motion of the system is:
 M x  Cx   K x  F (7)
where,  M  , C  and  K  are the mass, damping and stiffness matrices respectively, while {F} is force vector.
1012
Fig.1: Structural model of primary system with n numbers of MTMD
The matrices,  M  , C  and  K  are expressed as

 M   diag ms , m1, m2 , m3 ,......., mn  (8)
CS   CT CT 
C     where CT  C1 , C2 , C3 ,......., Cn  (9)
 CT  diag CT 
T

 K S   KT  KT 
K     where KT   K1 , K2 , K3 ,......., Kn  (10)
  KT  diag  KT 
T

The state space covariance matrix R can be solved from the Lyapunov equation [7, 8, 12]:
AR  RAT  B  0 (11)
where the state system matrix A is:
 0 I 
A
HC 
(12)
HK
and where the two sub-matrices H K and H C are respectively:
1013
  2f 
 
  2f 
 
   M 1 K  
HK    (13)
 
 
 f 
2
 
0 0  2f 
 2 f  f 
 2 f  f 

 
   M 1C  
HC    (14)
 
 
 2 f  f 
 
0 0 2 f  f 
All elements of matrix B are zero except the last element on the main diagonal,
0 0
 
B  2 S0   (15)
0 1 

Let, the TMD is placed on the primary structure with n0 degrees of freedom. Then, the root mean square
displacement  ys (rmsd) and the root mean square velocity  ys (rmsv) of the system are found as follows:
 y  R  n0  1, n0  1
s
(16)
and  y  R  n0  n  2, n0  n  2 
s
(17)
The system matrices  A, B  are considered as uncertain and the parameters of tuned mass damper are optimized
for minimizing the objective function. If structural parameters X   X1 , X 2 ,..., X q  ( q = no. of uncertain
parameters) are uncertain but bounded (UBB) parameters, then the better option is to define these parameters
using intervals. Let X   X1 , X 2 ,..., X q  be the corresponding box and X i   X i , X i  be the corresponding
interval. The mean value of X i is  i and the maximum deviation is  X i . Then we can write the UBB
parameter as follows:
X i   X i , X i    i  X i , i  X i   i  e X i where e   1,1 (18)
If uncertainty level is assumed small, we can extend the response in Taylor series about mean value
   1 ,..., q  in the first order terms of  X i  X i , X i  as:
q
f
f  X   f    (19)
i 1 X i X i  i
f
All derivatives X i are calculated at the mean value of the related uncertain parameter. With the help of
interval extension these UBB parameters are transformed into two functions, one is upper bound function and
another is lower bound function. These functions can be written as:
1014
q
f q
f
f  X   f    X i and f  X   f    X i (20)
i 1 X i i 1 X i
The system matrices A and B can be calculated using Taylor expansion. We can obtain the following
equations for ith UBB parameter by neglecting the higher order terms:
A   R     R    A    B     0
T
(21)
R R
A    A     Bu  0
T
(22)
X i X i
A A B
where Bu  R  R  (24)
X i X i X i
3 Optimization strategies
The TMD optimization problem consists in finding the parameters of TMD such as the optimal natural
frequency T  kT
mT and the damping ratio T  CT 2 kT mT
. The mass ratio   mT mS is considered as constant,
where mS = mass of primary system. A common formula [7, 8] of TMD optimization problem is defined as
follows:
Find d  T , T  To minimize OF (25)
where OF is the objective function. This function is correlated to the response of primary system. In this study
OF is correlated to the displacement response of primary system say, yS and the stochastic structural
optimization strategy is considered.
The optimal values of the TMD parameters are calculated by reducing the root mean square displacement
(rmsd) as defined in Eq. (16). The formulation of the problem can be written as follows [11, 12]:
Find d  T , T  To minimize OF1   ys (26)
The objective function OF1 is UBB parameters’ function. This function is approached and extended considering
Eqs. (19) and (20) as follows:
q  ys
 y  x   y      xi (27)
s s
i 1 xi
q  ys
 y  x   y      xi (28)
s s
i 1 xi
 y    can be calculated from Eq. (16) and the sensitivity of rmsd can be calculated as:
s
 R  n0  1, n0  1 
 ys 1  xi


 (29)
xi 2 R  n0  1, n0  1 
 
 
1015
The sensitivity of R is found from the solution of Eq. (22). The sensitivity of A and B are found with the help of
central finite difference method. The primary SSO problem is transferred into two deterministic independent
sub-problems leading to the lower and upper bound solutions. These two sub-problems are as follows:
Sub-problem 1 (lower bound): Find d  T , T  To minimize  ys  x  (30)
Sub-problem 2 (upper bound): Find d  T , T  To minimize  ys  x  (31)
4 Numerical Study
A numerical study has been considered to examine and discuss the efficiency of the presented technique of
optimization strategy. The optimized results for SSO problem is matched with respect to the deterministic
optimization. For this investigation the mean values of the uncertain parameters are taken.
For modelling the base acceleration Yb  t  , the stationary Kanai-Tajimi model [16, 17] is used in this
investigation:

Yb  t   Y f  t     t    2 f  f Y f   2f Y f  (32)
Y f  t   2 f  f Y f   2f Y f    t 
where ω(t) = stationary Gaussian zero mean white noise process (it’s intensity is S0 which representing the
excitation at bed rock),  f = frequency of filter,  f = damping ratio of filter.
In this study, natural frequency of primary system, damping ratio of the primary system, natural frequency of
elastic filter, damping ratio of elastic filter, intensity of the power spectral density of white noise excitation at
bed rock are taken as the uncertain parameters as usually considered by various researchers.
X  S , S ,  f ,  f , S0  (33)
Table-1 shows the mean values of uncertain parameters. Different results of the SSO strategy are shown in the
Figs. 2-9.
The effects of rmsd with respect to different uncertainties are shown in Fig.2-5. The mass ratio is taken as 0.02
and the uncertainty varies from 2.5% to 20%. Four cases of MTMD systems are considered where the primary
system is coupled with MTMD system with (i) two mass units (ii) three mass units (iii) five mass units and (iv)
ten mass units.
The effects of rmsd with respect to different mass ratios are shown in Fig.6-9. In this case the mass ration varies
from 0.02 to 1 and the uncertainty is taken as 10%.
From the numerical study, we can see that when the uncertainties are increasing for a fixed mass ratio, the
difference between the upper and lower bound of rmsd value is also increasing. But in case of fixed uncertainty,
if the value of mass ratio increases, the difference between the upper and lower bound of rmsd value does not
increase so much in compared to the previous case.
Table-1: Mean values of uncertain parameters
Uncertain  S (rad/s) S  f (rad/s) f S0 (m2S-3)

parameters
Mean values () 10 0.03 9 0.04 0.03
1016
n=2
σ + Δσ σ σ - Δσ
0.055
0.05
Rmsd () 0.045
0.04
0.035
0.03
0 5 10 15 20
ΔX (%)
Fig.2: Optimized root mean square displacement with respect to uncertainty (n=2)
n=3
0.055
0.05
Rmsd ()
0.045
0.04
0.035
0.03
0 5 10 15 20
ΔX (%)
n=5
0.055
0.05
Rmsd ()
0.045
0.04
0.035
0.03
0 5 10 15 20
ΔX (%)
1017
n=10
0.06
0.055
0.05
Rmsd ()
0.045
0.04
0.035
0.03
0 5 10 15 20
ΔX (%)
n=2
0.06
0.055
RMSD ()
0.05
0.045
0.04
0.035
0 2 4 6 8 10

Fig.6: Optimized root mean square displacement with respect to mass ratio (n=2)
n=3
0.065
0.06
0.055
RMSD ()
0.05
0.045
0.04
0.035
0 2 4 6 8 10

1018
n=5
0.075
0.07
0.065
RMSD () 0.06
0.055
0.05
0.045
0.04
0.035
0 2 4 6 8 10

n=10
0.15
0.13
0.11
RMSD ()
0.09
0.07
0.05
0.03
0 2 4 6 8 10

5 Conclusion
In this present work, the stochastic structural optimization of MTMD system considering uncertain bounded
structural parameters is studied. The optimum parameters of MTMD system are found in probabilistic
framework for different mass ratio and different numbers of MTMD-units. The method studied in this paper is
about Taylor expansion and it is followed by interval extension. This method has been examined with the
stochastic structural optimization strategy. The method allows the primary non deterministic problem to transfer
into two deterministic independent sub-problems in presence of UBB parameters.
From the obtained results, it is observed that when the uncertainties are increasing for a fixed mass ratio, the
difference between the upper and lower bound of rmsd value is also increasing. But in case of fixed uncertainty,
if the value of mass ratio increases, the difference between the upper and lower bound of rmsd value does not
increase so much in compared to the previous case. So, there is a certain change in the upper and lower bound
1019
optimum results, when considering the uncertainties. It is observed that SSO is suitable for higher levels of
uncertainties associated with structural parameters. It may be mentioned that such observations are found for
various numbers of mass-units of MTMD system.
6 References:
1. Den Hartog, J.P.: Mechanical Vibrations. New York: McGraw-Hill (1956)
2. Ormondroyd, J., Den Hartog, J.P.: The theory of the dynamic vibration absorber. Transactions of ASME,
Applied Mechanics. 50, 9–22 (1928)
3. Debnath, N., Deb, S.K., Dutta, A.: Multi-modal vibration control of truss bridges with tuned mass dampers
under general loading. Journal of Vibration and Control. 22(20), 4121–4140 (2015)
4. Rana, R., Soong, T.T.: Parametric study and simplified design of tuned mass dampers. Engineering
Structures. 20, 193–204 (1998)
5. Schueller, G.I., Jensen, H.A.: Computational methods in optimization considering uncertainties – an

overview. Computer Methods in Applied Mechanics and Engineering. 198, 2–13 (2008)
6. Marano, G.C., Greco, R., Sgobba, S.: A comparison between different robust optimum design approaches:
application to tuned mass dampers. Probabilistic Engineering Mechanics. 25, 108–118 (2010)
7. Rubio, L., Loya, J.A., Miguelez, M.H., Fernandez-Saez, J.: Optimization of passive vibration absorbers to
reduce chatter in boring. Mechanical Systems and Signal Processing. 41, 691–704 (2013)
8. Qiu, Z., Wang, X.: Comparison of dynamic response of structures with uncertain-but-bounded parameters
using non-probabilistic interval analysis method and probabilistic approach. International Journal of Solids and
Structures. 40, 5423–5439 (2003)
9. Kroese, D.P., Taimre, T., Botev, Z.I.: Handbook of Monte Carlo Methods. John Wiley & Sons, Inc.,
Hoboken, New Jersey (2011)
10. Muscolino, G., Sofi, A.: Bounds for the stationary stochastic response of truss structures with uncertain-but-
bounded parameters. Mechanical Systems and Signal Processing. 37, 163–181 (2013)
11. Liu, Z.S., Chen, S.H., Han, W.Z.: Solving the extremum of static response for structural systems with
unknown but bounded parameters. Computers & Structures. 50, 557–561 (1994)
12. Qiu, Z., Xia, Y., Yang, J.: The static displacement and the stress analysis of structures with bounded
uncertainties using the vertex solution theorem. Computer Methods Applied Mechanics Engineering. 196,
4965–4984 (2007)
13. Qiu, Z., Wang, X.: Vertex solution theorem for the upper and lower bounds on the dynamic response of
structures with uncertain-but-bounded parameters. Acta Mechanica Sinica. 25(3), 367–379 (2009)
14. Mrabet, E., Guedri, M., Ichchou, M.N., Ghanmi, S.: Stochastic structural and reliability based
optimization of tuned mass damper. Mechanical Systems and Signal Processing. 60-61, 437-451 (2015)
15. Ranganathan, R.: Structural reliability analysis and design. Jaico publishing house, Mumbai (1999)
16. Kanai, K.: Semi-empirical formula for the seismic characteristics of the ground. Bulletin of the Earthquake
Research Institute, University of Tokyo. 35, 309-324 (1957)
17. Tajimi, H.: A statistical method of determining the maximum response of a building during earthquake. In:
Proceedings of the of second world conference on Earthquake engineering (1960)
1020
Paper ID- 193
BEHAVIOUR OF UHPFRC UNDER EXPLOSIVE LOADING:

AN OVERVIEW
Rizwanullah1, Suruchi Mishra2, Dr. H.K. Sharma4
1, 2, 3
Department of Civil Engineering, National Institute of Technology, Kurukshetra
Haryana, India, 136119.
E. mail: rizwan_6170002@nitkkr.ac.in
ABSTRACT
In recent years, a number of explosion related problems occurred in India. Such activities causes
not only damages to the structures but also so many causalities occurred in the urban areas. In an
explosion, blast load is applied to the concrete structures as an impulsive load of extremely short
duration with very high pressure and heat. There have been so many types of blast loading either
free air blast or Air blast or ground blast. An explosion can be defined as a quick chemical reaction
in solids, dust or gases where an explosion of the matter occurs. The supersonic combustion, that
often has a duration of microseconds, is called detonation and is the type of explosion which occurs
when for example the explosive TNT is ignited. When an explosion occurs it will result in a
production of exceedingly high temperatures and pressures due to the expanding hot gases that are
created. The expansion leads to wave type propagation in the surrounding medium in a spherical
form and a blast wave is produced as the hot gases compresses the surrounding air.
KEYWORDS: UHPFRC, Explosion, TNT, Blast load, Blast wave propagation
INTRODUCTION
In the past decades, there has been numerous blast occurred because of accidental or intentional
activities related to important structures. The blast explosion is due to pressure or vehicle bomb
or quarry blasting. Lifeline structures like embassies, nuclear power plants, bridges, pipelines,
industrial plants, dams etc. protect from the blast shock.
An explosion is defined as the rapid release of large amount of stored energy in small volume so
as to generate a finite amplitude pressure wave in the surrounding media. The blast volume
changes from 10000-15000 and the temperatures increases near the area where explosion occurs
are 3000-4000 C.
Types of explosions (Occurrence)
 Natural
 Intentional
 Accidental
In an explosion, blast load is applied to the concrete structures as an impulsive load of extremely
short duration with very high pressure and heat.
Types of blast loading
 Free air blast
1021
 Air blast
 Ground/Surface blast
Fig 1: Types of blast loading
Fig. 2: Blast loads on a building
Table 1: Damage by Blast Pressure
Damage Incident overpressure(Psi)

Typical window glass breakage 0.15-0.22
Minor damage to some buildings 0.5-1.1
Panel of sheet metal buckled 1.1-1.8
Failure of concrete walls 1.8-2.9
Collapse of wood framed buildings Over 5.0
Serious damage of steel frame buildings 4-7
Severe damage of reinforced concrete structure 6-9
Probable total destruction of most buildings 10-12
1022
The blast wave, consisting of a shock front, travel faster than speed of sound and therefore does
not give any warning of its approach. The initial pressure rise crushes the object while velocity
behind leads to the disruption of the crushed object from the place. The blast wave is instantly
increased to higher pressure levels than the ambient atmospheric pressure. This high energy
intensity decreases as the blast wave moves further away from the explosion source and may after
a short time drop below ambient pressure.
Fig. 3 Pressure –time history (J.H.J. Kim, 2010)
Fig.3 shows a typical blast pressure profile. At the arrival time tA, following the explosion, pressure
at that position suddenly increases to a peak value of overpressure, Pso, over the ambient pressure,
Po. The pressure then decays to ambient level at time tA+to, then decays further to an under pressure
Pso (creating a partial vacuum) before eventually returning to ambient conditions at time tA+to+to.
The quantity Pso is usually referred to as the peak side-on over pressure, incident peak over pressure
or merely peak over pressure. The incident peak over pressures Pso are amplified by a reflection
factor as the shock wave encounters an object or structure in its path. Reflection factors depend on
the intensity of the shock wave, and for large explosives at normal incidence these reflection
factors may enhance the incident pressures by as much as an order of magnitude. Throughout the
pressure-time profile, two main phases can be observed; portion above ambient is called positive
phase of duration tA, while that below ambient is called negative phase of duration, t A‾. The
negative phase is of a longer duration and a lower intensity than the positive duration. As the stand-
off distance increases, the duration of the positive-phase blast wave increases resulting in a lower-
amplitude longer-duration shock pulse. Charges situated extremely close to a target structure
impose a highly impulsive, high intensity pressure load over a localized region of the structure but
charges situated further away produce a lower-intensity, longer-duration uniform pressure
distribution over the entire structure. Eventually, the entire structure is engulfed in the shock wave,
1023
with reflection and diffraction effects creating focusing and shadow zones in a complex pattern
around the structure. During the negative phase, the weakened structure may be subjected to impact
by debris that may cause additional damage.
Fig 4: Blast wave propagation (Ngo, 2007)
LITERATURE REVIEW
Several works were reported of blast analysis of structural members. This paper focusses to
understand the effect of blast loading on members. Columns are the members which affects the
most after the blast wave. Blast load though often ignored, have proved to be the vital
importance in the present day world.
C. Wu et al, 2009[1] conducted A series of tests to investigate the blast resistances of slabs
constructed with both plain ultra-high performance fiber concrete (UHPFC) and reinforced ultra-
high performance fiber concrete (RUHPFC), and slabs reinforced with externally bonded (EB)
fiber reinforced polymer (FRP) plates. LVDTs and pressure transducers were used to record
deflection histories, and pressure sensors located at the center and one edge of the slabs measured
airblast pressure. The performance of the RUHPFC slab was superior to all other slab types tested
in the program. Kim et al, 2010[2] suggested the Ultra High Strength Concrete (UHSC) and
Reactive Powder Concrete (RPC) have been actively developed to significantly improve concrete
strength. UHSC and RPC can improve concrete strength, member size and weight reductions and
workability improvement. High strength concrete usages in better earthquake resistance and
increase a building height and bridge span. Ultra High Strength Concrete (UHSC) and Reactive
Powder Concrete (RPC) RC slabs’ response induced by explosive of blast wave pressure are
evaluated to understand the blast resistance capacity blast resisting repair materials and retrofitted
structure. The failure patterns of both UHSC and RPC indicate that they are much more resistant
to blast loading and have higher blast resistance capacity than NSC. Aoude et al, 2015[3]
presented the results of a study examining the blast load performance of ultra-high performance
fiber reinforced concrete (UHPFRC) columns. They studied concrete type, fiber content, fiber
properties, transverse reinforcement spacing and longitudinal reinforcement ratio. The results
demonstrate that the use of UHPFRC significantly improves the blast performance of reinforced
1024
concrete columns by reducing maximum and residual displacements, enhancing damage tolerance,
and eliminating secondary blast fragments. Lai et al, 2015[4] researched the dynamic behaviour
of Ultra-high performance concrete (UHPC) under repeated penetration and different depth
explosion using 14.5 mm bullets and TNT explosives. The penetration depth of UHPC was
measured on different number of penetrations. The damage of UHPC was measured by the
ultrasonic wave velocity method and the penetration process of UHPC was observed by the high-
speed camera. The explosion damage of UHPC with TNT explosive embedded at different depths
was measured and the explosion process was simulated by the finite element method. Li et at,
2015[5] researched a series of reinforced concrete slabs are tested to determine their response
under explosive loading conditions. Concrete materials used in the slab construction are ultra-high
strength concrete (UHPC) and normal strength concrete (NSC). In total five slabs are tested
including four UHPC slabs with varying reinforcement ratios and one control NSC slab with
normal reinforcement. Explosive charges with TNT equivalent weights ranging from 1.0 to 14.0
kg at scaled distances ranging from 0.41 to 3.05 m/kg1/3 are used in the current experiments. Yoo
et al, 2016[6] studied the behavior of various UHPFRC structures under different loading
conditions, such as flexure, shear, torsion, and high-rate loads (impacts and blasts). The bond
performance between UHPFRC and reinforcements, which is fundamental information for the
structural performance of reinforced concrete structures, was investigated. The use of 2 % steel
fibers resulted in a higher post cracking stiffness and ultimate load capacity of steel bar reinforced
UHPFRC beams. With the inclusion of 2.5 % steel fibers, approximately 250 % higher shear
strength was obtained, compared to that without fibers. Remennikov et al, 2017[7] studied the
Detonations of nitro methane spherical charges in blast loading of steel plates and the effectiveness
of several protective solutions. Three types of bare steel plates named as mild steel, high-strength
steel, and stainless steel were subjected to explosive blast loading compared to the mild steel plates.
Yoo et al, 2017[8] studied comprehensively to investigates impact and blast resistances of ultra-
high-performance fiber reinforced concrete (UHPFRC) by considering various influential factors.
They examined the (1) UHPFRC is able to dissipate much higher energy by impact than ordinary
concrete with and without fibers, (2) the use of long straight steel fiber is effective in improving
the impact resistance of UHPFRC compared to that of deformed steel fibers at high volume
fractions, (3) fiber orientation significantly influences the impact resistance of UHPFRC: when
more fibers are aligned in the tensile load direction, better impact resistance is achieved, and (4)
size effect on the dynamic increase factor versus strain-rate relationship is insignificant. The
dynamic compressive, tensile, and flexural behaviors were improved by adding steel fibers, and
the effectiveness increased when the amount of fibers increased. The strain capacity and toughness
at impact loads decreased with increasing specimen size. The maximum moment capacity and
deflection of reinforced UHPFRC beams increased with increasing potential energy. Wang et al,
2017[9] presented the use of polyurea to reinforce clay brick masonry unit walls in dynamic event.
In order to investigate the peak pressure of damage, failure modes and failure mechanisms of
spray-on polyurea reinforced clay brick masonry unit walls subjected to blast, six tests were
performed. The primary damage of the clay brick masonry unit wall are the fracture of the brick
and the mortar crack that extended from top to bottom at the center of the wall. The polyuria layer
can significantly improve blast resistance, and turn the collapse of unreinforced wall into local
mortar joint separation and the development of flexure in the walls.
1025
CONCLUSION
After the detailed literature review it has been observed that progressive collapse of building after
the detonation of bomb is due to the weak beam column joint. Minimization of progressive collapse
by strengthening the beam column joint located at close to the detonation. Ultra high performance
concrete is best available concrete having outstanding material properties and blast resistance
capacities. UHPFRC indicates that they are much more resistant to blast loading and have higher
blast resistance capacity than NSC. Rebar and short steel fiber used in ultra high performance
concrete negate the brittle material characteristic of UHPC and confer outstanding energy
absorption and crack controlling capacities.
REFERENCE
1. C. Wu, D.J. Oehlers, M. Rebentrost, J. Leach, A.S. Whittaker, “Blast testing of ultra-high
performance fibre and FRP-retrofitted concrete slabs”, Engineering Structures 31 (2009),
pp. 2060-2069.
2. J.H.J. Kim, N.H. Yi, I.S. Oh, H.S. Lee, J.K. Choi, Y.G. Cho,” Blast loading response of
ultra high performance concrete and reactive powder concrete slabs”, Fracture Mechanics
of Concrete and Concrete Structures, pp. 1715-1722.
3. Hassan Aoude, Frederic P. Dagenais, Russell P. Burrell, Murat Saatcioglu, “Behavior of
ultra-high performance fiber reinforced concrete columns under blast loading”,
International Journal of Impact Engineering 80 (2015), pp. 185-202.
4. Jianzhong Lai, Xujia Guo, Yaoyong Zhu,” Repeated penetration and different depth
explosion of ultra-high performance concrete”, International Journal of Impact
Engineering 84 (2015), pp. 1-12.
5. Jun Li, Chengqing Wu, Hong Hao,” An experimental and numerical study of reinforced
ultra-high performance concrete slabs under blast loads”, Materials & Design 82 (2015),
pp. 64–76.
6. Doo-Yeol Yoo, Young-Soo Yoon,” A Review on Structural Behavior, Design, and
Application of Ultra-High-Performance Fiber-Reinforced Concrete”, International Journal
of Concrete Structures and Materials, Vol.10, No.2, pp.125–142, June 2016.
7. Alex Remennikov, Tuan Ngo, Damith Mohotti, Brian Uy, Michael
Netherton,“Experimental investigation and simplified modeling of response of steel plates
subjected to close-in blast loading from spherical liquid explosive charges”, International
Journal of Impact Engineering 101 (2016), pp.78-89.
8. Doo-Yeol Yoo, Nemkumar Banthia, “Review Mechanical and structural behaviors of ultra-
high-performance fiber-reinforced concrete subjected to impact and blast”, Construction
and Building Materials 149 (2017), pp. 416–431.
9. Junguo Wang, Huiqi Ren, Xiangyun Wu, Canliu Cai, “Blast response of polymer-
retrofitted masonry unit walls”, Composites Part B 128 (2017), pp.174-181.
1026
Paper ID- 195
Experimental Study on Liquid Sloshing Dynamics in a Sway Excited Rectangular Tank
Sahaj K V1* and T Nasar 1
1*
Research scholar, National Institute of Technology Karnataka, Surathkal, India
1 Assistant Professor, National Institute of Technology Karnataka, Surathkal, India
Email: sahaj30.k.v@gmail.com
Email: t.nasar@gmail.com
Abstract presence of the internal obstructions with the analytical and

