Construction and Building Materials 51 (2014) 179–186

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier .com/locate /conbui ldmat

Mechanical performance of self-compacting concrete reinforcedwith steel fibers

0950-0618/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.conbuildmat.2013.10.054

⇑ Corresponding author. Tel.: +98 (912) 146 2590; fax: +98 (21) 6601 4828.E-mail address: e.molaei.r@gmail.com (E. Molaei Raisi).

Alireza Khaloo, Elias Molaei Raisi ⇑, Payam Hosseini, Hamidreza TahsiriCenter of Excellence in Structures & Earthquake Engineering, Sharif University of Technology, Azadi Ave., Tehran, Iran

h i g h l i g h t s

�Workability of SCC decreases considerably by adding steel fibers.� Compressive strength loss in HS-SCC class was lower than that of MS-SCC class.� Flexural toughness enhances significantly when steel fibers were utilized.

a r t i c l e i n f o

Article history:Received 3 August 2013Received in revised form 10 October 2013Accepted 31 October 2013Available online 26 November 2013

Keywords:Self-compacting concreteRheological propertiesMechanical performanceSteel fibersStrength class

a b s t r a c t

Self-compacting concrete (SCC) is a highly-workable concrete that without any vibration or impact andunder its own weight fills the formwork, and it also passes easily through small spaces between rebars. Inthis paper, the effect of steel fibers on rheological properties, compressive strength, splitting tensilestrength, flexural strength, and flexural toughness of SCC specimens, using four different steel fiber vol-ume fractions (0.5%,1%,1.5%,and 2%), were investigated. Two mix designs with strengths of 40 MPa (med-ium strength) and 60 MPa (high strength) were considered. Rheological properties were determinedthrough slump flow time and diameter, L-box, and V-funnel flow time tests. Mechanical characteristicswere obtained through compressive strength and splitting tensile strength tests with standard cylindricalspecimens of 150 � 300 mm, and flexural strength and flexural toughness tests were performed by usingbeams of 100 � 140 � 1200 mm.

The results revealed that the workability of medium and high strength SCC classes is reduced byincreasing the steel fiber volume fraction, and using high percentages of fibers led to decrease ofother rheological characteristics that have been specified by EFNARC and ACI 237R. On the contrary,splitting tensile strength, flexural strength, and flexural toughness are increased by increasing thepercentage of fibers; however compressive strength is decreased by increasing the percentage offibers.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

In the recent two decades, Self-Compacting Concrete (SCC) hasbeen identified as one of the important achievements in the con-crete industry. Due to its high workability, SCC is compacted underits own weight without any vibration. To obtain the desired work-ability in SCC, more fine aggregates and superplasticizer admixtureare to be used. Besides, in comparison to the conventional con-crete, a smaller maximum size of aggregate should be utilized.When the superplasticizer is used, creep and shrinkage cracks in-crease, and it leads to more segregation. The use of filler and flyash increases the mortar volume of SCC, which in turn, improvesthe workability [1–5].

Hardened SCC has similar mechanical properties compared toconventional concrete [6]. Before 1970s, a kind of concrete wasused in Europe which needed less vibration; SCC, however, hadnot been developed until late 1980s. Initial ideas of SCC weregenerated in Japan, though Sweden was the first Europeancountry which built transportation structures with SCC in1990s [6,7].

Plain concrete is brittle, thus it seems essential to use materialsthat can fix this problem. Using fibers in the concrete decreasesbrittle fracture of the concrete significantly, and under variousloads, especially the compressive loads, tensile loads, and blastloads, the behavior of fiber-reinforced concrete will be ductile. Bybridging between sides of cracks, fibers tend to preserve the inte-gration of concrete until high deformation and therefore preventbrittle failure. Nowadays, fiber-reinforced concrete is utilized invarious areas including road pavements, sidewalks, bridges, liningof tunnel segments, and slabs [8].

Fig. 1. Steel fibers.

Table 2Characteristics of used steel fibers.

