Fracture Mechanics of Concrete and Concrete Structures High Performance, Fiber Reinforced Concrete, Special Loadings and Structural

Applications- B. H. Oh, et al. (eds) 2010 Korea Concrete Institute, ISBN 978-89-5708-182-2

Effects of fluidity and placing method of HPFRCC on tensile performance test results
N. Morii & S-C. Lim
Deros-Japan, Kanazawa, Japan

Y. Yamada & K. Rokugo

Department of Civil Engineering, Gifu University, Gifu, Japan

ABSTRACT: Tension tests were conducted on HPFRCC dumbbell-shaped specimens fabricated by shotcreting or casting to investigate the effects of their fluidity and placing method on the tension test results. Five HPFRCC mixtures with different fiber contents (2% and 1.7 % by volume) and fluidities were used. The differences of the placing methods and flow values scarcely affected the tensile strength. The ultimate tensile strain of a HPFRCC with a lower fiber content was more strongly affected by the fluidity and placing method, leading to a larger scatter. 1 INTRODUCTION The tensile performance of a high performance fiberreinforced cementitious composite (HPFRCC) is mostly evaluated in terms of the tensile stress-strain relationship and cracking properties. A higher ultimate tensile strain with a smaller scatter, as well as a smaller scatter of tensile strength, is desirable for a HPFRCC from the aspect of the durability and aesthetics of structures (JSCE 2007). A capability of dispersing fine cracks over the tension zone is also desirable. Uniaxial tension testing on dumbbell-shaped specimens is mostly selected to measure the tensile performance of a HPFRCC. Its fluidity during placing is evaluated by flow values. It is mostly placed by shotcreting or normal placing (casting). When applying a HPFRCC to actual structures, it is important to elucidate the effects of its fluidity and placing method (method of fabricating specimens) on its tension test results. Tension tests were conducted in this study on HPFRCC dumbbell-shaped specimens with different fluidities fabricated by shotcreting or casting to experimentally investigate the effects of their fluidity and placing method on the tension test results. 2 OOTLINE OF EXPERIMENT 2.1 Mix proportions and materials Five HPFRCC mixtures with different fiber contents (2% and 1.7 % by volume) and fluidities were used for the tests. Different fluidities were obtained by changing the dosage of a water-reducing admixture for 2% fiber mixtures (A) and both the dosages of the water-reducing admixture and an air-entraining admixture for 1.7% fiber mixtures (B). Polymer cement mortar made by premixing normal portland cement and a polymer was used as the matrix for the HPFRCCs. Water-based acrylic and a redispersible powder-type vinyl acetate/vinyl versatate copolymer were used as polymers for A and B mixtures, respectively. While the water-binder ratio was kept constant at 34% for all mixtures, the binder content and unit water content for A mixtures were higher than for B mixtures. Short fibers were a blend of two types: high strength polyvinyl alcohol (PVA) fibers 12 mm in length and 40 m in diameter and ultrahigh strength polyethylene (PE) fibers 9 mm in length and 12 m in diameter. 2.2 Methods of placing HPFRCC Dumbbell-shaped specimens were fabricated by two placing methods, shotcreting and casting, to investigate the effects of the fluidity and placing methods of HPFRCCs on their tensile performance. For specimens by shotcreting, the material conveyed by a squeeze pump was pneumatically (0.7 N/mm2) shotcreted onto the molds. Pneumatic shotcreting expeled part of the air entrained in HPFRCCs, generally reducing the air content from the as-mixed value. For this reason, the material for casting was shotcreted onto a container before placing to equalize the air contents by both methods. Table 2 gives the air content and flow value, which is an index to fluidity, after mixing and after

Table of 20 types of dumbbell-shaped specimens. explicitly J = 1. D (Test h, T )conditions h (1)