numerical techniques (Saoudi et al., 2013, Wang 2014 and
Liquid in partially filled tanks tends to slosh when subjected to Neves et al. 2016). Jin et al., 2014, Cho et al., 2017, and Xue et
external disturbances. Liquid sloshing constitutes a broad class al., 2017 reported the impact of the obstructions on the sloshing
of problems of great practical importance with regard to the frequency with the hydraulic tests.
safety of liquid transportation system. Seismic design of liquid Present study focuses on finding out the sloshing oscillation
storage tanks requires knowledge of sloshing frequency of and sloshing force in sway excited tanks of rectangular shapes.
liquid and hydrodynamic force on the wall. An experimental
study is carried out to study the phenomena of liquid sloshing in
a partially filled rectangular tank. Liquid fill level with liquid
depth, hs to length of tank, l ratio (hs/l) of 0.325 is studied. Four
The set up for the experimental work is shown in Fig. 1
run up probes are used at distinct locations. Time histories of
comprises of a hydraulic shake table along with a digital
sloshing oscillation are measured along the length of the tank at
amplifier. The shake table is driven by a signal conditioner
distinct locations. The behaviour of sloshing oscillation is
which is attached to a computer shown in Fig. 1. The frequency
observed for the excitation. Sloshing oscillation for the
and amplitude of excitation are controlled with the help of a
excitation frequency f1, f2, f3, f4 and f5 are observed and
Scientech (wave transformer). Scaled model of tank are made
analysed. Excitation amplitude of 2.5 mm is embraced.
up of acrylic plate, which is a transparent material. The liquid
Sloshing force is captured by load cells and it is also analysed.
filled tank is mounted on the shake table and shake table is
excited harmonically at a particular frequency. The sloshing
Keywords: Sloshing Oscillation; Excitation Amplitude;
motion of liquid is observed. The excitation amplitude is kept
Excitation Frequency;
varying for all the frequencies of excitations. If the excitation
frequency is very close to the sloshing frequency of liquid, then
1. INTRODUCTION the liquid sloshes with large amplitude for the same amplitude
of excitation. The excitation frequency at which the liquid
Liquid storage tanks are commonly used in water distribution sloshes with large amplitude is taken as the sloshing frequency
systems, and in industries for storing liquids. Importance of of liquid. It is noted that at frequencies on lower and higher side
seismic analysis of liquid storage tanks does not need any over of sloshing frequency, the amplitude of sloshing is less. Using
accent. During the past earthquakes, tanks have suffered this procedure, the sloshing frequency of liquid is obtained.
varying degree of damages. During lateral base excitation
seismic ground acceleration causes hydrodynamic force on the
tank wall which depends on the geometry of tank, height of
liquid, properties of liquid and fluid-tank interaction. Proper
estimation of hydrodynamic force requires a rigorous fluid-
structure interaction analysis. The impulsive liquid moves along
with the tank wall, as it is rigidly connected and the convective
and sloshing liquid moves relative to tank wall as it under goes
sloshing motion. The mechanical model is quantified in terms
of impulsive mass, convective mass, and flexibility of
convective liquid. The hydrodynamic liquid forces are obtained
using the frequency of sloshing and impulsive modes.
From the past three decades, significant amount of work has
been carried on the fluid tank interaction problems with the
numerical and experimental approaches. Nakayama and Fig. 1: Schematic diagram of the shake table
Washizu (1980) reported the unsteady surface motions of liquid
in a rectangular tank using the finite element method (FEM).
Few of the researches addressed the sloshing phenomena in the
1
1027
National
Table 2: Liquid oscillation in Institute of Technology
the rectangular tank Silchar
3. RESULTS AND DISCUSSION
Sloshing frequency is an influencing parameter in the seismic Sl. No. Frequency (Hz) Excitation Amplitude (a) mm
analysis of tank-liquid system. The results are presented for 1 0.7755 2.5
50% fill levels of rectangular tanks. Natural frequency will be 2 1.2287 2.5
found using mathematical expression. Ibrahim (2005) proposed 3 1.5270 2.5
a linear approximation of resonant liquid sloshing frequencies 4 1.7666 2.5
for the nth mode as: 5 1.9756 2.5
1 n g  n hs 
fn  tanh   n =1, 2, 3…… (1) The liquid oscillation of the tank is studied by considering
2 l  l  excitation amplitude of 2.5mm for different frequencies as
shown in Table 2. From the experimental study, increase in the
where, fn = sloshing frequency (Hz), l is the length of frequency shows the variation in the sloshing inside the tank.
rectangular tank along the direction of excitation (m), hs is the For first mode of natural frequency, the sloshing inside the tank
static liquid depth (m), g is acceleration due of gravity (m/s2) is observed to be high and for rest of the frequencies the
and n is the surface mode number. Sloshing of 50% filled liquid sloshing occurred inside the tank is moderate. The sloshing
tank is mounted on shake table is considered for the experiment oscillation are explored for the aspect ratio (hs / l) of 0.325 that
(Fig. 2). Table 1 shows the fundamental frequency of sloshing correspond to 50% fill level of liquid in the tank. Surface
tank of dimension (inner) of 1.0m x 0.4m x 0.65m (Fig.3). elevation spectrum and force spectrum are snatched by run up
probes and load cells respectively which are mentioned in detail
shown in Fig. 4, Fig. 5, Fig. 6, Fig. 7 and Fig. 8.
Figure 2: Liquid sloshing tank with 50% liquid fill condition

fitting on to the shake table.
Fig. 3: Details of rectangular tanks in cm
Table 1: The fundamental frequency of sloshing tank
Sl. Fill f1(Hz) f2(Hz) f3(Hz) f4(Hz) f5(Hz)

No. level
1 50% 0.7755 1.2287 1.5270 1.7666 1.9756
2 fe /f1 1.0 1.5844 1.9691 2.2780 2.5475
1028
Fig. 4: sloshing oscillation for amplitude a/l= 0.0025 and h s/l = Fig. 5: sloshing oscillation for amplitude a/l= 0.0025 and h s/l =
0.325: (a) surface elevation at the end wall fe / f1=1.0; (b) 0.325: (a) surface elevation at the end wall fe / f1 =1.5844; (b)
surface elevation spectrum corresponding to (a); (c) sloshing surface elevation spectrum corresponding to (a); (c) inertia
force; (d) sloshing force spectrum corresponding to (c). force; (d) inertia force spectrum corresponding to (c).
The sloshing oscillation f1 corresponding to the excitation The sloshing oscillation f2 corresponding to the excitation
frequency of 0.7755 Hz (fe / f1 =1.0) is illustrated in Fig. 4. frequency of 1.2287 Hz (fe / f1 =1.5844) is illustrated in Fig. 5.
From Fig. 4(b) it can be inferred that dominant peak (fe) is near A single dominant peak is observed at f2 from Fig. 5(b). From
to f1. However, the excitation frequency (fe) close to secondary the force spectral plot (Fig. 5d), it is observed that higher force
parametric resonance of the third mode approximately equal to is acting at second mode frequency.
half of the third mode (f3/2), secondary spectral peaks are
observed at mode f3. It can be deduced that higher force is
acting at first mode frequency from (Fig. 4d).
1029
Fig. 6: sloshing oscillation for amplitude a/l = 0.0025 and h s/l Fig. 7: sloshing oscillation for amplitude a/l = 0.0025 and h s/l
= 0.325: (a) surface elevation at the end wall fe / f1 =1.9690; (b) = 0.325: (a) surface elevation at the end wall fe / f1 =2.2780; (b)
surface elevation spectrum corresponding to (a); (c) inertia surface elevation spectrum corresponding to (a); (c) inertia
force; (d) inertia force spectrum corresponding to (c). force; (d) inertia force spectrum corresponding to (c).
The sloshing oscillation f3 corresponding to the excitation The sloshing oscillation f4 corresponding to the excitation
frequency of 1.5270 Hz (fe /f1 =1.9690) is illustrated in Fig. 6. frequency of 1.7666 Hz (fe / f1 =2.2780) is illustrated in fig. 7.
Single dominant peak at f3 is observed shown in Fig. 6(b). The The dominant peak is at f3 and secondary spectral peaks are at f4
higher force is acting at third mode frequency (Fig. 6d). The and f1 from (Fig. 7b). From the force spectral, it is observed that
excitation frequency at f3 excites sloshing at f3. higher force is acting at fourth mode frequency (Fig. 7d). The
excitation frequency at f4 excites sloshing at f4.
1030
References
1. Cho IH, Choi JS, Kim MH (2017) Sloshing reduction in

a swaying rectangular tank by an horizontal porous baffle.
Ocean Engg. 138:2334.
doi:10.1016/j.oceaneng.2017.04.005
2. Tsukasa Nakayama, Kyuichiro Washizus (1980)
Nonlinear analysis of liquid motion in a container
subjected to forced pitching oscillation. International
journal for numerical methods in engg 15:1207-1220.
doi:10.1002/nme.1620150808
3. Wang W, Xiong YP (2014) Minimising the sloshing
impact in membrane LNG tank using a baffle. University
of Southampton. ISSN: 2311-9020; ISBN: 978-972-752-
165-4
4. Jin H, Liu Y, Li HJ (2014) Experimental study on sloshing
in a tank with an inner horizontal perforated plate. Ocean
Engg 82:75-84
5. Neves MAS, Fonfach JM, Manderbacka T (2016)
Numerical sloshing simulations: Comparison between
lagrangian and lumped mass models applied to two
compartments with mass transfer. Ocean Engg 114:168-
184. doi:10.1016/j.oceaneng.2016.01.023
6. Saoudi Z, Hafsia Z, Maalel K (2013) Dumping effects of
submerged vertical baffles and slat screen on forced
sloshing motion. J Water Resource. Hydraulics. Engg.
2:51-60
7. Xue MA, Zheng J, Lin P, Yuan X (2017) Experimental
study on vertical baffles of different configurations in
suppressing sloshing pressure. Ocean Engg.136:178-189.
Fig. 8: sloshing oscillation for amplitude a/l = 0.0025 and h s/l doi:10.1016/j.oceaneng.2017.03.031
= 0.325: (a) surface elevation at the end wall fe / f1 =2.5475; (b) 8. Ibrahim RA (2005) Liquid sloshing dynamics: theory and
surface elevation spectrum corresponding to (a); (c) inertia applications. Cambridge University.
force; (d) inertia force spectrum corresponding to (c). doi:10.1017/CBO9780511536656
The sloshing oscillation f5 corresponding to the excitation

frequency of 1.9756 Hz (fe / f1 =2.5475) is illustrated in Fig. 8.
The dominant peak is at f4 and secondary spectral peak is
observed at mode f1 and mode f5. It is observed that higher force
(Fig. 8d) is acting close to fifth mode frequency. The excitation
frequency at f5 excites sloshing at f5.
4. CONCLUSION
The behaviour of sloshing oscillation is observed at amplitude
of 2.5mm for five different modes of frequencies. The sway
responses of rectangular tank corresponding to the excitation
frequency of 0.7755 Hz and for the aspect ratio of 0.325, the
spectral peaks are observed at excitation harmonic fe and at
higher harmonic (f3). It can be inferred third mode plays a major
role than fourth mode and fifth mode since third mode natural
frequency is secondary dominant in some frequencies.
Acknowledgements. The authors are grateful to the SERB,

