Construction and Building Materials 152 (2017) 964–977

Contents lists available at ScienceDirect

Construction and Building Materials

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

Influence of steel fibers on the mechanical properties and impactresistance of lightweight geopolymer concrete

http://dx.doi.org/10.1016/j.conbuildmat.2017.06.0920950-0618/� 2017 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.E-mail address: johnson@um.edu.my (U.J. Alengaram).

Azizul Islam, U. Johnson Alengaram ⇑, Mohd Zamin Jumaat, Nurasyiqin Binti Ghazali, Sumiani Yusoff,Iftekhair Ibnul BasharCentre for Innovative Construction Technology (CICT), Department of Civil Engineering, Faculty of Engineering, University of Malaya, Kuala Lumpur 50603, Malaysia

h i g h l i g h t s

� Effect of uncrushed and crushed OPS on mechanical properties reported.� Impact behavior of geopolymer concrete is investigated.� Significant improvements in tensile and flexural strengths of FROPSGC observed.� Uncrushed OPSGC has better impact resistance compared to crushed OPSGC.� Significant improvement in impact energy of FROPSGC.

a r t i c l e i n f o

Article history:Received 18 September 2016Received in revised form 11 June 2017Accepted 16 June 2017

Keywords:Fiber-reinforced lightweight geopolymerconcreteDrop hammer impactImpact energyCrack growth resistanceMechanical propertiesOil palm shell

a b s t r a c t

The influence of fiber on the mechanical properties and impact resistance of oil palm shell geopolymerconcrete (OPSGPC) prepared with ground granulated blast-furnace slag (GGBS) and palm oil fuel ash(POFA) as binders is reported. The mechanical properties of OPSGPC, namely compressive, flexural, split-ting tensile strengths, and modulus of elasticity were investigated; impact resistance was found throughdrop hammer test. The addition of 0.5% steel fibers enhanced the splitting tensile and flexural strengths offiber reinforced OPSGPC (FROPSGPC) by about 19–38% and 13–44%, respectively compared to the non-fibrous OPSGPC. The FROPSGPC with uncrushed OPS developed higher initial and final impact resistancecompared to specimens with crushed OPS. With the addition of 0.5% steel fiber, the first crack load of thegeopolymer concrete increased by 1.5–3.5 times compared to the corresponding mixes of OPSGPC with-out fiber. The ultimate impact energy of most of the OPSGPC and FROPSGPC with uncrushed OPS wasfound 15–152% higher compared to the corresponding mixes with the crushed OPS aggregate.

� 2017 Elsevier Ltd. All rights reserved.

1. Introduction

Rapid industrialization during the last 100 years broughtremarkable changes to method of construction and in materials;over 11 billion tons of concrete are being used annually [1] makingit as one of the most widely used construction materials. However,the developmental activities were accompanied by exploitation ofnatural resources in the production of concrete. The realization ofoveruse of natural resources as construction materials in concreteproduction had alarmed the entire world to minimize the exploita-tion of natural materials and this led to search for alternate con-struction materials towards achieving sustainability. In relationto this, the usage of recycled aggregates is vital in developing

sustainable concrete through effective inclusion of industrial by-products considered as waste materials and this in turn wouldreduce the exploitation of natural resources. The optimum usageof recycled aggregate was found 30% with 0.6% steel fiber in termsof strength and cost benefit analysis [2]. One of the major con-stituent materials in concrete is cement and its production andenergy demand is well established. The emission of carbon di-oxide (CO2) during cement production is a major concern and pres-surized researchers to look for alternative of binding material inconcrete. Geopolymer concrete, which excludes conventionalcement as binder, is considered as one of the potential alternativesto cement based concrete. The use of industrial by-products suchas fly ash (FA), slag (GGBS), rice husk ash (RHA), metakaolin(MK), palm oil fuel ash (POFA), etc. as partial and whole cementreplacement in conventional and geopolymer concrete has beenreported [3–6]. In addition, another industrial by-product from

A. Islam et al. / Construction and Building Materials 152 (2017) 964–977 965

palm oil industry, known as oil palm shell (OPS) was used as wholereplacement for conventional crushed granite aggregate as sustain-able material in lightweight concrete and geopolymer lightweightconcrete production and owing to its ductility and energy absorb-ing characteristics OPS has been successfully tested for blast loads[7,8]. In Malaysia, 4.5 million tons of OPS and 0.516 million tons ofPOFA are produced annually from the palm oil industry as wastematerial [9].

The replacement of conventional aggregate by OPS and usage ofPOFA in geopolymer concrete (GPC) would reduce the environmentpollution as these two materials are dumped in the factory yardscausing land and air pollution. Thus, this work focusses on theuse of POFA and GGBS as binding materials and OPS as wholereplacement for crushed granite aggregates in the developmentof geopolymer concrete.

Fibers are usually used in concrete to control cracking due toplastic and drying shrinkages; they also reduce the permeabilityof concrete by reducing bleeding water. Bernal et al. [10] carriedout a study on the effect of steel fiber on the mechanical propertiesof slag-based GPC and reported that utilization of steel fiberreduces the compressive strength but largely improve splittingtensile and flexural strengths. They reported that alkali-activatedfiber reinforced slag concrete shows better mechanical perfor-mance than Portland cement concrete. Aldahdooh et al. [11] inves-tigated that the ultrafine POFA with micro steel fiber significantlyimprove the compressive strength of mortar. In their study, theyused micro-steel fibers (6 and 13 mm) with diameters of0.16 mm and tensile strength of up to 2850 MPa and achievedthe compressive strength up to 158 MPa, a direct tensile strengthof 13.78 MPa.

Puertas et al. [12] conducted a study on polypropylene (PP)fiber reinforced GPC and found no reduction in compressivestrength. In their study, different types of source materials suchas slag, fly ash and slag/fly ash combination were used. Thepolypropylene (PP) fibers of 0.5% and 1% by volume of mortar wereused. The addition of 0.5% and 1% PP fiber did not affect the com-pressive strength of slag based FRGPC at 2- and 28-day. Though, infly ash based FRGPC the 2-day compressive strength was increaseddue to increase of PP fiber contents but a slight reduction wasobserved at 28 days in the same composite. In the case of com-bined slag/fly ash based FRGPC, slight increase in compressivestrength was noticed by increasing the PP fibers from 0.5% to1.0% at both ages. Shaikh and Hosan [13] studied the mechanicalproperties of steel FRGPC at elevated temperatures. In their study,they investigated two types of alkali activators (Na and K-based),and reported that Na-based activators showed much higher com-pressive and indirect tensile strengths in steel fiber reinforcedGPC. Another study [14] conducted on the effects of micro steelfibers on the mechanical properties of fly ash based geopolymercomposites and reported that addition of micro steel fibers signif-icantly improve flexural strength and energy absorption capacity.On the contrary, Yu et al. [15] investigated the effects of singlesized and hybrid steel fibers in the ultra-high performance conven-tional concrete and reported that the hybrid steel fibers is moreefficient for improving the energy dissipation capacity of concreteunder impact load.

It is reported [16] that the addition of GGBS improves the qual-ity of OPS concrete (OPSC) by reducing water absorption and theuse of 40% GGBS performed the best due to decrease in permeablepores. Another study shows [17] that even though an increase inslag content led to the reduction in the strength, the OPSC withGGBS as high as 60% cement replacement satisfied the minimumstipulated strength required for structural lightweight concrete(LWC). Ahmmad et al. [18] carried out a study by using two indus-trial wastes from palm oil mill, OPS and palm oil clinker (POC) andreported that the replacement of OPS by POC as coarse aggregate

has positive impact on the compressive strength. A review [19]on green concrete for the potential usage of waste materials inthe form of cement replacement, aggregate replacement as wellas fiber reinforcement reported that if proper treatments andselection of materials are accomplished, these waste materialscould be incorporated in concrete to improve mechanical anddurability performances. A recent microstructural analysis hasbeen investigated for the potential usage of POFA and the specimencontained POFA showed less surface water absorption and higherdurability under acid and sulfate attack [20]. Islam et al. [21]reported that the OPSC with POFA as a 10% cement replacementexhibited the most optimum sustainability performance in termsof both cost and eco-efficiency of OPSC. Apart from this, anotherstudy [22] shows the comparison of chemical composition ingeopolymer mortar among the binders GGBS, POFA and FA interms of strength; this study also reported a possible estimationof the concrete strength by modifying the chemical compositionand varying binder content.