Specimen Water-binder Unit water Fiber Water-reduced

The proportionality coefficient D(h,T) called ratio(%) content (W) volume (%) is agent ratio Monolithic AHN moisture permeability and it is a nonlinear function Casting Cut-and-mold AHNC of the humidity h and temperature (Baant None AH relative 34% 336.3 2.0vol% T 1.0(AH/AH) Monolithic AHS Shotcreting & Najjar 1972). The moisture mass balance requires w (h ) = G ( )Cut-and-mold AHSC + c s c s e that the variation in time of the water mass per unit Monolithic )h (g ALN Casting c c Cut-and-mold ALNC e volume (water content 2.0vol% w) be equal to the (4) 0.33(AL/AH) None AL of concrete 34% 336.3 Monolithic ALS divergence of the moisture flux J Shotcreting ALSC Cut-and-mold )h (g
1 , , , 1 1 10 1

accounts for the evolution of hydration reaction and SF content.Shaping This sorption isotherm AE agent Placing Group name reads ratio methods methods of specimen

the first term (gel isotherm) represents the Casting The water 34% content w can be expressed as the sum where BM 0.5(BM/BH) physically 1.5(BM/BH) bound (adsorbed) water and the second 301.5 1.7vol% of the evaporable water we (capillary water, water term (capillary Shotcreting isotherm) represents the capillary vapor, and adsorbed water) and the non-evaporable water. This expression is valid only for low content Casting (chemically bound) water wn (Mills 1966, of SF. The coefficient G represents the amount of 0.5(BL/BH) 1.0(BL/BH) BL 34% Pantazopoulo & Mills 301.5 1995). It 1.7vol% is reasonable to water per unit volume held in the gel pores at 100% Shotcreting assume that the evaporable water is a function of relative humidity, and it can be expressed (Norling relative humidity, h, degree of hydration, c, and Mjornell 1997) as degree of silica fume reaction, Table 2. Air content and flow values. s, i.e. we=we(h,c,s) = age-dependent sorption/desorption isotherm c c + k s s Air content (%) (Norling Mjonell 1997). Under this assumption G ( c s ) = k vg (5) Flow value and (mm) c vg s by 1 shotcreting into Equation 2 one After shotcreting Mix substituting proportions Equation Before Before After Before dropping After 15 strokes Before dropping After 15 strokes shotcreting shotcreting obtains c s AH 147 170 135 where k vg and 169 k vg are material 19 parameters. 3.2 From the AL 100 132 100 maximum amount 129 20 5.6 that can of water per unit volume w w w h BH 110 e & 153 121 152 18 4.1 e e (both capillary pores and gel pores), one & & (3) 105 fill all pores128 c + + ( D h ) = s + wn 15 2.2 h 106 134 h BM t can calculate K1 as one obtains s 120 BL 101 c 100 118 5.6 2.6
, 1

BH = J
w t

34%

301.5

1.7vol%

1.0(BH/BH)

(2)

1.0(BH/BH)

c K ( c s )Monolithic e Casting
10 , 1 1

Shotcreting

Cut-and-mold Monolithic Cut-and-mold Monolithic Cut-and-mold Monolithic Cut-and-mold Monolithic Cut-and-mold 1 Monolithic Cut-and-mold

c BHN
BHNC BHS BHSC BMN BMNC BMS BMSC BLN BLNC BLS BLSC
1

where we/h is the slope of the sorption/desorption isotherm (also called moisture capacity). The governing equation (Equation 3) must be completed by appropriate boundary and initial conditions. 60 30 The relation between the amount of evaporable water and relative humidity is called adsorption 40 80 85 15 isotherm if85measured with 40 increasing relativity mm) humidity and desorption isotherm(Unit in the opposite case. Neglecting their difference (Xi et al. 1994), in Figure 1. Dumbbell-shaped specimens. the following, sorption isotherm will be used with reference to The bothspread sorption desorption conditions. shotcreting. of and mortar was measured imBy the way, if the hysteresis of the moisture mediately after lifting the flow cone (before jigging) isotherm would be taken into account, different and after 15 jigging strokes based on thetwo mortar flow relation, evaporable water vs relative humidity, must test method specified in JIS R 5201 to determine the be used according tocontent the signwas of the variation of the flow values. The air measured using an relativity humidity. The shape of the sorption air meter for mortar. isotherm for HPC is influenced by many parameters, especially those that influence extent and rate of the 2.3 Methods of shaping chemical reactions and,specimens in turn, determine pore structure and pore size distribution Dumbbell-shaped specimens shown (water-to-cement in Figure 1 were ratio, cement SF cut-andcontent, shaped by twochemical methods:composition, monolithic and curing time and method, temperature, mix additives, mold. Monolithic specimens were shaped by shotetc.). In the literature various formulations can be creting the HPFRCC onto, or casting it in, dumbbellfound to describe the sorption isotherm of normal shaped molds. concrete (Xi et al. 1994). However, the present Cut-and-mold specimens (Rokugoin2007) were paper the semi-empirical expression proposed by shaped into dumbbell-shaped specimens as follows: Norling Mjornell (1997) is adopted because it Prepare HPFRCC boards 300 by 330 by 15 mm by
Proceedings of FraMCoS-7, May 23-28, 2010