Length (mm) Width (mm) Thickness (mm) Aspect ratio

20.6 1.8 0.5 20

180 A. Khaloo et al. / Construction and Building Materials 51 (2014) 179–186

There have been conducted a number of studies on the SCC.El-Dieb [9,10] studied mechanical and durability properties ofultra-high strength fiber-reinforced concrete (UHS-FRC) withself-compacting characteristics and the influence of fibers onrheological properties. Siddique [11] investigated the propertiesof SCC made with different amounts of fly ash. According to Favaet al. [12], in SCC with ground-granulated blast furnace slag(GGBFS), the strength can be increased. Flexural performances ofreinforced, pre-stressed and composite self-compacting concretebeams were also studied by Cattaneo et al. [13]. Soutsos et al.[14] also scrutinized the flexural performance of fiber-reinforcedconcrete made with steel and synthetic fiber.

According to the above mentioned studies, the present researchis aimed at investigating the influence of steel fibers on rheologicaland mechanical properties of SCC with different strength classes.To determine the rheological properties of SCC, slump flow timeand diameter, V-funnel flow time, and L-box tests were used. In or-der to investigate the mechanical properties of the fiber-reinforcedSCC, compressive strength, splitting tensile strength, flexuralstrength, and flexural toughness of SCC beams were obtained.

Fig. 2. Gradation curves of filler, fine and coarse aggregates.

2. Experimental plan

2.1. Materials

The cement used was commercially available ASTM Type II Portland cementwith a specific gravity of 3.12 and a fineness of 295 m2/kg. In addition, silica fumewas used as a reactive pozzolanic material in conjunction with lower W/B ratio toproduce high strength SCC class. Table 1 summarizes the chemical composition ofcement and silica fume. Silica fume particles size falls in the range of 0.05 to0.2 lm. Steel fibers used had hooked shape. Fig. 1 illustrates the image of the steelfibers and the physical characteristics of the steel fibers are given in Table 2.

Superplasticizer was used for improving the workability of concrete mixtures isbased on high performance poly-carboxylic, with the specific gravity of 1.1. Besides,a polysaccharide based viscosity modifying agent (VMA) was utilized with a specificgravity of 1.01 ± 0.01 VMA is used for retaining the integrity of the SCC compositionand it also decreases the bleeding phenomenon.

It has to be noted that aggregates are of different parts, namely the coarse andfine aggregate as well as the filler. Coarse aggregate with a maximum size of12.5 mm, a specific gravity of 2.65, and water absorption of 1.43% was used. The fineaggregate is natural river sand with a specific gravity of 2.61, absorption of 1.95%,and fineness modulus of 3.11. Both the coarse and fine aggregates met the require-ments of ASTM C 33 [15]. Potable water was used for preparation of concretemixtures.

Fig. 2 provides the gradation curves for the coarse aggregates, fine aggregates,and the filler, as obtained through sieve analysis. The coarse and fine aggregatesused in this research were in the form of saturated surface dry (SSD).

2.2. Mixing and testing procedures

In the current study, various fiber volume fractions of 0.5%, 1%, 1.5%, and 2%were used. Also, it should be mentioned that the fiber volume fractions were basedon the mortar volume of SCC. In addition, the two strength classes of SCC (40 MPaand 60 MPa) were investigated. The concrete mix designs are given in Table 3. Over-all, 10 mix designs were made, two of which were plain and without any fiberswhile the others had fibers. MS-SCC-FX and HS-SCC-FX are the two kinds of speci-mens terms used for coding. MS-SCC-FX stands for medium-strength self-compact-ing concrete and X is the steel fiber volume fraction in percent. HS-SCC-FX standsfor high-strength self-compacting concrete with the steel fiber volume fraction ofX percent.

The process of making SCC with fibers is the same as that of fiber-free conven-tional concrete, with the fibers being added during the mixing process. The SCCmixture was made in 3 steps. First, the powder materials and aggregates weremixed in dry form for 1 min. Then half of the water containing the whole superp-

Table 1Chemical analysis of cement and silica fume (values are in percent).

Chemical analysis SiO2 Fe2O3 CaO Al2O3 MgO SO3 Na2O + K2O

Cement 20.4 3.9 63.0 4.9 1.7 2.0 0.9Silica fume 93.2 1.5 0.4 0.7 0.3 0.1 1.4

lasticizer was poured and mixed for 3 min. Following that, a 1 min rest was allowedand finally the rest of the water containing VMA was added to the mixture andmixed for another 2 min.