DST, Govt. of India Sponsored R&D project and NITK for
providing the research fellowship for pursuing Ph.D. and lab
facilities.
1031
Paper ID- 196
Hearth Monitoring of Blast Furnace using Finite Element Analysis and Artificial
Intelligence
Debi Prasad Ghosh*, Bhaskar Sengupta, Shyam Krishna Maitra
Larsen & Toubro Limited, Kolkata, India
* Corresponding Author: debiprasadghosh@lntecc.com
Abstract
Hearth wear profile determines the life of a Blast furnace, which is generally monitored by an inverse heat
conduction analysis. Numerous thermocouples are embedded in the refractory lining and the temperature readings
are used to access the wear profile of the hearth, which provides warnings about the infiltration of liquid iron
through the refractory mass to the shell of the Blast Furnace. Because of the fatal nature of the consequences,
accurate predictions about the profile of the inner refractory of the blast furnace are useful for initiating corrective
action.
This paper describes a method whereby the left out refractory lining profile is estimated as well as the thickness
of the protective skull using Finite Element Analysis and Artificial Intelligence. It serves as a guide to continue
operation of the furnace, to direct repair work to the critical areas, as well as to suggest replacement of faulty
sensors. In the effect, this method serves to increase the working life of Blast Furnaces.
Keywords
Hearth Monitoring. Blast Furnace. Finite Element Analysis. Artificial Intelligence
Terminologies
 Campaign: The duration of operation of a Furnace, for one lining.
 Breakout: An accident, when the hot metal infiltrates the refractory layer through to the steel shell of the
Furnace.
 Dead Man Zone: An unreactive body of coke at the base of the hearth
 Ramming mass: A refractory mix used to fill gaps
 Hot metal: The metal produced in Blast Furnaces. Iron with 4%C, at 1450 Degree C.
 Hearth: The base of the Blast Furnace holding hot metal formed with refractory material. Subject to
severe thermal and abrasive wear conditions.
 Tap-hole: An opening through which the hot metal (liquid iron) is extracted.
 Shell: Cylindrical steel lining, which forms a container for the Blast Furnace.
 Skull: A hard growth of refractory material and charge ore, on the lining.
 RMS: Root Mean Square of the deviation between calculated and measured temperature values in
sensors.
 1150 Degree C solidus surface: A temperature contour at 1150 Degree C, considered to be the
solidification location of liquid iron (iron-ceramic solidus). Also, considered as the refractory boundary.
 Forward/Inverse Finite Element (FE) problem: Forward refers to the standard FE technique, where the
thermal properties are used to solve for the temperature field. Inverse refers to the estimated values of
the thermal properties that result in a temperature field that best matches measurements at a few specified
locations.
1032
Introduction
Blast Furnaces are the primary route through which iron is produced. It is a counter current reactor, comprising a
vertical steel shell lined internally with refractories. A preheated hot blast of air is forced into the base, hence the
name "Blast Furnace". Raw material as Iron ore together with fuel in the form of coke and fluxes as limestone are
charged from the top. Complex reactions in the solid liquid and gas phases occur, with temperatures in the region
of 2000 Degree C. Liquid iron known as Hot metal drips down in the lower third of the furnace at a temperature
of 1450 Degree C, and collects in the lower region known as the Hearth, from where it is tapped out through
openings called "Tap-holes".
Typically numerous thermocouples are embedded in the refractories at insertion depths of varying depth [Fig. 1],
and the temperatures are watched for sudden & unexplained increases. A linear interpolation from dual inline
thermocouples is sometimes used to obtain the 1150 Degree C solidus line [Fig. 2], as 1150 Degree C is the
temperature at which liquid iron solidifies, and can be considered to be the refractory boundary [Fig. 3]. The
common practice is to derive the remnant refractory thickness based on a linear interpolation of the temperature
drop in a pair of inline Thermocouples (example: the temperatures are 150 & 50 Degree C in a pair of inline
thermocouples placed 100 mm apart, then the linear interpolation would place a point at 1000 mm inline inwards
from the inner sensor to be at 1150 Degree C). This simple method is flawed in that the heat flow in a solid body
is a 3 dimensional phenomenon, wherein the thermal properties of the neighbouring bricks influence the heat flux
field. It frequently happens that the outer sensor of a pair of thermocouples records a higher temperature than the
inner sensor. Linear interpolation of the pair in such a situation would place the 1150 Degree C solidus contour
line outside the shell of the Blast Furnace. The liquid Hot metal must not touch the steel shell of the Blast Furnace,
where it would melt and pierce the structure, and lead to fatal accidents. This is known as a "Break out", and is a
major concern in the iron making process.
The Finite Element Analysis is an established method of calculating the heat flux & temperature fields, and has
been employed by many researchers to model the Hearth of Blast Furnaces. A simplified 2D Axis symmetric
Mesh [Fig. 4] are considered. The boundary conditions comprise the Hot Metal temperature [1450 Degree C] in
the interior of the furnace, and the ambient temperature [40 Degree C say] at the exterior of the shell. The
temperature & heat fluxes at the mesh are readily computed based on standardized software [1]. In ideal conditions
the calculated and the measured temperatures should match. This is found to be so in the initial de novo state of
the Blast Furnace. But as time progresses, the measured temperatures tend to increase.
The heat in the body of the Blast Furnace will seek to flow out to the cooler periphery, giving rise to a heat flux.
Normally there would be a smooth gradient of temperatures, assuming isotropic material properties. In practice
discrepancies occur. Some of the reasons are as follows:
 Cracks in the refractories: Hot metal enters these spaces. The thermal expansion induces progressive
increase in the cracks. The increase in the conductivity causes an increase in the heat flux and so to an
increase in the temperature.
 Gaps develop between the refractory and the steel shell: The discontinuity provides an impediment to
the heat flow, and leads to a rise in the refractory temperature.
 The Hot metal fluid flow in the vicinity of the Tap holes carries an enormous amount of heat flux. This
serves to pump up the temperatures where the fluid impinges on the refractory wall.
1033
 The coke body comprising the "Dead Man Zone" influences the metal flow, and directs the fluid stream
towards the refractory wall in an unpredictable manner.
Because of these effects, the thermocouple temperatures change dynamically, and differ from the static predictions
of the FE calculations. The problem is to estimate the thermal conductivity of the refractory bricks. As explained
above, the properties of the refractories degrade during the course of their service life [Fig. 5].
Cracks in the refractory bricks would impede the heat flow. Conversely if the interstitial spaces become filled
with hot metal which subsequently solidifies, the conductivity would increase. The hot metal is at 1450 Degree
C. At such high temperatures, there is considerable thermal expansion. The outer regions of the furnace being
cooled to ambient temperatures, would mean that these regions are not subjected to thermal expansion. To
compensate for these unequal expansions, a region of about 50mm is loosely packed with a mixture known as the
"ramming mass". The steel shell would serve to withstand the thermal stresses induced in the inner layers in excess
of the compensating ramming mass layers.
While it is theoretically possible to withdraw core samples from the different regions of the refractory bed and
experimentally determine the thermal conductivities, such invasive methods would ruin the refractories. The best
alternative is to estimate the thermal properties, in a manner that predicts a temperature field that matches the
measured values. This is the opposite of the conventional Finite Element thermal analysis technique, which starts
with the thermal properties, to deduce a temperature field. In a sense this is an "Inverse" application of the Finite
Element technique, in that the thermal properties are iteratively searched to provide a matching temperature field.
This modified mesh [Fig. 6] is then used in a 2D axis symmetric FE thermal analysis to derive the refractory wear
profile.
The subject of this paper is to monitor the thermocouple temperature readings, and interpret the temperature field
through a 2D axis symmetric Finite Element Analysis technique, to provide an estimate of the degradation of the
refractory layers. Further this paper derives a technique to solve the inverse problem based on global search
criterion using artificial intelligence.
Methodology
To determine the position of the left out refractory lining in the hearth of a Blast Furnace, it requires physical
measurements of the temperatures inside the refractory lining. The location of these sensors is an input, although
the method places no constraints that the thermocouples are to be placed in pair inline. Any arbitrary configuration
of thermocouples can be entertained, and equal credence is placed on all the sensors when deriving the refractory
profile. Typically about 350 thermocouple sensors are used in hearths of large modern Blast Furnaces. For this
purpose, the structure of the Blast Furnace hearth is modeled in 6 sectors. The asymmetry introduced by the tap
holes is ignored in this description, although it will be apparent that the tap holes can be readily included in a 3D
representation. The 6 sectors are individually analyzed as axisymmetric 2D FE thermal analysis problems. There
are several sensors at the same elevation and insertion depth, spread out circumferentially at different angular
locations. In the 3D representation, they can be individually considered, but in the 2D representation, the ones in
the same sector are grouped together and their mean temperature value is taken.
Although, no major iron making chemical reactions are involved (since the air from the tuyeres enters at a higher
level), a complex set of phenomena occur in the hearth of the Blast Furnace. Several simplifications are made in
order to keep the problem tractable. One of these is that once a refractory means is worn off, it is removed from
the model. Since the calculations are based on the measured temperatures, this implies that the thermocouples
1034
should measure monotonically increasing values. To achieve this, the temperatures are stored in a data base, and
the campaign maxima values are used in the refractory wear calculations. A temporary protective layer forms over
the remnant refractory lining, known as the skull. This is also estimated based on the 1150 Degree C solidus
contour line. The difference is that whereas the refractory wear profile is estimated based on the campaign
maxima, leading to a monotonically increasing wear profile, the skull is inferred from the current temperatures.
If the current temperatures fall, it is reflected in an increased skull layer thickness over the refractory lining. Thus,
to calculate the left out refractory thickness, necessary steps are:
(a) Determining maximum campaign temperature of individual sensor of individual sector;
(b) Receiving the temperature data from the sensor and recording and storing maximum campaign temperature
of individual sensor of individual sector in the memory of the computer for providing the stored data when
needed during simulation; Or optionally recording and storing maximum campaign temperature of individual
sensor of individual sector in an external memory device or a repository which when loaded/installed with
the computing device for providing the stored data when needed during simulation;
(c) Estimating conductivity values of the defined refractory bricks from predefined maximum and minimum
limit;
(d) Determining temperature at the location of the sensor using finite element method;
(e) Selecting conductivity values, whereby the campaign maximum temperature values matches the computed
values at the location of the sensor ; and
(f) Determining about 1150°C isotherm representing the left out refractory lining profile in the hearth of the blast
furnace.
Method also estimate the skull profile over left out refractory lining profile in the hearth of an operating blast
furnace in the following steps:
(a) Estimating conductivity values of the defined refractory bricks from predefined maximum and minimum
limit;
(b) Determining temperature at the location of the sensor using finite element method;
(c) Determining about 1150 Degree C isotherm representing the skull profile over the left out refractory lining
profile in the hearth of the blast furnace.
The initial conductivity values are recorded from the refractory brick manufacturer's specifications. The procedure
outlined above has to be executed individually for each of the sectors twice, once with the campaign maxima
temperatures to derive the refractory wear profile, and a second time with the current temperatures, to derive the
skull profile over the worn refractory lining.
Values for conductivity of bricks is initially assumed through a heuristic method. Later artificial intelligence is
used to derive the same and improvements in RMS error reduction.
Finite Element Formulation
1D Heat conduction equation of a heat guide of length L is as follows:

d dT
(Ak ) + Q = 0, 0 ≤ x ≤ L
dx dx
And boundary conditions are Temperature and Temperature gradient (flux)
𝑇|𝑥=0 = 𝑇0
1035
𝑑𝑇
| = 𝑇𝐿′
𝑑𝑥 𝑥=𝐿
Where,
A is area of the 1D heat guide, k is conductivity of the material, Q is heat supply in the heat guide, T is temperature
in the heat guide.
2D Heat conduction equation is as follows:

𝜕 𝜕𝑇 𝜕 𝜕𝑇
(𝑘 )+ (𝑘 )+𝑄 =0
𝜕𝑥 𝑥𝑥 𝜕𝑥 𝜕𝑦 𝑦𝑦 𝜕𝑦
And boundary conditions are Temperature and Temperature gradient (flux)
𝑇|𝑏𝑇 = 𝑇𝑏𝑇
𝜕𝑇 𝜕𝑇
𝑛𝑥 𝑘𝑥𝑥 + 𝑛𝑦 𝑘𝑦𝑦 | = 𝑞𝑏𝑞
𝜕𝑥 𝜕𝑦 𝑏𝑞
Where,
kxx and kyy are thermal conductivities of the material in x and y direction respectively, nx and ny are
component of boundary in x and y direction respectively, Q is internal heat supply, qbq is flux boundary
condition. TbT is temperature boundary condition.
2D Axis symmetric Heat conduction equation is as follows:

1 𝜕 𝜕𝑇 𝜕 𝜕𝑇
(𝑘𝑟 𝑟 ) + (𝑘𝑧 ) + 𝑄 = 0
𝑟 𝜕𝑟 𝜕𝑟 𝜕𝑧 𝜕𝑧
And the boundary conditions are temperature and temperature gradient (flux) as
𝑇|𝑏𝑇 = 𝑇𝑏𝑇
𝜕𝑇 𝜕𝑇
𝑛 𝑟 𝑘𝑟 + 𝑛𝑧 𝑘𝑧 | = 𝑞𝑏𝑞
𝜕𝑟 𝜕𝑧 𝑏𝑞
Where,
kr and kz are thermal conductivities of the material in radial and axial direction respectively, nr and nz are
component of boundary in radial and axial direction respectively, Q is internal heat supply, qbq is flux
boundary condition. TbT is temperature boundary condition.
The Galerkin method for axisymmetric equations results in the following integral form
1 𝜕 𝜕𝑇 𝜕 𝜕𝑇
∫ 𝑁𝑖 [ (𝑘 𝑟 ) + (𝑘𝑧 ) + 𝑄] 𝑑Ω = 0
Ω 𝑟 𝜕𝑟 𝑟 𝜕𝑟 𝜕𝑧 𝜕𝑧
Subsequently finite element formulation will result in the familiar final form of the matrix equation as
[𝑲]{𝑻} = {𝒇}
Where
[𝑲] = ∫ [𝐵]𝑇 [𝐷][𝐵]𝑑Ω

Ω
𝜕𝑁𝑖 𝜕𝑁𝑗 𝜕𝑁𝑘

…
[𝑩] = [ 𝜕𝑟 𝜕𝑟 𝜕𝑟 ]
𝜕𝑁𝑖 𝜕𝑁𝑗 𝜕𝑁𝑘
…
𝜕𝑧 𝜕𝑧 𝜕𝑧
1036
𝑘𝑟 0
[𝑫] = [ ]
0 𝑘𝑧
{𝒇} = ∫ 𝑄[𝑵]𝑇 𝑑Ω − ∫ 𝑞[𝑵]𝑇 𝑑𝛤

Ω 𝛤
{𝑻} are nodal temperatures, which can be obtained by solving above equation.
Artificial Intelligence
Artificial intelligence is used for search of correct value of conductivities of refractories through surrogate models
and genetic algorithm. Genetic algorithm is used for global search in the conductivity space and surrogate model
is able to approximate input/output behaviour of complex systems. Out of various types of surrogate model, deep
learning based “deep-surrogate” is considered in this study to explore input vs. output results of FE analysis,
which can learn from result without depending on hand-engineering features. Deep learning is a recently-
developed field belonging to artificial intelligence and probably the hottest topics in machine learning research
today. It tries to mimic the human brain, which is capable of processing the complex input data, learning different
knowledges intellectually & fast, and solving different kinds of complicated tasks well. Switching these features
of human brain to a learning model, the model can deal with the high-dimensional data, support a fast and
intellectual learning algorithm and perform well in the complicated tasks. Specially, if original domain problem
(computational solid mechanics, computational fluid dynamics, computational granular mechanics, computational
electro-magnetic etc.) is computationally expensive such deep-surrogate model can reduce the computational
burden through a representative input vs output map of the domain.
Conclusion
The present paper relates to an artificial intelligence for providing estimates and knowledge/status of the condition
of the refractories in the hearth region, particularly the left out lining thickness in a blast furnace. The method also
determines the temporary protective skull layer thickness over the remnant refractory. This method is applicable
to running Blast Furnaces in the later stages of their operating life. In case the estimate shows a dangerously thin
refractory layer remaining in a specific location, it alerts the operators to take remedial action by directing repairs
of the refractories at the relevant areas, to improve cooling arrangements to withdraw heat fluxes, or to close
specific tuyeres to run the Blast Furnace at reduced capacity. Failing these, this method informs the management
that the time has come to shut down the Blast Furnace for relining the refractory layers.
References
1. “Ansys Parametric Design Language”: Ansys Inc
1037
Figures
Fig. 1. Typical thermocouple layout in a new blast furnace lining for 6 sectors in different view.
Fig 2. 1150 Degree C Isotherm; Computed with Coarse Mesh and Fine Mesh
1038
Fig. 3 Refractory boundary and Skull iso-surface 3D rendered. Juxtaposition of physical picture of an actual hearth.
Fig. 4 Fine Mesh of new blast furnace in 6 different sectors
1039
Fig 5. Conductivity Plot of Liquid Iron, Refractory, Steel Shell and Ramming mass
Fig. 6 Remeshing of refractory profile after knowing size of elephant's foot
1040
Paper ID- 201
COCONUT FIBRE REINFORCED CONCRETE

a*
Sonali Upadhyaya, bBharadwaja Nanda
a*
Lecturer, Dept. of Government college of Engineering, Keonjhhar, Odisha, India
b
Assistant Proessor, Dept. of Structural Engineering, Veer Surendra Sai University of
Technology, Burla Odisha, India
1
cvrce15sonali@gmail.com, 2bnanda_ce@vssut.ac.in
Abstract
Coconut fibres are the natural light weight fibres which have very high tensile strength. These are widely
available in the agricultural countries and can be used as a reinforcing material for low cost concrete structures.
This experimental work is aimed to study the mechanical effect of coconut fibre reinforced concrete. Various
strength aspects along with the workability were studied by varying the coconut fibre content of 0.5%, 1% and
3% (by weight of cement). When the control concrete is compared with reinforced fibre concrete, it was found
that strength increases up to certain percentage of addition of coconut fibre and then decreased. It was also seen
that when subjected to ultimate load, coconut fibre reinforced concrete doesn’t collapse completely like plain
concrete.
Introduction
A sustainable reinforced concrete can be developed by using the short, discontinuous natural coconut fibre in
concrete. Also concrete is weak in tension, so these fibre can be used as reinforcing materials to minimising its
cracking. These fibres are the waste generated from coconut shell which can be easily collected from local
temples or from coir industries. Addition of fibre in concrete make the fresh concrete possible to mould in
simpler way as well as it helps in increasing the toughness which will increase load carrying capacity. So this
natural fibre can be used as a potential material in agricultural countries where cultivation of coconut tree is
more.
Ramli et al. (2013) observed the performance of coconut fibre under three types of aggressive
environments: air environment in a tropical climate (A-series), alternate air and seawater
environments in a 14-day cycle (4 days wetting + 10 days drying) (N-series), and continuous
immersion in seawater (W-series). The experimental results indicated that the compressive and
flexural strengths of the structures improved up to 13% and 9%, respectively, with the incorporation
of coconut fibers. However, in terms of durability, the chloride penetration, intrinsic permeability, and
carbonation depth increase with the increase in fiber content. Most importantly, in the intrinsic
permeability, the plain specimen in the N-series showed a sudden increase in intrinsic permeability
when the exposure period increased from 365 days to 546 days. This result signifies that the fibers
1041
play a role in restraining the development of cracks. In general, the deleterious effects brought about
by aggressive environments can be suppressed with fiber-reinforced concrete. However, the dosage of
coconut fiber should be low, not exceeding 1.2% of the binder volume, due to the drawback of its
natural degradation.
Ali et al. (2012) studied the mechanical and dynamic properties i.e. damping ratio and fundamental frequency of
simply supported CFRC beams experimentally. A comparison between the static and dynamic moduli was
conducted. The influence of 1%, 2%, 3% and 5% fibre contents by mass of cement and fibre lengths of 2.5, 5
and 7.5 cm is investigated. To evaluate the effect of coconut fibres in improving the properties of concrete, the
properties of plain concrete are used as a reference. Results showed that damping of CFRC beams increases
while their fundamental frequency decreases with structural damage. CFRC with higher fibre content has a
higher damping but lower dynamic and static modulus of elasticity. It is found that CFRC with a fibre length of
5 cm and a fibre content of 5% has the best properties.
Ali et al. (2013) investigated the tensile strength and pull out load carrying capacity of coconut fibres in coconut
fibre reinforced concrete. The factors considered include rope embedment length, rope diameter, pre-treatment
condition, concrete mix design ratio, fibre content and knot in the material matrix. The tensile strength and
elongation of coconut-fibre ropes were determined considering the parameters i.e. rope diameter and pre-
treatment. To increase the pullout energy, bond strength and tensile strength of the rope, the boiling treatment
was found to be beneficial compared to chemical treatment. The pullout energy increased with an increase in
embedment length, rope diameter, cement and fibre content in the matrix. With the knowledge obtained,
empirical equations were proposed to determine the pullout energy, bond strength and tensile strength of the
rope.
Mazid Ali (2014) studied seismic performance of three different reinforcement patterns of coconut-fibre ropes
in columns. Coconut-fibre-reinforced-concrete was used to cast columns. Overall ductile behaviour of structure
is a basic requirement; therefore coconut fibres were selected because of their highest toughness amongst all
natural fibres. Incremental ground motion excitations were applied to the columns using a shake table. The
excitations were time-scaled to produce damage in columns. This was to match the frequency of excitation with
that of structure. The dynamic properties of coconut fibre and rope reinforced concrete columns were
determined by an impact test after each excitation. A change in natural frequency was observed, showing that
there was some non-visible material degradation in the specimen before cracking. Out of three patterns, column
with multiple central ropes performed well showing rocking phenomenon.
Experimental programme
Materials
2.1.1 Portland cement
Ordinary Portland cement (OPC) of 43 Grade, confirming to BIS: 8112-2013 was procured from local supplier.
The physical properties of cement are provided in Table1.
Table1. Physical properties of cement

Properties Results IS: 8112-2013
Fineness1 4% <10%
Specific gravity2 3.14 >3.0
Consistency3 33% -
Initial setting time1 90 min >30 min
Final setting time1 265 min <600 min
compressive strength1 45.8 MPa >43MPa and <58MPa
1042
Soundness 1.0 10 maximum
2.1.2 Aggregate
Coarse aggregate of granite stones of size 10 mm and 20 mm was used with a proportion of 60:40. In fine
aggregate natural sand of 4.75 mm from Mahanadi River bed and granite dust from locally available granite
industry of Sambalpur, Odisha was used in this study. The physical properties of aggregates are provided in
Table 2.
Table 2.Physical properties of aggregates

Properties Coarse aggregate Natural Sand
Zone - II
Water Absorption 0.4% 0.6%
3
Bulk density (kg/m ) - 1605.6
120
100
80
% passing
60
40
20
0
Pan 150 µ 300 µ 600 µ 1.18 mm 2.36 mm 4.75 mm
IS sieve size in mm
Fig.1. Particle size distribution curve of sand
2.2 Concrete mix proportions
Four mix proportions were prepared along with control concrete. Coconut fibre was added at a percentage of
0.5%, 1% and 3% by weight of cement. The mix calculation was in accordance with BIS: 10262 (2009).
Concrete mix proportioning was done with 0.5 w/c ratio and 390 kg/m3 cement content for design of M20 grade
taking severe environmental condition. Table-3 represents detailed design mix.
2.3 Specimen preparation and testing
A drum mixture was used for the preparation of concrete. The aggregates were used in surface saturated dry
(SSD) condition. All the materials were first dry mixed in mixture machine and then water was added to it.
Immediately after preparation of concrete, slump test was carried out as per BIS 1199:1959 to measure
workability. The specifications of the slump cone apparatus is as shown in Fig.2 (a). Cubes of size
(150×150×150) mm, cylinder of 150 mm diameter and 300 mm height and beam of size (150×150×750) mm
were casted for all the design mixes. Total of 24cubes, 12 cylinders and 24 beams were casted for compressive,
split tensile and flexural strength test. All the specimens were demoulded after 24 hours and cured in water for
28 days under normal room temperature. After completion of curing period cubes and cylinder specimens were
tested for 7 days and 28 days strength by compression testing machine (CTM) and beams were tested in
universal testing machine (UTM) under one point loading to check its flexure behaviour.
1043
3.1 Slump
The slump value and the percentage of coconut fibre are shown in Fig.2 (b). It is observed from the figure that,
with increase in percentage of the coconut fibre, the slump value decreases. The reason behind the decrease is
due to high water absorption of coconut fibre which reduces the workability. With 3% use of fibre, slump
decreased to 21.05%.
100
95
90
Slump value in mm
85
80
75
70
65
60
55
50
0 12 3 4
% of coconut fibre
Fig.2 (a). Slump test apparatus Fig.2 (b). Slump value of coconut fibre reinforced concrete
3.2.1 Non destructive test (Rebound hammer test)
Rebound hammer test is conducted on the 12 cube specimens after 28 days curing. It is observed from
Fig.3.that, with increase in fibre percentage, the compressive strength increases up to 1%. But it decreased at
3% strength.
36
35
34
MPa
33
32
31
30
0 0.5 1 1.5 2 2.5 3
% Fibre
Fig.3. Variation of compressive strength with respect to coconut fibre percentage
3.2.2 Destructive test (Cube compressive strength test)
Compressive strength of 150 mm cubes are tested in compression testing machine at 7 and 28 days as shown in
Fig.4 (a). When compared to control concrete it can be visualised from Fig.4 (b). that the compressive strength
of coconut fibre reinforced concrete shows higher strength up to 1% addition of fibre content. 0.5% coconut
fibre enhances strength 1.75% and 3.2% at 7 days and 28 days. When 1% coconut fibre was added, the
1044
compressive strength increased by 19% and 3% respectively for 7 and 28 days. The fibre seizes the small pores
in concrete and resists the fine cracks which ultimately result higher strength.
7 days strength 28 days strength