g h shotcreting and casting; cut out sticks 29.5 by c c 300 by w s + s G e c s a 15 mm from the boards using concrete cutter; place (6) the in dumbbell-shaped molds; and pour K ( cut = c s )sticks h to form shoulders g parts the HPFRCC in the enlarged c c e (avoid the central control zones). In the shotcreting process,c the HPFRCC was apksvg and g Thefor material parameters and k vg and 1 can plied, both monolithic cut-and-mold specibe calibrated by fitting experimental data relevant to mens, with a spray gun in passes in the direction of free (evaporable) water testing. contentAsin at tensile forces to act during for concrete casting, the various ages (Di Luzio & Cusatis 2009b). HPFRCC was placed in the center of each specimen and allowed to flow in the direction of tensile forces to act during testing. 2.2 Temperature evolution As given in Table 1, 20 types of dumbbell-shaped Note that, at early age, since chemical five reactions specimens were fabricated bythe combining mixassociated with cement hydration and SF reaction tures, two placing methods, and two shaping methare exothermic, thespecimens temperature field is not uniform ods. The names of were given in the colfor non-adiabatic systems if the environmental umn on the right side of theeven table.
10 0 0.188 0.22 1 1 1 , 1 10 1 1

temperature is constant. Heat conduction can be described in concrete, at least for temperature not 2.4 Tension100C test method exceeding (Baant & Kaplan 1996), by Fouriers law, which Figure 2 shows the reads tension test setup. Upper and
lower catches set in a steel frame (mass: approxiq = T (7) mately 30 kg, external size: 250 by 250 by 500 mm) held the shoulders of each dumbbell-shaped specimen to transfer forces. where q the is tensile the heat flux, T is the absolute A simple and hydraulic loading apparatus (a in handtemperature, is the heat conductivity; this cranked BRI-type bond test apparatus) was fixed to

2 (N/mm ) (N/mm )

Load cell
Tensile stress

6 5 4 3 2 1 0 6 4 2 0 6 4 2 0 6 4 2 0 6 4 2 0 0 1

= D ( h , T ) h

Jack Pin Rod

Clamp Specimen Displacement tranceducers

Tensile stress

Base plate

Guide pins

2 (N/mm ) (N/mm )

The proportionality coefficient D(h,T) moisture permeability and it is a nonlinea of the relative humidity h and temperature AHN AHS & Najjar 1972). The moisture mass balanc AHNC AHSC that the variation in time of the water mas ALS volume of concrete (water content w) be eq ALSC divergence of the moisture flux J
ALNC = J
w t

Fixer

ALN

2 (N/mm ) (N/mm )

Base plate

Clamp

the top of the steel frame and pulled up the upper catch through a loading bar along two guide pins to apply the tensile forces to the specimen. The lower catch was fix-supported with the bottom of the steel frame, whereas the upper catch was pin-supported with a hinge provided in the loading bar. The displacement and load during tension testing were measured using displacement gauges directly set to the specimens and a load cell fixed on top of the loading apparatus. 3 AIR CONTENT AND FLUIDITY TEST RESULTS Table 2 gives the air content and flow, which was an index to fluidity, of five HPFRCC mixtures. All mixtures contain polymers. The B mixtures also contain an air-entraining admixture. The as-mixed air contents of all mixtures excepting BL were 15% or more. The reason for the low as-mixed air content of BL, as well as its small flow, was not clear. The air contents of all mixtures after shotcreting were2 to 6%, being lower than immediately after mixing, due to losses of air from the material during shotcreting. The flow values after shotcreting did not appreciably differ from the as-mixed values, showing little effect of the reductions in the air content. The flow values of AH and BH having a high water-reducing admixture dosage (high water-reducing admixture ratio) were large. 4 TENSILE PERFORMANCE TEST RESULTS AND DISCUSSIONS Figure 3 shows the stress-strain diagrams of dumbbell-shaped specimens (five specimens for each set of

(N/mm (N/mm ) )

Tensile stress

(N/mm ) (N/mm )

Figure 2. Tension test setup.