After the materials were mixed, fresh concrete tests were performed to deter-mine the rheological properties of the SCC. The flow rate of SCC depends on the vis-cosity of the concrete. SCC must have four main characteristics. First, it should beable to fill out the form with its weight. Besides, it should be of an acceptable levelof resistance against segregation. Yet another important characteristic of SCC is itsability to pass through spaces between rebars, and finally, it needs to have a smoothsurface after demoulding. In order to achieve these characteristics, there are sometests in EFNARC and ACI 237R such as slump flow time and diameter, V-funnel flowtime, visual stability index, J-ring, and L-box [1,6].

According to Nagataki and Fujiwara’s research [16], slump flow time and diam-eter tests are two common methods to determine the flow characteristics of unob-structed concrete in horizontal surface. In these tests, the fresh concrete is pouredinto a slump cone. When the cone is withdrawn upwards, the time it takes from thebeginning of the upward movement to when the concrete has flowed to a diameterof 500 mm is measured, called the T50 time. The largest diameter of the flow spreadof the concrete and the diameter of the spread at right angles to it are then mea-sured and the mean is the slump-flow diameter.

The L-box test is used to assess the passing ability of self-compacting concreteto flow through tight openings including spaces between reinforcing bars and otherobstructions without segregation or blocking. There are two variations of the test,namely the two-bar test and the three-bar test. The three-bar test, which was alsoused for the purpose of the present study, simulates more congested reinforcement.Hence, the concrete is poured from the container into the filling hopper of the L-box. Then the gate is raised so that the concrete flows into the horizontal sectionof the box. When the movement is ceased, the vertical distances are measured, atthe end of the horizontal section of the L-box, between the top layer of the concreteand the top of the horizontal section of the box, and at three positions equallyspaced across the width of the box. Differing from the height of the horizontal sec-tion of the box, these three measurements are used to calculate the mean depth ofconcrete as H2. The same procedure is followed to calculate the depth of concreteimmediately behind the gate as H1. The value of H2/H1 as blocking ratio is thenreported.

Table 3Mix proportions and characteristics of SCC reference mixtures.

Specimen W/B ratio Cement(kg/m3)

Silica fume(kg/m3)

Water(kg/m3)

Filler(kg/m3)

Fine aggregate(kg/m3)

Coarse aggregate(kg/m3)

SP(kg/m3)

VMA(kg/m3)

Mortar volume(l/m3)

MS-SCC 0.48 400 0 192 200 977 651 8 3.2 761HS-SCC 0.38 500 50 209 200 861 574 7.5 4 795

Fig. 4. Test set-up and monitoring.

A. Khaloo et al. / Construction and Building Materials 51 (2014) 179–186 181

For measuring the V-funnel flow time test a V-shaped funnel is filled with freshconcrete and the time taken for the concrete to flow out of the funnel is measuredand recorded as the V-funnel flow time. In this research, to determine the rheolog-ical properties of SCC, the measuring apparatus and minimum and maximumacceptable value are according to EFNARC [6].

Immediately after the completion of fresh concrete tests, the fresh concrete waspoured into the oiled molds to form 150 � 300 mm cylinders for compressive andsplitting tensile strengths testing at 7, 28, and 91 days and into100 � 140 � 1200 mm prisms for flexural strength and toughness testing at28 days. The samples were de-molded after 24 h and then cured in a water tank(at 20 ± 2C) for 7, 28, and 91 days. The compressive and splitting tensile strengthswere determined according to ASTM C 39 [17] and ASTM C 496 [18], respectively,at 7, 28, and 91 days.

In the present research, the 3-point bending test was run on100 � 140 � 1200 mm beam specimens. The specimens were tested in the struc-tural dynamic strong floor lab at Sharif University of Technology by using the dis-placement control method, and also the pace of loading was 1 mm/min. Fig. 3illustrates the set-up of testing and Fig. 4 provides a picture of a lab, showing thetest set-up and monitoring procedure. Flexural moment in the middle of the spanwas also obtained, and the flexural strength, i.e. the maximum tensile stress in max-imum bending load, was calculated according to ASTM C78 [19].

