Construction and Building Materials 55 (2014) 2028

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

The effect of steel bres on the enhancement of exural and compressive

toughness and fracture characteristics of oil palm shell concrete
Kim Hung Mo, Kathy Khai Qian Yap, U. Johnson Alengaram , Mohd Zamin Jumaat
Department of Civil Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia

h i g h l i g h t s
 Addition of 1.0% steel bre enhanced the exural toughness of OPSC by up to 16 times.
 Compressive toughness improved by 6 times with 0.75% of steel bre in SFOPSC.
 Fracture toughness and fracture energy increased with the addition of steel bres.
 Direct tensile strength increased up to 41% with the addition of 1.0% steel bre.

a r t i c l e

i n f o

Article history:
Received 29 September 2013
Received in revised form 28 December 2013
Accepted 30 December 2013
Available online 1 February 2014
Keywords:
Steel bre
Oil palm shell concrete
Toughness
Fracture energy
Flexural toughness
Compressive toughness
Mechanical properties
Direct tensile strength

a b s t r a c t
This effect of steel bre on the toughness characteristics such as exural toughness, fracture parameters
and compressive toughness of steel bre oil palm shell concrete (SFOPSC) is described in this paper. In the
post-peak regions of the bending specimens, the addition of steel bres signicantly increased both the
fracture energy and exural toughness of the SFOPSC up to 16 times. The compressive toughness of
SFOPSC specimens with 0.75% steel bre of about 864 kN mm was 6 times higher than the specimens
without bres. The pre-peak response through the mechanical property tests shows an increase of up
to 178%, 88% and 41% for splitting tensile, exural and direct tensile strengths, respectively.
2014 Elsevier Ltd. All rights reserved.

1. Introduction
Oil palm shell (OPS) is considered as a waste material produced
from the extraction of palm oil in South East Asian countries, such
as Indonesia, Malaysia, and Thailand. Most of the wastes that emanate from the production of palm oil are dumped in the vicinity of
the factories. The excessive dumping of these wastes including OPS
leads to pollution in the surrounding environment. Research works
on utilizing the oil palm wastes such as OPS, palm oil fuel ash
(POFA) and palm oil clinker paved way to develop sustainable
materials. The utilization of OPS as lightweight coarse aggregates
resulted the development of structural grade lightweight oil palm
shell concrete (OPSC) with a reduction in the density of about
1533% [13] compared to the normal weight concrete (NWC) of
density of about 2400 kg/m3. Thus the reduction of concrete

Corresponding author. Tel.: +60 379677632; fax: +60 379675318.

E-mail address: johnson@um.edu.my (U.J. Alengaram).
0950-0618/$ - see front matter 2014 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.conbuildmat.2013.12.103

density signicantly increases the strength to density ratio of the

concrete; this allows the design exibility of structural members
and further reduces transportation and fabrication costs.
The cube compressive strength of the OPSC varies in the range
of 1545 MPa; however, one of the shortcomings of the OPSC is
the low tensile strength. As known, most of the lightweight aggregate concrete is inferior in the tensile strength [4] as well as higher
brittleness compared to NWC of the same strength [57]. Recently,
Shagh et al. [8] have researched on enhancing the compressive
strength of OPSC by varying the size of the OPS and developed high
strength OPSC with a cement content of 550 kg/m3. The increase in
the compressive strength, however, will be offset by the increased
brittleness of the concrete [610]. The brittleness of concrete is
more prone towards catastrophic failure that occurs without much
warning, and could be hazardous to the surroundings. In addition,
the OPSC has a relatively low modulus of elasticity [4,9,11]
compared to NWC which might lead to acceleration of crack
development [12]. The cracks in concrete provide easy routes for
deleterious agent and that could lead to durability issues. The

K.H. Mo et al. / Construction and Building Materials 55 (2014) 2028

addition of bres in the concrete enhances both the tensile

strength and the ductility of the concrete.
It has been reported that the addition of specied quantity of
steel bres is known to increase the tensile capacity of the bre
reinforced concrete (FRC) along with its post-failure ductility
[1315]. The post-failure ductility is extremely useful in cases
where tensile strength is not adequate to characterize the mechanical response of concrete [16]. After the cracking of matrix, the
steel bres function as a crack bridging mechanism, in which the
bres undergo bre pull-out, thus delaying the crack formation
and limit the crack propagation [17,18]. De-bonding and pulling
out of bres from FRC require higher amount of energy, resulting
in the increased toughness and ductility of concrete. Since the
toughness characteristic is essential for FRC, codes of practice in
some countries incorporated this aspect to measure this property.
The commonly used codes of practice for the measurement of exural toughness tests of FRC are ASTM C1018-92 and JCI SF-4 and
these are widely reported in literatures [5,1921]. However, the
toughness indices as proposed in these standards have been questioned as these are not material properties [22].
On the other hand, the fracture behaviour of concrete is the fundamental to understand the basics of reinforced concrete structures through the application of fracture mechanics. One of the
vital fracture properties for the design of concrete structures is
the fracture energy (Gf) and this is the most accepted and commonly used in numerical models [23]. Another characteristic
namely, fracture toughness (Kic) has also been found to be a fracture mechanics parameter to describe the resisting property of
concrete to cracking [24].
There have been many research works carried out on the
mechanical, thermal, durability and structural behaviour of OPSC
[9,25]. In addition, a low cost house along with a foot bridge constructed using OPS as a lightweight aggregate in Malaysia is being
monitored for structural performance [26]. A low aggregate impact
value (AIV) implies high resistance to impact. Thus, owing to its high
impact resistance, OPS could be used in road blocks, crash barriers,
etc. In addition, pavement blocks, drain culverts, etc. are some other
possible applications. Research works on potential structural application of OPSC in reinforced concrete beams and impact-resistant
panels had been carried out. Alengaram et al. [27] reported that
the ductility of reinforced OPSC beam was about 2 times higher than
the corresponding reinforced NWC beam; Mo et al. [28] investigated
the impact resistance of OPSC panels due to the improved AIV of OPS
compared to conventional granite aggregates; the AIV of the OPS was
found 2 times lower than the crushed granite aggregate. In order to
promote the potential application of the OPSC, the structural behaviour with respect to fracture behaviour needs to be investigated.
Thus the study on the inuence of steel bre in OPSC (SFOPSC) would
pave way in understanding the material properties, toughness and
fracture energy of the concrete for further application.
In this investigation, the relationship of steel bre content and
post crack toughness as well as the mechanical properties of SFOPSC
is investigated and reported. In addition, the post-crack response of
SFOPSC with varying steel bre content (01.0%) was also compared
through exural toughness test; the fracture parameters of SFOPSC
are quantied via a fracture energy test. Another aim of this investigation on the emphasis of a greener and sustainable concrete