In another study reported the early improvement of compres-sive strength of PP based FRGPC compared to plain concrete [23].The source material used in that study was FA and calcined kaolin.The compressive strength of FRGPC containing 0.5% PP fiber (bywt.) reached about 52 MPa at the age of 3 days and fiber contentbeyond 0.5% reduced the compressive strength. Yap et al. [24] per-formed a study on the torsional behavior among the normal weightconcrete and OPSC using steel fiber and reported that the addi-tional of steel fiber reduces crack width of both the normal weightconcrete and OPSC specimens by about 30–43% and 42–60%,respectively. In their investigation it was also reported that theOPSC is more ductile than normal weight concrete. Yoo et al. [25]carried out a study on concrete beam to enhance its energy absorp-tion capacity under impact load by incorporating steel fiber. Yooet al. [25] conducted drop-weight impact test to evaluate theimpact capacity of fiber reinforced concrete beam and reportedthat the first crack was mainly influenced by the matrix crackingrather than the fiber bridging effect but the final crack is stronglyinfluenced by the fiber content.

This research focusses on the development of sustainable GPCusing POFA and GGBS as binders and OPS as whole replacementof conventional coarse aggregates. In order to enhance the impactresistance of POFA-GGBS based lightweight OPS geopolymer con-crete (OPSGPC), steel fibers were added and the impact resistanceof the OPSGPC was investigated by drop-hammer method.

2. Experimental program

The main objective of the experimental study was to evaluate the impact resis-tance of steel fiber reinforced OPSGPC. The other tests include compressivestrength, splitting tensile strength, flexural strength, and modulus of elasticity.

2.1. Materials

2.1.1. BinderPOFA with a specific surface area and specific gravity of 1720 m2/kg and 2.14,

respectively, was used in this investigation. GGBS was used along with POFA asthe source material in the development of GPC. The specific surface area and speci-fic gravity of GGBS were 3200 m2/kg and 2.9, respectively. The binder contents fornormal weight geopolymer concrete (NWGPC) and lightweight geopolymer con-crete (OPSGPC) were 308 kg/m3 and 400–454 kg/m3, respectively.

2.1.2. Fine and coarse aggregateManufactured sand (MS) with specific gravity and fineness modulus of 2.60 and

3.19, respectively, was used as fine aggregate. The fine aggregate content forNWGPC was kept constant at 618 kg/m3, while it varied from 998 to 1134 kg/m3

in case of OPSGPC.Conventional crushed granite aggregate as shown in Fig. 1 (a) was used for com-

parison purpose. The other coarse aggregate, OPS used in this study was collectedfrom the local palm oil factory with maximum size of 14 mm (Table 1). Generally,the raw OPS collected from the factory had oily surfaces that could hamper the

Table 1Physical properties of OPS and crushed granite [26].

Physical property UncrushedOPS

CrushedOPS

Crushedgranite

Maximum size (mm) 14 9 20Compacted bulk density (kg/m3) 633 655 1468Specific gravity 1.34 1.33 2.6224 h water absorption (%) 25.7 24.7 0.95Moisture content (%) 10.5 10.2 –Aggregate impact value (AIV) (%) 2.63 3.13 11.9

966 A. Islam et al. / Construction and Building Materials 152 (2017) 964–977

bond. Hence, the OPS was washed and then air-dried in the laboratory to a satu-rated surface dry (SSD) condition; for the crushed OPS, the cleaned OPS was crushedto the required size.

The outer convex surfaces of uncrushed OPS (Fig. 1(b)) have smoother surfacecompared to concave surfaces. The crushed OPS as shown in Fig. 1(c) has morespiky edges than the uncrushed OPS (Fig. 1(b)) [27]. The physical properties ofthe uncrushed and crushed OPS, along with the conventional crushed granite aggre-gate are given in Table 1. Both the crushed and the uncrushed OPS have a loweraggregate impact value (AIV) and bulk density than the crushed granite aggregate.The OPS content in OPSGPC mixes was varied from 181 to 319 kg/m3; while for theNWGPC, crushed granite of 1235 kg/m3 was used. The 24 h water absorption of OPSwas found to be about 25%.

2.1.3. Alkaline activators and waterA combination of sodium silicate (Na2O = 12%, SiO2 = 30%, and water = 57% by

mass) and sodium hydroxide solution (NaOH) was used as alkaline activator. Thesolution of 12 molarity (M) NaOH prepared with 99% purity such that 361 g of pel-lets was dissolved in 1 kg of solution [28]. The ratio of Na2SiO3/NaOH was kept con-stant at 2.5 for all the mixes and the mixture contained additional water.

Potable tap water was used in the mixing of concrete and the constant a waterto binder ratio of 0.25 for all the mixes.

2.1.4. FibersThe hooked-end steel fiber (specific gravity of 7.9) with the aspect ratio of 65

and length of 35 mm was used in this study (Fig. 2).

2.2. Mixing procedure

A total of fourteen (14) mixes were prepared using variables of three differentOPS as coarse aggregates of which the OPS to binder weight ratio 0.4, 0.6, 0.8 (forOPS both in crushed (C) and uncrushed (UC) condition) (Table 2). In addition,0.5% of steel fiber was added and the effect of varying OPS contents on the freshand hardened geopolymer concrete (GPC) properties were investigated; the com-parison between the mechanical properties and impact behavior with and withoutfiber was also performed. A normal weight geopolymer concrete (NWGPC) usingcrushed granite aggregate was prepared as a control mix to compare the mechan-ical properties and impact behavior of the OPSGPC.

Fig. 1. Coarse aggregates (a) crushed granite;

The mix proportions and experimental parameters are shown in Table 2 andTable 3, respectively, while the casting specimens and the graphical representationof mix proportions are presented at Fig. 3 and Fig. 4, respectively. Initially, thecoarse and fine aggregates were mixed in the rotary mixer followed by GGBS andPOFA for about 5 min. Those mixes contained fiber, it was added during the mixtureof coarse and fine aggregates. This was followed by the alkaline activators and addi-tion of water and the mixing continued for another 4 min. Free water was added toenhance the workability. The alkaline solution to binder ratio (s/b) of 0.45 andwater to binder ratio (w/b) of 0.25 was kept constant for all the mixes.

2.3. Specimen preparation and testing

The concrete was cast in 100 mm cubes, Ø150 � 300 mm cylinders,Ø100 � 200 mm cylinders, 100 � 100 � 500 mm prisms and 600 � 600 � 50 mmpanel for testing the compressive strength, modulus of elasticity, splitting tensilestrength, flexural strength and drop hammer impact test, respectively. The com-pressive strength, modulus of elasticity, splitting tensile strength and flexuralstrength tests were done in accordance to BS 1881: Part 118, ASTM: C469/C469M, BS EN 12390-6:2009 and BS EN 12390-5:2009, respectively. The cube com-pressive test was carried out at 3-, 7- and 28-days, while the modulus of elasticity,splitting tensile strength and flexural strength were tested at the age of 28-day.Steel fiber was added to investigate the impact resistance of the concrete. The pan-els were demoulded after 24 h of casting and left in the laboratory till the age oftesting. The impact capacity of panels were tested at 28-day.

(b) uncrushed OPS; (c) crushed OPS [26].

35 mm

Fig. 2. Steel fibers.

Table 2Mix proportion of geopolymer concrete with and without fiber (kg/m3).

