See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/251540427

Investigation on the flexural behaviour of

reinforced concrete beams using phyllite
aggregates from mining waste
ARTICLE in MATERIALS AND DESIGN DECEMBER 2011
Impact Factor: 3.17 DOI: 10.1016/j.matdes.2011.05.043

CITATIONS

DOWNLOADS

VIEWS

229

78

2 AUTHORS:
Mark Adom-Asamoah

Russell Owusu Afrifa

Kwame Nkrumah University Of Science and

10 PUBLICATIONS 21 CITATIONS

28 PUBLICATIONS 44 CITATIONS

SEE PROFILE

SEE PROFILE

Available from: Mark Adom-Asamoah

Retrieved on: 28 June 2015

MATERIALS & DESIGN, 32(10), 5132-5140

Investigation on the flexural behaviour of reinforced concrete beams using

phyllite aggregates from mining waste
Mark Adom-Asamoah, Russell Owusu Afrifa
Department of Civil Engineering, Kwame Nkrumah University of Science and Technology, College of
Engineering, Kumasi, Ghana

*Corresponding author: Tel. +2335160226, Fax +2335160235

Email: markadomasamoah@gmail.com; madom-asamoah.coe@knust.edu.gh

(Dr Mark Adom-Asamoah)

Abstract
This paper investigated the flexural behaviour of 12 reinforced concrete (RC) beams made of
phyllite coarse aggregates produced as by-product of underground gold mining activity. The
beams were tested to failure under four point tests. Collapse of the beams which were
adequately designed against shear failure occurred mostly through either flexural-shear
failure and/or diagonal tension failure. The experimental failure loads averaged
approximately 115% of the theoretical failure loads. It was observed that the beams
developed early shear cracks and higher flexural crack widths than allowable at service
loads. Deflections compared reasonably well with the design code requirement but
displacement ductility was low. It is recommended that British Standard (BS) 8110 design
concrete shear stress values be multiplied by 0.8 to assure that the predicted shear capacity
of phyllite concrete would be low and reasonable as compared to flexural capacity. In that
case, BS 8110 can be used to provide adequate load factor against flexural failure for underreinforced RC beams made of phyllite coarse aggregates.

1. Introduction
Coarse aggregate is a major component of concrete as it takes a high percentage of either the mass or
volume of concrete based on any standard mix design. The incidence of increasing rate of natural
resource depletion coupled with the high cost associated with some of the traditionally used
aggregates calls for the use of new, abundant and cheap materials. This has led to work by several
researchers on either the mechanical properties of plain concrete or the bending/shear strengths of
beams made from non-traditional aggregate sources.

Waste aggregates obtained from four granite pits in Turkey were used to produce concrete. Physical
and chemical properties of the aggregates were found to be adequate when compared with code
standards. Tests on concrete made from the granite waste aggregates indicated that the results of
compressive strength, split-ting tensile strength, static modulus of elasticity, water absorption rate and
1

MATERIALS & DESIGN, 32(10), 5132-5140

density are comparable to Turkish and European standards [1]. The mechanical properties of concrete
with recycled coarse aggregate were also investigated for ten mixes of concrete with target
compressive cube strength ranging from 20 to 50 Mpa. The trends in the development of compressive
and shear strength and the strain at peak stress in recycled aggregate concrete were similar to those in
natural aggregate concrete [2]. The shear behaviour of four recycled aggregate concrete beams was
compared to those of corresponding normal concrete beams. The results showed that whereas the
deflections and the ultimate loads were little affected by the type of concrete, the recycled beams
developed cracking at lower loads than the corresponding normal concrete beams [3]. In areas where
high quality aggregates are not available, silica fume has been incorporated in concrete to enhance its
engineering properties. Researchers [4] characterized the properties of recycled aggregates by the
addition of silica fume waste. Tests were conducted on concrete made from recycled concrete
aggregates with silica fume waste to determine the properties both physical (density and water
absorption) and mechanical (compressive and tensile splitting strength and static modulus of
elasticity). The compressive strength of recycled concrete with silica fume was similar to that of
conventional concrete with silica fume. The optimum percent of silica fume for compressive strength
of the concrete was 8%. The addition of the silica fume was found beneficial in controlling the
performance levels in terms of sustained loading as compared to that of recycled concrete. In further
experimental research, the behaviour of reinforced concrete beams made from recycled concrete
aggregate with 8% silica fume [5] improved in terms of ultimate shear loads and cracking when
compared with those of recycled concrete aggregates.