Table 1
Physical properties of OPS.
Physical properties

OPS

Specic gravity (saturated surface dry)

Bulk density (compacted) (kg/m3)
Bulk density (loose) (kg/m3)
Water absorption (24 h) (%)

1.30
635
538
24.4

21

through 50% replacement of ordinary Portland cement (OPC) using

ground granulated blast furnace slag (GGBS) as binder for all the
SFOPSC mixes. The comparison of the mechanical properties between the OPSC with 100% OPC and OPSC with a replacement of
50% OPC replacement through GGBS is also analysed and reported.

2. Experimental program
2.1. Materials
2.1.1. Binder
Ordinary Portland Cement (OPC) with specic gravity and specic surface area
of 3.10 and 352 m2/kg, respectively was used in all mixes. The ground granulated
blast furnace slag (GGBS) supplied by YTL Cement Sdn Bhd was used as partial cement replacement; the specic gravity and specic surface area of the GGBS used
was 2.90 and 405 m2/kg, respectively.

2.1.2. Aggregates
Manufactured sand passing through 5 mm and retained on 300 lm was used as
ne aggregate. OPS with specic gravity of 1.30 were used as coarse aggregates,
with size ranging between 2.36 and 14 mm. The OPS collected from a local crude
palm oil mill were washed, air-dried and used in saturated surface dry (SSD) condition. Physical properties of OPS are listed in Table 1.

2.1.3. Superplasticizer and water

A polycarboxylic-ether (PCE) based superplasticizer (SP), commercially known
as Glenium Ace 388 and supplied by BASF (Malaysia) Sdn Bhd was used to reduce
the water and enhance the workability of the mixes. Potable water, free from contaminants and impurities was used for mixing.

2.1.4. Steel bre

Hooked-end type steel bres of length 60 mm with aspect ratio of 80 were used
as steel bres with high aspect ratio were reported to have better exural performance [29]. The steel bres used had minimum tensile strength of 1100 MPa as
specied by the manufacturer.

2.2. Mix proportion and procedure

A total of ve mixes were prepared with binder: water: sand: aggregate ratio of
1:0.33:0.65:1.70 for all mixes. The binder content was xed at 550 kg/m3 with 50%
GGBS replacement (by mass) used for K series mixes (K1, K2, K3 and K4) while 100%
OPC was used for mix N1. The steel bre content was varied at 0% (K1), 0.5% (K2),
0.75% (K3) and 1.0% (K4) volume fractions. The SP used in all the mixes was kept
constant at 1.0% by mass of binder.
The OPS and manufactured sand were mixed in the rotary drum mixer for about
3 min, followed by the addition of OPC and GGBS for another 6 min. After the addition of mixing water along with the SP, the mixing was continued for another 8 min.
Finally, steel bres were evenly distributed into the mixture and mixed. The concrete was then poured into moulds and compacted. All the specimens were demoulded after 24 h and moist cured until the day of testing. The testing plan was
presented in Table 2.

2.3. Test methods

2.3.1. Mechanical properties
The mechanical properties tests on compressive strength (BS EN 12390-3:
2002), splitting tensile (BS EN 12390-6: 2009) and exural strength (BS EN
12390-5: 2009) tests were performed on 100 mm cube, 100 mm /  200 mm
height cylinder and 100 mm  100 mm  500 mm prism at the age of 28-day
respectively. The static modulus of elasticity test was conducted on 150 mm /
 300 mm height cylinders at the age of 28-day based on ASTM C469-10. In addition, the compressive strength test results performed at the ages of 1-, and 7-day
are reported.

2.3.2. Direct tensile test

The direct tensile test was performed based on RILEM TC 162-TDF. In this modied test, notched specimen with 100 mm / and 200 mm height cylinder was used
(Fig. 1). The depth of the notch was sawn to 7.5 mm in order to ensure the failure
occurred in the mid-section of the specimen. The effective cross-sectional area was
therefore reduced from 7854 to 5675 mm2. In the report by Barragan et al. [30], no
signicant inuence of geometry characteristics of the test specimen was noticed in
the direct tensile test.