40
35
Compressive strength in MPa

30
25
20
15
10
0
0 0.5 1 3
% Fibre
Fig.4 (a). Cube in CTM Fig.4 (b). Strength of coconut fibre reinforced concrete at
different percentage of fibre content
But with increase in fibre content up to 3 %, both 7 days and 28 days strength decreased by 4% and 17%. The
decrease in strength may be due to the presence of more heterogeneous materials which may lead to improper
bonding of fibres with cement. Thus, 1% coir fibre volume reinforced concrete had the optimum set of
mechanical properties in comparison with other fibre volume reinforced concrete. Another thing was observed
that conventional concrete specimens were fully crashed when reached their ultimate failure load but the
specimens with 1% and 3% of coconut fibre did not crash when reached their ultimate failure load. Thus,
coconut fibre reinforced concrete can enhance higher toughness.
3.3 Split tensile strength test
Split tensile strength test is conducted on the 12 cube cylinders (Fig.5) after 28 days curing. Table 3 shows the
28 days tensile strength of cylinder specimens with different fibre percentage. It is observed that, with increase
in fibre percentage, the tensile strength of concrete increases 16 % and 18.66 % at 0.5% and 1%. At 3% addition
of coconut fibre the split tensile strength increased at a decreasing rate. Unlike compressive strength split tensile
strength increased at 3% of fibre content, because the fibre acts as a binder holding reinforcement.
Table 3. Split tensile strength at 28 days of concrete at different % of fibre content
% Fibre 28 days split tensile strength % Increase

0 3 -
0.5 3.48 16%
1 3.56 18.66%
3 3.1 3%
1045
Fig. 5(a). Cylindrical specimen under CTM Fig. 5 (b). Mechanism involved in split tensile test
3.4 Flexural strength test of beam
Fig.6. shows similar behaviour of coconut fibre reinforced concrete as found in compressive and split tensile
strength. The reason of increase in flexural strength due to addition of fibre is due to the bridging of cracks by
fibres and perfect bonding with aggregates. The flexural strength of beams when tested under UTM are
represented in Table 4. The optimum strength was at 1% fibre addition. But even at 3% addition of fibres,
flexural strength results more than control concrete.
Table 4. Flexural strength of coconut fibre reinforced beams
Sl no % of coconut fibre Flexural strength (28 days) in Mpa

1 0 4.59
2 0.5 5.03
3 1 5.54
4 3 4.61
1046
Fig.6. Flexural testing of Beam in UTM
Conclusion
• The cube compressive strength, split tensile strength of cylinder and flexural strength of beams were
casted and tested for 7 days and 28 days.
• The results reflected that the slump of fresh concrete decreases gradually with increase in percentage
addition of fibre in concrete as its water absorption is very high.
• The 28 days non-destructive test showed that up to 1 % use of coconut fibre increases the strength.
• Further, the 7 and 28 days cube compressive strength showed that 0.5% and 1 % use of coconut fibre
increased strength by 3.2% and 3%. The small discrete coconut fibres fill up the small pores and thus
increased the ability of resisting more loads.
• But with increase in percentage of fibre content the strength decreased due to the non homogeneity of
materials in concrete which leads in improper bonding of fibre with concrete and ultimately decreased
the strength.
• The split tensile strength of cylinder for a curing period of 28 days showed that when fibre content
increased from 0.5% to 3%, strength increased by 16%, 18.6% and 1%. The reason behind the decrease
in strength of concrete may be due to the presence of dust and impurities on the surface of fibres.
• Similarly for flexural strength of beam when tested under three point load showed that increase in fibre
content inhibit the cracks and hence the flexural behaviour showed improved results.
• The strength criterion in fibre reinforced concrete depends on the content of fibres, the orientation of
fibres and the bonding of fibres with the concrete.
Reference
1. Mulinari, D.R., Baptista, C.A.R.P.b, Souza, J. V. C.a, Voorwald, H.J.C.,” Mechanical Properties of
Coconut Fibers Reinforced Polyester Composites”(2011), procedia -10, pp. 2074-2079
1047
2. Bhutani s., Bansal S., Singh P., Dwivedi SB.,”Role of coir fibre in crack reduction in concrete”(2016),
International Journal of Mechanical And Production Engineering, Volume- 4, Issue-6, ISSN: 2320-
2092
3. Ali M., Liu A., Sou H., Chouw N., “Mechanical and dynamic properties of coconut fibre reinforced
concrete” (2012), Construction and building materials 30, 814-825
4. Ali M., “Seismic performance of coconut-fibre-reinforced-concrete columns with different

reinforcement configurations of coconut-fibre ropes” (2014), Construction and building materials
70, 226-230
5. Ali M., Chouw N., “Experimental investigations on coconut-fibre rope tensile strength and pullout
from coconut fibre reinforced concrete” (2013), Construction and building materials 41, 681-690
6. Ramaswamy H.S., Ahuja B.M., Krishnamoorthy S., “Behavoiur of concrete reinforced with jute, coir
and bamboo fibres” (1983), International journal of cement composite and lightweight concrete, vol. 5
7. Ramli M., Kwan H.W., Abas F.N., Strength and durability of coconut-fiber-reinforced concrete in
aggressive environments”(2013), Construction and building materials 38, 554-566
1048
Paper ID- 202
VIBRATION OF SIMPLY SUPPORTED LAMINATED COMPOSITE

PLATES
Leena Sinha1*, Amar Nath Nayak2, Sishir Kumar Sahu3

1,2
Department of Civil Engineering, VSS University of Technology, Burla, Sambalpur, Odisha, India
E-mail: sinha.leena@gmail.com, nayakan1964@gmail.com
3
Department of Civil Engineering, National Institute of Technology, Rourkela, Odisha, India
E-mail: sksahu.nitrkl@gmail.com
Abstract
This paper gives the experimental and numerical findings related to free vibration characteristics of
woven glass fibre epoxy laminated composite plates.The effects of aspect ratio, number of layers and
different fiber orientations on the natural frequencies of these plates are studied. The natural frequencies
obtained from the experimental analysis using FFT analyser are compared with those obtained from the
finite element analysis considering nine-noded isoparametric plate element. There is good agreement
between these results. The natural frequencies increase with increase in layer numbers & decrease with
increase in aspect ratio. The natural frequency of fiber orientation 45 degrees comes higher compared to 0
and 30 degrees. This study will help in providing adequate knowledge and confidence to the designer to
design the structures with sufficient accuracy.
Key words: FFT analyser, finite element analysis, free vibration, laminated composite plates
1. INTRODUCTION
In recent years, composite plates are utilized in a variety of very light weight and load carrying structural
parts of various aerospace, automotive and civil engineering applications because of their very higher
stiffness and strength to weight ratio , high resistance to impact damage, longer durability as compared to
other structures. A majority of structural elements used in these structures are plates/panels and most of
them are subjected to dynamic loadings during their service period where due to the resonant vibrations
maximum damage occurs. Therefore analysis of vibration is very important for the design of structures to
get its response in advance and accordingly the structural vibrations can be controlled by taking necessary
steps. So the free vibration study of composite plates has been a very useful problem while designing
mechanical, aerospace, civil applications.
A number of numerical and analytical researches on vibration analysis of composite plates are present in
existing study. But experimental researches are very less in literature. Bert et al. [1] analysed the free
vibration characteristics of anisotropic rectangular plates in clamped boundary condition by Rayleigh Ritz
energy method and experimentally. Clary et al. [2] investigated the vibration behavior of aluminum plates
reinforced with boron-epoxy composite materials experimentally and analytically. Cawley and Adams [3]
studied the vibration of free-free square and trapezoidal laminated composite CFRP plate experimentally
and using FEM. Crawley [4] studied the vibration characteristics of graphite-epoxy and graphite-epoxy-
aluminum hybrid plates and shells experimentally and numerically using FEM. Chai [5] analysed the free
vibration behavior of symmetrical stacked composite plates considering Rayleigh-Ritz method &
experimentally in clamped-simply supported boundary conditions. Chakraborty et al. [6] investigated the
vibration analysis of composite plates experimentally and numerically with FEM. Mishra and Sahu [7]
1049
studied the free vibration of Glass/Epoxy composite plates experimentally and numerically. Ratnaparkhi
and Sarnobat [8] studied vibration behavior of woven fiber glass/epoxy composite plate in free free
boundary condition experimentally and considered the effect of aspect ratio and fiber orientation.
Augustine et al [9] calculated the fundamental frequency of unidirectional and woven fabric composites
with both experimentally and numerically. Biswas & Ray [10] compared various shear deformation
theories with experimental modal analysis of hybrid laminates. It is found that experimental investigations
are very few in literatures along with various parametric studies. Therefore, to fill up the lacuna of the
research, the present paper describes the free vibration characteristics of composite plates along with
variation of no. of layers, weight fraction, and aspect ratio in simply supported boundary condition.
2. THEORY AND FORMULATION
A nine-noded isoparametric plate element is used to calculate the element stiffness matrix and
mass matrix for the plate. The plate element used for the problem has five degree of freedom. i.e.
for each node u, v, w, α andβ. In x, y and z-directions the displacements are u, v and w
respectively and along x and y-axes α,β denotes the rotations, respectively. By considering the
procedure available in literature (Nayak and Bandyopadhyay, 2002), the element stiffness matrix [K] is
given as follows:
[K]= (1)
The element mass matrix [M] is given by,
(2)
Where, [B] is the local derivatives of shape functions, the matrix corresponds to shape function is [N] and
the matrix representing inertia is [m]. The elasticity matrix of the element is [D]. The element stiffness
and mass matrices are obtained by using numerical integration using Gaussian quadrature with 2x2
reduced integration method. The global stiffness matrix and global mass matrix are obtained by
assembling the element stiffness and element mass matrix and the equation of undamped free vibration
problem takes the following Eigen value form:
(3)
(22 )
Where, ω is the natural frequency in rad/sec and {d} is the mode shape. The solution of equation (3) is
done by a subspace iteration technique to obtain the natural frequencies and mode shapes.
3.1. Materials Used & Casting Procedure of Laminated Composite Plates
The materials utilized for fabricating the laminated composite plates are woven fabric glass fiber, Epoxy,
Hardener, polyvinyl alcohol, Plastic sheet, and flat rigid platform. The composite plates are prepared from
woven glass fiber as reinforcement and epoxy as matrix. Hand layup technique is used for casting of
laminated composite plates.
3.2. Determination of Material Properties
The elastic properties are calculated experimentally by conducting tensile tests. For testing purpose
coupons are cut in longitudinal direction and at 45° to this direction, as specified in ASTM standard: D
3039/D 3039 M (2008). By the help of diamond cutter or hex blade as per requirement, the coupons are
cut. 3 samples of each coupon specimens are tested and the mean value is taken for elastic constants. The
coupon is fitted in INSTRON 3382 UTM to measure the elastic modulus. The rate of loading is 0.2
1050
mm/minute. The readings of load and the extension are taken digitally by a load cell and an extensometer.
From the recorded data, a graph of engineering stress and Strain is obtained and modulus of elasticity is
found out from this curve. Poisson’s ratio is calculated by the ratio of lateral strain to longitudinal stain
directly by using bidirectional strain gauge. The modulus of rigidity is obtained from the following
formula by Jones (1975) as follows:
The material properties of laminated composite plates obtained as above are presented in Table 1.
Table 1. Material properties of glass/epoxy laminated composite plates
Weight E1 GPa E2 GPa E45 GPa G12 = G13 = G23 υ12 ρ(kg/m3)
fraction GPa
50:50 16.07 16.07 8.915 2.814 0.25 1664
3.3. Modal Testing Procedure
First the composite plates are attached in simply supported boundary condition with the frame made of
iron as shown in Fig 1. The FFT analyzer, transducers, laptop, modal hammer, and cables are connected.
The pulse lab shop version-10 software key is attached to the laptop. Pulse lab shop is the platform used
in the computer to search the information from the FFT Analyzer and generate the required Frequency
response. For the excitation of the plate at a particular chosen point an impact hammer used. For FRF,
modal hammer was struck five times at a solitary point on the specimen. Perpendicular strokes should be
applied to the plates. The work of accelerometer, which is placed on the plate, by the help of honey bees
wax, is to receive the response. Then it sends the response to the FFT analyser. FFT analyser is utilized to
get a time varying signal from the accelerometer and convert it into a frequency based signal or FRF. FFT
analyzer is associated with a computer platform where we can see FRF. The pulse lab software is used to
get average output from the analyzer and that can be shown on the laptop. The response curves are plotted
on the screen and then the FRF are specifically estimated. After that by taking the cursor on the highest
point of the graph of FRF, the frequencies are seen. The FRF (pulse report) and coherence of test samples
obtained from FFT analyser are shown in Figs.2 & 3, respectively. In Fig 2, the various peaks represent
the various vibrations modes. The exactness of the measurement is obtained by making coherence nearly
equal to 1 which is shown in Fig 3.
4. RESULTS AND DISCUSSIONS
4.1 Convergence Study
The convergence study of the FEM formulation is done for composite plate with respect to the mesh size
for eight layered simply supported plates. The variations of frequency versus mesh size of the problem of
chakraborty (2000) are presented in Fig 4. From the figure, it is shown that the mesh converges at 8 X 8
for four side’s simply supported square plates.
4.2 Validation of FEM Model
The accuracy of the present FEM code is first developed by the comparison of the present vibration
results of laminated composite plate with those obtained by Crawley (1979). The first, second and third
modes of frequencies of the composite plates are got from the present code are compared with the FEM
results given by Crawley (1979) and are presented in Table 2.
1051
Table 2. Natural frequencies (Hz) of Cantilevered laminated composite Graphite/epoxy plate of present
FEM with Crawley. (1979).
Fiber orientations Mode of Observed Calculated Present FEM
frequency frequency frequency
0/45/-45/90/90/- 1st 48.6 55.58 55.44
45/45/0 2nd 169 175.4 174.13
3rd 303 345.3 344.03
0/0/30/-30/- 1st 234.2 261.9 260.82
30/30/0/0 2nd 362 363.5 360.71
3rd 728.3 761.8 752.69
4.3 Results of natural frequency obtained experimentally and numerically
In this section, there is comparison of experimental result with numerical result for free vibration analysis
of composite plates with the effect of no. of layers, aspect ratio, fiber orientation are studied in simply
supported boundary condition.
4.3.1 Effect layer numbers of laminated composite plates
For this study, 8, 12 and 16 layered composite plates are tested. Experimental and numerical frequencies
of three modes are considered. The experimental and numerical frequencies of 1st, 2nd and 3rd modes with
increase in no. of layers are presented in Fig 5. It is found out that the with the increase of number of
layers the frequencies of laminated composite plates increase.
4.3.3 Effect of fiber orientation
3 types of angle of orientations of fiber, i.e. (0/90)2s, (30/-30)2s, (45/-45)2s are taken. The experimental and
numerical frequencies of first three modes with different angle of orientations of fiber are shown in Fig
6.For the 1st, 2nd & 3rd mode, as the angle of orientation increases from o degrees to 45 degrees the natural
frequency increases. But for the 3rd mode frequency, the variation of frequency between 30 degree to 45
degree is less.
4.3.2 Effect of aspect ratio
Four values of aspect ratios i.e. a/b value (0.5, 1.0, 1.5 and 2.0) are taken for laminated composite plates.
The experimental and numerical first three mode frequencies with different aspect ratio are shown in Fig
7. It is found that as the aspect ratio increases the natural frequency decreases both experimentally and
numerically from aspect ratio 0.5 to 2.0 for all the three modes.
Fig. 1.Laminated Composite plates in Fig. 2. FRF of test sample

Simply supported boundary condition
1052
300
280
260
240
220 %1st mode
200 %2nd mode
frequency in Hz
180 %3rd mode
160
140
120
100
80
60
40
20
0
0 2 4 6 8 10 12 14 16 18
mesh size
Fig. 3. Typical coherence of test specimen Fig. 4. Convergence study of the simply supported
Laminated Composite Plate element
1200
650
1100 % 1st mode frequency exp
% 1st mode frequency num 600
1000 % 2nd mode frequency exp
% 2nd mode frequency num 550 % 1st mode frequency exp
900 % 3rd mode frequency exp % 1st mode frequency num
% 3rd mode frequency num 500 % 2nd mode frequency exp
% 2nd mode frequency num
800 % 3rd mode frequency exp
450
% 3rd mode frequency num
frequency
700
frequency
400
600
350
500 300
400 250
300 200
200 150
100 100
8 10 12 14 16 0/90 30/-30 45/-45
number of layers fiber orientation
Fig. 5. Graph of natural frequencies with respect to no. of layers Fig. 6. Graph of natural frequencies with respect to fiber
Orientation
900
850
% 1st mode frequency exp
800 % 1st mode frequency num
750 % 2nd mode frequency exp
% 2nd mode frequency num
700 % 3rd mode frequency exp
650 % 3rd mode frequency num
600
550
frequency
500
450
400
350
300
250
200
150
100
50
0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2
aspect ratio
Fig. 7. Graph of natural frequencies with respect to aspect ratio
5. CONCLUSION
The present study includes the free vibration characteristics of laminated composite plates using
experiment and numerical methods using FEM. The effect of various parameters like no. of layers, aspect
ratio and fiber orientations in simply supported boundary condition is investigated. From the above study
the following conclusions are drawn.
1. Natural frequencies obtained from both experimental study and numerical methods using FEM
are in good agreement.
2. The natural frequency increases with the increase of the number of layers due to increase in
stiffness.
3. The natural frequency decreases with the increase in aspect ratio from 0.5 to 2.0.
1053
4. With the increase of orientation of fiber from 0 degrees to 45 degrees the natural frequency
increases for all the three modes.
REFERENCES
[1] Bert, C. W. & Mayberry, B. L., “Free vibrations of unsymmetrical laminated isotropic plates with
clamped edges”, J. Compos. Mater. 3, 1969, 282–293.
[2] Clary, R. R., & Cooper, P. A., “Vibration characteristics of aluminum plates reinforced with boron-
epoxy composite material”. Journal of Composite Materials, 7(3), 1973, 348-365.
[3] Cawley, P., & Adams, R. D., “The predicted and experimental natural modes of free-free CFRP
plates”. Journal of Composite Materials, 12(4), 1978, 336-347.
[4] Crawley, E. F., “The natural modes of graphite/epoxy cantilever plates and shells”, J. Compos. Mater.
13, 1979, 195–205.
[5] Chai, G. B., “Free vibration of generally laminated composite plates with various edge support
conditions”, Compos. Struct. 29,1994, 249–258
[6] Chakraborty, S., Mukhopadhyay, M.and Mohanty, A. R., “Free vibration responses of FRP composite
plates: Experimental and numerical studies”, J. Reinf. Plast. Compos. 2000,535–551
[7] Mishra, I., & Sahu, S. K., “An experimental approach to free vibration response of woven fiber
composite plates under free-free boundary condition”. International Journal of Advanced Technology
in Civil Engineering, 2012, 1(2).
[8] Ratnaparkhi, S. U., & Sarnobat, S. S., “Vibration analysis of composite plate”. International Journal
of Modern Engineering Research (IJMER), 2013. 3(1).
[9] Alexander, J. and Augustine, B. S. M., “Free vibration and damping characteristics of GFRP and
BFRP laminated composites at various boundary conditions”, Indian J. Sci. Technol., vol. 8, 2015,
no. 12,
[10] Biswas, D. & Ray, C., “Comparative perspective of various shear deformation theories with
experimental verification for modal analysis of hybrid laminates” Journal of Vibration and Control,
Vol. 23(8), 2017, 1321–1333
[11] ASTM D 3039/D 3039 M (2008), Standard Test Method for Tensile properties Polymer Matrix
Composite Materials.
[12] Jones, R.M., (1975) “Mechanics of Composite Materials”, Mc Graw Hill, New York.
[13] Nayak, A.N, & Bandyopadhyay, J.N., “Free vibration analysis and design aids of stiffened
conoidal shells”. J Eng Mech, 2002, 128:419–27.
[14] Nayak, A. N, & Bandyopadhyay, J.N., “On the free vibration of stiffened shallow shells”. J
Sound Vib 2002, 255:357–82
1054
Paper ID- 205
Effect of Temperature Differential on Fatigue Performance of Marble