Mix No. Mix label Binder Fine Aggregate Coarse Aggregate Activator Added water Steel fiber

GGBS POFA MS OPS (UC) OPS (C) Granite (%) volume

OPSGPC with uncrushed OPS (UC)A1 OPSGPC4-NF-UC 227 227 1134 181 – – 204 114 –A2 OPSGPC4-F- UC 227 227 1134 181 – – 204 114 0.5A3 OPSGPC6-NF- UC 212 212 1061 255 – – 191 106 –A4 OPSGPC6-F- UC 212 212 1061 255 – – 191 106 0.5A5 OPSGPC8-NF- UC 200 200 998 319 – – 180 100 –A6 OPSGPC8-F- UC 200 200 998 319 – – 180 100 0.5OPSGPC with crushed OPS (C)B1 OPSGPC4-NF-C 227 227 1134 – 181 – 204 114 –B2 OPSGPC4-F-C 227 227 1134 – 181 – 204 114 0.5B3 OPSGPC6-NF-C 212 212 1061 – 255 – 191 106 –B4 OPSGPC6-F-C 212 212 1061 – 255 – 191 106 0.5B5 OPSGPC8-NF-C 200 200 998 – 319 – 180 100 –B6 OPSGPC8-F-C 200 200 998 – 319 – 180 100 0.5NWGPC with crushed graniteC1 NWGPC-NF 154 154 618 – – 1235 139 77 –C2 NWGPC-F 154 154 618 – – 1235 139 77 0.5

Legend: OPSGPC8 — NF — UC, OPSGPC6 — F — C(80%OPS) (no fiber) (uncrushed OPS) (60%OPS) (fiber) (crushed OPS).

Table 3Experimental parameters of geopolymer concrete with and without fiber.

Label Binder: Fine aggregate:Coarse aggregate (wt. ratio)

Binder (kg/m3) s/b (wt/wt) Activators (kg/m3) w/b (wt/wt) Added water (kg/m3)

NaOH solution (12 M) Na2SiO3 solution

(1: 2.5)

OPSGPC 1:2.5:0.4 454 0.45 58 145 0.25 1141:2.5:0.6 424 55 137 1061:2.5:0.8 400 51 128 100

NWGPC 1: 2: 4 308 40 100 77

Fig. 3. Casting specimens for impact test, compressive strength, splitting tensile strength, flexural strength, and Young’s modulus test.

A. Islam et al. / Construction and Building Materials 152 (2017) 964–977 967

Fig. 4. Graphical representation of fresh concrete mix ratio by volume (a) 1:2.5:0.4 (OPSGPC); (b) 1:2.5:0.6 (OPSGPC); (c) 1:2.5:0.8 (OPSGPC); and (d) NWGPC.

968 A. Islam et al. / Construction and Building Materials 152 (2017) 964–977

The drop hammer impact test was done based on modification of the recom-mendations by ACI Committee 544 in which an impact specimen is subjected torepeated blows on the same spot. In this modified impact test, a 10 kg drop hammerwas released from a height of 300 mm on the panel specimen (Fig. 5.). The numberof blows to cause the first visible crack and failure was observed and used to calcu-late the first crack and failure impact energy of the concrete, respectively. Theimpact energy is given in the following equation:

Eimpact ¼ m � g � h� N ð1Þ

where, Eimpact = impact energy in Joule (J); m = mass of drop hammer = 10 kg;g = 9.81 m/s2; h = releasing height of drop hammer = 300 mm; N = number of blows.

The ratio of the number of blows to cause failure, Nf to the number of blows tocause the first crack, Nc is defined as impact ductile index, mi = Nf/Nc [27]. Crackwidths of all the geopolymer panel were measured using a high magnification crackmicroscope, immediately after the first crack development and during the propaga-tion of cracks in the panel. Fig. 6. shows the crack width measurement system usingmagnification microscope.

2.4. Curing

Only ambient curing was adopted. After casting, all the test specimens werekept in laboratory at temperature and humidity of 26–29 �C and 75–80%, respec-tively till the age of testing.

3. Results and discussion

The specimens with MS produced comparable results to that ofNS [26] and hence its use as sustainable material was furtherinvestigated using fibers. The variables investigated in this sectionare OPS to binder contents with crushed and uncrushed OPS; thefiber content was kept constant at 0.5% on volume of concrete.MS was used as fine aggregate and comparison of mechanicalproperties of NWGPC with and without fiber was also done.

Fig. 5. Impact test arrangement.

A. Islam et al. / Construction and Building Materials 152 (2017) 964–977 969

3.1. Workability

3.1.1. Slump valuesThe slump values of the OPSGPC and NWGPC with and without

fibers mixes are shown in Table 4. In general, the addition of fibersin concrete caused significant loss in the workability of the con-crete. The large surface area of fibers absorbed more binder mortararound the fibers and hence the viscosity of the concrete increased,resulting in low slump values [29]. In this study, those mixes con-tained OPS less than 40% of binder (mix A1, A2, B1 and B2)observed no slump. Apart from this, the slump values of FROPSGPCin the range of 15–25 mm was found 33–38% and 25–43% lowerthan control mixes for uncrushed and crushed OPS, respectively.However in the case of LWC, the slump values in the range of0–25 mm produced satisfactory compaction [30]. Further, themixes with normal weight aggregate, mix C2 (0.5% steel fiber) pro-duced 33% lower slump compared to mix C1 (no fiber).

3.1.2. Effect of uncrushed and crushed OPS on workabilityThe mixes B3 and B5 using crushed OPS showed 33–13% low

slump values compared to the mixes A3 and A5 with uncrushed

Fig. 6. Crack width

OPS. The crushed OPS has larger surface area compared to theuncrushed OPS [31]. Hence, the crushed OPS requires more quan-tity of mortar and for a given mortar content, it led to lowworkability.

3.2. Density

The 28-day density of oil palm shell geopolymer concrete(OPSGPC) and normal-weight geopolymer concrete (NWGPC) withand without fiber are shown in Table 4. The density of OPSGPC andfiber reinforced oil palm shell geopolymer concrete (FROPSGPC)ranges 1820–1940 and 1873–1994 kg/m3, respectively, and thus,it could be categorized as LWC [32]. Fig. 7 shows the relationshipbetween density and the compressive strength. As seen in Fig. 7,the compressive strength increases proportionally with thedensity.

The comparison between the compressive strength with andwithout fiber and the density are shown in Fig. 8. The mixes B3and B4 show the higher value of compressive strength with lowdensity compared to the corresponding mixes. The density ofOPSGPC decreases as the OPS content increases. The density ofthe mix A1 that contained 181 kg/m3 of OPS (OPS to binder weightratio of 0.4) was found as 1940 kg/m3. The effect of increase in theOPS content was evident as the mix A5 produced lower density of1843 kg/m3 compared to A3 of 1925 kg/m3 (for an increase in theOPS content of 74 kg/m3). The addition of 0.5% of steel fiber on vol-ume of concrete, the density of OPSGPC increases about 2–3%. Therate of reduction in the density for OPSGPC mixes with crushedOPS is similar to that of mixes with uncrushed OPS.

3.3. Development of compressive strength in steel fiber reinforcedOPSGPC

The development of compressive strength of uncrushed andcrushed OPSGPC with varying OPS contents and NWGPC withand without fibers is shown in Table 4.

3.3.1. Effect of uncrushed and crushed OPS on compressive strength ofsteel fiber reinforced OPSGPC

The 28-day compressive strength of the mixes B1, B3 and B5were 27.85, 28.33 and 22.50 MPa, respectively (Table 4). These val-ues were 0.6%, 17.1% and 11.3% higher compared to the corre-sponding mixes using uncrushed OPS. The little difference ofcompressive strength between the mixes B1 and A1 is due to lessquantity of OPS and the effect of OPS is less on the developmentof compressive strength. The non-fibrous OPSGPC with crushedOPS aggregate produced higher compressive strength compared

measurement.

Table 4Workability, dry density, and compressive strengths of OPSGPC and NWGPC with 0% and 0.5% steel fiber.