22

K.H. Mo et al. / Construction and Building Materials 55 (2014) 2028

Table 2
Testing plan for various specimens at 28-day.
Mix

Mechanical properties test

Direct tensile test

Fracture and exural toughness

test

Compressive
toughness test

K1
K2
K3
K4
N1

100 mm cube (compressive strength test), 100 mm /  200 mm height

cylinder (splitting tensile strength test), 100 mm  100 mm  500 mm
prism (exural strength test), 150 mm /  300 mm height cylinders
(modulus of elasticity test)

100 mm /  200 mm
height cylinder with notch
depth of 7.5 mm

100 mm  100 mm  500 mm

prism with bottom notch depth
of 30 mm

100 mm /
 200 mm
height cylinder

2.3.3. Ultra pulse velocity (UPV) test

The UPV test was conducted based on ASTM C597-09. The UPV of concrete was
obtained by direct transmission using a portable ultrasonic non-destructive digital
indicating tester (PUNDIT). The time taken for pulse to travel through a 100 mm
cube from the pulse generator to the receiver was measured. The UPV values were
taken by dividing the length of specimens tested to the time recorded.
2.3.4. Fracture and exural toughness test
Three-point load was applied on notched 100 mm  100 mm  500 mm prism
(Fig. 2) based on RILEM 50-FMC to determine the fracture energy (Gf) and exural
toughness of concrete. A notch of depth of 30 mm was sawn using diamond cutter.
The evaluation of fracture toughness and characteristic length is described in the
analysis section.
2.3.5. Compressive toughness
Compressive toughness is a measure of energy absorbed by concrete when subjected to compressive load. In this test, 100 mm /  200 mm height cylinder was
subjected to a compressive load by a universal testing machine of 600 kN capacity.
Three LVDT were placed around the specimen at equal distances to measure the
plate-to-plate deformations. The plate-to-plate deformation measurement by LVDT
was preferred over the conventional extensometer; the gauge reading of the extensometer might get affected due to the cracks and dilation of the specimen [31]. The
tests on direct tension, fracture and exural toughness were conducted for 0%, 0.5%,
0.75% and 1.0% of steel bres; however, the compressive toughness test was done
on specimens with 0%, 0.5%, and 0.75% of steel bres.
2.4. Calculations
2.4.1. Flexural toughness
Based on ASTM C1609-12, the toughness of FRC was determined using the area
under the loaddeection curve of specimen subjected to bending; the code species the end point of the deection can be taken based on the span length/150,
or a larger value if applicable. Since lightweight concrete usually undergoes larger
deection due to its lower modulus of elasticity, the end point was chosen at an
arbitrary point of span length/60, which is 5 mm. The test results from the fracture
energy test are to be extracted and used for the analysis of exural toughness of
SFOPSC. The ratio of the exural toughness of specimens with bre and without bre (i.e. the ratio of exural toughness of SFOPSC and OPSC) is taken as the exural
toughness ratio.
2.4.2. Fracture energy
The fracture energy (Gf), was calculated using the following equation as specied in RILEM TC 50-FMC:

Fracture energy; Gf

W 0 mgd0
N=m
BD  a

where W0 = area under the loaddeection curve (N m), m = mass of sample (kg),
g = acceleration due to gravity (m/s2), d0 = maximum deection (m) at nal fracture
(taken as 5 mm in this study), B = beam thickness (m), D = depth of beam (m),
a = depth of notch (m).
2.4.3. Characteristic length
The characteristic length (lch) is a fracture parameter used to determine the brittleness of concrete and determined by the following equation:

Characteristic length; lch EGf =ft2 mm

where Gf = fracture energy (N/m), ft = tensile strength (splitting tensile strength is

used similar to work done by Sahin and Koksal [18]).
2.4.4. Fracture toughness
The fracture toughness (Kic) is calculated in accordance with RILEM TC-89 FMT
using the equation as given below:

Fracture toughness; K ic 3P0

p
S paga
2D2 B

MPa m1=2

in which

ga

1:99  a1  a2:15  3:93a 2:7a2

p
p1 2a1  a3=2

where a = a/D, P0 = maximum load + self-weight of specimen (kg), S = span length

(m).
2.4.5. Compressive toughness
The compressive toughness is calculated from the total area under the loaddeformation curve of concrete specimen subjected to compressive load.