Dust added Concrete Pavement
Milind V. Mohod 1, Swati S. Nibhorkar2
1
Assistant Professor, Department of Civil Engineering, Prof Ram Meghe Institute of Technology & Research,
Badnera , P.O. Box 444701, Badnera, Amravati,(M.S.) ; e-mail: mvmohod@gmail.com
2
Assistant Professor, First Year Engineering Department, S.B. Jain Institute of Technology Management &Rsearch,
Nagpur, P.O. Box441501, Nagpur,(M.S.); e-mail: swatinibhorkar12@gmail.com
Abstract. Concrete pavement has proved to be a versatile solution in terms of

long term life, low maintenance while meeting the need of present world
without compromising the ability of the future demand to meet its own need.
With reference to the focus on sustainable construction practices, researchers
around the globe are keenly interested in motivating the use of waste/recycled
materials in concrete to reduce dumping of CO2 into the environment. Marble
dust is an industrial waste which is developed while cutting of marble blocks
into smaller blocks/slates to obtain the desired shape. During this process, about
25% original mass is lost in the form of dust/powder. This marble dust has
become a threat to agriculture and human health as it is being dumped in open
environment. Hence, utilization of marble dust in construction practices has
become essential to prevent environmental pollution. Many researchers have
recommended the use of marble dust in the concrete pavement as it enhances
mechanical properties of concrete. As concrete pavements are subjected to
dynamic loading conditions, the feasibility of marble dust in concrete pavements
has not been reported. Hence, an attempt is made to evaluate the performance of
marble dust added concrete pavement with reference to fatigue damage
guidelines provided by IRC 58:2015.Mechanical properties and realistic
temperature differentials are obtained through experimental approach for
evaluating fatigue performance. Present research reveals that the realistic
temperature differentials are not in agreement with the temperature differential
recommended in IRC 58:2015, which leads to overestimation/underestimation in
design thickness of pavement slab.
Keywords: Concrete pavement, Marble dust/powder, Fatigue performance, IRC

58:2015, Temperature Differential.
Introduction
In recent years, cement concrete pavements are being adopted in many new road projects in India
in view of their longer services lives, lesser maintenance requirement and smoother riding surface. Indian
road network of almost 3.5 million km comprising both paved and unpaved surfaces is the world's second
largest. A pavement is the layered structure on which vehicles travel. Rigid pavements are helpful to
reduce stresses on underlying soils. In this research paper, we have focused on a sustainable construction
practice in pavement by using waste/ recycled material (marble dust) in concrete. We explored the effect
of temperature differential in concrete and allied effect with reference to fatigue performance of marble
dust added concrete pavements. There are various methods/guidelines available to design rigid pavements
among which most of methods/guidelines are related to computer programs or FEA codes, where analysis
can do by inputting the various loading, environmental parameter and properties of material. IRC 58 2015
method/guidelines have been adopted in present study to perform cumulative fatigue damage study for
obtaining safe thickness of pavement slab. Input data related to material plays a vital role while deciding
the safe thickness. If these properties are not known at design stage, the design and analysis done by any
analysis and design tool will be approximate. Hence, initially properties of marble dust added concrete
and their effect on mechanical properties have been obtained, so that they may be used for deciding safe
PQC (Pavement Quality Concrete). Analysis and design theory of rigid pavements are presented followed
by utilization of measured properties of marble dust in concrete. Fatigue analysis guidelines provided by
IRC 58 2015 have been adopted to demonstrate the effectiveness of marble dust added concrete in
reducing PQC thickness considering the effect of temperature differential available with us, the one
1055
recommended by IRC 58 2015 and the other realistic temperature differential recorded from temperature
in concrete pavements experimentally.
Marble dust
Marble has been commonly used as a building material since the ancient times. The industry’s
disposal of the marble powder material, consisting of very fine powder, today constitutes one of the
environmental problems around the world. Marble blocks are cut into smaller blocks in order to give them
the desired smooth shape. During the cutting process about 25% the original marble mass is lost in the
form of dust [1]. Marble dust is settled by sedimentation and then dumped away which results in
environmental pollution, in addition to this forming of dust in summer is threatening both agriculture and
public health. Therefore, utilization of the marble dust in various industrial sectors especially the
construction, agriculture, glass and paper is important. By considering all the aspects related to safe
disposal of marble dust and effective utilization in construction sector, this research is dedicated to
prepare a mix of M30 grade concrete obtain various mechanical properties and obtaining the realistic
temperature on top and bottom of concrete pavement. It is desirable that use of marble dust in concrete
will lead to economize the safe PQC thickness of the slab.
Analysis and Design theory

Highway pavements are divided in two broad categories: Flexible pavement and Rigid pavement.
Flexible pavement is a multilayered structure resting over the soil subgrade, with inferior quality material
at the base. In flexible pavements, wheel loads are transferred by grain-to-grain contact of the aggregate
through the granular structure (e.g. bituminous road). On the other hand rigid pavement is made up of
concrete, occasionally over a base course. In rigid pavements, wheel loads are transferred to sub-grade
soil by flexural strength of the pavement and the pavement acts like a rigid plate (e.g. cement concrete
roads). The rigid pavement because of its rigidity and high modulus of elasticity tends to distribute the
load over a wide area of a soil by slab made by concrete. Thus, concrete flexural strength is of prime
importance in the design of rigid pavement to enhance its structural capacity.
Forces Acting on Rigid Pavement

Various stresses act on rigid pavement due to wheel loads, cyclic changes in the temperature
(warping and shrinkage or expansion), changes in moisture and volumetric changes in the subgrade or
base course. Due to all these changes intensity of stresses varies widely and deform slab. The stresses
acting on a rigid pavement are, i) Wheel load stresses and ii) Temperature stresses.
i. Wheel load stresses: Wheel load of vehicle acting on a concrete slab of pavement resting in
a perfect contact with an elastic subgrade will cause increase in flexural stress of slab. The
slab produces tensile and compressive forces due to wheel load. As concrete is weak in
tension and strong in compression, usually tensile stresses is critical due to load on slab.
Equations for calculations of stresses in such a slab were first developed by Westergaard and
later extended by other researcher [2]. Westergaard considered three different loading
conditions for calculation of stresses, Interior loading, Edge loading and Corner loading.
ii. Temperature Stresses: Daily changing temperature causes significant thermal stress and
deformations in concrete pavement with laterally fixed planes. These stresses can exceed
permissible values in the case of incidental strong thermal differences. Due to the
temperature differential between the top and bottom of the slab, curling stresses (similar to
bending stresses) are induced at the bottom or top of the slab. Due to the contraction of slab
due to shrinkage or due to drop in temperature tensile stresses are induced at the middle
portion of the slab [3]. Temperature thus tends to produce two types of stresses in a concrete
pavement. These are warping stresses and frictional stresses.
Design Technique Based on Fatigue Cumulative Damage:

Westergaard gave a conventional theory to analyse rigid pavement. New design technique for
rigid pavement is developed on the basis of Westergaard findings with certain modification, IRC 58:2015
method/guidelines are drafted. The severest combination of different factors that induce the maximum
stress in the pavement will give the critical stress condition. Designing of pavement thickness, flexural
stress due to combine action of wheel loads and temperature differential between top and bottom fibers of
concrete pavement is also considered. 1056
a) Bottom- up cracking
In rigid pavement flexural stresses occur at the bottom layer is maximum during day hours. When
the axle loads act midway on the pavement slab while there is positive temperature gradient. Single
axles cause highest stress followed by tandem and tridem axles respectively. Spacing between
individual axles for tandem and tridem axles varies from 1.30 m to about 1.40 m (IRC 58: 2015).
Figure 1: Bottom up cracking(IRC 58 : 2015)
b) Top-down cracking
During the night hours, the top surface is cooler than the bottom surface and the ends of the slab
curl up resulting in loss of support for the slab. Temperature tensile stress are caused at the top because of
the restraint provided by the self- weight of concrete and by the dowel connections Axle loads when
placed close to transverse joint when there is negative temperature gradient during night hours , causes
high flexural stresses in the top layer leading to top- down cracking.(IRC 58: 2015)
Figure 2: Top- down cracking (IRC 58: 2015)

Experimental Data:
Cement:
The cement used was Pozzolona Portland cement (PCC) grade 53 and confirming to IS 1489-1-
1991. Initial and final settling times of cement were 158 min and 345 min, respectively.
Aggregate:
Aggregates are the important constituents in concrete. They give body to the concrete, reduce
shrinkage and affect economy. One of the most important factors for producing workable concrete is a
good gradation of aggregates.
I. Fine Aggregate:
Those fractions from 4.75 mm to 150 micron are termed as fine aggregate. The river sand is
washed and screened, to eliminate deleterious materials and over size particles. Good quality WARDHA
river sand was used as a fine aggregate.
1057 3: Wardha sand

Figure
II. Coarse Aggregate:

The material whose particles are of size as retained on I.S Sieve No. 480 (4.75 mm) is termed as
coarse aggregate .Coarse aggregate used were 15mm downgraded and 30 mm downgraded.
Figure4: Coarse Aggregate

Marble dust
In India, marble processing industry generates around 7 million tons of wastes mainly in the form
of powder during sawing and polishing processes. These are dumped in the open which pollute and
damage the environment. In this research we used marble dust from Amravati MIDC area which is
dumped as a waste material.
Figure 5: Marble waste sludge dumped

Super plasticizer
Plasticizer is used for better workability of concrete. During the trial mix the workability
was found to be very less so we used plasticizer. It reduces the w/c ratio and ultimately increases the
strength of concrete. Plasticizer is added at rate 5ml per litre of water. The plasticizer we used wasAC-
MENT-BV-430-A3, Super Plasticizer for pump able concrete on PC Base, an Apple chemie industries
product.
TESTS ON MATERIAL:
For mix design calculations we required certain data of materials like specific gravity, sieve
analysis and water absorption. For which we had conducted the different tests on material which are
described below.
Fineness Test:
The fineness of cement has an important bearing on the rate of hydration and hence on the
rate of gain of strength and also on the rate of evolution of heat.
By Sieving:-
Table No.1 Fineness of cement
Sr. Weight sample Weight sample % Weight retained
No taken (gm) retained on sieve
1 100 gm 0 0
2 100 gm 2 2
3 100 gm 1 1
Average retained weight = 1

Standard Consistency Test
For finding out initial setting time, final setting time and soundness of cement, and strength a
parameter known as standard consistency has to be used.
Table No 2: Standard Consistency
Sr. Water added for different trials in Penetration of plunger
No. Percentage from bottom in mm
1 30 19
1058
2 35 12
3 40 5
Standard Consistency in percentages = (total water for standard consistency)×100/400

= 160 × 100/400
=40
Standard Consistency is 40%
Specific Gravity Test

Table no 3. Specific Gravity of Coarse Aggregate
Test particulars sample sample 2 sample 3
1(gm) (gm) (gm)
Empty weight of pycnometer W1(gm) 596 596 596
Weight of pycnometer oven dry soil W2 (gm) 1206 1221 1191
Weight of oven dry silts=W2-W1(gm) 610 625 595
Weight of pycnometer +water +soil W3(gm) 1904 1895 1870
Weight of pycnometer + water W4 (gm) 1502 1502 1502
Specific gravity 2.66 2.69 2.63
Average value of specific gravity 2.66
Similarly Average value of specific gravity fine aggregate observed is 2.96
Sieve Analysis Test

This is the name given to the operation of dividing a sample of aggregate into various
fractions each consisting of particles of same size. The sieve analysis conducted to determine the particle
size distribution in a sample of aggregate, which we called gradation.
Table No.4 Sieve analysis

Sieve Retained on each sieve Cumulative Retained Passing through
Wt (gm) % Wt (gm) %
80mm 0 0 0 2000 100
40mm 0 0 0 2000 100
20mm 0 0 0 2000 100
10mm 0 0 0 2000 100
4.75mm 278 13.9 13.9 1722 86.1
2.36mm 358 17.9 31.8 1364 68.2
1.18mm 620 31 62.8 744 37.2
600µ 360 18 80.8 384 19.2
300µ 332 16.6 97.4 52 2.6
150µ 84 4.2 101.6
Pan 18 0.9 102.5
Total 2000 gm 100% 490.8
Setting Time Test

The initial and final setting time is calculated by using vicat’s apparatus
Table No.5 Setting time of cement
Sr.No Sample no. Results
1. Weight of sample 400 gm
2. Water added in (ml) 160
3. Time when water added 12.50 PM
4. Time of initial set 3:28 PM
5. Time of final set 6:35 PM
6. Initial set in min(4-3) 158 min
7. Final set in min1059
(5-3) 345 min
MIX DESIGN
The method we used is popularly known as 'AMBUJA METHOD'. This method has been adopted
on several sites in and around Mumbai and has given technically and economically good results. The
study carried on the basis of these following objectives:1) To determine the mechanical properties of
recycled marble dust in concrete.2) To determine the optimum value of marble dust needed for
improvement in compression, flexural and split tensile strength over the plain traditional mix 3) To
develop the cost saving concrete, savings fortune in large projects. 4) To develop feasible waste
management for waste material.
In this method, for different concrete specification, tables from IS 456-2000 were referred. The
concrete was designed and tested for different proportion of marble dust individually. The different
proportions for testing for marble dust were selected upon reviewing various research papers as available
in literature. The selected dosage for marble dust was 5%, 10%, 15% and 20% to be replaced with fine
aggregate. The specimens were casted in batches individually for marble dust. The aim is to calculate the
optimum dosage for improvement of qualities in concrete. The compressive, flexural and split tensile
strength is calculated for specimens for 7 days, 14 days, and 28 days. Design mix for M30 grade concrete
was prepared by partially replacing fine aggregate with marble dust is replaced by weight of fine
aggregate.
For calculating the various strength and properties of concrete tests were performed on the
specimens viz., cubes of size 150mm × 150mm, beams of length 500mm and cross section of 100mm, and
cylinder of diameter 150mm of length. Another batch of slabs/sections consisting of marble dust added
concrete mix M30 was prepared of sizes having cross section 150 × 150mm having varying heights of
150mm, 200mm, 250mm, 300mm, 350mm and 400mm. Temperature at various depth was calculated with
the aim to get our experimental temperature differential for pavement design and the results of CFD are
compared with IRC given temperature differential values.
TESTS FOR PROPERTIES ON CONCRETE SPECIMEN

Compressive Strength Test
From the result it is observed that Marble dust with pozzolanic properties impart technical advantages
along with larger quantities of cement replacement to be achieved. The compressive strength is found
increasing with the addition of marble dust, the optimum dosage for marble dust replacement is found to
be 15%. The compressive strength of concrete can be calculated using the following formula
Compressive strength = (Load in N/ Area in mm2 ) =…N/mm2 ...(eq. no. 1)
Figure 6: Compression testing machine

Split Tensile Strength Test
The test was performed on CTM (compression testing machine). The figure 7 shows the setup for
split tensile strength test on CTM. The task of this test was performed to find the increase and differences
of strength of concrete according the increasing percentage of marble dust %, 10%, 15% and 20%. in the
concrete. The indirect tensile strength test was conducted in the laboratory after the concrete specimens
were cured for 7, 14 and 28 days respectively. The optimum dosage of marble dust, for best split tensile
strength is 15%.
1060
Figure7: Cylinder cracked during split tensile strength test, Split tensile test on cylinder
Flexural Strength Test

Flexural strength of concrete is a measure of its ability to resist bending. Flexural strength can be
expressed in terms of 'modulus of rupture'. Types of loadings are Third -point loading and Centre-point
loading. Centre-point loading is when the entire load is applied at the center of span. The maximum stress
is present only at the center of the beam. Third-point loading is the when half the load is applied at each
third of span length. Maximum stress is present over the center 1/3 portion of beam. We have applied
loading in this manner. Use of three-point loading is preferred for determining flexural strength of
concrete is shown in Fig.9. It is observed that with the addition of marble dust optimum dosage is found
out to be 15%.
Figure 8: Central-Point Loading Figure 9: Third-Point Loading
Result and Discussion

The experimental tests were carried out to obtain the mechanical properties of marble dust
added concrete to use them in the pavements. Effect of replacing the fine aggregate by locally
available marble dust on concrete was studied. The replacement of materials was increase in
percentages by weight of fine aggregate and cement respectively. Observation for 7, 14 & 28 days
curing period were recorded and presented in the form of tables and graphs. Cumulative fatigue
damage (CFD) was also performed on slabs/section cased of marble dust added concrete. The
temperature variations of 24 hours were recorded after each hour and temperature at top, mid and
bottom were recorded to calculate the temperature differential. After incorporating the temperature
variation, flexural strength & compressive strength results in CFD data analysis chart/guidelines
provided in IRC 58 2015, we get the result regarding obtaining safe pavement thickness.
For Marble dust added concrete (M30 mix): Marble dust replacement with fine aggregate is done in
increasing percentages in concrete and from the tests results obtained for compressive strength, flexural
strength and split tensile strength. The percentage replacement was 0, 5, 10, 15 & 20 with fine aggregate.
Compressive strength for marble dust added concrete
Table no.6 Compressive strength for marble dust added concrete M30
% of marble dust replacement Compressive strength
with fine aggregate (Mpa) 28 days
0 40
5 44.44
10 46.22
15 48
20 1061 39.11
Figure 10: Compressive strength for marble dust added concrete

Flexural strength for marble dust added concrete
Table no.7 Flexural strength for marble dust added concrete M30
% of marble dust replacement Flexural strength (Mpa) 28
with fine aggregate days
0 12.7
5 16.76
10 18.07
15 19.27
20 17
Flexural Strength
25
20
Flexural Strength
15
10 Flexural…
5
0
0 5 10 15 20
% of Marble Dust
Figure11: Flexural strength for marble dust added concrete

Table no.8 Split tensile strength for marble dust added concrete M30
% of marble dust replacement Tensile strength (Mpa)28 days
with fine aggregate
0 3.11
5 3.67
10 3.96
15 4.25
20 3.25
1062
Figure 12: Split tensile strength for marble dust added concrete
Deflection of beam for marble dust added concrete for 28 days
Figure13: Deflection of beam for marble dust added concrete
From the result obtained experimentally different graphs are plotted like compressive strength, flexural
strength, split tensile strength and deflection of beam. It is observed that from result 15% of marble dust
added in concrete as a fine aggregate improve strength and lowers the deflection.
VARIATION OF TEMPERATURE IN CONCRETE PAVEMENT

For 150 mm depth concrete slab/section
Temperature Variation for 150 mm depth

45
Temperature in Celsius
40
35
30
25
20
15
10
12 pm - 0th hr
12 am 12th hr
12 pm - 24th hr
1
2
3
4
5
6
7
8
9
10
11
13
14
15
16
17
18
19
20
21
22
23
Time in hours
0 mm 50 mm 100 mm 150 mm
Figure 14: Temperature Variation for 150 mm Thick Slab
Temp. Gradient = Highest temperature at top for 24 hours – Lowest temperature at bottom for
24 hours
Positive temperature gradient for 150mm slab = 43° - 24° = 19°C
1063

60
55
50
45
40
35
30
25
20
12 pm - 0th hr
12 am 12th hr
12 pm - 24th hr
1
2
3
4
5
6
7
8
9
10
11
13
14
15
16
17
18
19
20
21
22
23
Time in hours
0 mm 65 mm 130 mm 200 mm
Figure15: Temperature Variation for 200 mm Thick Slab
24 hours

60
55
50
45
40
35
30
25
20
12 pm - 0th hr
12 am 12th hr
12 pm - 24th hr
1
2
3
4
5
6
7
8
9
10
11
13
14
15
16
17
18
19
20
21
22
23
Time in hours
0 mm 85 mm 170 mm 250 mm
Figure16: Temperature Variation for 250 mm Thick Slab
24 hours
1064

60
50
40
30
20
12 pm - 0th hr
1
2
3
4
5
6
7
8
9
12 am 12th hr
12 pm - 24th hr
10
11
13
14
15
16
17
18
19
20
21
22
23
Time in hours
0 mm 100 mm 200 mm 300 mm

24 hours

60
55
50
45
40
35
30
25
20
12 pm - 0th hr
12 am 12th hr
12 pm - 24th hr
1
2
3
4
5
6
7
8
9
10
11
13
14
15
16
17
18
19
20
21
22
23
Time in hours
0 mm 115 mm 230 mm 350 mm
24 hours
Positive temperature gradient for 150mm slab = 56° - 32.5° =23.5°C
1065

60
55
50
45
40
35
30
25
20
12 pm - 0th hr
12 am 12th hr
1
2
3
4
5
6
7
8
9
10
11
13
14
15
16
17
18
19
20
21
22
23
12 pm - 24th hr
Time in hours
0 mm 130 mm 260 mm 400 mm

24 hours
Positive temperature gradient for 150mm slab = 56° - 32.5° =23.5°C
Table no. 9 Temperature Differentials in C

Thickness in mm/
150 200 250 300 350 400
C
Exp. Data result/
Temperature 19 22 23 22 23.5 23.5
gradient
IRC 58-2015 17.3 19 20.3 21 21 21
Difference 1.7 3 2.7 1 2.5 2.5
CFD Analysis of Normal Concrete

300
CFD Value for different temperature
250
0 Temp IRC
200
17.3 Temp IRC
150
19 Temp IRC
100 20.3 Temp IRC
50 21 Temp IRC
0
140 150 160 170
Thickness of Slab in mm
Figure 20: CFD Analysis of Normal Concrete under recommended temperature gradient by IRC 58-2015
1066
CFD Analysis of Marble Dust Added Concrete