Tensile stress

BHSC The water content w can be expressed a of the evaporable water we (capillary wa vapor, and adsorbed water) and the non-e BHN (chemically bound) water wn (Mil BHNC Pantazopoulo & Mills 1995). It is reas assume that the evaporable water is a fu relative humidity, h, degree of hydration BMS degree of silica fume reaction, BMSC s, i.e. we=w BMN = BMNC age-dependent sorption/desorption (Norling Mjonell 1997). Under this assum by BLN substituting Equation 1 into Equati BLS obtains BLNC BLSC

Tensile stress

BHS

w h e + ( D h) = we h h t
5 6 0 1 2 3 4 (%) Strain (%)

where we/h is the slope of the sorption/ capac governing equation (Equation 3) must be by appropriate boundary and initial test conditions) determined by tension tests. Table 3 conditi The relation between the amount gives their average tensile strength and average ulti- of e water and relative humidity is called mate tensile strain. In this study, the maximum tensile isotherm if measured with increasing stress on the tensile stress-strain diagram is defined humidity andtensile desorption isotherm as the tensile strength, and the strain at the in- in th case. Neglecting their difference (Xi et al. flection point immediately before the tensile stress fithe following, sorption isotherm nally stops increasing is defined as the ultimate ten-will be reference to both sorption and desorption c sile strain. Therefore, the ultimate tensile strain was By the way, if the hysteresis of the in most cases greater than the tensile strain at the isotherm would be taken into account, two point of tensile strength. relation, evaporable water vs relative humi be used according to the sign of the varia 4.1 Tensile performance monolithic The specimens relativityof humidity. shape of the isotherm for HPC is influenced by many p especially those that influence extent and 4.1.1 Tensile strength and ultimate tensile strain chemical reactions and, in turn, The average tensile strength of A specimens con- determ structure and pore2,size distribution taining 2.0% fibers was 5.5 N/mm being higher by(waterratio, composition, SF around 20% than thosecement of B chemical specimens (4 to 5 curing time and method, temperature, N/mm2) containing 1.7% fibers. The differences of mix etc.). In thevalues literature various formulatio the placing methods and flow scarcely affected found to describe the sorption isotherm the average tensile strength. concrete (Xi etthe al.average 1994). ultimate However, in th In regard to A specimens, the4%, semi-empirical expression pro tensile strain was paper 3.5% to with the value of deNorling Mjornell (1997) is adopted b posited specimens being greater than that of shotFigure 3. Stress-strainisotherm diagrams of dumbbell-shaped (also called specimens. moisture
Proceedings of FraMCoS-7, May 23-28, 2010

2 3 4 (%) Strain (%)

w e & & c + s + w c s
5 6

Table J = 3. D (T hensile , T )hstrength and ultimate tensile strain.

explicitly accounts for the evolution of hydration reaction and SF content. This sorption isotherm Monolithic specimen Cut-and-mold specimen / reads The proportionality coefficient D(h,T) is called Placing Tensile Ultimate Tensile Ultimate Tensile Ultimate Specimen moisture permeability and it is a nonlinear methods strength strain function strength strain strength rario strain ratio of the relative humidity h(N/mm2) and temperature (%) T (Baant (N/mm2) (%) & Najjar 1972). The moisture Casting 5.72 mass balance 4.13 requires 4.34 ( we (h0.48 c s ) = G 0.76 c s ) 0.12 + AH that the variation in time of the water3.42 mass per unit Shotcreting 5.54 3.96 0.80 0.72 0.24 (g c )h e 0.28 c volume of concrete content w ) be equal to the Casting (water 5.78 3.92 4.34 1.09 0.75 (4) AL divergence of the moisture 5.30 flux J Shotcreting 3.54 5.30 1.84 1.00 0.52 )h (g (1)
1 , , , 1 1 10 1