3. Results and discussion

3.1. Oven dry density (ODD)
In this study, the highest oven dry density (ODD) of SFOPSC of
2040 kg/m3 was produced with the addition of 1.0% steel bre.
The higher amount of steel bre content, as expected, led to increase in the ODD of the SFOPSC. However, the addition of low
quantity of steel bre volume fraction (<0.5%), in the OPSC produced the ODD less than 2000 kg/m3 and this is categorized as
lightweight concrete in accordance to EN 206-1. In general, the values between the measured 28-day ODD and the predicted values of
the SFOPSC show insignicant difference (Fig. 3). This could be
attributed to the effective compaction in the SFOPSC that could reduce the voids and enhance the density of concrete. Further, the
slight difference in the measured and the predicted ODD could
be attributed to uniform distribution of steel bres in the SFOPSC
mixes.
3.2. Ultrasonic pulse velocity (UPV)
The UPV test is done to assess the quality of the concrete; the
higher the UPV, the superior the quality of concrete. The measurement of UPV is therefore useful for assessing strength of concrete,
particularly on-site where destructive strength test is not applicable. The relationship between the UPV and the cube compressive
strength measured at the ages of 1-, 7-, 28-day is shown in
Fig. 4. It can be seen that the UPV of the SFOPSC generally increased
with the increase in the compressive strength and the values were
found within the range of 3.04.25 km/s (Fig. 4). The UPV measurement is also indicative of presence of cracks, voids and inhomogeneity of concrete. Although the addition of large quantity of steel
bres with high aspect ratio was known to cause clogging of bres,
such issue was not encountered in this investigation for the addition up to 1.0% bre volume. The addition of steel bres up to
1.0% slightly increased the UPV (Fig. 5), and this justied that sufcient compaction and uniform mix could be produced using the
steel bres. Gencel et al. [32] reasoned that the minor variation
in the UPV is an indication of uniformity of the concrete mix.
3.3. Mechanical properties
3.3.1. Compressive strength
As shown in Table 3, the increase in the steel bre content
improved the compressive strength of the OPSC. The addition of

K.H. Mo et al. / Construction and Building Materials 55 (2014) 2028

Fig. 1. Direct tensile strength test set-up.

steel bres of 0.5%, 0.75% and 1.0% resulted in the increase of the
compressive strength of about 10%, 30% and 37%, respectively. In
the case of lightweight aggregate concrete, the cracks occur in
the coarse aggregate before extending into the cement paste.
When sufciently good bond exists between bres and cement
paste, the steel bres act as crack growth arrestors, thus increasing
the ultimate compressive strength of concrete [19]. Further, in the
case of bres with high aspect ratio where bres are stiffer, there is
a better control of micro-crack development [33]. This could lead
to a delay in coalescence of micro-cracks to form macro-cracks,
and might enable the concrete to sustain additional loads. The
ability of steel bres to arrest both micro and macro-cracks in
the pre-peak region could therefore be the contributing factor to
the improvement in the compressive strength of the SFOPSC. The
increase in the compressive strength of steel bre reinforced

concrete using OPSC was also reported by Shagh et al. [34]. In

their studies using other lightweight expanded clay as coarse
aggregates, Chen and Liu [10] and Campione et al. [35] reported
that the addition of steel bre resulted in the increase in the compressive strength. Wang and Wang [15] found that the increase of
compressive strength up to 23% with the addition of 2.0% steel
bre volume using lightweight shale aggregate. Lee and Song
[36] found that the 1.0% addition of steel bre increased the compressive strength of lightweight cellular expanded shale concrete
by 37%. On the other hand, Balendran et al. [5] and Kayali et al.
[37] reported that there was no signicant effect of steel bres
on the compressive strength for Lytag and sintered y ash
aggregate concrete. The inuence of steel bre and GGBS on the
compressive strength of SFOPSC can be seen from Fig. 6. The 28day compressive strength was found about 23% higher than that
of 7-day for 1.0% of bre volume. This could be attributed to the
presence of GGBS in addition to the increase in bre content. The
hydration of GGBS is slower at the earlier ages, and it could have
contributed to weaker bonding of steel bres with the binder
matrix at 7 days. The increased in the GGBS hydration at 28 days
could have led to stronger brematrix bond, and thus effect of
steel bre was more pronounced at this stage.
The comparison of the compressive strength up to 90 days
between the OPSC with 100% OPC and the OPSC with 50% GGBS
as cement replacement is presented in Table 4. The 1-day compressive strength of the mix N1 was almost 83% higher compared to the
mix K1. However, the difference in the strength between the two
mixes reduced to 52%, 27%, 26% and 18% at 3-, 7-, 28-, and
90-day respectively. The reduction was due to the delayed hydration of GGBS, and as the by-product of calcium hydroxide from
hydration of OPC was required for the GGBS to show its binding
property [38]. The mix N1 had only 1% increment in compressive
strength whereas 8% increase was found for the mix K1 and this
showed that the hydration of OPC was almost completed at this
stage while the GGBS hydration process still continues in the mixes
with GGBS.
3.3.2. Modulus of elasticity (MOE)
The modulus of elasticity (MOE) is one the most important
material properties used in the design of concrete structures as it
provides useful information on the ability of concrete to deform
elastically. The 28-day MOE of the OPSC was found to increase with
the addition of steel bre as shown in Table 2 and similar nding is
reported by other researchers [19,32,3941]. The enhancement of
MOE with the addition of steel bre may be attributed to the effect
of bres in arresting the original shrinkage cracks in the concrete
[19] and hence the bres enhanced the stress redistribution and reduced the localized strain, thus giving a steeper slope in the stress
strain curve [32]. The increase of the MOE for the mixes with 0.5%,
0.75% and 1.0% of steel bre was found to be about 10%, 52% and
58%, respectively. As seen from this result, the increase of the steel
bre from 0.75% to 1.0% has negligible effect in the enhancement of
the MOE of SFOPSC. It could be inferred that the use of 0.75%
volume of steel bre has the maximized potential in the stress
transfer during the initial elastic stage of the SFOPSC. The effectiveness of 0.75% of steel bre volume in the enhancement of MOE was
also reported by Kurugol et al. [41], albeit using pumice stone as
lightweight aggregate. As the MOE is a function of compressive
strength of concrete, as seen from Fig. 7, a good correlation is
established between these two parameters. The following equation
is proposed to relate the cube compressive strength and MOE of
SFOPSC:

E 6:31fc1=2  26:90
Fig. 2. Fracture and exural toughness test set-up.