Other researchers have also used non-conventional aggregates in characterizing concrete beam
behaviour. A comparative study of concrete properties using coconut shell and palm kernel shell as
substitutes for conventional coarse aggregates showed that at least 30% cost reduction could be
achieved for similar compressive strength of conventional concrete [6]. Others [7] showed that the
structural bond properties of lightweight concrete incorporating solid waste oil palm shells (OPS) as
coarse aggregates was much higher than the theoretical bond strength as stipulated by codes of
practice. The durability performance of OPS concrete revealed that its water absorption and water
permeability were also comparable to other lightweight concretes. Elsewhere, the flexural behaviour
of 6 under-reinforced OPS concrete beams were investigated [8]. The investigations revealed that the
flexural behaviour of OPS concrete beams compared well with that of beams made from other
lightweight aggregates and experimental results compared reasonably well with codes of practice. As
a further encouragement to regional building authorities in Singapore, the overall flexural response of
lightweight RC beams made from expanded clay aggregates compared well with those of
conventional normal weight concrete beams [9]. Use has also been made of sandstone aggregates to
produce high-performance reinforced concrete (HPC) beams [10] which exhibited flexural strengths
higher than predicted ultimate moment capacities. Due to lower stiffness of sand-stone aggregates, the
2

MATERIALS & DESIGN, 32(10), 5132-5140

beams resulted in excessive deflection under service loads. The crack widths under service loads were
within acceptable limits. The tests results indicated that it is possible to produce HPC using sandstone
aggregates with silica fumes and super plasticisers.

The use of granite and sandstone as coarse aggregates is very popular with the Ghanaian
construction industry, yet the sources of these materials are facing the problem of depletion and
substitute for granite as coarse aggregate has been a challenge whereas conservation of natural
resources is very essential in any modern development. Phyllites are foliated rocks consisting of very
fine-grained phyllosilicates and quartz. The abundance of fine-grained phyllosilicates gives them a
soft feel and the existence of preferential cleavage gives them the property of easily breaking up into
thin slabs [11]. Work has been done on the characterization of phyllite rocks by several researchers.
The characterization is necessary in evaluating the suitability of using phyllite in a wide range of
earthen construction works such as liners (in waterproofing roofs and water reservoirs), road
subgrades, embankments and core material in zoned dams. An experimental programme was carried
out to characterize the physical, micro-structural and geotechnical properties of clay phyllites. Despite
the presence of active clay minerals, the material exhibited good compaction properties and,
consequently, low water permeability plus a stiff response on loading [12]. The presence of phyllites
has been reported in several countries such as Spain [12], Venezuela [13], south western Germany and
eastern France [14] and the Himalayan region [15]. In Ghana, phyllite rocks are obtained as a byproduct of underground mining activity.
1.1. Aims, scope and significance of research
Since aggregate use and cost of concrete production is related to availability and haulage distance,
the use of phyllite aggregates is one of such attempt to substitute for crushed granite as coarse
aggregates in concrete production. The continuous accumulation of phyllite aggregates is an
environmental issue if no use is found for this waste material. Environmental regulations have also
be-come more stringent, causing waste to become increasingly expensive to dispose. Subsequently,
exploitation of the waste material as sustainable building material in the construction industry helps
preserve the natural resources such as sources of coarse aggregates. Earlier work [16] on plain phyllite
concrete cubes and modulus of rupture beams made from phyllite aggregates obtained from mining
waste showed low compressive and bending strengths as compared to those of granite aggregate
concrete. This was probably because the flakiness and elongation properties coupled with reactive
materials in phyllite aggregates affect the water absorption and bond characteristics of its concrete.
This results in the high water absorption of the phyllite aggregates which could affect the strength
properties of the plain concrete. The phyllite aggregates in concrete required more water than
expected since the increased surface area by virtue of their flaky and elongated nature increased the
3

MATERIALS & DESIGN, 32(10), 5132-5140

watercement ratio demand which affects compressive and flexural behaviour of the plain concrete.
This led to seemingly dry concrete mixes with low slump values which affected workability.
Phyllite aggregates are already in use as coarse aggregates in concrete production by small scale
contractors. Ghana is yet to experience the use of phyllite aggregates in large scale structural
applications. First the observation of partial and complete collapse of structural concrete elements
made from phyllite aggregates do not convince the general public of the performance of phyllite
concrete compared to normal (granite) concrete. Secondly, there is insufficient information on the
structural performance of this material in terms of laboratory tests to provide adequate guidance and
confidence to local designers.

In order to address these limitations, a comprehensive research program is being undertaken at the
Department of Civil Engineering, Kwame Nkrumah University of Science and Technology in Ghana.
This paper deals with the overall flexural response of twelve reinforced concrete beams made from
phyllite aggregates. The choice of beam characteristics including concrete mix ratio were made to
reflect local construction practices. The significance of this study is to increase knowledge of the
material behaviour and therefore increase the utilization of the aggregates produced as waste from
underground mining activity.
2. Experimental program
Twelve (12) beams of design mix ratio 1:1.5:3 (cement: sand: coarse aggregates) and watercement
ratio of 0.5 were cast for experimental work. All beams are regular in cross-section, 110 mm wide and
225 mm in overall depth. Four different beam lengths (1400 mm, 1700 mm, 2000 mm and 2400 mm)
were used. Each length category of beam consisted of three different mild steel tension reinforcement
ratios of 1% (2R12 bars), 1.5% (3R12 bars) and 2% (4R12 bars). The 2% tension reinforcement
beams were treated as singly reinforced but with nominal hanger bars (2R6) to be more representative
of the practical situation whilst the top bars in the other beams contributed to flexural compression
capacity. The choice of beam characteristics was done to give adequate coverage of the effects of
under-reinforced, over-reinforced and spandepth ratios on the flexural behaviour of reinforced
concrete beams made from phyllite aggregates. Transverse reinforcement consisting of R6 bars bent
into closed stirrups was used in clear concrete cover of 15 mm on all sides of the beams cross-section.
All the beams had the same spacing of stirrups of 200 mm in both the shear and flexural zones
sufficient enough to ensure flexural failure prior to any shearing stress. Companion concrete cubes
(100 _ 100 _ 100 mm) and modulus of rupture beams (100 _ 100 _ 500 mm) were cast and cured in
water to determine the compressive and bending strengths of the plain concrete respectively. The
specimen descriptions are provided in Table 1 and Fig. 1.