w the of the evaporable water we (capillary water, water vapor, and adsorbed water) and the non-evaporable creted specimens by around was evident (Mills 1966, (chemically bound) water10%. wn As from the results of AH and AL, the average ultimate Pantazopoulo & Mills 1995). It is reasonable to tensile strain affected byis the value. assume that was the scarcely evaporable water a flow function of In regard to Bhspecimens, average ultimate relative humidity, , degree ofthe hydration, c, and tensile strain was in the range of 1% 6%, we =we(with h,c,the s) degree of silica fume reaction, s, i.e.to value of deposited specimens being greater than that = age-dependent sorption/desorption isotherm of shotcreted specimens. Among speci(Norling Mjonell 1997). Under thisdeposited assumption and mens, the average ultimate strain of those with relaby substituting Equation 1 into Equation 2 one tively low fluidity of 130 mm or less (BM and BL) obtains
was approximately 5%, whereas that of specimens with high fluidity (BH)w was as low as around 2%. w w h e e e Among shotcreted specimens, however, the & & & average ) = + ( D h (3) c + s + wn h relatively h t strain of ultimate high fluidity specimens c s (BH) was greater (around 2%) than those of relatively fluidity specimens (BM and BL, around is the slope of the sorption/desorption where low we/h 1%). isotherm (also called moisture capacity). The Figure 3equation reveals that the scatters ofbe the ultimate governing (Equation 3) must completed tensile strain of B specimens tend to be higher by appropriate boundary and initial conditions. than those ofrelation A specimens. A specimens, the The betweenAmong the amount of evaporable scatter of the ultimate tensile deposited water and relative humidity is strain called of adsorption specimens with a large flow (AHN) was smallest. isotherm if measured with increasing relativity Among B specimen, the scatter of deposited humidity and desorption isotherm in the opposite specimens with atheir small flow (BMN) smallest, case. Neglecting difference (Xi etwas al. 1994), in while that of deposited a large the following, sorptionspecimens isothermwith will be used flow with (BHN) was reference tolargest. both sorption and desorption conditions. As stated above, higher ultimate tensile strains By the way, if thewere hysteresis of the moisture with a smaller scatter desirable as a tensile perisotherm would be taken into account, two different formance of HPFRCCs. Comparison between the rerelation, evaporable water vs relative humidity, must sults of A and B specimens with a fiber content be 1.7% used and according to the sign of the variation of the of 2.0 % by The volume demonstrated that the relativity humidity. shape of the sorption ultimate tensile strain of a HPFRCC with parameters, a lower fiisotherm for HPC is influenced by many ber content was more strongly affected by the fluidity especially those that influence extent and rate of the (flow value) and placing method (execution method), chemical reactions and, in turn, determine pore leading toand a larger scatter. In regard(water-to-cement to the A mixstructure pore size distribution tures with a fiber content of 2%, AH with a high fluratio,(a cement chemical composition, SF content, idity flow value of around 170 mm) among deposcuring time andand method, temperature, mix additives, ited specimens AL various with a formulations low fluidity (around etc.). In the literature canstabe 130 mm) among shotcreted specimens achieved found to describe the sorption isotherm of normal ble average ultimate strains with small scatters. In reconcrete (Xi et al. 1994). However, in the present gard to B mixtures with 1.7% fibers, deposited paper the with semi-empirical expression proposed by specimens a low fluidityis (BM and BL) achieved Norling Mjornell (1997) adopted because it high ultimate tensile strains with small scatters.
Proceedings of FraMCoS-7, May 23-28, 2010

Casting Shotcreting w = J Casting t BM Shotcreting Casting The water content BL Shotcreting BH

4.78 4.86 4.96 4.21 4.21 can be 4.78

2.23 1.89 5.88 1.05 4.55 expressed as 1.20

3.82 4.13 (2) 3.99 2.99 2.72 sum 3.06

t at tu / ft

term (capillary isotherm) represents the capillary water. This expression is valid only for low content ALS BLS G1 represents BMS BHS 1.05 the amount of of SF. The coefficient ALN BLN BMN BHN water per unit volume held in the gel pores at 100% AHS AHN and it can be expressed (Norling relative1 humidity, Mjornell 1997) as
0.95
,

0.47 0.31 0.61 0.23 where the 0.41 physically 0.32

0.21 c 1 1 c 0.17 0.80 0.10 0.71 0.22 first 0.65 term (gel isotherm) represents the 0.09 bound (adsorbed) 0.26 water and the second 0.64
( K 0.80 )e c s 0.85
, 1 10

G ( c0.9 )= k c c + k s s s vg c vg s
1

(5)

where kcvg and ksvg are material parameters. From the 0 amount2of water 4 6 maximum per unit volume that can (%) tensile strain (%)gel pores), one tu fill all poresUltimate (both capillary pores and as one obtains can calculate 1 tensile Figure 4. Ratio K of stress at point of ultimate tensile
strain to tensile strength.
0.188