23

where E = MOE (GPa), fc = cube compressive strength (MPa).

K.H. Mo et al. / Construction and Building Materials 55 (2014) 2028

3.3.3. Splitting tensile strength

The splitting tensile test is an indirect and easier method to
determine concrete tensile strength. The determination of concrete
tensile strength is necessary to provide information on the maximum tensile load that a concrete member can sustain before cracking. In this investigation, it was found that the addition of steel
bre greatly improved the splitting tensile strength of the concrete.
As found from this experimental investigation, the addition of 0.5%,
0.75% and 1.0% steel bre volume enhanced the splitting tensile
strength up to 93%, 133% and 178% respectively compared to the
control concrete. When signicant amount of tensile stress was
introduced in concrete, micro-cracks and subsequently macrocracks were formed. The increase in the load induces critical crack
growth at the tip of macro-cracks which eventually led to failure of
concrete [44]. The steel bres act as a means of stress transfer
whereby the tensile stress was transferred across bres, which
subsequently delayed the propagation of macro-cracks and improved the splitting tensile strength of concrete [32]. The relationship between the compressive strength and splitting tensile
strength for SFOPSC at the age of 28-day was correlated in Fig. 8
and an equation for the relationship is proposed as follow:

ft 4:22fc1=2  21:47

where ft = splitting tensile strength (MPa).

The replacement of 50% OPC with GGBS was found to decrease
the 28-day splitting tensile strength of OPSC. As seen from Table 3,
the splitting tensile strengths of the mixes K1 and N1 were found
as 2.49 and 3.07 MPa, respectively. Kou et al. [45] also found that
the use of GGBS decreased the splitting tensile strength due to delayed hydration that leads to the decrease in both the splitting and
compressive strengths.
3.3.4. Flexural strength
The use of steel bre with high aspect ratio is known to have
more pronounced effect on the exural performance of concrete
[19]; this is attributed to longer length of steel bres as it is more
effective in delaying the crack propagation [17]. It was found in
this study that the addition of steel bre of 0.5%, 0.75% and 1.0%

Oven dry density (kg/m3)

2060
2040
2020
2000

50

Compressive strength, MPa

The MOE of the control mix N1 without the addition of GGBS

was found to be about 35% higher than that of the mix K1 as indicated in Table 3. As known, the hydration of GGBS based concrete
is slow at the early age and pores are not lled with the CSH gel.
This might cause weak matrix with pores and results in low compressive strength. Therefore, the mix N1 was found to have higher
MOE than the mix K1 as lower porosity leads to higher MOE [42].
Since the MOE is a function of the compressive strength, the lower
MOE of GGBS-based OPSC could be due to the lower compressive
strength compared to OPSC with plain OPC [43].

40
30
y = 28.313x - 75.411
R = 0.7013

20
10
0
3.00

3.40

3.80

4.20

4.60

UPV, km/s
Fig. 4. Relationship between the UPV and compressive strength up to 28 days.

4.20
4.16

28 day UPV, km/s

24

y = 0.212x + 3.9363
R = 0.9547

4.12
4.08
4.04
4.00
0.25

0.5

0.75

1.25

Steel fibre volume, %

Fig. 5. Effect of steel bre volume on 28-day UPV values.

led to improvement in exural strength to about 25%, 45% and

88%, respectively. The exural strength as high as 8.90 MPa was recorded for SFOPSC with the addition of 1.0% steel bre, which was
signicantly higher than for the OPSC as reported by Shagh et al.
[34]. The steel bres with long length offer more resistance to pull
out of bres due to better bond characteristics [21]. Further, higher
volume of steel bre enables the bres to cross the exural cracks,
thus giving rise to the efciency of bres in arresting the cracks.
Thus, the failure load was increased to cause more cracks for failure, leading to a higher exural strength [46]. The relationship between the compressive strength and exural strength of SFOPSC is
illustrated in Fig. 8 and the following equation is proposed for the
relationship:

fflex 4:02fc1=2  18:63

where fex = exural strength (MPa).

Similar to the observation for splitting tensile strength, the mix
K1 showed the decrease in exural strength compared to corresponding control mix, N1 and the reasons attributed to low splitting tensile strength apply to exural strength as well. However,
it was worth noting that the ratio of exural to compressive
strength of the mix K1 was 6% higher than that for the mix N1. This
proved that the use of GGBS slightly increased the exural tensile
strength of OPSC at a given compressive strength.

Predicted

1980

3.4. Direct tensile strength

Experimental
1960
1940
0

0.2

0.4

0.6

0.8

1.2

Steel fibre content , %

Fig. 3. Measured and predicted values of oven dry density (ODD) of SFOPSC.

Direct tension test was conducted as it is the most direct method in evaluating the tensile strength of concrete. It is an established
fact that the tensile strength obtained from a direct tension test is
usually lower than that of the result from an indirect splitting tensile test, while the exural test gives the highest tensile strength

25

K.H. Mo et al. / Construction and Building Materials 55 (2014) 2028

Table 3
Mechanical properties of SFOPSC.
Mix

Compressive strength (MPa)

Splitting tensile strength (MPa)

Flexural strength (MPa)

Modulus of elasticity (GPa)

K1
K2
K3
K4
N1

33.51
36.78
43.57
45.96
42.28

2.49
4.80
5.80
6.94
3.07

4.73
5.91
6.88
8.90
5.62

9.71
10.65
14.79
15.31
13.16

Modulus of Elasticity (MOE), GPa

Compressive strength, MPa

50
y = 10.63x2 + 2.06x + 33.45
R = 0.996

45
40
35

28 days
7 days

30
y = 6.011x2 + 3.077x + 28.44
R = 0.997

25
20
0

0.25

0.5

0.75

1.25

Fig. 6. Relationship of steel bre volume with 7- and 28-day compressive strength.