MATERIALS & DESIGN, 32(10), 5132-5140

Tale 1 Specimen description

28 days
28 days
Percentage
Concrete
Concrete
tension
compressive
flexural
reinforcement
strength,
strength,
(%)
fcu (N/mm2) ft (N/mm2)
1.0
23.25
2.5
1.5
23.75
3.4
2.0
23.65
4.0
1.0
23.25
2.5
1.5
23.75
3.4
2.0
23.65
4.0
1.0
23.25
2.5

110 x 225 x2400

Span/eff.
depth
ratio
(le/d)
10.5

Shear
spacing
(mm)
200

B2

110 x 225 x 2400

10.5

200

B3

110 x 225 x2400

10.5

200

B4

110 x 225 x2000

8.9

200

B5

110 x 225 x2000

8.9

200

B6

110 x 225 x 2000

8.9

200

B7

110 x 225 x1700

7.5

200

B8

110 x 225 x1700

7.5

200

1.5

23.75

3.4

B9

110 x 225 x1700

7.5

200

2.0

23.65

4.0

B10

110 x 225 x 1500

6.0

200

1.0

23.25

2.5

B11

110 x 225 x1500

6.0

200

1.5

23.75

3.4

B12

110 x 225 x1500

6.0

200

2.0

23.65

4.0

BEAM
No.

BXDXL

B1

Fig.1. Beam details

MATERIALS & DESIGN, 32(10), 5132-5140

(a)

(b)

Fig.2. (a) Schematic sketch of experimental set-up. (b) Experimental set-up with instrumentation.

All the beams (B1B12) were cast and placed in wooden moulds and compacted with a shutter
vibrator. Curing was done using hessian mat laid on the beams in the atmosphere and watered
regularly. Beams were left to cure for 28 days after which they were tested. The beams were tested as
simply supported at the ends on steel beams that form part of a rigid steel frame. All the beams were
tested to failure under four point bending test as shown in Fig. 2a and b which ensured a constant
moment in the central region of the beam span assuming the self-weight of the beam is negligible.
Beam deflections at mid-span for a steady loading rate of 0.2kN/s were taken with the aid of a dial
gauge fixed at the bottom of each beam. Crack development on the concrete beam surface was closely
monitored to ascertain first flexural and shear cracks, and crack width at tension steel level. Crack
widths were measured at the load increments using a crack microscope of optical magnification X10
and reading to 0.02 mm. Crack patterns were marked on the beams after failure.

3. Flexural and shear strengths

The cracking moment Mcr of a plain prismatic concrete section, based on its modulus of rupture is
expressed as:

MATERIALS & DESIGN, 32(10), 5132-5140

(1)

where ft denotes the modulus of rupture of the concrete, and b and h the width and overall depth of
beam respectively. For a simply supported beam that is subjected to four-point bending test, the
ultimate flexural load Pult is given by:
(2)

where Mult denotes the ultimate moment of resistance of the beam and L is the span of the beam.
In accordance with British Standard (BS) 8110 [17] method, the theoretical shear strength of the
beams was calculated by considering contributions from the concrete section, the tension
reinforcement and the steel stirrups. In this case, the design shear strength V of the beam is given by:

(3)

where Vc which is the design concrete shear force carried by un-cracked concrete in compression and
the dowel action of the longitudinal reinforcement whilst Vs is the shear force carried by the steel
stirrups crossed by the diagonal crack.

4. Test results and discussions

4.1. Material properties of beams
Ordinary Portland cement conforming to BS 12:1989 [18] was used for concrete works. Natural
river sand graded according to BS 882:1990 [19] was used as fine aggregates. Phyllite aggregates
obtained from underground mining activities at the Ashanti Gold-field Limited, ObuasiGhana were
used as coarse aggregates; the aggregates were retained on sieve size 12 mm (1/2 in.) and passed 19
mm (3/4 in.). The physical and mechanical properties of the aggregates (Table 2) were considered
adequate taking into account the standards for aggregates suitable for concrete [20] as reported in an
earlier paper [16].
The 28 day compressive strength of the plain concrete cubes had values ranging from 23.25 to 23.75
N/mm2 whilst the flexural strength of plain concrete beams had values from 2.5 to 4.0 N/ mm2. The
average yield strength and percent elongation of the ribbed mild steel reinforcing bars used were
374.14 N/mm2 and 17% respectively.