0.85

w s+ s G e c s in a Based on the authors field experience, fibers (6) K ( c s ) = immediately after being sprayed into the HPFRCC air g h appeared to be surrounded by the (the binder c matrix e c and sand) but segregated from it. It is therefore dec s fluidity (softsirable the mixture hasksufficient The that material parameters vg and k vg and g1 can ness) to integrate fibers in the matrix onrelevant the shotbe calibrated by fitting experimental data to creted surface. free (evaporable) water content in concrete at
0 0.22

1 1

10

h g c c
1

10

4.1.2 Shape of stress-strain diagram As stated above, the ultimate tensile strain defined in 2.2 study Temperature evolution this was in most cases larger than the strain at the point of tensile strength, tensile reactions stress at Note that, at early age, sinceand the the chemical the point of the ultimate tensile strain was in most associated with cement hydration and SF reaction cases lower than the tensile strength. Therefore, the are exothermic, the temperature field is not uniform ratio of the tensilesystems stress ateven the point the ultimate for non-adiabatic if the of environmental tensile strain to the tensile strength was determined temperature is constant. Heat conduction can be as shown in Figure 4. When the ultimate tensile strain described in concrete, at least for temperature not exceeds 2%, this ratio was 0.94 or higher, demonexceeding 100C (Baant & Kaplan 1996), by strating that a which large tensile Fouriers law, reads stress was retained to a large strain level. The ratios of shotcreted specimens with 1.7% q= T fibers (BHS, BMS, and BLS) were low. (7) 4.1.3 State of cracking Figure 5 shows the state of cracking in monolithic where q is thespecimens heat flux, T with is the dumbbell-shaped along the absolute average temperature, and is the heat conductivity; inlarge this ultimate tensile strain values. Specimens having average ultimate strain show a large number of

various ages (Di Luzio & Cusatis 2009b).

= D ( h , T ) h

AHN(4.13%) AHN(4.13%)

AHS(3.42%) AHS(3.42%)

The proportionality coefficient D(h,T) moisture permeability and it is a nonlinea of the relative humidity h and temperature & Najjar 1972). The moisture mass balanc that the variation in time of the water mas volume of concrete (water content w) be eq divergence of the moisture flux J
= J
w t

ALN(3.92%) ALN(3.92%)

ALS(3.54%) ALS(3.54%)

BHN(2.23%) BHN(2.23%)

BHS(1.89%) BHS(1.89%)

The water content w can be expressed a of the evaporable water we (capillary wa vapor, and adsorbed water) and the non-e (chemically bound) water wn (Mil Pantazopoulo & Mills 1995). It is reas assume that the evaporable water is a fu relative humidity, h, degree of hydration degree of silica fume reaction, s, i.e. we=w = age-dependent sorption/desorption (Norling Mjonell 1997). Under this assum by substituting Equation 1 into Equati obtains
w h e + ( D h) = we h h t

where we/h is the slope of the sorption/ isotherm (also called moisture capac governing equation (Equation 3) must be by appropriate boundary and initial conditi The relation between the amount of e water and relative humidity is called isotherm if measured with increasing humidity and desorption isotherm in th BLN(4.55%) BLN(4.55%) BLS BLS(1.20 BLS(1.20 )) case. Neglecting their difference (Xi et al. the following, sorption isotherm will be reference to both sorption and desorption c By the way, if the hysteresis of the isotherm would be taken into account, two relation, evaporable water vs relative humi be used according to the sign of the varia relativity humidity. The shape of the isotherm for HPC is influenced by many p especially those that influence extent and Figure 5. Cracking in monolithic dumbbell-shaped specimens. chemical reactions and, in turn, determ structure and pore size distribution (waterratio,were cement chemical composition, SF cracks dispersed over a wide area. Mixtures that exceeded 3%, cracks found in areas outside of curing time and method, temperature, mix showed a large scatter of the ultimate tensile strain the control zones of specimens. In contrast, cracks In the literature various formulatio values in Figure 3 showed large differences among were localized in etc.). specimens with an average ultimate found to describe the sorption isotherm specimens in the number of cracks and cracking area strain of around 1%. concrete (Xi et al. 1994). However, in th in this photograph. When the average ultimate strain paper the semi-empirical expression pro Norling Mjornell (1997) is adopted b
Proceedings of FraMCoS-7, May 23-28, 2010

BMN(5.88%) BMN(5.88%)