Table 4
Strength gain for K1 and N1 mixes.
Compressive strength (MPa)
K1
1-day
3-day
7-day
28-day
90-day

11.97
20.89
28.47
33.51
36.29

N1

(74.5%)
(36.3%)
(17.7%)
(8.3%)

21.90
31.70
36.18
42.28
42.70

(44.8%)
(14.1%)
(16.9%)
(1.0%)

Note: Values in parentheses represent percentage in compressive strength gain.

[47]. The direct tensile strengths measured in this study for the
specimens with 0.5%, 0.75% and 1.0% steel bre were found as
1.451, 1.604 and 1.794 MPa, respectively. As seen from the results,
the increase in the steel bre volume enhanced the direct tensile
strength of the SFOPSC and it could be attributed to the likelihood
of effectiveness of bres crossing the initial crack in the specimen.
Moreover, the hooked-end steel bres provide anchorage and thus
the energy required to pulling out of bres gives rise to increase in
the tensile strength of the concrete.
It should be noted that the increase in the direct tensile strength
with the addition of 1.0% steel bre was only 41%, as compared to
178% and 88% for splitting tensile and exural strengths, respectively. One possible explanation for low tensile strength could be
the orientation of certain bres; some of the bres might be oriented parallel to the crack and reduced crack bridging effect in
the concrete [30]. The relationship between the direct tensile and
splitting tensile strength is characterised by a strong R2 value as
shown in Fig. 9 and represented by the following equation:

ft 9:43f dt  5:19

15
E = 6.31fc1/2 - 26.90
R = 0.919

10

0
5.6

5.8

6.0

6.2

6.4

6.6

6.8

7.0

Compressive strength, MPa 1/2

Steel fibre volume, %

Age

20

where fdt = direct tensile strength (MPa).

Fig. 7. The relationship of compressive strength and 28-day MOE.

values of energy absorption and is considered more suitable in

comparing the ductility performance of FRC. The loaddeection
curve for the SFOPSC specimens investigated in this study is shown
in Fig. 10 and the calculated exural toughness values are presented in Table 5. The control concrete produced the lowest nal
deection and hence the area under the loaddeection curve used
for the exural toughness was calculated based on the nal deection of about 2.15 mm. The addition of steel bre was found to
have signicant effect as it dramatically increased the toughness
of the OPSC, with increase of about 6, 9 and 17 times higher than
that of the control concrete for the specimens with the addition
of 0.5%, 0.75% and 1.0% of steel bres, respectively. As seen from
Fig. 10, not only the peak load was increased with the addition of
steel bres, but the post cracking response was found to be totally
different for the control specimens compared to the specimens
with bres. In the control concrete, after the attainment of the tensile capacity, the load dropped drastically, giving a steep falling
branch in the loaddeection curve. On the other hand, the postcracking falling branch of the SFOPSC specimens was relatively
horizontal and the specimen, SFOPSC with 1.0% steel bre exhibited the least steep falling branch. This progressive failure is attributed to the presence of steel bres as the cracks that start at the
soft of the prism extend slowly to the top of the specimen [48]
unlike abrupt failure in the plain concrete. The bond failure of
the SFOPSC specimens was due to the pulling out of bres from
the matrix as shown in Fig. 11 and this contributed to the increased
energy absorption, and hence ductility of the SFOPSC. Further,
based on the pattern of the curves of the SFOPSC specimens in
Fig. 10, the SFOPSC with addition of 1.0% steel bre exhibited a
smoother curve prior to peak load as compared to the SFOPSC with
the control, 0.5% and 0.75% concrete. This smoother curve could be
attributed to the more pronounced effect of strain hardening for
the given volume of steel bre.

3.5. Flexural toughness

3.6. Fracture energy, fracture toughness and characteristic length

One of the signicant benets on the addition of steel bre in

concrete is its improved post-cracking response. ASTM C1609-12
species the determination of exural toughness through absolute

Fracture energy, fracture toughness and characteristic length

are all the fracture parameters that are used as design criterions
for reinforced concrete structures. The fracture energy is the

26

K.H. Mo et al. / Construction and Building Materials 55 (2014) 2028

14000

10

Tensile strength, MPa

- 18.63
R = 0.974

Flexural load, N

fflex=

4.024fc1/2

6
4

splitting tensile

10000

flexural

0% SF

8000

0.5% SF

6000

0.75% SF
1.0% SF

4000
2000
0

ft = 4.224fc - 21.47
R = 0.933

12000

Deflection, mm
0
5.6

5.8

6.0

6.2

6.4

6.6

6.8

Fig. 10. Loaddeection curve of SFOPSC under three point bending.

7.0

Compressive strength, MPa

Fig. 8. Relationship between compressive strength and tensile strengths (splitting
tensile and exural strengths).