MATERIALS & DESIGN, 32(10), 5132-5140

Table 2 Experimental Aggregate Strength Indices [16]
Physical Property
Specific gravity
Water Absorption (%)
Aggregate impact (%)
Aggregate Crushing (%)
Ten (%) fines (KN)
Flakiness Index (%)
Elongation Index (%)
Los Angeles Abrasion

Phyllite Aggregates
2.72
1.80
9.80
18.64
255.75
28.00
25.00
17.50

4.2. Loaddeflection behaviour

Loads that are applied to beams may either cause the beam to fail in shear, flexural tension or by
flexural concrete crushing. In or-der for the reinforced concrete (RC) beam to attain its full flexural
strength, shear links are provided to support the shear capacity of the concrete. The compatibility of
the concrete and the reinforcement causes them to deform (deflect) together. In a beam designed to
fail in flexure, flexural cracks begin to develop within concrete at point of high moments (middle third
of simply supported beam) when loading exceeds the flexural strength of concrete. From that point,
the tensile reinforcement proceeds with load carrying. After flexural cracks are formed a change in
slope of the loaddeflection curve occurs. Afterwards, another fairly linear curve may be obtained
when tensile yielding of the longitudinal steel reinforcement yields. Further increase in load beyond
steel yielding would result in the crushing and spalling of concrete cover. By this time higher shear
stresses have developed and any increase in load causes the disintegration of the compressed concrete
and failure.

MATERIALS & DESIGN, 32(10), 5132-5140

70

70

60

60

B3

B5

B2

B6

50

Load (KN)

Load (KN)

50

40
B1
30

40

30

20

20

10

10

B4

0
0 2 4 6 8 10 12 14 16

Midspan Deflectiom (mm)

10

Midspan Deflection (mm)

(a)

(b)
100

80
70

90

B8

B11
B12

80
60

B7
70

B9

Load (KN)

Load (KN)

50
40
30

B10

60
50
40
30

20
20
10

10

0
0

10

Midspan Deflection(mm)

Midspan Deflection(mm)

(c)

(d)
Fig. 3 Load-deflection curves of test beams

MATERIALS & DESIGN, 32(10), 5132-5140

Typical experimental loaddeflection curves for 1%, 1.5% and 2% reinforced beams with different
span to effective depth ratios are shown (Fig. 3ad). Generally in all the beams (B1B12), the slope of
the loaddeflection curves are fairly linear before first crack is formed. The gradients of the load
deflection curves of each of the three beams in a group at this stage are almost the same. This is
because the flexural strengths of the concrete beams were similar as they were cast from the same
mix. The first crack is mainly influenced by concrete flexural strength. The beams deflected as load
increased and it was observed that beams with less reinforcement deflected more under smaller loads
than beams with more reinforcement.

The under-reinforced beams with the highest (%) tension reinforcement (i.e. 1.5%) in all the ranges of
span to depth ratios were expected to produce the largest deflections at failure as a result of yielding
of tension steel and its associated increase in post-yield deflection. This was the case in Fig. 3b and c
where the ultimate deflections of B5 and B8 were higher than the deflections of their corresponding
1% and 2% beams (B4, B6, B7 and B9). However, this expectation is valid providing the mode of
failure of the beam is by yielding of tension reinforcement. In some cases of flexural-shear and
diagonal tension failure mode depending on the predominant mode of failure, under-reinforced beams
may not produce the largest deflection. This explains why in Fig. 3a and d, the over-rein-forced beams
B3 and B12 produced ultimate deflections higher than those of corresponding under-reinforced beams
(B1, B2, B10, and B11).

4.3. Cracking loads and ultimate loads of beams

The theoretical flexural and experimental failure loads are presented in Table 3. The theoretical
flexural failure loads were calculated by considering (1) the yielding of the steel in tension and (2)
crushing of the concrete in compression. The lowest among these predicted values is the governing
theoretical failure load for each of the twelve (12) beams. The corresponding theoretical shear
capacities Vc, Vs and V are also provided in Table 4. The design shear force (V) according to BS 8110
is calculated as the shear contributions from the design concrete shear force (V c) and the shear
resistance from the vertical steel stirrups (Vs).

10

MATERIALS & DESIGN, 32(10), 5132-5140

Table 3 Experimental and theoretical failure loads
Experimental load, (kN)

Theoretical flexural
Strength, P'ult (kN)
Based
Based
on
on
Steel
Concrete
yielding crushing

First
flexural
crack,

First
shear
crack

Failure
load,

Pcr

Ps

Pult

B1

18

22

36

30.19*

B2

16

30

52

B3

14

30

B4

16

B5

Beam
No.

Pcr/Pult

Pult/P'ult

Ps/Pult

42.00

0.50

1.19

0.61

45.28*

52.50

0.31

1.15

0.58

60

60.38

42.72*

0.23

1.40

0.50

24

40

35.30*

49.10

0.40

1.13

0.52

18

24

62

52.95*

61.38

0.29

1.17

0.39

B6

16

28

52

70.59

49.95*

0.31

1.04

0.54

B7

18

36

60

44.12*

61.38

0.30

1.35

0.60

B8

18

36

70

66.18*

76.73

0.26

1.05

0.51

B9

14

30

54

88.24

62.44*

0.26

0.86

0.56

B10

22

38

62

50.98*

70.93

0.35

1.21

0.61

B11

22

44

90

76.48*

88.66

0.24

1.17

0.49

B12

20

40

82

101.97

72.15*

0.24

1.13

0.49

0.31

1.15

0.53

Average
*Governing theoretical flexural load according to BS8110

Table 4 Shear strength of beams according to BS 8110

Beam No.