Mix No. Mix label Slump 28-d density Mean compressive strength (MPa)a

(mm) (kg/m3) 3-day 7-day 28-day

OPSGPC with uncrushed OPS (UC)A1 OPSGPC4-NF-UC 0 1940 23.09 (83.40) 26.78 (97.73) 27.69A2 OPSGPC4-F- UC 0 1994 19.87 (70.64) 24.12 (85.73) 28.14A3 OPSGPC6-NF- UC 30 1925 19.56 (80.85) 23.47 (96.98) 24.20A4 OPSGPC6-F- UC 20 1965 20.94 (83.79) 24.77 (99.13) 24.99A5 OPSGPC8-NF- UC 40 1843 15.92 (78.79) 17.44 (86.32) 20.21A6 OPSGPC8-F- UC 25 1885 15.54 (75.27) 18.91 (91.62) 20.64OPSGPC with crushed OPS (C)B1 OPSGPC4-NF-C 0 1929 17.42 (62.54) 25.06 (89.99) 27.85B2 OPSGPC4-F-C 0 1978 18.44 (64.39) 26.97 (94.17) 28.64B3 OPSGPC6-NF-C 20 1910 21.33 (75.28) 25.59 (90.32) 28.33B4 OPSGPC6-F-C 15 1950 21.37 (76.32) 24.28 (86.74) 28.00B5 OPSGPC8-NF-C 35 1820 16.17 (71.88) 19.77 (87.88) 22.50B6 OPSGPC8-F-C 20 1873 18.05 (86.08) 19.67 (93.79) 20.97NWGPC with crushed graniteC1 NWGPC-NF 15 2296 28.42 (69.95) 35.16 (86.54) 40.63C2 NWGPC-F 10 2330 27.66 (69.97) 32.32 (81.77) 39.53

a The data in parentheses are percentages of 28-day compressive strength.

20

22

24

26

28

30

1800 1850 1900 1950 2000

Com

pres

sive

stre

ngth

(MPa

)

Density (kg/m3)

Fig. 7. Relationship of density and compressive strength of steel fiber reinforcedOPSGPC.

970 A. Islam et al. / Construction and Building Materials 152 (2017) 964–977

to the corresponding mixes using uncrushed OPS aggregate as dis-cussed in earlier studies [26]. However, when fibers were added,the OPSGPC with crushed OPS produced the compressive strengthclose to that of uncrushed OPS. Bernal et al. [10] reported thatutilization of steel fiber reduces the compressive strength of

0

5

10

15

20

25

30

[A1,A2] [A3,A4] [A5,A6] [B

Com

pres

sive

stre

ngth

(MPa

)

Series of m

Comp. strength (0% steel fiber)Density (0% steel fiber)

Fig. 8. Compressive strength with res

slag-based geopolymer concrete. In general, the addition of fibersslightly improves the compressive strength of Ordinary PortlandCement (OPC) concrete [29,33].

3.3.2. Effect of OPS content on compressive strength of steel fiberreinforced OPSGPC

The relationship of compressive strength and OPS contents isshown in Fig. 9. OPS had a significant effect on the compressivestrength, both in uncrushed and crushed conditions. The compres-sive strength decreases linearly as the quantity of OPS increases. Asseen from Table 4 and Fig. 9, the increase of OPS content from181 kg/m3 to 319 kg/m3 (an increase of about 138 kg/m3), the com-pressive strength decreases by about 27% for uncrushed OPS. Incontrast, the reduction in compressive strength for OPSGPC pre-pared using crushed OPS aggregate was about 19% for a similarincrease in the OPS content.

3.3.3. Development of compressive strength between 3 and 28-dayThe development of compressive strength between 3 and

28 day expressed as a percentage is shown in Table 4. The 28-day compressive strength was taken as the reference and thestrength achievement in 3- and 7-day was calculated. The develop-ment of compressive strength for OPSGPC and fiber reinforcedOPSGPC (FROPSGPC) produced using uncrushed and crushed OPS

1700

1750

1800

1850

1900

1950

2000

2050

1,B2] [B3,B4] [B5,B6]

Den

sity

(kg/

m3)

ixes

Comp. strength (0.5% steel fiber)Density (0.5% steel fiber)

pect to density and OPS contents.

Uncrushed OPS: y = -18.384x + 693.46

R² = 0.99

Crushed OPS: y = -17.111x + 700.42

R² = 0.64

150

200

250

300

350

20 22 24 26 28 30

OPS

(kg/

m3 )

28-day compressive strength (MPa)

Uncrushed OPS Crushed OPS

Fig. 9. Relationship of compressive strength and OPS contents of crushed anduncrushed OPSGPC.

15

20

25

30

35

40

45

A1 A3 A5 B1 B3 B5 C1Com

pres

sive

stre

ngth

(MPa

)

Series of mixes

3-day 7-day 28-day

Fig. 10. Development of compressive strength of OPSGPC with 0.0% steel fiber.

15

20

25

30

35

40

45

A2 A4 A6 B2 B4 B6 C2Com

pres

sive

stre

ngth

(MPa

)

Series of mixes

3-day 7-day 28-day

Fig. 11. Development of compressive strength of OPSGPC with 0.5% steel fiber.

Table 5Comparison of 28-day compressive strength, splitting tensile strength, flexural strength, a

Mix No. Mix label Compressive strength, fc Splitting tensile strength, ft(MPa) (MPa)

A1 OPSGPC4-NF-UC 27.69 2.28A2 OPSGPC4-F- UC 28.14 2.72A3 OPSGPC6-NF- UC 24.20 2.07A4 OPSGPC6-F- UC 24.99 2.64A5 OPSGPC8-NF- UC 20.21 1.64A6 OPSGPC8-F- UC 20.64 2.18B1 OPSGPC4-NF-C 27.85 2.05B2 OPSGPC4-F-C 28.64 2.62B3 OPSGPC6-NF-C 28.33 1.84B4 OPSGPC6-F-C 28.00 2.54B5 OPSGPC8-NF-C 22.50 1.55B6 OPSGPC8-F-C 20.97 2.12C1 NWGPC-NF 40.63 2.24C2 NWGPC-F 39.53 3.01

A. Islam et al. / Construction and Building Materials 152 (2017) 964–977 971

along with NWGPC are shown in Fig. 10 and Fig. 11, respectively.The 3-day compressive strength of non-fibrous A1, A3 and A5mixes with uncrushed OPS aggregate produced 23, 20 and16 MPa, respectively. These compressive strengths are 83, 81 and79% of the 28-day strength as shown in Table 4. Similarly, themixes B1, B3 and B5 prepared using crushed OPS aggregate pro-duced 17, 21 and 16 MPa, respectively, and these developed by62, 75 and 72% of the 28-day strength as shown in Table 4. Allthe non-fibrous OPSGPC achieved 62–83% of the 28-day compres-sive strength at 3-day. As reported earlier [26], the 3-day compres-sive strength of OPSGPC can be achieved 57–82% of the 28-daystrength cured in AD condition. The 7-day compressive strengthof OPGPC reached about 88–97% of 28-day compressive strength.The 3- and 7-day compressive strength of mix C1 reached 70%and 86% of the 28-day strength.

Fig. 11 shows the development of compressive strength ofFROPSGPC with 0.5% fibers by volume of concrete. As seen inFig. 10 and Fig. 11, the rate of development of compressivestrength of FROPSGPC is close to that of OPSGPC of the correspond-ing mixes. This could be attributed to the less volume of fibersadded to the concrete and the effect of fibers is not significant onthe development of compressive strength. It is shown in the previ-ous studies [29,33] that the addition of fibers slightly improves thecompressive strength of concrete. It can be seen in Fig. 10 andFig. 11, the rate of early strength (3- and 7-day) of OPSGPC ishigher compared to NWGPC, both in fibrous and non-fibrousspecimens.