3000

2500
2000 0 Temp IRC
1500 17.3 Temp IRC

19 Temp IRC
1000
20.3 Temp IRC
500
21 Temp IRC
0
100 110 120 130
Figure 21: CFD Analysis of Marble dust added concrete under recommended temperature gradient by IRC
58-2015
CFD Analysis of Normal Concrete

120
100
80 19 Temp Exp.
60 22 Temp Exp
40 23 Temp Exp
23.5 Temp Exp
20
0
150 160 170 180
Figure 22: CFD Analysis of Normal Concrete under observed temperature gradient (Experimental Value)
CFD Analysis of Marble Dust Added Concrete

25000
20000
19 Temp Exp.
15000
22 Temp Exp
10000 23 Temp Exp
5000 23.5 Temp Exp
0
100 110 120 130
Figure 23: CFD Analysis of Marble dust added concrete under observed temperature gradient
(Experimental Value)
1067
Conclusions
In this research paper we studied the usage of waste marble dust materials and their
behavior, different mechanical properties and their effects on the concrete and we also studied about
fatigue behavior of cement concrete pavement. The CFD analyses for various temperature differentials of
different replacement with respect to PQC have been made. By this analysis the comparison of PQC and
concrete having waste materials replacement has been made.
1. From the experimental results, we can conclude that at 15% replacement of marble dust
by fine sand shows the highest compressive strength,
2. The flexural strength is found to have highest value for, 15% replacement of marble dust
by fine sand
3. At 15% replacement of marble dust by fine sand shows the highest split tensile strength
4. From the experimental results we can conclude that, The optimum percentage of
replacement of fine aggregate with waste marble dust comes out to be 15% by weight.
5. Using waste marble dust in concrete mix proved to be very useful to solve environmental
problem and produce green concrete. Therefore, it is recommended to re-use waste marble
dust in concrete to move toward sustainable development in construction industry.
6. Due to high fineness of marble dust, it proves to be effective in assuring good
cohesiveness of concrete in the presence of super-plasticizing admixture (AC- MENT-
BV-430-A3), provide that water to cement ratio was just adequate.
7. By CFD analysis as marble dust shows desirable results for fatigue, the replacement of
fine sand with marble dust is better.
8. It has been observed that temperature gradient recommended by IRC 58 and those
obtained by experimentally differ owing to which safe slab thickness is affected. In
additions to this, marble dust added concrete in spite of having more temperature
differential gave lesser thick slabs as compared to conventional concrete.
References
1. Rishi Gupta et al, “Use of Polypropylene Fibre Reinforced Concrete as a construction Material
for Rigid Pavements: Part-II”, ICE journal , November 2016.
2. Teller, L. W. and E. C. Sutherland, “The structural design of concrete pavement part-II”, public
roads, volume 15, No. 9,1935.
3. PiotrMackiewicz “Thermal stress analysis of jointed plane in concrete pavements” Article in
Applied Thermal Engineering · December 2014 DOI: 10.1016/j.applthermaleng.2014.09.006
4. BaharDemirel “The effect of the using waste marble dust as fine sand on the mechanical
properties of the concrete Department of Construction, Faculty of Technical Education, Firat
University, Elazig, 23119, Turkey.
5. Ahmed N. Bdour 1) and Yahia A. Al-Khalayleh 2) “Innovative Application of Scrap-tire Steel
Cords in Concrete Mixes” Assistant Professor and Assistant Dean of Engineering College,
Department of Civil Engineering, The Hashemite University, P.O.Box:150459, Zarqa 13115,
Jordan. E-mail: bdour@hu.edu.jo(Corresponding Author) 2) Assistant Professor, Department of
Civil Engineering, The Hashemite University, P .O.Box:150459, Zarqa 13115, Jordan
6. Constantia Achilleos, DiofantosHadjimitsis, KyriacosNeocleous , KyprosPilakoutasPavlos O.
Neophytou and SteliosKallis ” Proportioning of Steel Fibre Reinforced Concrete Mixes for
Pavement Construction and Their Impact on Environment and Cost” Sustainability 2011, 3, 965-
983; doi:10.3390/su3070965
7. Nishant Malay, Manish Kumar and JituKujur, “effect of partial replacement of cement by marble
dust on compressive strength and carbonation of ppc concrete” Birla Institute of Technology,
Mesra, Ranchi, Jharkhand, India Conference Paper · November 2015DOI:
10.13140/RG.2.2.12150.40003
8. V. M. Sounthararajan and A. Sivakumar, “effect of the lime content in marble powder for
producing high strength concrete”Structural Engineering Division, SMBS, VIT University,
Vellore, Tamil Nadu, India VOL. 8, NO. 4, APRIL 2013 ISSN 1819-6608 ARPN Journal of
Engineering and Applied Sciences.
9. Hanifi Binici1, Hasan Kaplan2 and Salih Yilmaz2 “influence of marble and limestone dusts as
additives on some mechanical properties of concrete”KahramanmarasSutcu Imam University,
Department of Civil Engineering, Avsar campus, Kahramanmaras, TurkeyPamukkale University,
1068
Department of Civil Engineering,Denizli, Turkey. Scientific Research and Essay Vol. 2 (9), pp.
372-379, September 2007Available online at ISSN 1992-2248 © 2007 Academic Journals.
10. E. Bacarji , R.D. Toledo Filho , E.A.B. Koenders , E.P. Figueiredo , J.L.M.P. Lopes “E. Bacarji ,
R.D. Toledo Filho , E.A.B. Koenders , E.P. Figueiredo , J.L.M.P. Lopes ” Article in Construction
and Building Materials. August 2013 DOI: 10.1016/j.conbuildmat.2013.03.032
11. k.pulkit, h. agarwal, v. ganeshan “durability studies of concrete using partial replacement of
cement by stone slurry” proceedings of national conference : civil engineering conference –
innovation for sustainability (cec – 2016), 09h - 10thseptember, 2016
12. nishantmalay, manishkumar and jitukujur “effect of partial replacement of cement by marble dust
on compressive strength and carbonation of ppc concrete” birla institute of technology, mesra,
ranchi, jharkhand, india conference paper · november 2015doi: 10.13140/rg.2.2.12150.40003
13. sarbjeetsingh , anshumantiwari , ravindranagar “comparative assessment of effects of sand
&cement replacement in concrete by marble dust&intern deriving an optimum design mix for
concrete paver blocks” conference paper.December 2015
14. prof. p.a. shirulea*, ataurrahmanb ,rakesh d. gupta “partial replacement of cement with
marbledust powder” bcdept. of civil engineering, ssbt’scoet, bambhori, jalgaon, maharashtra –
425001 international journal of advanced engineering research and studies e-issn2249–8974
15. S. Flrat, G. Yllmaz, A. T. Cömert, and M. Sümer “Utilization of Marble Dust, Fly Ash and
Waste Sand (Silt-Quartz) in Road Subbase Filling Materials” Article inKSCE Journal of Civil
Engineering · November 2012 DOI: 10.1007/s12205-012-1526-4Received March 29, 2011 /
Revised July 18, 2011 / Accepted December 6, 2011
16. HuseyinY lmazAruntas, Metin Guru , Mustafa Day , _IlkerTekin “Utilization of waste marble
dust as an additive in cement production” Article in Materials and Design · September 2010DOI:
10.1016/j.matdes.2010.03.036
1069
Fatigue Resistance of Recycled Steel Fibres (Discarded Vehicle Tyre Steel

Fibres) Concrete Pavement
M.V.Mohod, PhD scholar1 and Dr K.N.Kadam2
1
Civil Engineering Department, PRMIT&R, Badnera, Amravati, Maharashtra, India,
Email id- mvmohod@gmail.com
2
Department of Applied Mechanics, GCOEN, Nagpur, Maharashtra, India,
Email id- kadamkshitija7@gmail.com
Abstract: Concrete pavements are exposed to repetitive (Cyclic traffic) loads

throughout its lifespan resulting in the instigation of cracks. Propagation of these
cracks leads to permanent failure of concrete pavement due to fatigue. The
fundamental cause of the occurrence of cracks in concrete pavement is the poor
resistance of the concrete pavement to bending, tension and cracking [1]. Many
design guidelines such as PCA (1984), IRC: 58(2015), NCHRP Report (2004)
considers fatigue cracking as one of the major structural distresses [2].
Industrially produced steel fibres are added to plain concrete to improve its post-
cracking flexural strength and fatigue resistance [3].Even though industrially
produced steel fibres safeguard concrete pavements by restricting cracks they are
not preferred due to higher material cost [4]. An alternative of using recycled
steel fibres recovered from discarded vehicle tyres is a cheaper fibre solution
with notable environmental benefits. The steel wires/fibres from worn out tyres
are used as reinforcement in concrete pavement, solves a dual purpose of
obtaining robust composite material at cheaper cost and reducing breeding sites
for mosquitoes in the environment. Research work pertaining to the use of
recycled steel fibres have been reported [5,3] but there is the lack of studies on
exploring fatigue performance of recycled steel fibres reinforced concrete
pavement with reference to IRC 58:2015 guidelines. Hence, this research work
aims to explore fatigue behaviour of recycled steel fibres reinforced concrete
pavement, which would be efficient, cost-effective and environmentally friendly.
Introduction
India is a developing country; causing extensive use of tyres in automobile industry has resulted in
accumulation of large quantities of used tyres that have to be disposed at the end of their useful life. The
use of recycled waste tyres is less in construction industry which needs to get improve. The steel fibres
present in tyres are useful which can be added to the concrete to improve its properties. On the other hand
the use of industrial steel fibres is increasing rapidly with different size and shape. Influence of steel fibres
on concrete when added to concrete, steel fibres improves its qualities by increasing its energy absorption
capacity. Also, steel fibres increase the flexural strength and compressive strength of concrete. A
comparison of concrete containing steel fibres with conventional concrete it is observed that concrete
containing steel fibres gives more strength. It has also been reported that adding steel fibres in concrete
prevents the macro-cracks as well as provides some resistance to dynamic and impact load. The main
objective is to find alternative source of raw products in concrete technology, as this would reduce an
adverse environmental impact of production of concrete component, while at the same time preserving
natural resources. In recent years a significant research effort focused on the suitability and efficiency of
using various recycled materials and industrial by product as sustainable concrete constituents. In this
regards, one of the most promising solution from both environmental and technical point of view is to
reuse waste tyres from which two main constituents can be potentially obtained from recycling. As the
matter of fact, approximately 1.4 billion tons vehicle pneumatics is sold annually in the world and,
consequently many of them can be categorized as "end of life" tires. Therefore, there are strong
motivations for investigating solutions capable to reduce the negative environmental impacts of waste
tyres after their service life. In this essence the abundant availability of waste tyre steel fibre can be
utilized as an effective additional or replaceable material for concrete which will be beneficial for all the
circumstances.
IRC 58 2015 - Design chart for CFD analysis

The IRC 58 aims at rationalizing the design procedure by bringing it at par with current trends in design
considering cumulative fatigue damage (CFD) due to the combined effect of load and pavement
1070
temperature. In IRC 58 code, region wise maximum temperature differentials are recommended for
various pavements thicknesses. Analysis is made for cumulative fatigue damage (CFD) by using
recommended and experimental temperature differential for PCC and modified concrete.
Methodology
The concrete is made with concrete wastes which are eco-friendly so called as Green concrete.
Green concrete is a revolutionary topic in the history of concrete industry. Considerable research has been
carried out on the use of various industrial by-products and micro-fillers in concrete. The main concern of
using pozzolanic wastes was not only the cost effectiveness but also to improve the properties of concrete,
especially durability. Using the waste products steel tyre fibres we have developed a new type of concrete
to effectively study the improvement in the mechanical properties of the new developed concrete,
planning, procedures and method must be wisely chosen. The criteria to assess the mechanical properties
are based on the activities to plan and preparation, which carried out before the testing of fresh and
hardened properties of fibre reinforced concrete. These activities are: a) Aggregate and cement testing b)
Sieve analysis (aggregate grading). c) Fibre volume dosage rate. d) Mix Design e) Preparation of test
specimen. f) Concrete mixing, Curing and temperature study
Material Properties/Data:
Cement:
The cement used was Pozzolona Portland cement (PCC) grade 53 and confirming to IS 1489-1-
1991. Initial and final settling time of cement was 158 min and 345 min, respectively.
Aggregate:
Aggregates are the important constituents in concrete. They give body to the concrete, reduce shrinkage
and economy. One of the most important factors for producing workable concrete is a good gradation of
aggregates.
I. Fine Aggregate:
Those fractions from 4.75 mm to 150 micron are termed as fine aggregate. The river sand is wash and
screen, to eliminate deleterious materials and over size particles. Good quality WARDHA river sand was
used as a fine aggregate.
Fig 1: Wardha sand

II. Coarse Aggregate:
The material whose particles are of size as retained on I.S Sieve No. 480 (4.75 mm) is termed as coarse
aggregate.
Fig 2: Coarse Aggregate

Tyre steel fibres:
Fibres vary in types, geometry, properties and availability in construction industry. Most
common types of fibres are steel fibres, glass fibres, and polypropylene fibres. There usage may alter in
concrete for different applications. The fibres are selected from their properties like, effectiveness, cost
and availability. We selected high tensile strength steel fibres obtained from dismantled scrap tyres. The
waste management of used tyres is of major concern for many environmental bodies and agencies
worldwide. Tyre production is increasing every year due to the increase of vehicle sales. The generation
and disposal of waste are inherent to life itself and have presented very serious problems to the human
community in Malaysia. Recently, some research has been done for various use of recycled steel. Shape
and impermeability of tyres helps to hold water for long periods thus provide breeding sites for mosquito.
Waste tyres also pose a serious fire hazard since they are easy to ignite.
1071
Fig. 3 waste tyre site Fig 4 Tyre steel fibres
Super plasticizer
Plasticizer is used for better workability of concrete. During the trial mix the workability was
found to be very less so we used plasticizer. It reduces the w/c ratio and ultimately increases the strength
of concrete. Plasticizer is added at rate 5ml per litre of water. The plasticizer we used wasAC-MENT-BV-
430-A3, Super Plasticizer for pumpable concrete on PC Base, an Apple chemical industries product.
Concrete mix design

Mix design for M30
1) Grade designation = M30
2) Type of cement = Pozzolana Portland Cement 53 Grade
3) Max nominal size of aggregate = 20 mm
4) Minimum cement Content = 240 kg/m3
5) Water-Cement Ratio = 0.45
6) Standard deviation = 5 N/mm2
7) Crushed Sand confirming to Zone 1
8) Exposure Condition = Moderate
9) Workability = Low (25 mm -75 mm )
10) Target mean Strength = 38 N/mm2
After various calculations and corrections of oversize, specific gravity the final quantity of materials are
calculated as
1) Cement = 377.07 kg
2) Water content = 170.1 kg/cum
3) Fine Aggregate = 1075 kg
4) Coarse aggregate = 20 mm = 671.60 kg/cum
= 10 mm = 193.45 kg/cum
By using this quantity of material a mix design of concrete is prepared .This mix design helps in achieving
a desire workability, strength and durability for concrete work. For making the mix design more workable
admixture is also used. The experimental tests were carried out to obtain the mechanical properties of tyre
steel fibre added concrete to use them in the pavements. Effect of replacing the cement by locally
available tyre steel fiber on concrete was studied. The replacement of materials was increase in
percentages by weight of cement respectively. Observation for 7, 14 & 28 days curing period were
recorded and presented in the form of tables and graphs. Cumulative fatigue damage (CFD) was also
performed on concrete. In CFD analysis concrete of size 0.3m X 0.3m with varying thicknesses of
150mm, 200mm, 250mm, and 300mm were prepared. The temperature variations of 24 hours were
recorded after each hour and temperature differential was calculated. After obtaining the temperature
variation, flexural strength & compressive strength results in CFD data analysis chart available in IRC 58
2015, we get the result regarding safe pavement thickness.
Test Result for Ingredients of Concrete

Specific gravity test results
Specific gravity of fine aggregate
Table no. 1 Specific gravity of fine aggregate
Type of material Fine aggregate
Sample 1 2 3
Wt of container 644 644 644
Wt of container + solid 1028 1028 1096
Wt of material + water + solid 1790 1790 1830
Wt of container + water 1552 1552 1552
Specific gravity of material 2.63 1072 2.63 2.6
Specific gravity of coarse aggregate (10mm)

Table no. 2 Specific gravity of coarse aggregate (10mm)
Type of material Coarse agg. 10mm
Sample 1 2 3
Specific gravity of material 2.76 2.76 2.93
Average specific gravity of coarse agg. 10mm 2.8
Specific gravity of coarse aggregate (20mm)

Table no. 3 Specific gravity of coarse aggregate (20mm)
Type of material Coarse agg. 20mm
Sample 1 2 3
Specific gravity of material 2.9 2.83 2.86
Average specific gravity of coarse agg. 20mm 2.86
Sieve analysis test results

Sieve analysis of fine aggregate
Table No. 4 Sieve analysis for fine aggregate
Total Weight of Fine Aggregate = 1000gm
Sieve Analysis Fine Agg.
Wt retained Wt. Retained
IS Sieve Size Cumulative Wt. Retained (%) % Passing
(gm) (%)
10 mm 0 0 0 100
4.75mm 91 9.1 9.1 90.9
2.36mm 250 25 34.1 65.9
1.18mm 336 33.6 67.7 32.3
600µ 150 15 82.7 17.3
300µ 76 7.6 90.3 9.7
150µ 35 3.5 93.8 6.2
Pan 15 1.5 95.3 4.7
944 47.2 Fineness modulus 4.7
Sieve analysis of coarse aggregate (10mm)

Table No. 5 Sieve analysis for coarse aggregate (10mm)
Total Wt. of Coarse Aggregate = 5000 gm

Sieve analysis Crushed coarse Agg. (10mm)
IS Sieve Wt. Retained Wt. Retained Cumulative Wt Retained Passing
Size (gm) (%) (%) (%)
20 mm 982 19.6 19.6 80
12.5 mm 0 0 19.6 80.36
10 mm 3498 70 89.6 10
4.75 mm 475 94 99 1
600 µ 0 0 99 1
300 µ 0 0 99 1
150 µ 0 0 99 1
Pan 4 0.1 99.1
4956 99.1 Fineness Modulus 6.2
1073
Sieve analysis of coarse aggregate (20mm)

Table No. 6 Sieve analysis for coarse aggregate (20mm)
Total Wt. of Course Aggregate = 5000 gm
Sieve Analysis Crushed CoarseAgg. (20mm)
Ind Cumulative Wt
Is Sieve Wt Retained Passing(%)
% Retained Retained (%)
20mm 982 19.6 19.6 80
12.5mm 0 0 19.6 80.36
10mm 3498 70 89.6 10
4.75mm 475 94 99 1
600µ 0 0 99 1
300µ 0 0 99 1
150µ 0 0 99 1
Pan 4 0.1 99.1
4956 99.1 Fineness modulus 6.2
Moisture content
Moisture content of fine aggregate
Table No. 7 Moisture content of fine aggregate
Description Determination no.
Sample 1 2 3
Weight of empty container (W1) in g 20.12 20.08 20
Weight of container + Wet soil (W2) in g 44.12 44.11 46.1
Weight of container + Dry soil (W3) in g 44.12 44.11 46.1
CALCULATION
Weight of water = W2-W3 0 0 0
Weight of solid = W3-W1 24.00 24.03 26.1
Water content w= W2-W3/W3-W1 x 100% 0 0 0
Average value 0%
Moisture content of coarse aggregate (10mm)

Table No. 8 Moisture content of coarse aggregate (10mm)
Sample 1 2 3
CALCULATION
Average value 0%

1 2 3

1074

CALCULATION

Average value 0%

Sample 1 2 3
CALCULATION
Average value 0%

Table No.11 Moisture content of coarse aggregate (20mm)
1 2 3
CALCULATION
Average value 0%
Water absorption
Water absorption of fine aggregate
Table No. 12 Water absorption of fine aggregate
Determination No. 1 2 3
Weight of saturated surface dried sample in g (A) 2039 2051 2096
Weight of oven dried sample in g (B) 2003 2014 2059
Water absorption = A- B / B x 100 % 1.797 1.837 1.796
Average value 1.81%
Water absorption of coarse aggregate (10mm) 1075

Table No. 13 Water absorption of coarse aggregate (10mm)

Average value 0.591%
Water absorption of coarse aggregate (20mm)

Table No. 14 Water absorption of coarse aggregate (20mm)
Average value 0.591
Test Performed on Concrete