B1
B2
B3
B4
B5
B6
B7
B8
B9
B10
B11
B12
Average

Shear
span/effective
depth ratio

3.80
3.80
3.80
3.25
3.25
3.25
2.60
2.60
2.60
2.25
2.25
2.25

Theoretical shear strength (kN)

Concrete
Steel
Concrete
including
stirrups
including
steel in
only
steel in
tension
tension and
(Vc)
(Vs)
stirrups
(V=Vc+Vs)
33.00
40.17
73.17
37.84
40.17
78.01
41.80
40.17
81.97
33.00
40.17
73.17
37.84
40.17
78.01
41.80
40.17
81.97
33.00
40.17
73.17
37.84
40.17
78.01
41.80
40.17
81.97
33.00
40.17
73.17
37.84
40.17
78.01
41.80
40.17
81.97

0.67
0.79
0.71
0.72
0.63
0.67
1.09
0.95
0.71
1.15
1.16
0.95
0.85

0.49
0.66
0.73
0.54
0.79
0.63
0.82
0.89
0.66
0.85
1.15
1.00
0.77
11

MATERIALS & DESIGN, 32(10), 5132-5140

Tables 3 and 4 show that the experimental failure loads Pult averaged 1:15Pult (the theoretical flexural
strength of the beams) and 0.77V (the theoretical shear strength including the effects of tension
reinforcement and steel stirrups). The first flexural cracks were observed at 0.31P ult. As observed by
other researchers [8,21], the modulus of rupture values for the reinforced concrete beams is not a true
indicator for the first flexural crack loads as it over-estimates the cracking moment. The load Ps at
which first diagonal shear cracks appeared was 0.53Pult and 0.85Vc (the design concrete shear force).
This indicates that the shear resistance of the concrete beams made from phyllite (including the effects
of tension reinforcement) was mobilized at an average of 85% of the expected theoretical code
capacity.

Fig. 4 illustrates the relationship between the span to effective depth ratios and the experimental
failure loads. It was expected that for a particular percentage of tension reinforcement, the
experimental failure loads increased as the span to effective depth ratio (L/d) decreased. Since the
effective depth was kept constant for all the beams, it implies that as the span increased, design
moment was expected to increase to cause failure.

100
90

Ultimate load (kN)

80
70
60
1%

50

1.5%

40

2%

30
20
10
0
10.5

8.9
7.5
Span to effective depth ratio

Fig.4 Relationship between span to effective depth ratios and ultimate loads
Generally, for the different L/d ratios, beams with 1.5% tension reinforcement gave optimum
strength with the exception of the L/d ratio of 10.5 where beam B3 with 2% reinforcement recorded
the highest strength. For the same span to effective depth ratios B1, B4, B7 and B10 failed at lower
loads than corresponding B2, B5, B8 and B11. However, beams B6, B9 and B12 which all had 2%
tension reinforcements failed at lower failure loads than their corresponding 1.5% tension
reinforcement beams (B5, B8 and B11). This could be explained by the fact that B6, B9 and B12 were
12

MATERIALS & DESIGN, 32(10), 5132-5140

over-reinforced and therefore had the tendency to undergo brittle failure mode of concrete
compression. From these limited experimental results (Table 3 and Fig. 3), it could be inferred that BS
8110 can be used to obtain conservative estimates of the ultimate strength capacity and also provide
adequate load factor against short term loading for under-reinforced RC phyllite aggregate beams.

4.4. Deflection under service load and displacement ductility

BS 8110 has a design limit for deflection in order to satisfy the appearance and safety criteria of
design. The interpretation of the code provisions on deflection under service load is that deflection is
noticeable if it exceeds span/250. Table 5 presents deflection of the tested beams under service loads
and compares it with this design code requirement. The service load was obtained based on the load
factor method of BS 8110 for reinforced concrete beams.

Table 5

Service load deflection of beams

Beam
No.

Service
load,
(kN)

Experimental
service
deflection,
Es(mm)

BS8110
limit
Span/ 250
Bs(mm)

Es/ BS

B1
B2
B3
B4
B5
B6
B7
B8
B9
B10
B11
B12

21.17
30.58
35.29
23.50
36.47
30.58
35.29
41.17
31.76
36.47
52.94
48.23

4.38
5.32
7.60
3.35
4.25
3.78
3.91
3.85
3.09
2.60
2.30
3.30

8.40
8.40
8.40
7.20
7.20
7.20
6.00
6.00
6.00
5.20
5.20
5.20

0.52
0.63
0.90
0.47
0.59
0.53
0.65
0.64
0.52
0.50
0.45
0.64

The results indicate that the theoretical deflection values com-pared reasonably well with the
experimental values. Moreover, all the beams satisfied the code requirement. The ratio of Es/ BS
varied from 47% to 90% which is an indication that the BS 8110 predicts deflection satisfactorily in
RC beams made from phyllite aggregates. It is worthy of note that the code requirement includes long
term effects such as creep and shrinkage. Since the experimental results were obtained only under
short term, there is the need to investigate the creep and shrinkage effect on the service deflections. It
appeared that the experimental service deflection to the code limit for deflection was not significantly
influenced by the different tension reinforcement ratios that were considered.