One of the characteristics of reinforced concrete with fibers is its high flexuraltoughness property. This property of concrete can decrease the risk of concrete ele-ments failure especially under dynamic load. Flexural toughness properties of fiber-reinforced concrete can be measured by a toughness test [20]. In this research, inorder to determine the flexural toughness, beams with dimensions of100 � 140 � 1200 mm were used. Flexural toughness is the area under the load–deflection curve of concrete in flexure up until a deflection of 1/150 times the span,which corresponds to 8 mm for the used specimens [21].

3. Results and discussion

3.1. Fresh concrete properties

The obtained results revealed that the slump flow times wereless than 3.4 s. According to EFNARC [6], this time must bebetween 2–5 s. The time for the two mixtures (i.e. HS-SCC andHS-SCC-F0.5) was also found to be less than 2 s. The slump flowdiameters of all mixtures were in the range of 640–800 mm. Theacceptable diameter commonly falls in the range of 650–800 mm(EFNARC [6]). The lowest diameter was 640 mm for the MS-SCC-F2, which is less than the minimum limitation of this test, andthe highest value belonged to HS-SCC.

Blocking ratios of all mixtures in the L-box test were in therange of 0.65–0.96. The lowest ratio was 0.65 for the MS-SCC-F2

Fig. 3. Test

and the highest value was 0.96 for the HS-SCC. The lowest recom-mended blocking ratio is 0.8 according to EFNARC [6]. Therefore,since the blocking ratios for MS-SCC-F1, MS-SCC-F1.5, and MS-SCC-F2 were less than 0.8, they are discarded as not being accept-able. The findings also indicated that V-funnel flow times were inthe range of 3–19.1 s. The lowest V-funnel flow time was 3 s asmeasured for HS-SCC mixture, while the MS-SCC-F2 mixture hadthe highest V-funnel flow time of 19.1 s. The values recommendedare in the range of 6–12 s, as it is mentioned in EFNARC [6]. Accord-ing to the results, two mixtures (MS-SCC-F0.5 and MS-SCC-F1) fellinto this range. The rheological tests indicated that the mediumstrength SCC specimens reinforced with high fiber volume frac-tions, especially 2%, cannot be considered as an acceptable mixturefor heavily reinforced sections due to difficulty of passing of freshconcrete through the rebars. In this regard, the L-box test resultswhich evaluate the passing ability of SCC showed the blocking ra-tios of some mixtures were under the minimum limitation.

Visual inspection of fresh concrete did not prove the segrega-tion, yet a little bleeding was observed in some mixtures includingthe MS-SCC and MS-SCC-F0.5. Incorporating fibers made the SCCgenerally less viscous, and addition of fibers to SCC led to a de-crease in the workability, indicating that fibers do not allow aggre-gates to move freely. Referring to Figs. 5–8, it can be assumed thatfibers decrease the workability of SCC and high steel fiber volumefractions decreases the workability to values lower than the mini-mum acceptable values recommended by EFNARC [6].

set-up.

Fig. 5. Slump flow time vs. different percentages of fibers.

Fig. 6. Slump flow diameter as function of fibers content.

Fig. 7. Variation of blocking ratio with increasing the percentages of steel fibers.

Fig. 8. V-funnel time of different SCC mixtures.

Table 4Compressive strength results of SCC specimens.

Compressive strength (MPa)

Specimen 7 days 28 days 91 days

MS-SCC 29.4 39.7 47.3MS-SCC-F0.5 26.7 38.0 46.3MS-SCC-F1.0 24.2 35.1 43.2MS-SCC-F1.5 23.2 33.9 40.3MS-SCC-F2.0 22.4 32.3 40.2HS-SCC 47.4 59.5 66.3HS-SCC-F0.5 46.0 59.1 65.4HS-SCC-F1.0 42.2 58.2 61.8HS-SCC-F1.5 42.1 57.0 60.5HS-SCC-F2.0 40.4 55.2 59.1

Table 5Splitting tensile strength results of SCC specimens.