3.4. Splitting tensile strength

The splitting tensile strength of fibrous and non-fibrous OPSGPCand NWGPC are shown in Table 5. The relationship between split-ting tensile and compressive strength of fibrous and non-fibrousOPSGPC are shown in Fig. 12. It can be observed that the splittingtensile strength increases with the increases of compressivestrength. The experimental 28-day splitting tensile strength ofthe mixes A1, A3 and A5 were found 2.28, 2.07 and 1.64 MPa,respectively. It is evident that the OPS reduces the tensile strength;higher the OPS contents lower the tensile strength. It could beattributed to the bond failure between the OPS and the matrixoccurred along with failure of the OPS itself as reported earlier[26]. From this study, it was found that the OPSGPC prepared usinguncrushed OPS aggregate produced about 11%, 12% and 6% highercompared to the corresponding mixes prepared using crushed OPSaggregate at 28-day.

Fibers enhance the splitting tensile strength significantly [33].The addition of 0.5% steel fibers, improved the splitting tensile

nd elastic modulus for OPSGPC with and without steel fibers.

Flexural strength, fr Elastic modulus, E f tf c� 100

� �% f r

f c� 100

� �% f r

f t

� �(MPa) (GPa)

2.95 3.54 8.25 10.67 1.294.11 5.74 9.65 14.61 1.513.11 3.45 8.55 12.85 1.503.60 4.68 10.57 14.41 1.361.89 2.94 8.12 9.35 1.152.51 3.15 10.56 12.16 1.152.77 5.85 7.36 9.96 1.354.01 6.37 9.15 14.00 1.533.04 5.82 6.49 10.75 1.653.45 5.78 9.07 12.32 1.361.80 3.87 6.89 8.00 1.162.25 3.53 10.08 10.71 1.063.68 11.39 5.50 9.05 1.644.20 10.15 7.60 10.63 1.40

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

20 22 24 26 28 30

Split

ting

tens

ile st

reng

th (M

Pa)

Flex

ural

stre

ngth

(MPa

)

Compressive strength (MPa)

Flexure (0% steel fibre) Flexure (0.5% steel fibre)splitting (0% steel fibre) Splitting (0.5% steel fibre)

Fig. 12. Relationship between 28-day splitting tensile, flexural, and compressive strengths of OPSGPC with and without fibers.

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

0

5

10

15

20

25

30

35

40

45

A1 A2 A3 A4 A5 A6 B1 B2 B3 B4 B5 B6 C1 C2 Flex

ural

& sp

littin

g te

nsile

stre

ngth

s (M

Pa)

Com

pres

sive

stre

ngth

(MPa

)

Mix designation

Compressive strength, fc Flexural strength, fr Splitting tensile strength, ft

Fig. 13. Flexural and splitting tensile strength with respect to compressive strength and OPS contents.

2

3

4

5

6

7

20 22 24 26 28 30

MO

E (E

), G

Pa

Compressive strength (MPa)

0% steel fibres 0.5% steel fibres

Fig. 14. The relationship between MOE and compressive strength (at 28-day) ofOPSGPC with and without steel fibers.

972 A. Islam et al. / Construction and Building Materials 152 (2017) 964–977

strength of POFA-GGBS based OPSGPC of about 19–38% (Fig. 12)compared to the non-fibrous OPSGPC. In this study, the splittingtensile strength for OPSGPC and FROPSGPC was about 6.5–8.6%and 9–12.3% of the compressive strength, respectively. Previousstudies [33] found in the range of about 7.2% and 9.6–12.3% ofthe compressive strength for non-fibers and fibers oil palm shellconcrete, respectively, using Portland cement and conventionalmining sand (NS).

3.5. Flexural strength

The flexural strength of fibrous and non-fibrous OPSGPC andNWGPC are shown in Table 5. The relationship between the flexu-ral, splitting tensile and compressive strength of fibrous and non-fibrous OPSGPC and NWGPC are shown in Fig. 13. As seen inFig. 12, the flexural strength increases with the increase in com-pressive strength. The experimental 28-day flexural strength ofthe mixes A1, A3 and A5 were found 2.95, 3.11 and 1.89 MPa,respectively, and these strengths are about 6.6%, 2.2% and 5%higher compared to the corresponding mixes prepared usingcrushed OPS aggregate. The lower size of crushed OPS particlesin OPSGPC attributed to the large quantity of OPS used in thesemixes. For a given weight of OPS content in a mix, due to the smal-ler size of crushed OPS a larger number of OPS particles are presentcompared to that of uncrushed OPS [27]. It was found from mixesA1 and A3 that there was a slight improvement of flexural strengthby increasing the OPS contents as OPS to binder weight ratio from0.4 to 0.6 and further increase in OPS (mix A5) reduced thestrength. Similarly, the flexural strength varies for the specimensprepared using crushed OPS aggregate to the corresponding mixes.It could be due to the very less quantity of OPS presented to theformer that did not affect much on the flexural strength. But excess

(a) Tahenni et al. [39] (b) Dawood and Ramli [40]

65

64.2

64

63.1

62.2

42.3

41.6

41.6

42.1

41.3

5.9

5.3 6.6 7.4 8.2

NWC FRC/0.5 FRC/1.0 FRC/2.0 FRC/3.0

Compressive strength, fc Modulus of elasticity, Ec

Flexural strength, fr

63.6

50.4

55.8

64.4 67

.3

66.1

40.1

49.6

51.1

54.3

58.3

53

7.65

7.6 9.55

10.2

5

10.9

10.3

5

4.22

4.97

5.25

5.67

5.98

6.36

NWC FRC/1.0 FRC/1.25 FRC/1.5 FRC/1.75 FRC/2.0

Compressive strength, fc Modulus of elasticity, Ec

Flexural strength, fr Splitting tensile strength, ft

Fig. 15. Contributions of steel fibers to the mechanical behavior of conventional concrete. Note: NWC- Normal weight concrete, FRC/0.5- Fiber reinforced concrete/percentage(%) of steel fiber.

(a) Bashar et al. [38] (b) Ganesan et al. [37]

27.6

2

29.1

7

30 31

.35

30.9

31.3

22.3

5

22.7

6.25

6.56

8.69

8.72

3.41

3.61

4.33

4.69

4.72

4.86

2.31

2.62

2.22

2.81

2.88

2.93

Compressive strength, fc Modulus of elasticity, Ec

Flexural strength, fr Splitting tensile strength, ft45

.37

46.8

3

47.5

5

48.7

4

49.2

3

2.15

2.45

2.94

2.9 3.55

5 5.47

5.51

5.6 6.2

2.58

3.18

3.85

3.93

4.17

Compressive strength, fc Modulus of elasticity, Ec

Flexural strength, fr Splitting tensile strength, ft

Fig. 16. Contributions of steel fibers to the mechanical behavior of GPC.

A. Islam et al. / Construction and Building Materials 152 (2017) 964–977 973

OPS could cause bond failure between the OPS and the matrixoccurred along with failure of the OPS itself as reported earlier[26].

The addition of steel fibers increases both the splitting tensileand flexural strength. The addition of 0.5% steel fibers, enhancedthe flexural strength of POFA-GGBS based OPSGPC of about 13–44% (Fig. 12 and Fig. 13) compared to the non-fibrous OPSGPC. Thiscould be attributed to the strong bond and the matrix between theaggregates and the steel fibers. In this study, the flexural strengthfor OPSGPC and FROPSGPC was about 8–13% and 11–15% of thecompressive strength, respectively and in average, the flexuralstrength of lightweight OPSGPC and FROPSGPC was found approx-imately 34% higher than the splitting tensile strength; generally itis 35% as reported earlier investigations [34].

3.6. Modulus of elasticity (E-value) for OPSGPC with and without steelfibers

The values of static modulus of elasticity (E) of all the mixeswith and without fiber are shown in Table 5. As seen in Table 5

and Fig. 14, the addition of steel fibers in OPSGPC does not havea significant effect on the (E) value. It is reported [1] that the inclu-sion of steel fibers in concrete has little effect on the (E) value. Itwas found from this study that the E-values of NWGPC are higherthan the values obtained for OPSGPC. Generally, the (E) value ofOPSC is lower than the other types of lightweight aggregate con-crete (LWAC) [35]. The moduli of LWA particles are generally lowerthan NWA and though most LWA concretes contain higher cementcontents it follows that the overall moduli of lightweight aggregateconcretes will be lower than normal weight concretes [36]. Gener-ally, for LWAC with natural and artificial LWA, the value of the sta-tic MOE ranges between 10 and 24 kN/mm2 [37].