13

MATERIALS & DESIGN, 32(10), 5132-5140

The capability of structural elements to undergo inelastic behaviour and absorb energy is known as
ductility. Ductility may be measured in terms of curvature, rotation or displacement. The
displacement ductility ratio is employed in this research. The dis-placement ductility was calculated
as the ratio of ultimate load deflection to deflection at steel yielding load estimated from the load
deflection behaviour of the beams. The steel yielding load deflections were obtained from the load
deflection curves at the second stage where the curve changes in slope. The results are presented in
Table 6.

Beams
No.
B1
B2
B3
B4
B5
B6
B7
B8
B9
B10
B11
B12

Table 6
Deflection ductility ratios of beams
Yield Stage
Ultimate Stage
Displacement
Load
(kN)
30
45
50
35
53
48
44
55
45
50
74
72

Deflection,
y (mm)
6.00
7.35
10.00
5.00
5.50
5.00
4.00
5.00
4.50
3.00
2.45
4.00

Load
(kN)
36
52
60
40
62
52
60
70
54
62
90
82

Deflection, ductility ratio

(u/y)
u (mm)
8.01
1.33
10.05
1.36
15.23
1.52
5.78
1.15
9.25
1.60
7.40
1.48
6.30
1.58
7.65
1.53
6.49
1.44
4.98
1.66
4.45
1.82
6.55
1.63

It was observed that keeping other parameters constant, ultimate displacement increased with
increasing span/ effective depth ratio. The displacement ductility ratios ranged from 1.15 to 1.82
which indicates low ductility behaviour. Contrary to observation by other researchers [810,21,22]
that a decrease in tensile reinforcement improved ductility, decreasing or increasing tensile
reinforcement in this case did not seem to have any notice-able influence on ductility. In general, high
ductility ratios indicate that a structural member is capable of undergoing large deflections prior to
failure. Structural members of ductility ratio between 3 and 5 are deemed to be of adequate ductility
and can be considered for structural members subjected to large displacements such as sudden forces
caused by earthquakes [21,22]. Factors that contribute to good ductility behaviour of aggregates are
toughness and good shock absorbance nature [8,16] indicated by the aggregate crushing value (ACV)
and aggregate impact value (AIV) which were both deemed adequate in Table 2. Furthermore, the
other condition necessary for good ductility behaviour of RC beams is the per cent elongation [23] of
the reinforcing steel rods which was greater than the minimum requirement of 12%. The reason for
the low ductility behaviour exhibited by all the beams could be as a result of non-ductile failure
14

MATERIALS & DESIGN, 32(10), 5132-5140

modes such as flexural-shear and diagonal tension failure modes instead of the expected pure flexural
failure modes.

4.5. Cracking behaviour and modes of failure

The appearances of reinforced concrete elements may greatly be detracted if cracks develop and more
so when cracks are prominent. Excessive cracks and deep cracks affect the durability of concrete.
Although strength may not be directly affected, the corrosion of reinforcement within the concrete can
eventually cause the structures to fail. As beams deflect during loading, cracks occur first in the
tension zone. Further loading tend to increase the width of crack which gradually lead to durability
problems over time and may cause final failure. During the loading of the beams, it was observed in
all the beams that vertical cracks known as flexural cracks propagated first in the pure bending region
(middle third of loading arrangement). On further load increment, as shear stresses increased, the
crack formed migrated to the support where shear concentrations are high. The major controllers of
crack development in this case are concrete flexural strength and reinforcement ratios [9,10]. The
effects of tensile reinforcement on the development of maximum crack width at the level of tension
reinforcement are presented in Fig. 5ad. The trend of crack width increment as loading increased
(Fig. 5ad) was similar to that of loaddeflection curves as shown in Fig. 3ad. It is evident that the
parameter that influences the slower growth of cracks is the amount of tension reinforcement, higher
amount being more effective at a given load level. Since all the beams have similar concrete flexural
strengths, the tension reinforcement ratio was the major factor in flexural crack development. Table 7
presents the crack widths at service loads of all twelve beams.

Table 7 Crack width at service loads

Service
Average
Beams
load (kN) service load
No.
crack width
(mm)
B1
21.17
0.10
B2
30.58
0.24
B3
35.29
0.20
B4
23.50
0.10
B5
36.47
0.36
B6
30.58
0.45
B7
35.29
0.27
B8
41.17
0.28
B9
31.76
0.20
B10
36.47
0.25
B11
52.94
0.40
B12
48.23
0.42

15

MATERIALS & DESIGN, 32(10), 5132-5140

80
60
Load (KN)

B3

40

B2

B1

20
0
0

0.2

0.4

0.6
0.8
1
Crack width (mm)

1.2

1.4

1.6

1.8

(a)

80

Load (KN)

60

B5

B6
B4

40
20
0
0

0.3

0.6
0.9
1.2
Crack width (mm)

1.5

1.8

2.1

(b)

80

Load (KN)

60

B9

B8

B7

40
20
0
0

0.3

0.6
0.9
Crack width (mm)