Splitting tensile strength (MPa)

Specimen 7 days 28 days 91 days

MS-SCC 2.20 3.37 3.46MS-SCC-F0.5 2.26 3.40 3.59MS-SCC-F1.0 2.43 3.71 3.80MS-SCC-F1.5 2.48 3.97 4.01MS-SCC-F2.0 2.75 4.33 4.43HS-SCC 3.16 5.02 5.28HS-SCC-F0.5 3.44 5.14 5.30HS-SCC-F1.0 3.61 5.21 5.38HS-SCC-F1.5 3.88 5.41 5.70HS-SCC-F2.0 4.33 5.88 6.02

182 A. Khaloo et al. / Construction and Building Materials 51 (2014) 179–186

3.2. Mechanical performance

3.2.1. Compressive strengthThe results obtained from the compressive strength test at 7,

28, and 91 days, and with different fiber volume fractions are givenin Table 4. As can be observed, compressive strength decreases byincreasing the percentages of fibers. Strength reduction might bedue to decreasing of workability of the concrete [22]. Increasingthe percentage of steel fibers results in decreasing the workabilityof concrete which in turn causes reduction in compaction levels ofvibrated concrete [22]. This issue can be highlighted in SCC mix-tures when no compaction method is applied for molding of them

and the compaction only performed by their own weights. In thisregard, according to higher reduction in compressive strength athigher steel fiber volume fractions, one should be careful aboutthe application of these types of SCCs for heavily reinforced struc-tural sections.

For medium strength SCCs, at different ages, as can be seen inTable 4, addition of fiber volume fractions from 0.5% to 2% causesthe compressive strength to decrease 4.3%, 11.6%, 14.6%, and18.6%, respectively, for 28-day specimens with respect to the plainconcrete (MS-SCC).

Also Table 4 demonstrates the compressive strength of highstrength SCCs, at different ages. As can be seen, compressivestrength decreases by adding more fiber volume fractions from0.5% to 2%. The decrease in compressive strength for 28-day spec-imens is 0.7%, 2.2%, 4.2%, and 7.5%, respectively, with respect to theplain SCC specimen (HS-SCC).

The decrease in compressive strength for HS-SCC specimens areless than MS-SCC specimens, and it may be due to HS-SCC

Table 6Flexural strength and maximum bending load of beams at 28 days.

Specimen Maximum load (kN) Flexural strength (MPa)

MS-SCC 4.81 4.41MS-SCC-F0.5 5.00 4.59MS-SCC-F1.0 5.32 4.88MS-SCC-F1.5 5.45 5.00MS-SCC-F2.0 7.02 6.44HS-SCC 5.90 5.41HS-SCC-F0.5 6.23 5.72HS-SCC-F1.0 6.44 5.91HS-SCC-F1.5 6.91 6.34HS-SCC-F2.0 8.11 7.44

A. Khaloo et al. / Construction and Building Materials 51 (2014) 179–186 183

specimens are workable than MS-SCC specimens. Also, the addi-tion of silica fume in HS-SCC mixtures can influence on bondingbetween cement matrix and steel fibers which leads to form stron-ger interface structure and improve steel fibers-cement matrixinteractions [23].

The compressive strength increases with increasing age, yet therate of this increase descends when the age of the specimensascends.

Fig. 11. Load–deflection diagram for reinforced beams with 1% steel fiber volumefraction.

3.2.2. Splitting tensile strengthSplitting tensile strength of the specimens at ages of 7, 28 and

91 days, and with different fiber volume fractions are shown in

Fig. 9. Load–deflection diagram for plain concrete beams.

Fig. 10. Load-deflection diagram for reinforced beams with 0.5% steel fiber volumefraction.

Table 5. As can be seen, splitting tensile strength increases by theaddition of fibers. Adding more fiber volume fractions causes moreincrease in splitting tensile strength.

Fig. 12. Load–deflection diagram for reinforced beams with 1.5% steel fiber volumefraction.

Fig. 13. Load–deflection diagram for reinforced beams with 2% steel fiber volumefraction.

Fig. 14. Absorbed energy–deflection diagram for beams with 0.5% steel fibervolume fraction.