3.7. Comparison between the contributions of steel fibers to themechanical behavior of GPC as compared to that of conventionalconcrete

The comparison between the contribution of steel fibers to themechanical behavior of conventional concrete as compared to thatof geopolymer concrete (GPC) are shown in Fig. 15 and Fig. 16,

Table 6Impact test results of OPSGPC with and without steel fiber.

Mix No. Mix label No. of blows tocause first crack

Impact energy (first crack),Eimpact,1st,cr (J)

No. of blows to causespecimen failure

Impact energy (specimenfailure), Eimpact, fail (J)

Impact Strength, mi

A1 OPSGPC4-NF-UC 2 58.86 13 382.59 6.5A2 OPSGPC4-F- UC 4 117.72 164 4826.52 41.0A3 OPSGPC6-NF- UC 2 58.86 33 971.19 16.5A4 OPSGPC6-F- UC 7 206.01 202 5944.86 28.9A5 OPSGPC8-NF- UC 3 88.29 41 1206.63 13.7A6 OPSGPC8-F- UC 6 176.58 110 3237.30 18.3B1 OPSGPC4-NF-C 2 58.86 8 235.44 4.0B2 OPSGPC4-F-C 3 88.29 65 1912.95 21.7B3 OPSGPC6-NF-C 2 58.86 17 500.31 8.5B4 OPSGPC6-F-C 5 147.15 195 5738.85 39.0B5 OPSGPC8-NF-C 2 58.86 31 912.33 15.5B6 OPSGPC8-F-C 3 88.29 95 2795.85 31.7

012345678

[A1,A2] [A3,A4] [A5,A6] [B1,B2] [B3,B4] [B5,B6]

No

of b

low

s to

caus

e fir

st

crac

k

Series of mix designation

0% steel fiber 0.5% steel fiber

Fig. 17. Number of blows to cause first crack under impact test.

0

50

100

150

200

250

[A1,A2] [A3,A4] [A5,A6] [B1,B2] [B3,B4] [B5,B6]No

of b

low

s to

cau

se sp

ecim

en

failu

re

Series of mix designation

0% steel fiber 0.5% steel fiber

Fig. 18. Number of blows to cause specimen failure under impact test.

974 A. Islam et al. / Construction and Building Materials 152 (2017) 964–977

respectively. Many researchers [13,38–41] reported that the effectof fibers on the compressive strength and modulus of elasticity ofconcrete is really not apparent and a marginal increase or decreasecould be recorded both in conventional concrete as well as GPC. Ingeneral, the addition of fibers slightly improves the compressivestrength of OPC concrete [29,33] as compared to GPC. As seen inFig. 15 and Fig. 16, fiber is more effective in the development ofconcrete ductility (flexural and splitting tensile strength) butincrease in fiber beyond 0.75% is not significant in the improve-ment of mechanical properties of GPC as compared to conventionalconcrete. The viscosity of geopolymer concrete is much higher thanOPC concrete and the steel fibers were in bundled form. Therefore,the separation of steel fibers might be affected in the high viscousmedium. However, more research needs to be carried out to reducethe viscosity of geopolymer matrix and other volume fraction offiber in the improvement of mechanical properties of GPC.

3.8. Impact energy

The aggregate impact value (AIV) of OPS is much lower than thecrushed granite aggregate and this shows the OPS have highimpact energy absorption capacity. Thus, the impact energy ofthe OPSGPC was investigated using standard panels of600 mm � 600 mm � 50 mm.

3.8.1. First crack impact energyThe impact energy and the number of blows to cause first crack

are shown in Table 6 and Fig. 17, respectively. The first cracks inthe specimens without fiber was visible and the number of blowsto cause the first crack was lower compared to specimens withfibers. However, the number of blows to cause first crack wasnot much different due to low fiber content. The effect of OPS con-tent to cause first crack in the specimens without fiber is not sig-

A. Islam et al. / Construction and Building Materials 152 (2017) 964–977 975

nificant due to increase of OPS content from 0.6 to 0.8. The additionof steel fibers, however, there is slight improvement in the numberof blows to cause first crack. The effect of steel fiber in both theimpact energy and the first crack is quite significant due to brittlenature of lightweight concrete (LWC). Generally LWC is brittle [29],but OPSC has ductility characteristic [42].

The effect of steel fiber to the first crack impact resistance offiber reinforced oil palm shell geopolymer concrete (FROPSGPC)on the addition of steel fibers was observed. Table 6 shows, afteradding 0.5% steel fiber, the first crack strength of the geopolymerconcrete increases by 1.5–3.5 times compared to the correspond-ing mixes of OPSGPC without fiber. The steel fibers were foundhighly effective in preventing the growth of micro-cracks anddiminishing the propagation of these cracks before the cracksjoined up to form macro-cracks. The first crack impact strengthof uncrushed OPS mixes was found higher than that of the crushedOPS mixes. This could be attributed to the low AIV of uncrushedOPS as this aggregates resist impact due to their shape and orien-tation of the aggregate during the impact test [27].

3.8.2. Ultimate impact energyTable 6 and Fig. 18 show the number of blows and impact

energy to cause specimen failure under impact test. The impactstrength (mi) describes the ability of a material to absorb shockand impact energy without breaking. This ratio offers a good indi-cation to the ductility of the concrete subjected to impact load. Itwas found that the ultimate impact energy of OPSGPC andFROPSGPC was significantly higher than the first crack impactenergy. The mixes without fiber (A1, A3, A5, B1, B3, B5) show the

Fig. 19. Origination process of crack through OPS aggregates at late age [44].

Fig. 20. (a) Primary cracks in OPSGPC panel; and (b) p

increase in the ultimate impact energy as the OPS contentincreases (Table 6). But in the case of FROPSGPC, the optimumOPS to binder (OPS/b) ratio was found 0.6 (mixes A4 and B4). Afterthe formation of the first cracks, the FROPSGPC was able to sustainlarge amount of impact load before it failed. This could be attribu-ted to the hooked-ends steel fibers has high tensile strength andalso better cohesion due to their hooked-ends [43].

The effect of uncrushed OPS aggregate became more significantduring the post-crack stage when subjected to impact load. Table 6represents that the impact ductile index (mi) of all uncrushed OPSmixes were significantly higher compared to the correspondingcrushed OPS mixes. The ultimate impact energy of most of theOPSGPC and FROPSGPC with uncrushed OPS was 15–152% highercompared to the corresponding mixes with the crushed OPS aggre-gate. The uncrushed OPS with lower aggregate impact value(approx. AIV = 2.63) compared to the crushed OPS (approx.AIV = 3.13). When the cracks originated and encountered withthe uncrushed OPS aggregates, more energy is required to forcethe cracks through the aggregates (Fig. 19) compared to thecrushed aggregates.

The impact strength (mi) of FROPSGPC was higher than the cor-responding mixes without fibers by 1.3–6.3 times (Table 6). It isreported [45] that fiber significantly improve the impact and mul-tiple cracking behavior. The highest final impact energy of 5945 Jwas obtained for FROPSGPC (mix A4)) with the combination ofthe OPS to binder weight ratio of 0.6 and 0.5% steel fibers.

3.8.3. Failure modeFailure pattern of the OPSGPC is shown in Fig. 20. Two different

types of failure pattern were found in the OPSGPC panel speci-mens. For the plain non-fibrous OPSGPC, the concrete panel brokeinto four pieces upon failure (Fig. 20a). The OPSGPC lost its struc-tural integrity and geometry upon reaching the impact energycapacity. Nevertheless, the failure of the FROPSGPC (Fig. 20b) wasdue to perforation of the panels by the drop weight hammer andthe specimen was not broken into pieces, unlike plain OPSGPC pan-els. This behavior indicated that the FROPSGPC panels remainedstructurally integral, and also ductile. The failure pattern ofFROPSGPC also shows with a significant number of secondarycracks.