1.2

1.5

1.8

(c)

100

Load (KN)

80

B12

60

B10

B11

40
20
0
0

0.2

0.4
0.6
Crack width (mm)

0.8

(d)

Fig.5 Load-crack width relationship

Beams with 1% tension reinforcement ratio recorded very low average crack widths with an average
value of 0.18 mm. This could be explained by the fact that, these beams failed at small loads and
underwent very small deflections before failure. The average crack width at service load in the 1.5%
and 2% tension reinforcement beams was 0.32 mm which is greater than the code limit of 0.3 mm.
Therefore eight (8) of the twelve (12) beams did not satisfy the BS 8110 requirement for maximum
crack width limit. Durability problems may be encountered in such beams over time.
16

MATERIALS & DESIGN, 32(10), 5132-5140

The mode of failure of structural reinforced concrete beams de-pends on the amounts of flexural
tension reinforcement, flexural compression reinforcement, shear reinforcement, concrete
compressive strength and span to effective depth ratio. Table 8 shows the predicted and actual modes
of failure; and the type and number of cracks that developed. Fig. 6 shows the crack patterns on the
beams after failure. Although theoretical analyses (Tables 3 and 4) had indicated that all the 12 beams
were designed to fail in pure flexural mode by either steel yielding or concrete crushing, col-lapse
occurred mostly through combined diagonal tension and/or flexural-shear failure of the concrete in the
shear span.

Table 8

Predicted and experimental failure modes

Beam No. Predicted

Failure modes
Actual

B1

Steel yielding

Flexural-shear

B2

Steel yielding

Flexural-shear/Diagonal tension

B3

Concrete crushing

Flexural-shear/Diagonal tension

B4

Steel yielding

Flexural-shear

B5

Steel yielding

Flexural-shear/Diagonal tension

B6

Concrete crushing

Diagonal tension

B7

Steel yielding

Flexural-shear/Diagonal tension

B8

Steel yielding

Flexural-shear/Diagonal tension

B9

Concrete crushing

Flexural-shear/Diagonal tension

B10

Steel yielding

Flexural-shear/Diagonal tension

B11

Steel yielding

Flexural-shear/Diagonal tension

B12

Concrete crushing

Flexural-shear/Diagonal tension

17

MATERIALS & DESIGN, 32(10), 5132-5140

Fig.6 Crack configuration of beams after failure

18

MATERIALS & DESIGN, 32(10), 5132-5140

Nine (9) of the beams (B2, B3, B5 and B7B12) failed in combined diagonal tension/flexural-shear
failure modes whilst two
(2) Of the beams (B1 and B4) failed in flexural-shear mode. Only B6 failed in diagonal tension mode
alone. The observation of failure modes other than the predicted could be due to the fact that BS 8110
can only predict either pure flexural (steel yielding and/or concrete crushing) or shear failure but
cannot predict diagonal tension/flexural-shear failure unless the theoretical pure flexural and shear
capacities are close for which a flexural-shear failure could be predicted. The general lack of
agreement on shear transfer mechanisms in structural engineering could also be responsible for the
lack of precision in failure mode prediction for flexural beams that are not over-reinforced in steel
stirrup shear capacity. The design shear capacity of concrete beams is related to the compressive and
flexural strengths of the plain concrete, percentage longitudinal tension reinforcement, and the size
and spacing of the steel stirrups.

The type of aggregate also affects the shear strength mainly through its effect on aggregate-interlock
capacity [2426]. In this experimental study, the concrete beams made from phyllite aggregates
developed early shear cracks and failed prematurely in shear as observed by others in the study of
beams made of recycled concrete aggregates [3]. This could be attributed to the flaky nature of the
phyllite aggregates which appears to affect the effective mobilization of the aggregate interlock
contribution to shear capacity of the phyllite concrete.

Notwithstanding the fact that the experimental failure loads averaged 77% of the theoretical shear
capacity according to BS 8110 provisions, most of the beams failed in shear. More so, even though
the phyllite aggregate concrete is not lightweight concrete, the first diagonal shear crack in the beams
occurred at 0.85Vc (Table 4). On the evidence of the experimental/theoretical failure loads (Tables 3
and 4) and failure modes/crack configuration (Table 8 and Fig. 6), it is recommended that the code of
practice design shear stress values should be multiplied by 0.8 to assure that the predicted shear
capacity would be low and reasonable as com-pared to flexural capacity.

5. Conclusions
The experiment investigated the flexural behaviour of twelve RC beams made of phyllite aggregates
all of which were designed with adequate shear resistance as compared to flexural capacity. The
ultimate failure loads increased as the span to effective depth ratio decrease. The RC beams behaved
similarly as other RC beams made of other conventional aggregates materials with respect to load
deflection behaviour. Deflections compared reasonably well with BS 8110 requirements. However,
the average crack widths under service loads were greater than the limit permitted by the code
indicating that the beams may encounter durability problems with time. Crack development and
19

MATERIALS & DESIGN, 32(10), 5132-5140

patterns seemed to be influenced by percentage tension reinforcement. The beams exhibited low
ductility implying that such beams would not provide adequate warning prior to failure. It is
recommended that the BS 8110 code of practice design concrete shear stress values be multiplied by
0.8 due to the flaky nature of the phyllite aggregates which affects the aggregate interlock
contribution to shear capacity. In this case predicted shear capacity would be low and reason-able as
compared to flexural capacity.
It is concluded from the results that BS 8110 can be used to pro-vide adequate load factor against
flexural failure for under-rein-forced RC beams made of phyllite aggregates but the early
development of shear cracks, higher crack widths than allowable at service loads, the susceptibility of
the beams to shear failure mode for beams adequately designed against shear failure and low ductility
behaviour are of concern.