BMS(1.05%) BMS(1.05%)

w e & & c + s + w c s

explicitly accounts for the evolution of hydration reaction and SF content. This sorption isotherm AHNC(0.48%) AHSC(0.80%) The proportionality coefficient D(h,T) is called reads moisture permeability and it is a nonlinear function of the relative humidity h and temperature T (Baant & Najjar 1972). The moisture mass balance requires w (h ) = G ( ) + c s c s e that the variation in time of the water mass per unit (g c c )h e volume of concrete (water content w) be equal to the (4) divergence of the moisture flux J )h (g
J

= D ( h , T ) h

(1)

10

the first term (gel isotherm) represents the The water content w can be expressed as the sum where physically bound (adsorbed) water and the second of the evaporable water we (capillary water, water term (capillary isotherm) represents the capillary vapor, and adsorbed water) and the non-evaporable water. This expression valid only for low content (chemically bound) water wn (Mills 1966, of SF. The coefficient is G1 represents the amount of Pantazopoulo & Mills 1995). It is reasonable to water per unit volume held in the gel pores at 100% assume that the evaporable water is a function of relative humidity, and it can be expressed (Norling BHNC(0.47%) , and Mjornell 1997) as relative humidity, h, degree of hydration, cBHSC(0.31%) degree of silica fume reaction, s, i.e. we=we(h,c,s) = age-dependent sorption/desorption isotherm c c+ ks s (Norling Mjonell 1997). Under this assumption and G ( c s ) = k vg (5) c vg s by substituting Equation 1 into Equation 2 one obtains where kcvg and ksvg are material parameters. From the maximum amount of water per unit volume that can w w w h e e & & & w + + + ( D h ) = e (3) fill all pores (both capillary pores and gel pores), one c s n h h t can calculate K1 as one obtains BMNC(0.61%) BMSC(0.23%) c s
, 1

ALNC(1.09%) = J
w t

ALSC(1.84%)

(2)

K ( c s )e
, 1

10

where we/h is the slope of the sorption/desorption g h c c w s + s G e isotherm (also called moisture capacity). The c s governing equation (Equation 3) must be completed K ( ) = (6) c s by appropriate boundary and initial conditions. g h The relation between the amount of evaporable e c c water and relative humidity is called adsorption isotherm if measured with increasing relativity The material parameters kcvg and ksvg and g1 can BLNC(0.41%) BLSC(0.32%) humidity and desorption isotherm in the opposite be calibrated by fitting experimental data relevant to case. Neglecting their difference (Xi et al. 1994), in free (evaporable) water content in concrete at the following, sorption isotherm will be used with various ages (Di Luzio & Cusatis 2009b). reference to both sorption and desorption conditions. By the way, if the hysteresis of the moisture 2.2 Temperature evolution isotherm would be taken into account, two different relation, evaporable water vs relative humidity, must Note that, at early age, since the chemical reactions be used according to the sign of the variation of the associated with cement hydration and SF reaction relativity humidity. The shape of the sorption are exothermic, the temperature field is not uniform Figure 6. Cracking in cut-and-mold dumbbell-shaped specimens. for non-adiabatic systems even if the environmental isotherm for HPC is influenced by many parameters, especially those that influence extent and rate of the temperature is constant. Heat conduction can be chemical reactions and, ofin turn, determine pore also described in concrete, at least for temperature not 4.2 Tensile performance cut-and-mold specimens smaller by more than 70%. The tensile strength structure and pore size distribution (water-to-cement exceeding 100C (Baant &A Kaplan 1996), by and ultimate tensile strain of specimens were ratio, cement chemical composition, SF content, slightly Fouriers law, which reads greater than those of B specimens. Those curingFiber time and method, temperature, mix additives, of A specimens 4.2.1 content and tensile performance with a smaller flo-w tended to be q = T etc.). In the literature various formulations be greater. As shown in Figure 3 and Table 3, the average can tensile (7) The ultimate tensile strain of ALSC, which strengths of all cut-and-mold excepting found to describe the sorptionspecimens isotherm of normal was made by shotcreting mixture AL with a fiber ALSC, which was of However, AL by shotcreting, were concrete (Xi et al.made 1994). in the present where q 2% is the heat flux, T is being the the absolute content of by volume, was 1.8%, larglower of monolithic specimens by 15% by to paper than the those semi-empirical expression proposed temperature, and is the heat conductivity; in this 35%, with their average ultimate tensile strain being Norling Mjornell (1997) is adopted because it est among cut-and-mold specimens.
10 0 0.188 0.22 1 1 1 , 1 10 1 1