Splitting tensile strength, MPa

8
ft = 9.43fdt - 5.19
R = 0.997

0
0.9

1.1

1.3

1.5

1.7

1.9

Direct tensile strength, MPa

Fig. 9. Relationship between the direct tensile and splitting tensile strength of
SFOPSC.

energy required to create a unit area of crack surface, measured by

the area under the loaddeection graph and calculated based on
Eq. (1). The fracture energy for the control specimen, K1 with a
28-day compressive strength of 35 MPa was found as 555 N/m.
Cui et al. [49] used area under loadCMOD curve to determine
the fracture energy for NWC and lightweight expanded shale
concrete with the compressive strength of 55 MPa and 45 MPa,
respectively and reported the values as 1680 and 935 N/m. The
comparison between the fracture energy of the OPSC and the
lightweight expanded shale concrete shows that the former with
a lower compressive strength of 35 MPa produced 60% of the fracture energy of the latter.
In contrast to the plain concrete specimens, the addition of steel
bre was found to enhance the fracture energy of SFOPSC. This
could be attributed to the role of steel bres that divert the crack
paths and also prevent direct propagation of the cracks. And this
increases the fracture area and the work of fracture. The increase
of steel bre content from 0.5% to 0.75% and 0.5% to 1.0% enhanced
the fracture energy by approximately 1.5 and 2.9 times, respectively. Sahin and Koksal [18] used NWC of grade 45 and steel bres
of aspect ratio of 80, and found the increase of fracture energy by

Fig. 11. Progressive pulling out of steel bres from cement matrix.

about 2.2 and 3.6 times for the addition of steel bres from
0.33% to 0.67% and 0.33% to 1.0%, respectively.
The fracture energy alone is not sufcient to characterize
brittleness or ductility of concrete. In addition, the characteristic
length is used as a measure of brittleness since it includes the inuence of material and size of structure [16]. The characteristic
length is proportional to fracture energy; higher characteristic
length leads to tougher concrete. It can be seen from Table 5 that
the characteristic length of OPSC increased from 869 to 1288,
1914 and 2568 mm with the addition of 0.5%, 0.75% and 1.0% steel
bre, respectively. Hence, it can be shown that the addition of steel
bre greatly improved the toughness and ductility of OPSC,
especially with 1% of steel bre.

Table 5
Flexural toughness and fracture parameters of the SFOPSC.
Mix

Flexural toughness (N mm)

Toughness ratio

Fracture energy, Gf (N/m)

Fracture toughness, Kic (MPa m1/2)

Characteristic length, lch (mm)

K1
K2
K3
K4

3364
18,983
29,956
56,032

5.64
8.90
16.66

555
2786
4354
8080

0.563
0.865
1.266
1.877

869
1288
1914
2568

K.H. Mo et al. / Construction and Building Materials 55 (2014) 2028

27

toughness was investigated. The following conclusions are drawn

from the investigation:

300
250

Load, kN

0% SF

200

0.5% SF
0.75% SF

150
100
50
0
0

Deformation, mm
Fig. 12. Compressive load-deformation graph of SFOPSC.

The fracture toughness, Kic is determined by the value of critical

stress intensity factor, as this is measure of the crack propagation.
The Kic obtained for the control OPSC is 0.563 MPa m1/2, which is
slightly lower than the range of 0.610.82 MPa m1/2 obtained by
Eskandari et al. [50] using self-compacting concrete with compressive strength of 45 MPa. It should be noted that the use of OPSC
with lower compressive strength of 35 MPa managed to achieve
similar Kic value to those of self-compacting concrete with higher
compressive strength. However, with the addition of steel bres,
the fracture toughness improved to 0.865, 1.266 and
1.877 MPa m1/2 for 0.5%, 0.75% and 1.0% volume fraction, respectively. The enhancement of the fracture toughness in the SFOPSC
could be due to the mechanism of bres undergoing de-bonding,
sliding against matrix and stabilization of the matrix cracks by
exerting closure traction on crack surfaces [51].
3.7. Compressive toughness
The load-deformation graphs for the mixes K1, K2 and K3 are
presented in Fig. 12. It was observed that the addition of steel bre
slightly increased the peak compressive load of the OPSC, but
greatly enhanced the post-peak compressive ductility of OPSC.
The compressive toughness values calculated using the area under
the load-deformation curves by integration were found to be 147,
433 and 864 kN mm, respectively for the mixes K1, K2 and K3. This
shows that the addition of 0.5% and 0.75% steel bre improved the
compressive toughness of the OPSC by almost 3 and 6 times,
respectively. It can be seen from Fig. 12 that the area under the
peak load of the SFOPSC specimens was higher than the corresponding area of the control specimen and this is an indication
of higher energy absorption of SFOPSC even before the peak load.
One possible explanation for high energy absorption of the SFOPSC
specimens could be due to the effective micro-cracks bridging of
steel bre that increases the stress and deformation capacity of
concrete prior to peak load. In the post-peak region, the control
mix K1 exhibited much steeper falling branch compared to the
mixes K2 and K3 as in the SFOPSC, the presence of steel bre delayed the propagation of micro-cracks to form macro-cracks
through its crack bridging mechanism as explained earlier. Hence,
this mechanism allowed the SFOPSC to sustain more load with
higher deformations, leading to an increased energy absorption
capacity and compressive toughness of concrete.