Compressive strength, flexural strength and split tensile strength and CFD analysis results for mix
M30 concrete.
For conventional concrete mix M30
Tests results obtained for compressive strength, flexural strength and split tensile strength.
Compressive strength for conventional mix M30
50
Strength (N/mm²)
40 40.89
Compressive
35.55
30
20 26.67
10
0
7days 14days 28 days
Period of Curing (Days)
Fig: 5Compressive strength for conventional mix M30
From the above figure it can be seen that the conventional concrete has achieved satisfying results
according to days of curing. i.e. at 7 days period of curing the concrete has achieved above 60% results, at
14 days curing above 80% and at 28 days achieved 100% results.
Flexural strength for conventional mix M30
14
Flexural strength
12
13.24
10
(N/mm²)
8 11
9.23
6
4
2
0
Days of curing
Fig:6 Flexural strength for conventional mix M30

From the above figure, it can be seen that the flexural strength for the M30 mix of 7 14 & 28 days curing
shows good gain in strength
.
1076
Split Tensile strength for conventional mix M30

3.5
Tensile Strength
3
2.5 2.97 2.99
(N/mm²)
2 2.48
1.5
1
0.5
0
days of curing
Fig:7 Split tensile strength for conventional mix M30
From the above figure it can be seen that the split tensile strength for 7, 14, 28 days shows satisfying
results.
For Tyre steel fibre added concrete (M30 mix)

Tyre steel fibre replacement with cement is done in increasing percentages in concrete and from the tests
results obtained for compressive strength, flexural strength and split tensile strength. The percentage
replacement was 0%, 0.5%, 0.1%, 0.15% & 0.2% with cement.
Compressive strength for tyre steel fibre added concrete
50
46.67
40.88 41.77 42.67

40
35.56
30
(N/mm²)
20
10
0
0% 0.50% 1% 1.5% 2%
% of tyre steel fiber added
Fig: 8 Compressive strength for tyre steel fibre added concrete
From the fig above plotted graph for the variation in compressive strength for mix proportion M30 with
respect to varying tyre steel fibre content (0.5%, 1%, 1.5%, 2%) shows the drop of strength after 1% tyre
steel fibre addition. 2% tyre steel fibre replacing to cement shows lowest strength. So we conclude that the
optimum value of tyre steel fibre is 1% of replacement of cement for compressive strength.
Flexural Strength for tyre steel fibre added concrete
15
13.24 14.12
11.26
Strength
10 10.36
9.81
0
(N/mm²)
Flexural
0% 0.50% 1% 1.5% 2%
Fig:9 flexural strength for tyre steel fibre added concrete
From the fig above, plotted graph for the variation in Flexural strength for mix proportion M30 with
respect to varying tyre steel fibre content (0.5%, 1%, 1.5%, 2%) shows the drop of strength after 1% tyre
steel fibre addition. The strength for conventional concrete is less than 1% replacement of fibres. 2% tyre
steel fibre replacing to cement shows lowest strength. So we conclude that the optimum value of tyre steel
1077
fibre for flexural strength is 1% of replacement of cemen
Split tensile strength for tyre fibre steel added concrete
split tensile Strength

3 2.79 2.99
2.62
2 2.06
1.41
1
(N/mm²)
0
0% 0.50% 1% 1.5% 2%
Fig:10 Split tensile strength for tyre fibre added concrete
From the fig above plotted graph for the variation in split tensile strength for mix proportion M30 with
respect to varying tyre steel fibre content (0.5%, 1%, 1.5%, 2%) shows the drop of strength after 0.5%
tyre steel fibre addition. The strength for conventional concrete is less than 0.5% replacement of fibres.
2% tyre steel fibre replacing to cement shows lowest strength. So we conclude that the optimum value of
tyre steel fibre for flexural strength is 0.5% of replacement of cement.
For Industrial steel fibre added concrete (M30 mix)

Industrial steel fibre replacement with cement is done in increasing percentages in concrete and from the
tests results obtained for compressive strength, flexural strength and split tensile strength. The percentage
replacement was 0%, 0.5%, 1%, 1.5% & 2% with cement.
Compressive strength for industrial steel fibre added concrete

45
40.88 41.33
Compressive Strength (N/mm²)
40 38.22
34.67
35 33.77
30
25
20
15
10
5
0
0% 0.50% 1% 1.5% 2%
Fig:11 Compressive strength for industrial steel fibre added concrete
From the fig above plotted graph for the variation in compressive strength for mix proportion M30
with respect to varying industrial steel fibre content (0.5%, 1%, 1.5%, 2%) shows the drop of strength
after 0.5% industrial steel fibre addition. The strength of conventional concrete is slightly less than
optimum percentage i.e. 0.5% replacement. 2% industrial steel fibre replacing to cement shows lowest
strength. So we can conclude that the optimum value of industrial steel fibre is 0.5% of replacement of
cement for compressive strength.
1078
Flexural Strength for industrial steel fibre added concrete
16
Flexural Strength (N/mm²) 14 13.24 14.52
12.26
11.93
12 9.18
10
8
6
4
2
0
0% 0.50% 1% 1.5% 2%
% of industrial steel fiber added
Fig:12 flexural strength for industrial steel fibre added concrete
From the fig above plotted graph for the variation in Flexural strength for mix proportion M30 with
respect to varying tyre steel fibre content (0.5%, 1%, 1.5%, 2%) shows the drop of strength after 0.5%
tyre steel fibre addition. The strength for conventional concrete is less than 0.5% replacement of fibres.
2% tyre steel fibre replacing to cement shows lowest strength. So we conclude that the optimum value of
tyre steel fibre for flexural strength is 0.5% of replacement of cement.
Split tensile strength for tyre steel fiber added concrete

3.5
splittensileStrength
3 2.79 2.99
2.5 2.62
2 2.06
1.5
(N/mm²)
1.41
1
0.5
0
0% 0.50% 1% 1.5% 2%
% of industrial steel fiber added
Fig:13 Split tensile strength for industrial fibre added concrete
From the fig above plotted graph for the variation in split tensile strength for mix proportion M30 with
respect to varying industrial steel fibre content (0.5%, 1%, 1.5%, 2%) shows the drop of strength after
0.5% industrial steel fibre addition but also increases at 1% replacement. The strength for conventional
concrete is less than 0.5% replacement of fibres. 2% industrial steel fibre replacing to cement shows
lowest strength. So we conclude that the optimum value of industrial steel fibre for flexural strength is
0.5% of replacement of cement.
For Hybrid fibre concrete (combined tyre steel fibre & industrial steel fibre added in concrete) M30
mix.
Combined Industrial steel fibre and Tyre steel fibre replacement with cement is added in
increasing percentages (0%, 0.5%, 1%, 1.5%, 2%) in concrete and from the tests results obtained for
compressive strength, flexural strength and split tensile strength. The replacement was 50% of industrial
steel fibres and 50% of tyre steel fibres for each replacement percentages (0%, 0.5%, 1%, 1.5%, 2%)) for
Industrial steel fibre & tyre steel fibre respectively.
1079
Compressive Strength for Hybrid Fibre concrete
Compressive Strength 44 43.11

42 40.88
40 39.11 38.22
(N/mm²)
38
36 36
34
32
0% 0.50% 1% 1.5% 2%
% of hybrid fibers added
\
Fig:14 Compressive strength for combined hybrid fibre added concrete
From the above figure plotted graph for the variation in compressive strength for mix proportion
M30 hybrid fibres replacement with cement is added in increasing percentages (0%, 0.5%, 1%, 1.5%, 2%)
in concrete and the tests results obtained for compressive strength. The replacement was 50% of industrial
steel fibres and 50% of tyre steel fibres for each replacement percentages (0%, 0.5%, 1%, 1.5%, 2%) for
Industrial steel fibre & tyre steel fibre respectively. The graph shows the drop of strength after 1% Hybrid
fibre content. Here the 1% shows highest compressive strength which is more than conventional mix. So
we can conclude that the optimum value of hybrid fibre is 1% for compressive strength
Flexural Strength for hybrid fibre added concrete
20
Flexural Strength
15 13.87 14.78
13.24 11.93
(N/mm²)
10 8.75
5
0
0% 0.50% 1% 1.5% 2%
% of hybrid fibers added
Fig:15 Flexural strength for combined hybrid fibre added concrete
From the above figure plotted graph for the variation in compressive strength for mix proportion M30
hybrid fibres replacement with cement is added in increasing percentages (0%, 0.5%, 1%, 1.5%, 2%) in
concrete and the tests results obtained for compressive strength. The replacement was 50% of industrial
fibre content. Here 1% replacement shows highest compressive strength which is more than conventional
mix. The lowest flexural strength observed was of 2% replacement. So we can conclude that the optimum
value of hybrid fibre is 1% for flexural strength.
Split tensile strength for Hybrid fibre concrete
4
3.53
Split tensile Strength
3.13
3 2.79 2.57
2.29
(N/mm²)
0
0% 0.50% 1% 1.5% 2%
%1080
of hybrid fibers added
Fig:16 Split tensile strength for hybrid fibre fibre
From the above figure plotted graph for the variation in compressive strength for mix proportion M30
hybrid fibres replacement with cement is added in increasing percentages (0%, 0.5%, 1%, 1.5%, 2%) in
concrete and the tests results obtained for compressive strength. The replacement was 50% of industrial
fibre content. Here 0.5% replacement shows highest compressive strength which is more than
conventional mix and industrial fibre concrete but less than tyre fibre concrete. The lowest flexural
strength observed was of 2% replacement. So we can conclude that the optimum value of hybrid fibre is
1% for flexural strength.
Variation in the optimum percentages of industrial fibre concrete, tyre steel fibre, Hybrid fibre
concrete & conventional concrete
Compressive strengths at optimum percentages
48
46
(N/mm2)
44
42
40
38
Industrial fiber tyre fiber concrete hybrid fiber conventional
concrete concrete concrete
Fig:17 compressive strength at optimum percentages
Bar graph from the figure above shows the variation in compressive strength for the type of materials used
in concrete mix. Here cement replaced with tyre steel fibre shows comparatively good strength than that
of industrial fibre, conventional concrete and hybrid fibre concrete. The strengths for industrial fibre
concrete, tyre steel fibre and hybrid fibre concrete all of them satisfying the target strength.
Flexural strengths at optimum percentages
15
flexural strength
14.5
(N/mm2)
14
13.5
13
12.5
12
Industrial fiber tyre fiber hybrid fiber conventional
concrete concrete concrete concrete
Fig:18 Flexural strength at optimum percentages
Bar graph from the figure above shows the variation in compressive strength for the type of materials used
in concrete mix. Here cement replaced with hybrid fibre shows comparatively good strength than that of
industrial fibre, conventional concrete and tyre fibre concrete. The strengths for industrial fibre concrete,
tyre steel fibre and hybrid fibre concrete all of them satisfying the target strength
Split tensile strengths at optimum percentages
1081
4
split tensile strength
(N/mm2) 3
2
1
0
Industrial fiber tyre fiber concrete hybrid fiber conventional
concrete concrete concrete
Fig:19 Split tensile strengths at optimum percentages
From the figure above split tensile strength for, tyre steel fibres shows highest results than industrial fibre
concrete, hybrid fibre concrete and conventional concrete.
Cumulative Fatigue Damage (CFD) Analysis, Variation of Temperature in Concrete Pavement

Table no.15 Temperature Differentials in C
Thickness in mm/
150 200 250 300
C
Exp. Data result 21.9 22.5 21.5 20.4
IRC 58-2015 17.3 19 20.3 21
Difference 4.6 3.5 1.2 0.6
According to IRC 58-2015 Cumulative fatigue damage analysis for bottom-up cracking
In this section, various steps and calculations involved in designing a rigid pavement are presented. It
shows the axle load spectrum used in this example and referred from IRC 58-2015. The following graphs
shows CFD analysis for standard IRC values and performed Experimental values for 150mm, 200mm,
250mm, 300mm thick concrete slab of tyre steel fibres.
Temperature Variation for 150 mm Thick Slab

60
55
50
Temperature
45 Top
40 Middle
35 Bottom
30
25
10.00 AM
3.00 PM
4.00 PM
7.00 PM
9.00 PM
11.00 PM
1.00 AM
2.00 AM
3.00 AM
5.00 AM
6.00 AM
7.00 AM
8.00 AM
9.00 AM
12.00 PM
3.00 PM
5.00 PM
6.00 PM
8.00 PM
10.00 PM
11.00 AM
1.00 PM
2.00 PM
4.00 AM
12.00 AM
Time
Fig:20 Temperature Variation for 150 mm Thick Slab
From the above temperature results the temperature gradient is calculated as-
Highest temperature at top for 24 hours – Lowest temperature at bottom for 24 hours
For ex. :- Temperature gradient for 150mm slab = 54.6° - 32.7° = 21.9
1082

60
55
50
Temperature
45
Top
40
Middle
35
Bottom
30
25
10.00 AM
3.00 PM
4.00 PM
7.00 PM
9.00 PM
11.00 PM
1.00 AM
2.00 AM
3.00 AM
5.00 AM
6.00 AM
7.00 AM
8.00 AM
9.00 AM
12.00 PM
3.00 PM
5.00 PM
6.00 PM
8.00 PM
10.00 PM
11.00 AM
1.00 PM
2.00 PM
4.00 AM
12.00 AM
Time
Temp. gradient = Highest temperature at top for 24 hours – Lowest temperature at bottom
for 24 hours
Temperature gradient for 150mm slab = 54.8° - 32.3° = 22.5°C

60
55
50
Temperature
45
Top
40
Middle
35
Bottom
30
25
3.00 PM
4.00 PM
7.00 PM
9.00 PM
11.00 PM
1.00 AM
2.00 AM
3.00 AM
5.00 AM
6.00 AM
7.00 AM
8.00 AM
9.00 AM
10.00 AM
12.00 PM
3.00 PM
5.00 PM
6.00 PM
8.00 PM
10.00 PM
11.00 AM
1.00 PM
2.00 PM
4.00 AM
12.00 AM
Time
Temp. gradient = Highest temperature at top for 24 hours – Lowest temperature at bottom for
24 hours
1083

60
55
50
Temperature
45 Top
40
Middle
35
Bottom
30
25
10.00 AM
3.00 PM
4.00 PM
7.00 PM
9.00 PM
11.00 PM
1.00 AM
2.00 AM
3.00 AM
5.00 AM
6.00 AM
7.00 AM
8.00 AM
9.00 AM
12.00 PM
3.00 PM
5.00 PM
6.00 PM
8.00 PM
10.00 PM
11.00 AM
1.00 PM
2.00 PM
4.00 AM
12.00 AM
Time
Temp. gradient = Highest temperature at top for 24 hours – Lowest temperature at bottom for
24 hours
Performance of concrete slabs on IRC-58 2015 fatigue analysis for Hybrid fibre concrete &
conventional concrete
CFD for Conventional Concrete
From Table No.15 the IRC-58 2015 Values & Experimental data result values are observed and CFD
analysis is performed for conventional concrete at flexural strength 13.24Mpa & compressive strength
40.89Mpa. The following graphs show the behavior of fatigue damage
250.00
Temp.
200.00
CFD Value
Gradient 0°C
150.00 133.35
100.00
50.00
9.86
0.00 0.60
130 140 150 160
Thickness Of Slab (mm)
Fig:24 CFD Analysis for 0°C of conventional concrete
For 0°C the slab thickness is safe at 140mm thickness for conventional concrete at flexural strength
13.24Mpa & compressive strength 40.89 Mpa. For 17.3°C IRC value the slab thickness 150mm is safe at
160mm thickness for conventional concrete at flexural strength 13.24Mpa & compressive strength 40.89
Mpa.
1000.00 Temp. Gradient 150.00 Temp.

919.56 123.29
CFD Value
800.00 17.3°C Gradient21.9°C

CFD Value
(IRC value) 100.00 (exp value)

600.00
400.00 50.00
15.512.160.13
200.00 73.56 8.88 0.95 0.00
0.00 140 150 160 170 180
130 140 150 160 170 Thickness Of Slab (mm)
Fig:25 CFD Analysis for 17.3°C&21.9°C of 150 mm conventional concrete
For 17.3°C experimental value the slab thickness 150mm is safe at 160mm thickness for conventional
1084 strength 40.89 Mpa. For 19°C IRC value the slab
concrete at flexural strength 13.24Mpa &compressive
thickness 200mm is safe at 170mm thickness for conventional concrete at flexural strength 13.24Mpa &
compressive strength 40.89 Mpa.
100.00 89.02 Temp Gradient 150.00 Temp. Gradient

131.84
CFD Value
80.00
CFD Value
19.0°C 22.5°C
60.00 (IRC value) 100.00 (exp value)
40.00
50.00
20.00 10.94 0.04 16.66 0.15
1.31 2.37
0.00 0.00
140 150 160 170 180 140 150 160 170 180
Thickness Of Slab (mm) Thickness Of Slab (mm)
Fig: 26 CFD Analysis for 19.0°C&22.5°C of 200 mm of conventional concrete
concrete at flexural strength 13.24Mpa & compressive strength 40.89 Mpa.For 20.3°C IRC value the slab
compressive strength 40.89 Mpa
.
150.00 Temp. Gradient 150.00 Temp.
CFD Value
20.3°C 117.89 Gradient

CFD Value
100.00 103.03
(IRC value) 100.00 21.5°C…
50.00 50.00
12.811.65 0.07 14.79 0.11
0.00 2.02
0.00
140 150 160 170 180 140 150 160 170 180
Fig:27 CFD Analysis for 20.3°C&21.5°C of 250 mm of conventional concrete
concrete at flexural strength 13.24Mpa & compressive strength 40.89 Mpa.For 21°C IRC value the slab
compressive strength 40.89 Mpa.
120.00 111.47 Temp. Gradient 120.00 Temp.
104.22 Gradient
100.00
CFD Value
100.00 21°C
CFD Value
80.00 (IRC value) 80.00 20.4°C…

60.00 60.00
40.00 40.00
20.00 12.96 0.07
20.00 13.93 0.09 1.68
1.86 0.00
0.00
140 150 160 170 180
140 150 160 170 180
Fig:28 CFD Analysis for 21°C & 20.4°C of 300 mm for conventional concrete
concrete at flexural strength 13.24Mpa & compressive strength 40.89 Mpa.
CFD for Hybrid Fiber Concrete

From Table No. 15 the IRC-58 2015 Values & Experimental data result values are observed and CFD
analysis is performed for Hybrid fiber concrete at flexural strength 14.78Mpa & compressive strength
43.11 Mpa. The following graphs show the behaviour of fatigue damage.
1085
250.00 Temp. Gradient
CFD Value
200.00 203.07
0°C
150.00
100.00
50.00 12.42 0.66
0.00
120 130 140 150
Fig:29 CFD Analysis for 0°C for Hybrid fibre concrete

For 0°C the slab thickness is safe at 140mm thickness for Hybrid fibre concrete at flexural strength
14.78Mpa & compressive strength 43.11 Mpa
80.00 74.50 Temp. 150.00 Temp.

Gradient Gradient
CFD Value
60.00 118.33
CFD Value
17.3°C… 21.9°C
100.00
40.00 (exp value)
20.00 7.47 50.00
0.57 0.03
0.00 12.41
1.32
130 140 150 0.00
Thickness Of Slab (mm) 130 140 150 160 170
Fig:30 CFD Analysis for 17.3°C&21.9°C of 150mm for Hybrid fibre concrete
For 21.9°C experimental value of 150mm thickness the slab thickness is safe at 160mm thickness for
hybrid fiber concrete at flexural strength 14.78Mpa & compressive strength 43.11Mpa.For 19°C IRC-58
value of 200mm the slab thickness is safe at 150mm thickness for hybrid fiber concrete at flexural
strength 14.78Mpa & compressive strength 43.11 Mpa.
100.00 Temp. 140.00 Temp.