Fig. 15. Absorbed energy–deflection diagram for beams with 1% steel fiber volumefraction.

Fig. 16. Absorbed energy–deflection diagram for beams with 1.5% steel fibervolume fraction.

Fig. 17. Absorbed energy–deflection diagram for beams with 2% steel fiber volumefraction.

184 A. Khaloo et al. / Construction and Building Materials 51 (2014) 179–186

For medium strength class SCCs, as can be observed in Table 5,the addition of fiber volume fractions 0.5%, 1%, 1.5%, and 2% causesthe splitting tensile strength to increase 0.9%, 10.1%, 17.8%, and28.5%, respectively, with respect to the plain specimen (MS-SCC)at 28 days of curing.

Table 5 also illustrates the splitting tensile for high strength SCCclass. It can be observed that the fiber volume fractions influencethe splitting tensile strength, in which an increase in fibers per-centage leads to a rise in splitting tensile strength. Increasing thefiber volume fractions from 0.5% to 2% causes splitting tensilestrength to increase 1.2%, 3.8%, 7.8%, and 17.1%, respectively, withrespect to the plain concrete (HS-SCC) at 28 days of curing.

3.2.3. Flexural strengthThe maximum bending load and flexural strength of beams are

shown in Table 6. According to Table 6, by increasing the fiber vol-ume fractions, the maximum bending load and flexural strengthincreased.

As can be seen from Table 6, for medium strength SCC class, theaddition of fiber volume fractions from 0.5% to 2% causes the flex-ural strength increases by 4.1%, 10.6%, 13.4%, and 46%, respectively,with respect to the plain SCC specimen (MS-SCC) at 28 days. More-over, the addition of fiber volume fractions from 0.5% to 2% for highstrength SCC class causes the flexural strengths increase 5.7%, 9.2%,

17.2%, and 37.5%, respectively, with respect to the plain specimen(HS-SCC) at the age of 28 days.

3.2.4. Load–deflection relationshipsThe diagrams for all the beam specimens are shown in Figs. 9–13.

The solid lines represent the load–deflection curves for specimenswith high strength SCC, and the dashed lines indicate the valuesobtained from testing of medium strength SCC beam specimens.

For beams with medium strength SCC class, the addition of steelfibers increases the maximum bending load. Addition of 0.5%, 1%,1.5%, and 2% steel fiber volume fractions causes the maximumbending loads to increase by 3.9%, 10.6%, 13.3%, and 45.9%, respec-tively, with respect to the plain SCC (MS-SCC). For high strengthSCC class, maximum bending loads of beam specimens containing0.5%, 1%, 1.5%, and 2% fiber volume fractions increase 5.5%, 9.1%,17.1%, and 37.4%, respectively, in comparison with the plain beamspecimen (HS-SCC). The main reason for this increase is the perfor-mance of randomly distributed steel fibers which provide bridgingforces across micro-cracks and thus prevents them from growing[24,25]. As a result, by increasing the fiber volume fractions themaximum bending load of beam specimens increases.

At a constant fiber volume fraction, a beam specimen with highstrength SCC class had higher maximum bending load than a beamspecimen with medium strength SCC. Addition of steel fiber vol-ume fractions 0%, 0.5%, 1%, 1.5%, and 2% causes the maximum

A. Khaloo et al. / Construction and Building Materials 51 (2014) 179–186 185

bending loads for high strength SCC beam specimens, 22.6%, 24.6%,21%, 26.8%, and 15.5%, respectively, more than corresponding med-ium strength SCC beam specimens. Indeed, due to the lower W/Bratio, the maximum bending loads of beam specimens with highstrength SCC class were higher. Moreover, silica fume in HS-SCCmixtures causes better bonding between cement matrix and steelfibers which leads to improve steel fibers-cement matrix interac-tions [23].

3.2.5. Flexural toughnessFlexural toughness shows the ability of concrete to absorb en-

ergy. Flexural toughness, in fact, refers to the area under load-deflection curve. The amount of flexural toughness of a concretebeam is known as the absorbed energy of the concrete. The curvesin Figs. 14–17 display the absorbed energy in term of beamdeflection.