References
[1]
Demir I. Investigation of mechanical properties of concrete produced with waste granites
aggregates. Sci Res Essay 2009;4(4):26774.
[2]
Rahal K. Mechanical properties of concrete with recycled coarse aggregates. Build Environ
2007;42:40715.
[3]
Gonzalez-Fonteboa B, Martinez-Abella F. Shear strength of recycled beams. Constr Build
Mater 2007;21:88793.
[4]
Gonzalez-Fonteboa B, Martinerz-Abella F. Concrete waste aggregates from demolition waste
and silica fumes. Materials and mechanical properties. Build Environ 2008;43:42937.
[5]
Gonzalez-Fonteboa B, Martinez-Abella F, Martenez-Lage I, Javier EL. Structural shear
behaviour of recycled concrete with silica fume. Constr Build Mater 2009;23(11):340610.
[6]
Olanipekun EA, Olusola KO, Ata O. A comparative study of concrete properties using
coconut shells and palm kernel shells as coarse aggregates. Build Environ 2006;41:297301.
[7]
Teo DCL, Mannan MA, Kurian VJ, Ganapathy C. Lightweight concrete made from oil palm
shell (OPS): structural bond and durability properties. Build Environ 2007:261421.
[8]
Teo DL, Mannan MA, Kurian JV. Flexural behavior of reinforced lightweight concrete beams
made with oil palm shell (OPS). J Adv Concr Technol 2006;4(3):110.
[9]
Lim HS, Wee TH, Mansour MA, Kong KH. Flexural behaviour of reinforced lightweight
aggregate concrete beams. In: Proceedings of the 6th Asia-Pacific structural engineering and
construction conference. Kuala Lumpur; 2006. p.
A6882.
[10]
Kumar PS, Mannan MA, Kurian VJ, Achuytha H. Investigation on the flexural behaviour of
high-performance reinforced concrete beams using sandstone aggregates. Build Environ
2007;42:26229.
[11]
Valera TS, Ribeiro AP, Valenzuela-Diaz FR, Yoshiga A, Ormanji W, Toffoll SM. The effect
of phyllite as a filler for PVC plastisols. Annu Tech Conf Soc Plast Eng 2002;60(3):394953.
[12]
Garzon E, Sanchez-Soto PJ, Romero E. Physical and geotechnical properties of clay phyllites.
Appl Clay Sci 2010;48:30718.
20

MATERIALS & DESIGN, 32(10), 5132-5140

[13]
Vazquez J, Garzon E, Romerosa A, Serrano-Ruiz M, Sanchez-Soto P, Galera M. South Spain
soils containing phyllites improved by using cement and lime. Concr Sust Agric 2005:299304.
[14]
Montenari M, Servais T, Paris F. Palynological dating (acritarchs and chitinozoans) of Lower
paleozoic phyllites from the black forest/ southwestern Germany. Ser IIA Earth Planet Sci
2000;330(7):4939.
[15]
Ramamurthy T, Venkatappa RG, Singh J. Engineering behaviour of phyllites. Eng Geol
1993;33(3):20925.
[16]
Adom-Asamoah M, Afrifa Owusu R. A study of concrete properties using phyllite as coarse
aggregates. J Mater Des 2010;31(9):45616.
[17]

British Standards Institution. Structural use of concrete. BS 8110: Part 1; 1985.

[18]

British Standard Institute. Specification for ordinary Portland cement. BS12; 1989.

[19]

British Standard Institute. Specification of aggregates for concrete. BS882; 1983.

[20]

British Standard Institute. Specification of aggregates for concrete. BS812: Part 1; 1975.

[21]
Ashour SA. Effect of compressive strength and tensile reinforcement ratio on flexural
behaviour of high-strength concrete. Eng Struct 2000;22(5):41323.
[22]
Shuab HA, Ray B. Flexural behaviour of high-strength light weight concrete beams. ACI
Struct J 1991;88(1):6677.
[23]
Kankam CK, Adom-Asamoah M. Strength and ductility characteristics of reinforcing steel
bars milled from scrap metals. J Mater Des 2002;23:53745.
[24]
Taylor HPJ. The fundamental behaviour of reinforced concrete beams in bending and shear.
In: Proceedings ACI-ASCE shear symposium, Ottawa, 1973 (ACI Special Publication SP42).
ACI, Detroit; 1974. p. 4377.
[25]
Kong, Evans, Reinforced and prestressed concrete. 3rd ed. London: Chapman & Hall; 1994.
[26]
Jumaat MZ, Alengaram UJ, Mahmud H. Shear strength of oil palm shell foamed concrete
beams. J Mater Des 2009;30:222736.

21