88.43 Gradient 125.68 Gradient
80.00 120.00
19.0°C 22.5°C
CFD Value
CFD Value
100.00 (exp
60.00 (IRC value)
80.00 value)
40.00 60.00
40.00
20.00 9.04 13.24 0.04
0.80 20.00 1.45
0.00 0.00
130 140 150 160 130 140 150 160 170
Fig:31 CFD Analysis for 19° & 22.5°C of 200 mm for Hybrid fibre concrete
For 22.5°C experimental value of 200mm the slab thickness is safe at 160mm thickness for hybrid fiber
concrete at flexural strength 14.78Mpa & compressive strength 43.11 Mpa.For 20.3°C IRC-58 value of
250mm the slab thickness is safe at 160mm thickness for hybrid fiber concrete at flexural strength
14.78Mpa & compressive strength 43.11Mpa
140.00 Temp. 120.00 Temp.
125.68 113.68
120.00 Gradient Gradient
100.00
CFD Value
100.00 20.3°C 21.5°C

CFD Value
80.00 (IRC value) 80.00 (exp

60.00 60.00 value)
40.00 40.00
20.00 13.24 0.04 20.00 11.89
1.46 0.03
0.00 1.23
0.00
130 140 150 160 170
130 140 150 160 170
Fig:32CFD Analysis for 20.3°C &21.5°C of 250 mm for Hybrid fibre concrete
1086
concrete at flexural strength 14.78Mpa & compressive strength 43.11 Mpa.For 22.5°C IRC-58 value of
300mm the slab thickness is safe at 160mm thickness for hybrid fiber concrete at flexural strength
14.78Mpa & compressive strength 43.11 Mpa.
120.00 Temp. Gradient 120.00 Temp. Gradient

108.12 101.80
100.00 21°C 100.00 20.4°C
CFD Value
CFD Value
(IRC value) 80.00 (exp value)

80.00
60.00 60.00
40.00 40.00
20.00 11.26 0.02 20.00 10.54 0.01
1.13 1.02
0.00 0.00
130 140 150 160 170 130 140 150 160 170
Fig:33 CFD Analysis for 21 & 20.4°C of 300mm for hybrid fibre concrete
concrete at flexural strength 14.78Mpa & compressive strength 43.11 Mpa
Comparing Hybrid fiber concrete & conventional concrete

175
170
165
160
IRC Value
155
Exp. Value
150
145
140
Hybrid fiber concrete conventional concrete
The graph clearly indicates that the thickness of Hybrid fiber concrete slab is less as compared to
conventional concrete slab. Thus, in case of hybrid fiber concrete pavement, The thickness will be less as
compared to conventional concrete. So ultimately there is no need to provide 200 mm thick slab. Instead
by Using hybrid fibers 170 mm thickness of slab is sufficient. Eventually 30 mm concrete thickness can
be saved.
Cost analysis
• For 200 mm conventional concrete slab, 170 mm of Hybrid fiber concrete slab is found safe.
200 mm – 170 mm = 30 mm saving of concrete
• According to Current Schedule Rate, cost of concrete / m3 = 7500 Rs/-
• For Road pavement of 3.75 m wide, 1 km long, 3.75 X 1000 X 0.03
= 112.5 m3
• Saving of concrete / kilometer = 112.5 X 7500
= 8,43,750 Rs / km
Conclusions
In this research paper we have studied the usage of waste Tyre steel fibres their behaviour
on different mechanical properties and their effects on the concrete and we also have studied about fatigue
behaviour of cement concrete slabs. The research utilizes waste tyre steel fibre in concrete which
enhances the mechanical properties of concrete. The CFD analysis for various temperature differentials of
different replacement has been made. By this analysis the comparison of different thicknesses are made.
1) From the experimental results, we can conclude that at 1% replacement of waste tyre steel fibre
by cement shows the highest compressive strength.
2) The flexural strength is found to have highest value for 1% replacement of waste tyre steel fibre
by cement.
3) At 0.5% replacement of waste tyre steel fibre by cement shows the highest split tensile strength
and for hybrid steel fibres. 1087
4) From the experimental results we can conclude that, The optimum percentage of replacement of
cement with tyre steel fibres comes out to be 1% by weight for compressive and flexural
strength, for split tensile strength is 0.5% by weight.
5) Using waste tyre steel fibre in concrete mix proved to be very useful to solve environmental
problem and produce green concrete. Therefore, it is recommended to re-use tyre steel fibre in
concrete to move toward sustainable development in construction industry.
6) On the other hand for 200mm thickness of slab 160mm is safe thickness against fatigue. i.e. no
need to increase the thickness upto 200 mm. similarly in case of 250mm & 300mm thick slab,
160mm thickness is safe thickness against fatigue.
7) After studying CFD analysis with IRC-58 2015 for worst condition (as the experimental results
were observed at peak of summer) it can be concluded that concrete thickness can be reduced
rather than providing higher thickness and which ultimately results in saving of concrete
materials especially natural constituents such as sand, aggregates and most importantly water.
References
1. Choi S.Y., Park J.S., Jung W.T., A Study on the Shrinkage of Fibre Reinforced Concrete Pavement, The 12th
East Asia-Pacific Conference on Structural Engg. & Construction. ProcediaEngg 14 (2011),2815-2822
2. Maitra S.R., Reddy K.S., Ramachandra L.S., Numerical Investigation of Fatigue Characteristics of Concrete
Pavement.,Int. J.Fract. (2014) 189: 181-193
3. Graeff A.G., Pilakoutas K., Neocleous K., Peres M.V.N.N., Fatigue resistance & cracking resistance of concrete
pavements reinforced with recycled steel fibres recovered from post-consumer tyres, Engg Structures 45(2012),
385-395.
4. Shah S.P., Rangan B.V., Fibre Reinforced Concrete Properties. ACI J Proc 1971,68, 126-37
5. Kwadwo A., Obeng N., Abeka H.K., Concrete pavements slabs reinforced with steel fibres from discarded
vehicle tyres, Sunyani Polytechnic Journal ,No.1, Vol.1, April 2016, ISSN 2508-0997
6. Anne & Russ evans “Steel & fiber from scrap tyre processing” Published by: the waste & resources action
programme www.wrap.org.uk
7. Gregory Marvin Garrick “analysis and testing of waste tire fiber modified concrete” (2005) LSU Master’s
theses.217, Louisiana state university and agriculture and mechanical college, ggarril@lsu.edu
8. Marijana serdar, ana baricevic, stjepan lakusic, dabravka bjegovic “Special purpose concrete products from tyre
recyclates” faculty of civil engineering, university of zegrab, gradevinar 65 (2013)9, 793-801.
9. Marko bartolac, domagoj damjanovi , joško krolo, ivan duvnjak and ana bari evi “punching of slabs reinforced
with recycled steel fibers from used tyres” university of zegrab, faculty of civil engineering Croatia, research gate
publication.
10. Khaleel H. Younis “Mechanical Performance of Concrete Reinforced with Steel Fibres Extracted from Post-
Consumer Tyres” Lecturer at Erbil Polytechnic University (EPU), Erbil, Kurdistan-Iraq khaleelyounis@epu.edu.krd
11. Kypros Pilakoutasa, Kyriacos Neocleousb, Houssam Tlematc, Beratende Ingenieur “Reuse of tyre steel fibres as
concrete reinforcement” Department of Civil & Structural Engineering The University of Sheffield, Sir Frederick
Mappin Building Mappin Street, Sheffield S1 3JD, UK, Tel, Fax +44-114-2225700. Research gate.
12. IRC: 58:2015 (2015) “Guidelines for the Design of Plain Jointed Rigid Pavements for Highways” IRC, New Delhi
1088
Paper ID- 206
Innovative Techniques to Improve Strength and Sustainability of Concrete by

Shock Wave Applications
Dr.P.V.Selvana 1 https://orcid.org/0000-0003-1715-3962
, Dr.P.Eswaramoorthib, V Gayathric https://orcid.org/0000-0002-7838-3481
a,c Associate Professor, b Professor, Department of Civil Engineering,Kumaraguru College of Technology,Coimbatore-641022,India
Abstract
Concrete is a versatile material and can be moulded to any shape and size when it is green. The compressive
strength of concrete depends on various factors like the water- binder ratio, degree of compaction, quality
and quantity of fine and coarse aggregates, curing and climatic condition etc. Among all, one of the important
factor on which the strength of concrete depends is the degree of compaction. Conventionally, concrete is
compacted by hand compaction , or vibrator in the field. In this paper, an innovative method is used to
compact the concrete. A better efficiency of the compressive strength on the concrete is focused here. This will be
more useful to the Construction Industries and society, both economically and sustainability will be the major output
of this paper. The shock waves are made to pass through the concrete and the same is tested for compressive
strength at the age of 14 and 28 days. As a result of this innovative technique, we shall expect the increase in
compressive strength of concrete upto 18-20 %which will be more beneficial to the Construction Industries and to
the common public or society. For the experimental study and investigation we use M25concrete which is
extensively used for various construction applications viz., beams, columns, roof slabs etc. of residential and
industrial structures.
Keywords: Shock Wave, Concrete, Compressive Strength, Sustainability, Grade of Concrete.
1 Introduction of Shock Wave Application in Concrete
Concrete is a material used in building construction, consisting of a hard, chemically inert substance, known as
aggregate. Concrete is a composition of many materials composed primarily of aggregates, cement and water. There
are many formations that have varied properties. The aggregate is generally a coarse gravel or crushed rock such as
limestone or granite along with fine aggregate. Compressive strength is the capacity of a material or structure to
withstand loads tending to reduce size. For designers, compressive strength is one of the most important engineering
properties of concrete. It is a standard industrial practice that the concrete is classified based on grades. This grade is
nothing but the Compressive Strength of the concrete cube or cylinder. Cube or Cylinder samples are usually tested
1089
under a compression testing machine to obtain the compressive strength of concrete. The test requisites differ
country to country based on the design code. The compressive strength of concrete is given in terms of
the characteristic compressive strength of 150 mm size cubes tested at 28 days. The characteristic strength is defined
as the strength of the concrete below which not more than 5% of the test results are expected to fall.Compressive
strength of concrete depends on many factors such as water-cement ratio, cement strength, quality of concrete
material, quality control during production of concrete etc.
2 Experimental Testing and Results
Nomenclature
A M25-Grade of Concrete
B fck-Compressive Strength of Concrete at 28days
C fb -Flexural Strength of Concrete
2.1 Shock Tube
The shock tube is an instrument used to replicate and direct blast waves at a sensor or a model in order to simulate
actual explosions and their effects, usually on a smaller scale. A shock wave inside a shock tube may be generated
by a small explosion (blast-driven) or by the build up of high pressures which cause diaphragm(s) to burst and a
shock wave to propagate down the shock tube (compressed-gas driven).A simple shock tube is a tube, rectangular or
circular in cross-section, usually constructed of metal, in which a gas at low pressure and a gas at high pressure are
separated using some form of diaphragm. The diaphragm suddenly bursts open under predetermined conditions to
produce a wave propagating through the low pressure section. The shock that eventually forms increases the
temperature and pressure of the test gas and indicates a flow the direction of the shock wave.
Table 1. Experimental Results for Compressive Strength of Concrete Cubes with and without Shock waves:
S.No Age of Specification of the Specime Load at Ultimate Ultimate Average

concret cube n first crack load stress compressiv
e e strength
2
Unit days 15cmx15cmx15cm - KN KN N/mm N/mm2
Without Shock waves I 425 483 21.46
II 432 490.3 21.79 21.76
1 14 III 440 496 22.04
With shock waves i) 540 585 26
ii) 542 589 26.17 26.08
iii) 543 587 26.09
1090
Without Shock waves I 546 567.9 25.24

II 538 629.1 27.96 25.83
2 28 III 522 546.75 24.3
With Shock i) 688 723 32.13
waves ii) 683 719.3 31.96 31.98
iii) 670 717 31.86
2.2 Experimental Procedure
Cube Casting
• Measure the dry proportion of ingredients (Cement, Sand & Coarse Aggregate) as per the design requirements.
The Ingredients should be sufficient enough to cast test cubes
• Thoroughly mix the dry ingredients to obtain uniform mixture
• Add design quantity of water to the dry proportion (water-cement ratio) and mix well to obtain uniform texture
• Fill the concrete to the mould with the help of vibrator for thorough compaction
• Finish the top of the concrete by trowel & tapped well till the cement slurry comes to the top of the cubes.
• Six cubes were chosen for applying shock waves and 80 shock waves are applied for each cube continuously.
Curing
• After some time the mould is kept undisturbed for 24 hours at a temperature of 27 ° Celsius ±2
• After 24 hours remove the specimen from the mould.
• Keep the specimen submerged under fresh water at 27 ° Celsius. The specimen should be kept for 7 or 28 days.
Every 7 days the water should be renewed.
• The specimen should be removed from the water 30 minutes prior to the testing.
• The specimen should be in dry condition before conducting the testing.
• The Cube weight should not be less than 8.1 Kilograms
Experimental Investigation
• Now place the concrete cubes into the testing machine. (centrally)
• The cubes should be placed correctly to the machine plate (check the circle marks on the machine). Carefully align
the specimen to the spherically seated plate.
• The load will be applied to the specimen axially.
• Now slowly apply the load at the rate of 140kg/cm2 per minute till the cube collapse.
• The maximum load at which the specimen breaks is taken as a compressive load.
Table 2. Percentage Increase in Mean Compressive Strength
1091
S.No Age of Concrete Specification of the Average Compressive Percentage increase of

(days) cube Strength Compressive Strength
(MPa)
1 14 Without Shock wave 21.76 19.88 %
With Shock Waves 26.08
2 28 Without Shock wave 25.83 23.80 %

With Shock waves 31.98
Table 3. Flexural Strength of Concrete

S.No Age of concrete Specification of the specimen Flexural
(days) Strength (MPa)
1 14 Without shock wave 3.26
With shock waves 3.57
2 28 Without Shock wave 3.55
Without shock waves 3.95
1. 3 Images of Experimental Work on Concrete using Shock Waves
Fig. 1. Shock Tube Fig. 2. Concrete Cube Preparation
1092
Fig. 3. Compression Testing Machine Fig. 4. Concrete Cube after Testing
Fig. 5. Mixing of Concrete
1093
Graph. 1.Compressive strength of concrete in 14 and 28 days without shock wave
Graph. 2.Compressive strength of concrete in 14 and 28 days with shock waves
1094
Graph. 3. Mean compressive strength of concrete in 14 and 28 days without and with shock waves
Graph. 4..Increase in Percentage of compressive strength of concrete in 14 and 28 days without and with shock waves
1095
4 Equations
The flexural strength of concrete also arrived from the compressive strength as per IS 456:2000 are as follows:
Flexural strength of concrete (fb) =( 0.7 x √fck) in MPa.
Model calculation
Compressive strength of concrete for 28 days :
1) Without shock wave
Ultimate load = 629.1 KN

Area = (150x150) = 22500 Sq.mm.
Compressive strength = (629.1x1000) / 22500 = 27.96MPa.
2) With Shock wave
Ultimate load= 719.3 KN

Area = (150x150) = 22500 Sq.mm.
Compressive Strength = (719.3x1000) / 22500 = 31.96 MPa.
5 Conclusion
From the experimental investigation of compressive strength ofM25 concrete by application of shock waves, it is
concluded that there is a significant increase in compressive strength of 6.15 MPa (23.80%) at the age of 28 days. It
is also found that there is a significant increase in compressive strength of concrete by 4.32 MPa(19.88%) at the age
of 14 days. The flexural strength of concrete increases with shock waves by 11.275 % compared to normal concrete
.The observed vital increase in Compressive strength of concrete by the application of shock waves can be utilized
enormously by the construction industries to obtain the better grade(M30) with the existing grade of M25 which will
be more beneficial to the common public or society in terms of both economy and sustainability. In future, the same
techniques can be extended to the higher grades of concrete for various applications.
Acknowledgements
We are indebted to thank our guide Dr.Selvan P.V.(SREC,Asso.Prof/Aero Engg), Head of the Department
Dr.DavidRathnaraj J.(Aeronautical Engineering,Sri Ramakrishna Engineering College,Coimbatore-22) and
Department of Civil Engineering for their valuable support and guidance in terms of laboratory facilities . We
wholeheartedly thank our Management, Principal and Director for the support extended to carry out the R&D
project.
References
1. El-Dieb.A.S.(2007), Self-Curing Concrete: water retention, hydration and moisture transport”, Construction
and Building Materials vol. 21, pp. 1282–1287.
2. PietroLura, Ole Mejlhede Jensen and Shin-Ichi Igarashi, Experimental observation of internal water curing
of concrete, Materials and Structures 40 pp.211-220,2007.
1096
3. IS 10262: 2009, Concrete mix proportioning-guidelines, Bureau of Indian Standards, New Delhi, India. [4]
IS: 516-1959, Methods of tests for strength of concrete, Bureau of Indian Standards, New Delhi, India.
4. Ambily.P.S and Rajamane N.P, (2009), Self curing concrete – An Introduction, Concrete Composites Lab,
Structural Engineering Research Centre, Chennai.
5. ManojKumar.M and Maruthachalam. D(2013), Experimental investigation on self-curing concrete,
International Journal of Advanced Scientific and Technical Research vol. 2, pp. 300-306.
6. Shock properties of Conventional and high strength Concrete : Experimental and mesomechanical analysis
, International Journal of Impact Engineering , Volume 36 , Issue 3,March 2008.
7. xperimental Study on Torsion of Reinforced Concrete Slab with Various end Connditions – published .
(Feb-2014) 52-58 , International Journal of Engineering Science and Engineering, Volume-2,Issue-4 ,
IJESE Dr.Selvan P.V. , Dr.Sundararajan R.
8. Experimental Study on Torsion reinforcement in R.C.Solid slab with four edges discontinuos , Vol. 3 ,
Issue No.3,Feb-2014.International Journal of Engineering and Advanced Technology(IJEAT),page no65-
72 ,Dr.Selvan P.V. , Dr.Sundararajan R.
9. Behaviour of Torsion reinforcement in R.C.Slab with two ends discontinuous and other ends continuous,
published in International Journal of Engineering Technology Science and Research, ISSN 2394-3386,
Volume 4, Issue 12, December 2017 ,page 1342-1348Dr.Selvan P.V. , Dr.Sundararajan K.
10. Geetha.M and Malathy.R(2011), Comparative study of strength and durability properties of polymeric
materials as self curing agents, International Journal of Engineering Science and Technology vol. 3, pp.
771-776.
1097
National Conference
on
CoAST-2019
Feb 1 to 3, 2019
Organized by
NIT Silchar
Acknowledgement
The conference has been funded by TEQIP-III, NIT

Silchar.
Springer Publications has agreed to publish Select

Papers in Scopus indexed Springer Proceedings.
These are thankfully acknowledged.

CoAST Full Paper With Acknowledgement PDF

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

CoAST Full Paper With Acknowledgement PDF

Caricato da

Copyright:

Formati disponibili

Proceedings

Advances in Structural Technologies

Department of Civil Engineering

National Institute of Technology Silchar

Advances in Structural Technologies

Department of Civil Engineering

National Institute of Technology Silchar

Advances in Structural Technologies

Advances in Structural Technologies

Prof. Biswajit Bhattacharjee

Prof. Nirjhar Dhang

Prof. Siddhartha Ghosh

Dr. Balaji Raghavan

Dr. Debiprasad Ghosh

Message from the Director

Prof. Sivaji Bandyopadhyay

Message from the HoD, Civil Engineering

I am very happy to know that Structural Engineering Division is organizing a

Prof. Parthasarathi Choudhury

A.I. Laskar S. Choudhury

Biswajit Bhattachrjee is a Professor at the Department of Civil Engineering, Indian

Nirjhar Dhang (born 1962) is a structural engineer and an educator. He is known

Siddhartha Ghosh is currently a Pro fessor in the Department of Civil Engineering,

Balaji Raghavan is a tenured Associate Professor of Civil and Mechanical

He graduated from IIEST, Shibpur (formerly known as Bengal Engineering

He has published many papers on subjects ranging from smart materials to

Presently, he is exploring quantum structural health mo nitoring, and studying

A National Conference in Structural Engineering gives us an opportunity to exchange ideas

A.I. Laskar S. Choudhury

Paper Id Title Page No.

Finite Element Simulation of Low Velocity Impact of a

Hrishikesh Dev Sarma1

Abstract. Honeycomb sandwich structures are widely used in the aerospace

Keywords: Sandwich Structures, Honeycomb Structures, Impact Mechanics,

2 Quasi-static crushing of an aluminum honeycomb sandwich

2.1 Experimental investigations and observations

Table 1 geometric specifications of the honeycomb used in the crushing test.

s (mm) d (mm) h (μm) hc (mm) LL (mm) LW (mm)

Table 2 peak and plateau stresses at various rates of loading

2.2 Numerical modeling and simulation

3 Modeling impact of the conical impactor

3.1 Numerical modeling of the impactor, sheets and core

Table 3 material Properties used for the finite element model

Material Part Young’s Yield stress Density Poisson’s

3.2 Results of numerical simulations and observations

Reaction Force (kN)

Reaction Force (kN)

Fig 5 plots showing deflection-time, force-time and force-deflection relationships as the

4 Analytical method to find deflection and reaction force

4.1 Contact law for local indentation

-0.8 actual profile

This can be simplified by substituting for Df, a and α to obtain, ,

The work done by crushing of a perfectly plastic core is ∫

Specifically, for a Poisson ratio of ν = 0.3, the above relationship becomes

4.2 Global panel deflection

5 Comparison of FEA with analytical results

Effect of Wind and Seismic forces on different Components of Cable

Fig: 1 Components of Suspension Bridge

2. Structural Behaviour of Different Components of Cable Suspension Bridge under

2.1 Cable and Hanger

Figure. 2. Effect of different wind speeds in lateral and vertical displacement

[44] A. A. Dumanoglu, R. T. Severn (1989) Seismic Response Of Modern Suspension Bridges To

REPLACEMENT OF SAND WITH STONE DUST AS A FINE