Proceedings of FraMCoS-7, May 23-28, 2010

The tensile performance of monolithic specimens was high, because fibers were oriented along the molded surfaces in the control zone in the center of each specimen. On the other hand, the tensile performance of cut-and-mold specimens was low, presumably because fibers crossing the sawed surfaces in the control zones were cut off by a concrete cutter. It is inferred that actual slab-shaped HPFRCC members will demonstrate an intermediate tensile performance between those of monolithic and cutand-mold specimens. 4.2.2 Shape of stress-strain diagram Figure 6 shows the state of cracking in cut-and-mold dumbbell-shaped specimens along with the average ultimate tensile strain values. Though cracks in ALSC specimens with an ultimate tensile strain of 1.84% spread over a wide range, cracks in the other specimens with an ultimate tensile strain of less than 1% were localized to at most two locations. 5 CONCLUSIONS Tension tests were conducted on HPFRCC dumbbell-shaped specimens fabricated by shotcreting or casting to experimentally investigate the effects of their fluidity and placing method on the tension test results. Five HPFRCC mixtures with different fiber contents (2% and 1.7 % by volume) and fluidities were used. Polymer cement mortar was used as the matrix. The material for casting was shotcreted onto a container before casting to equalize the air contents by both methods.

J =after D (h,shotcreting T ) h The flow values did not appreciably differ from the as-mixed values, showing little effect of the reductions the air content. coefficient The differ- D(h,T) Theinproportionality ences of the placing methods and flow moisture permeability and it values is a nonlinea scarcely affected the average tensile strength. of the relative humidity h and temperature The average ultimate was 3.5% to balanc & Najjartensile 1972).strain The moisture mass 4% for A specimens containing 2.0% fibers, and that the variation in time of the water mas was in the rangevolume of 1% of toconcrete 6% for (water B specimens content w) be eq containing 1.7% divergence fibers. The of scatters of the flux ultimate the moisture J tensile strain of B specimens tended to be higher than those of A specimens. The ultimate tensile wwith = aJ lower fiber content was strain of a HPFRCC t more strongly affected by the fluidity and placing method, leading to aThe larger scatter. Specimens having water content w can be expressed a large average ultimate strain showed a large number of the evaporable water we (capillary wa of cracks dispersed over and a wide area. water) and the non-e vapor, adsorbed The average tensile strengthsbound) of most of cut-and(chemically water wn (Mil mold specimens were lower than those of monolithic Pantazopoulo & Mills 1995). It is reas specimens by 15% to 35%, average ulti- is a fu assume thatwith the their evaporable water mate tensile strain being also smaller by more than relative humidity, h, degree of hydration 70%. degree of silica fume reaction, s, i.e. we=w

REFERENCES

JSCE. 2007. Recommendations for Design and Construction of High Performance Fiber Reinforced Cement wComposites w w h (HPFRCC). Concrete with Multiple Fine Cracks e Library e w e + ( D h ) = c s 127, JSCE. /Concrete Series 82. h http://www. hEngineering t c s jsce.or.jp/committee/cocrete/e/hpfrcc_JSCE.pdf Rokugo, K., Nishimatsu, H., Kato, H. & Uchida, Y. 2007. Direct tension testing method for is strain-hardening fiberthe slope of the sorption/ where we/h reinforced cement-based composites (SHCC) using dumbisotherm (also called moisture capac bell specimens, Nonlocal Modelling of Failure of Materigoverning equation (Equation 3) must be als, Aedificatio Publishers, 163-171.

= age-dependent sorption/desorption (Norling Mjonell 1997). Under this assum by substituting Equation 1 into Equati obtains

&+

&+

by appropriate boundary and initial conditi The relation between the amount of e water and relative humidity is called isotherm if measured with increasing humidity and desorption isotherm in th case. Neglecting their difference (Xi et al. the following, sorption isotherm will be reference to both sorption and desorption c By the way, if the hysteresis of the isotherm would be taken into account, two relation, evaporable water vs relative humi be used according to the sign of the varia relativity humidity. The shape of the isotherm for HPC is influenced by many p especially those that influence extent and chemical reactions and, in turn, determ structure and pore size distribution (waterratio, cement chemical composition, SF curing time and method, temperature, mix etc.). In the literature various formulatio found to describe the sorption isotherm concrete (Xi et al. 1994). However, in th paper the semi-empirical expression pro Norling Mjornell (1997) is adopted b
Proceedings of FraMCoS-7, May 23-28, 2010