3.8.4. Crack development resistanceTable 7 shows the crack width and number of secondary cracks

prior to failure of geopolymer concrete panel. The initial crackwidth was used as a comparative study to determine the effective-ness of the steel fiber in bridging micro-cracks in the geopolymerconcrete. Initial crack widths of the control mixes A1 and B1 werefound to be around 0.240 and 0.252 mm, respectively. OPS was

rimary and secondary cracks in FROPSGPC panel.

Table 7Crack widths and number of secondary cracks prior to failure of all mixes.

Mix No. Mix label Crack width (mm) Number of secondary cracks

First crack Final crack

A1 OPSGPC4-NF-UC 0.240 1.037 0A2 OPSGPC4-F- UC 0.110 0.470 6A3 OPSGPC6-NF- UC 0.190 0.730 3A4 OPSGPC6-F- UC 0.099 0.440 9A5 OPSGPC8-NF- UC 0.155 0.860 5A6 OPSGPC8-F- UC 0.095 0.930 17B1 OPSGPC4-NF-C 0.252 1.212 0B2 OPSGPC4-F-C 0.112 0.509 5B3 OPSGPC6-NF-C 0.211 0.790 4B4 OPSGPC6-F-C 0.105 0.428 8B5 OPSGPC8-NF-C 0.169 0.968 5B6 OPSGPC8-F-C 0.100 1.010 15

976 A. Islam et al. / Construction and Building Materials 152 (2017) 964–977

found to be effective in diminishing the propagation of micro-crack. By increasing OPS content as OPS to binder weight ratio from0.4 to 0.8, reduction in the initial crack width was found to beabout 16–20%, in both specimens with uncrushed and crushedOPS aggregates. The reduction in the initial crack width in theFROPSGPC specimens was found to be about 54–39% and 55–40%in the uncrushed and crushed OPS specimens, respectively, com-pared to the corresponding plain OPSGPC. The lowest crack widthof 0.095 mm was found in the specimens with 0.5% of steel fiber ofmix A6.

The secondary cracks initiated to visible just before to the fail-ure in the FROPSGPC specimens as shown in Fig. 20b. The plainOPSGPC specimens had no secondary cracks upon failure (Fig. 20a).The formation of the secondary cracks is an indication of the effectof fibers in arresting and preventing the crack growth.

4. Conclusion

Based on the experimental work reported in this study, the fol-lowing conclusions could be drawn:

a) The density of OPSGPC decreases as the OPS contentincreases.

b) The compressive strength of OPSGPC decreases as the OPScontent increases.

c) The addition of 0.5% of steel fiber on volume of concreteincreases the density of OPSGPC about 2–3%. The rate ofreduction in density for OPSGPC with crushed OPS is similarto that of uncrushed OPS.

d) When fibers were added, the OPSGPC with crushed OPS pro-duced the compressive strength close to that of uncrushedOPS.

e) The rate of early strength (3- and 7-day) of OPSGPC is highercompared to NWGPC, both in fibrous and non-fibrousspecimens.

f) The splitting tensile and flexural strength of geopolymerconcrete increases with the increases of compressivestrength.

g) The splitting tensile strength of OPSGPC prepared usinguncrushed OPS aggregate produced about 6–12% highercompared to the corresponding mixes prepared usingcrushed OPS aggregate at 28-day.

h) The flexural strength is about 33–35% higher than splittingtensile strength.

i) The compressive strength of FROPSGPC is close to the non-fibrous plain OPSGPC.

j) The addition of 0.5% steel fibers enhanced the splitting ten-sile and flexural strength of POFA-GGBS based OPSGPC byabout 19%–38% and 13%–44%, respectively compared to thenon-fibrous OPSGPC.

k) The addition of 0.5% steel fibers in OPSGPC does not have asignificant effect on the (E) value.

l) The effect of steel fiber in both the impact energy and thefirst crack is quite significant due to brittle nature of LWC.

m) Though the OPSGPC specimens with crushed OPS aggregatesproduced higher compressive strength, its impact resistancewas lower compared to the corresponding mixes of theuncrushed OPS aggregates. This could be attributed to thelow AIV of uncrushed OPS as these aggregates resist impactdue to their shape and orientation of the aggregate duringthe impact test.

n) All the FROPSGPC had higher first crack impact energy com-pared to the plain OPSGPC due to the micro-crack bridging ofthe fibers.

o) All the FROPSGPC specimens resisted high impact loadsbefore failure and produced smaller crack widths comparedto the OPSGPC.

p) The uncrushed OPS aggregate produced higher final impactenergy due to its energy absorption capability compared tothe corresponding crushed OPS specimens. The impactenergy distribution in the FROPSGPC specimens was clearlyevident due to the formation of more number of secondarycracks in the specimens.

Since geopolymer is a high viscous material and the distributionof fiber plays vital role in the enhancement of concrete perfor-mance, more investigation is required.

Acknowledgement

The authors are grateful to University of Malaya for the financialsupport through the University of Malaya Research Project-GC003A-15SUS ‘‘Development of Effective Repair and RehabilitationMaterial Consisting Local Waste Materials for Affordable Housing”.

References

[1] P.K. Mehta, P.J. Monteiro, Concrete: microstructure, properties, and materials,McGraw-Hill, New York, 2006.

[2] S. Senaratne, D. Gerace, O. Mirza, V.W.Y. Tam, W.-H. Kang, The costs andbenefits of combining recycled aggregate with steel fibres as a sustainable,structural material, J. Clean. Prod. 112 (Part 4) (2016) 2318–2327.

[3] M.F. Alnahhal, U.J. Alengaram, M.Z. Jumaat, M.A. Alqedra, K.H. Mo, M. Sumesh,Evaluation of industrial by-products as sustainable pozzolanic materials inrecycled aggregate concrete, Sustainability 9 (5) (2017) 767.

[4] M. Sumesh, U.J. Alengaram, M.Z. Jumaat, K.H. Mo, M.F. Alnahhal, Incorporationof nano-materials in cement composite and geopolymer based paste andmortar–A review, Construct. Build. Mater. 148 (2017) 62–84.

[5] A. Sharmin, U.J. Alengaram, M.Z. Jumaat, M.O. Yusuf, S.M.A. Kabir, I.I. Bashar,Influence of source materials and the role of oxide composition on theperformance of ternary blended sustainable geopolymer mortar, Construct.Build. Mater. 144 (2017) 608–623.

A. Islam et al. / Construction and Building Materials 152 (2017) 964–977 977

[6] I.A. Shehu, A.S.M. Awal, Mechanical properties of concrete incorporating highvolume palm oil fuel ash, in: Advanced Materials Research, vol. 599, Trans TechPublications, 2012, pp. 537–540.

[7] S.A. Kabir, U.J. Alengaram, M.Z. Jumaat, A. Sharmin, I.I. Bashar, Performanceevaluation and some durability characteristics of environmental friendly palmoil clinker based geopolymer concrete, J. Clean. Prod. 161 (2017) 477–492.

[8] U.J. Alengaram, N.H. Mohottige, C. Wu, M.Z. Jumaat, Y.S. Poh, Z. Wang,Response of oil palm shell concrete slabs subjected to quasi-static and blastloads, Construct. Build. Mater. 116 (2016) 391–402.

[9] M.N. Borhan, A. Ismail, R.A. Rahmat, Evaluation of palm oil fuel ash (POFA) onasphalt mixtures, Aus. J. Basic Appl. Sci. 4 (10) (2010) 5456–5463.

[10] S. Bernal, R. De Gutierrez, S. Delvasto, E. Rodriguez, Performance of an alkali-activated slag concrete reinforced with steel fibers, Construct. Build. Mater. 24(2) (2010) 208–214.

[11] M.A.A. Aldahdooh, N. Muhamad Bunnori, M.A. Megat Johari, Development ofgreen ultra-high performance fiber reinforced concrete containing ultrafinepalm oil fuel ash, Construct. Build. Mater. 48 (2013) 379–389.