For beam specimens with medium strength SCC class, flexuraltoughness, which is equal to absorbed energy corresponding to8 mm deflection, for different fiber volume fractions of 0.5%, 1%,1.5%, and 2% was 2.9, 3.7, 4.5, and 6.7 times, respectively, higherthan the plain beam specimen (MS-SCC). For beam specimens withhigh strength SCC class, flexural toughness for different fiber vol-ume fractions from 0.5% to 2% was 1.6, 2.6, 5.2, and 6.1 times,respectively, more than the plain specimen (HS-SCC).

According to Fig. 10, it can be observed that the HS-SCC-F0.5collapsed early and its mid-span deflection was 7.6 mm at failure.Also, Flexural toughness for MS-SCC-F0.5 and MS-SCC-F1 was morethan HS-SCC-F0.5 and HS-SCC-F1, respectively; however the max-imum bending loads of MS-SCC-F0.5 and MS-SCC-F1 were less thanHS-SCC-F0.5 and HS-SCC-F1. As it is shown in Figs. 16 and 17, forspecimens with higher fiber volume fractions (1.5% and 2%), theflexural toughness of specimens with high strength SCC class ismore than that of the specimens with medium strength SCC class.

It should be noted that, high strength concrete is more brittlethan medium strength concrete, and also the addition of fibers im-proves concrete ductility. At low fiber volume fractions, fibers can-not enhance efficiently the ductility of concrete because theamount of fibers is low. Consequently, as it can be seen fromFigs. 14–17, at lower fiber volume fractions (0.5% and 1%), beamswith medium strength SCC class have higher flexural toughness.However, for specimens with higher fiber volume fractions (1.5%and 2%), beams with medium strength SCC class have lower flex-ural toughness in comparison with high strength SCC class beams.

4. Conclusions

An experimental investigation was performed to study the rhe-ological and mechanical behavior of two strength classes of theself-compacting concrete reinforced with steel fibers. The resultsobtained in this study can be summarized as follows:

� Addition of steel fibers decreases the workability of the SCC;especially when 2% fiber volume fraction are added to the mix-ture, the workability falls below the minimum limits specifiedin EFNARC. The results revealed that the addition of fibers tothe SCC reduces the passing ability of the SCC, and hence, SCCwith high fiber volume fraction cannot pass easily through therebars.

� Addition of steel fibers reduces the compressive strength of theSCC in both strength classes (MS and HS). It can be attributed todecrease in the workability of self-compacting concrete. There-fore, concrete with the lower fiber volume fraction is denserthan the one with high fiber volume fraction. For 28-day spec-imens, addition of 2% fiber volume fraction at medium strengthSCC class led to 18.6% decrease with respect to the plain

concrete (MS-SCC), as well as, for high strength SCC class thedecrease is about 7.5% with respect to the plain concrete (HS-SCC).

� The presence of steel fibers increased the splitting tensilestrength of the SCC specimens. Steel fibers enhance the splittingtensile strength through bridging the gap between two sides ofcrack opening. Addition of 2% fiber volume fraction improvedsplitting tensile strength by 28.5% and 17.1% with respect tothe medium and high strength plain specimens, respectively.

� Flexural strength was also improved by utilization of steelfibers. Addition of fibers improves the ultimate load capacityof the SCC beams, and it leads to an increase in the flexuralstrength.

� Flexural toughness of the SCC beams increases by increasing thepercentage of fibers. In fact, steel fibers tend to increase theflexural toughness and in turn enhance the ductility of concreteelements. In low fiber volume fractions (0.5% and 1%), beamsmade with medium strength SCC had more flexural toughnesscompared to beams made with high strength SCC class, becausehigh strength SCC is more brittle, and fibers cannot enhance theductility of SCC due to low amount of incorporated fiber in thematrix. However, for specimens with high fiber volume frac-tions (1.5% and 2%), beams with medium strength SCC classhad lower flexural toughness.

Acknowledgment

Special thanks of the authors are extended to Center of Excel-lence for Structures and Earthquake Engineering for partial supportof this research study. Cooperations of strong floor laboratory spe-cialists are also appreciated.

References

[1] ACI Committee 237. Self-consolidating concrete, ACI 237R-07. FarmingtonHills: American Concrete Institute; 2007.

[2] Bartos PJM. Self-compacting concrete. Concrete 1999;33(4):9–14.[3] Okamura HM, Ouchi M. Self-compacting concrete. J Adv Concr Technol

2003;1(1):5–15.[4] Collepardi M, Collepardi S, Ogoumah Olagat JJ, Troli R. Laboratory-test and

filled-experience SCC’s. In: Proc of the 3rd international symposium on self-compacting concrete. Reykjavik (Iceland); 17–20 August, 2003. p. 904–12.

[5] Bouzoubaa N, Lachemi M. Self-compacting concrete incorporating highvolumes of class F fly ash-preliminary results. Cem Concr Res 2001;31:413–20.

[6] EFNARC. Specifications and guidelines for self-compacting concrete. Englished. European Federation for Spec Constr Chem & Concr Syst; May 2005.

[7] Kim JK, Han SH. Mechanical properties of self-flowing concrete. High-PerformConcr: Des Mater Adv Concr Technol 1997;172:637–52.

[8] Khaloo AR, Afshari M. Flexural behavior of small steel fibre reinforced concreteslabs. Cem Concr Compos 2005;27:141–9.

[9] El-Dieb AS. Mechanical, durability and microstructural characteristics of ultra-high-strength self-compacting concrete incorporating steel fibers. Mater Des2009;30(10):4286–92.

[10] El-Dieb AS, Reda Taha MM. Flow characteristic and acceptance criteria of fiber-reinforced self-compacted concrete (FR-SCC). Constr Build Mater2012;27:585–96.

[11] Siddique R. Properties of self-compacting concrete containing class F fly ash.Mater Des 2011;32:1501–7.

[12] Fava et al. In: Proceedings of the 3rd international RILEM symposium on self-compacting concrete. Reykjavik: RILEM Publications S.A.R.L.; 2003. p. 628–36.

[13] Cattaneo et al. Flexural behaviour of reinforced, prestressed and compositeself-consolidating concrete beams. Constr Build Mater 2012;36:826–37.

[14] Soutsos et al. Flexural performance of fibre reinforced concrete made withsteel and synthetic fibres. Constr Build Mater 2012;36:704–10.

[15] ASTM C 33. Standard specification for concrete aggregates. AmericanStandards for Testing and Materials; 2003.

[16] Nagataki S, Fujiwara H. Self-compacting property of highly-flowable concrete.In: Malhotra VM, editor. Am Concr Inst SP 154 301-14 June; 1995.

[17] ASTM C 39. Standard test method for compressive strength of cylindricalconcrete specimens. American Standards for Testing and Materials; 2012.

[18] ASTM C 496. Standard test method for splitting tensile strength of cylindricalconcrete specimens. American Standards for Testing and Materials; 2011.

186 A. Khaloo et al. / Construction and Building Materials 51 (2014) 179–186

[19] ASTM C 78. Standard test method for flexural strength of concrete (usingsimple beam with third-point loading). American Standards for Testing andMaterials; 2010.

[20] ASTM C 1018. Standard test method for flexural toughness and first-crackstrength of fiber-reinforced concrete (using beam with third-point loading).American Standards for Testing and Materials; 1997.

[21] JSCE-SF4. Standard for flexural strength and flexural toughness, method oftests for steel fiber reinforced concrete. Concrete library of JSCE, JapanConcrete Institute (JCI); 1984. p. 58–66.

[22] Mohammadi Y, Singh SP, Kaushik SK. Properties of steel fibrous concretecontaining mixed fibres in fresh and hardened state. Constr Build Mater2008;22:956–65.

[23] Yan H, Sun W, Chen H. The effect of silica fume and steel fiber on the dynamicmechanical performance of high-strength concrete. Cem Concr Res1999;29:423–6.

[24] Rossi P. Steel fiber-reinforced concrete (SFRC): an example of French research.ACI Mater J 1994;91(3):273–9.

[25] Banthia N, Azzabi M, Pigeon M. Restrained shrinkage cracking in fiber-reinforced cementitious composites. Mater Struct 1993;26(161):405–13.