[12] F. Puertas, T. Amat, A. Fernández-Jiménez, T. Vázquez, Mechanical and durablebehaviour of alkaline cement mortars reinforced with polypropylene fibres,Cem. Concr. Res. 33 (12) (2003) 2031–2036.

[13] F.U.A. Shaikh, A. Hosan, Mechanical properties of steel fibre reinforcedgeopolymer concretes at elevated temperatures, Construct. Build. Mater. 114(2016) 15–28.

[14] N. Ranjbar, M. Mehrali, M. Mehrali, U.J. Alengaram, M.Z. Jumaat, High tensilestrength fly ash based geopolymer composite using copper coated micro steelfiber, Construct. Build. Mater. 112 (2016) 629–638.

[15] R. Yu, L. van Beers, P. Spiesz, H.J.H. Brouwers, Impact resistance of a sustainableUltra-High Performance Fibre Reinforced Concrete (UHPFRC) under pendulumimpact loadings, Construct. Build. Mater. 107 (2016) 203–215.

[16] K.H. Mo, U.J. Alengaram, M.Z. Jumaat, M.Y.J. Liu, J. Lim, Assessing somedurability properties of sustainable lightweight oil palm shell concreteincorporating slag and manufactured sand, J. Clean. Prod. 112 (Part 1) (2016)763–770.

[17] K.H. Mo, U.J. Alengaram, M.Z. Jumaat, S.P. Yap, Feasibility study of high volumeslag as cement replacement for sustainable structural lightweight oil palmshell concrete, J. Clean. Prod. 91 (2015) 297–304.

[18] R. Ahmmad, M.Z. Jumaat, U.J. Alengaram, S. Bahri, M.A. Rehman, Hashim Hb,Performance evaluation of palm oil clinker as coarse aggregate in highstrength lightweight concrete, J. Clean. Prod. 112 (Part 1) (2016) 566–574.

[19] K.H. Mo, U.J. Alengaram, M.Z. Jumaat, S.P. Yap, S.C. Lee, Green concretepartially comprised of farming waste residues: A review, J. Clean. Prod. 117(2016) 122–138.

[20] N. Ranjbar, A. Behnia, B. Alsubari, P. Moradi Birgani, M.Z. Jumaat, Durabilityand mechanical properties of self-compacting concrete incorporating palm oilfuel ash, J. Clean. Prod. 112 (Part 1) (2016) 723–730.

[21] M.M.U. Islam, K.H. Mo, U.J. Alengaram, M.Z. Jumaat, Mechanical and freshproperties of sustainable oil palm shell lightweight concrete incorporatingpalm oil fuel ash, J. Clean. Prod. 115 (2016) 307–314.

[22] A. Islam, U.J. Alengaram, M.Z. Jumaat, I.I. Bashar, The development ofcompressive strength of ground granulated blast furnace slag-palm oil fuelash-fly ash based geopolymer mortar, Mater. Des. 56 (2014) 833–841.

[23] Z.H. Zhang, X. Yao, H.J. Zhu, S.D. Hua, Y. Chen, Preparation and mechanicalproperties of polypropylene fiber reinforced calcined kaolin-fly ash basedgeopolymer, J. Central South Univ. Technol. 16 (2009) 49–52.

[24] S.P. Yap, K.R. Khaw, U.J. Alengaram, M.Z. Jumaat, Effect of fibre aspect ratio onthe torsional behaviour of steel fibre-reinforced normal weight concrete andlightweight concrete, Eng. Struct. 101 (2015) 24–33.

[25] D.-Y. Yoo, Y.-S. Yoon, N. Banthia, Flexural response of steel-fiber-reinforcedconcrete beams: effects of strength, fiber content, and strain-rate, Cem. Concr.Compos. 64 (2015) 84–92.

[26] A. Islam, U.J. Alengaram, M.Z. Jumaat, I.I. Bashar, S.M.A. Kabir, Engineeringproperties and carbon footprint of ground granulated blast-furnace slag-palmoil fuel ash-based structural geopolymer concrete, Construct. Build. Mater. 101(Part 1) (2015) 503–521.

[27] K.H. Mo, S.P. Yap, U.J. Alengaram, M.Z. Jumaat, C.H. Bu, Impact resistance ofhybrid fibre-reinforced oil palm shell concrete, Construct. Build. Mater. 50(2014) 499–507.

[28] S.E. Wallah, B.V. Rangan, Research Report GC2: Low-Calcium Fly Ash-BasedGeopolymer Concrete: Long-Term Properties, Faculty of Engineering, CurtinUniversity, Perth, Australia, 2006.

[29] B. Chen, J. Liu, Contribution of hybrid fibers on the properties of the high-strength lightweight concrete having good workability, Cem. Concr. Res. 35 (5)(2005) 913–917.

[30] S.P. Yap, U.J. Alengaram, M.Z. Jumaat, Enhancement of mechanical propertiesin polypropylene– and nylon–fibre reinforced oil palm shell concrete, Mater.Des. 49 (2013) 1034–1041.

[31] P. Shafigh, M.Z. Jumaat, H. Mahmud, U.J. Alengaram, A new method ofproducing high strength oil palm shell lightweight concrete, Mater. Des. 32(10) (2011) 4839–4843.

[32] EN 206-1, Concrete - Part 1: Specification, performance, production andconformity, European Committee for Standardization CEN, 2000.

[33] S.P. Yap, U.J. Alengaram, M.Z. Jumaat, K.R. Khaw, Torsional and crackingcharacteristics of steel fiber-reinforced oil palm shell lightweight concrete, J.Compos. Mater. 50 (1) (2016) 115–128.

[34] W. Zheng, A. Kwan, P. Lee, Direct tension test of concrete, ACI Mater. J. 98 (1)(2001) 63–71.

[35] U.J. Alengaram, H. Mahmud, M.Z. Jumaat, Enhancement and prediction ofmodulus of elasticity of palm kernel shell concrete, Mater. Des. 32 (4) (2011)2143–2148.

[36] J.L. Clarke, ed, (Eds.), Structural lightweight aggregate concrete, Taylor &Francis e-Library, 2005.

[37] FIP Manual, FIP Manual of Lightweight aggregate concrete, Surrey UniversityPress, London, UK, 1983.

[38] N. Ganesan, P. Indira, A. Santhakumar, Engineering properties of steel fibrereinforced geopolymer concrete, Adv. Con. Construct. 1 (4) (2013) 305–318.

[39] I.I. Bashar, U.J. Alengaram, M.Z. Jumaat, A. Islam, H. Santhi, A. Sharmin,Engineering properties and fracture behaviour of high volume palm oil fuelash based fibre reinforced geopolymer concrete, Construct. Build. Mater. 111(2016) 286–297.

[40] T. Tahenni, M. Chemrouk, T. Lecompte, Effect of steel fibers on the shearbehavior of high strength concrete beams, Construct. Build. Mater. 105 (2016)14–28.

[41] E.T. Dawood, M. Ramli, Mechanical properties of high strength flowingconcrete with hybrid fibers, Construct. Build. Mater. 28 (1) (2012) 193–200.

[42] U.J. Alengaram, M.Z. Jumaat, H. Mahmud, Ductility behaviour of reinforcedpalm kernel shell concrete beams, Eur. J. Sci. Res. 23 (3) (2008) 406–420.

[43] A.A. Nia, M. Hedayatian, M. Nili, V.A. Sabet, An experimental and numericalstudy on how steel and polypropylene fibers affect the impact resistance infiber-reinforced concrete, Int. J. Impact Eng. 46 (2012) 62–73.

[44] D.C.L. Teo, M.A. Mannan, V.J. Kurian, Structural concrete using oil palm shell(OPS) as lightweight aggregate, Turkish J. Eng. Environ. Sci. 30 (4) (2006) 251–257.

[45] F.U.A. Shaikh, Review of mechanical properties of short fibre reinforcedgeopolymer composites, Construct. Build. Mater. 43 (2013) 37–49.