4. Conclusions
The effect of steel bres addition by 0.5%, 0.75% and 1.0% in
SFOPSC on the density, UPV, mechanical properties, direct tensile
strength, fracture parameters, exural toughness and compressive

(i) The ODD of SFOPSC specimens that contained steel bres up

to 0.5% ODD was found to be within the limit of lightweight
concrete as stipulated in EN206-1. However, a slight
increase in the ODD was found for mixes with 0.75% and
1.0%.
(ii) The homogeneity of the concrete was not affected by the
addition of steel bres as UPV values were generally found
to increase with addition up to 1.0% steel bre. The UPV values of above 4.00 km/s for the mixes with steel bres show
that the concrete is of good quality and the bres had no
impact on the compaction of concrete that is generally
encountered while using bres.
(iii) The overall effect of bres in arresting the crack and its propagation enhanced the mechanical properties of OPSC such as
compressive strength, splitting tensile strength, and exural
strength. Further, the contribution of bres to stress redistribution caused lower strain and improved the MOE as for the
same stress level the strain produced in the OPSC was higher
compared to the SFOPSC.
(iv) The direct tensile strength obtained in this investigation
gave lower values of concrete as compared to the splitting
tensile and exural strengths. The direct tensile strength of
SFOPSC was found to increase with the increase in steel bre
volume. Despite the increase in steel bre content, the
enhancement in the direct tensile strength was found lower
compared to the splitting tensile and exural strengths.
(v) The addition of 1.0% of steel bre signicantly improved the
exural toughness of the SFOPSC by up to 16 times compared to OPSC. The improvement is attributed to better
post-crack response as the steel bres are effective in
imparting ductility to the concrete.
(vi) The fracture parameters such as fracture energy, fracture
toughness and characteristic length investigated for the
SFOPSC indicate improved performance due to addition of
steel bres. The fracture energy was enhanced by almost 3
times when steel bres were added from 0.5% to 1.0% volume fraction.
(vii) The ductility of the SFOPSC of cylindrical specimens under
compression measured as compressive toughness shows signicant improvement by about 3 and 6 times with the addition of 0.5% and 0.75% of steel bre, respectively.

Acknowledgements
This research work was funded by the University of Malaya under the High Impact Research Grant (HIRG) No. UM.C/HIR/MOHE/
ENG/02/D000002-16001 (Synthesis of Blast Resistant Structures).
The authors are grateful to YTL Cement Sdn Bhd for the supply of
GGBS used in this study.
References
[1] Mannan MA, Ganapathy C. Concrete from an agricultural waste-oil palm shell
(OPS). Build Environ 2004;39(4):4418.
[2] Basri HB, Mannan MA, Zain MFM. Concrete using waste oil palm shells as
aggregate. Cem Concr Res 1999;29(4):61922.
[3] Shagh P, Jumaat MZ, Mahmud H, Alengaram UJ. Oil palm shell lightweight
concrete containing high volume ground granulated blast furnace slag. Constr
Build Mater 2013;40:2318.
[4] Alengaram UJ, Mahmud H, Jumaat MZ. Comparison of mechanical and bond
properties of oil palm kernel shell concrete with normal weight concrete. Int J
Phys Sci 2010;5(8):12319.
[5] Balendran RV, Zhou FP, Nadeem A, Leung AYT. Inuence of steel bres on
strength and ductility of normal and lightweight high strength concrete. Build
Environ 2002;37(12):13617.

28

K.H. Mo et al. / Construction and Building Materials 55 (2014) 2028

[6] Hassanpour M, Shagh P, Mahmud H. Lightweight aggregate concrete ber

reinforcement a review. Constr Build Mater 2012;37:45261.
[7] Libre NA, Shekarchi M, Mahoutian M, Soroushian P. Mechanical properties of
hybrid ber reinforced lightweight aggregate concrete made with natural
pumice. Constr Build Mater 2011;25(5):245864.
[8] Shagh P, Jumaat MZ, Mahmud H. Oil palm shell as a lightweight aggregate for
production high strength lightweight concrete. Constr Build Mater 2011;25(4):
184853.
[9] Alengaram UJ, Muhit BAA, Jumaat MZ. Utilization of oil palm kernel shell as
lightweight aggregate in concrete a review. Constr Build Mater 2013;38:
16172.
[10] Chen B, Liu J. Contribution of hybrid bers on the properties of the highstrength lightweight concrete having good workability. Cem Concr Res 2005;
35(5):9137.
[11] Shagh P, Jumaat MZ, Mahmud H. Mix design and mechanical properties of oil
palm shell lightweight aggregate concrete: a review. Int J Phys Sci 2010;5(14):
212734.
[12] Altun F, Aktas B. Investigation of reinforced concrete beam behavior of steel
ber added lightweight concrete. Constr Build Mater 2013;38:57581.
[13] Banthia N, Sappakittipakorn M. Toughness enhancement in steel ber
reinforced concrete through ber hybridization. Cem Concr Res 2007;37(9):
136672.
[14] Soutsos MN, Le TT, Lampropoulos AP. Flexural performance of bre reinforced
concrete made with steel and synthetic bres. Constr Build Mater 2012;36:
70410.
[15] Wang HT, Wang LC. Experimental study on static and dynamic mechanical
properties of steel ber reinforced lightweight aggregate concrete. Constr
Build Mater 2013;38:114651.
[16] Rosello C, Elices M, Guinea GV. Fracture of model concrete: 2. Fracture energy
and characteristic length. Cem Concr Compos 2006;36(7):134553.
[17] Koksal F, Altun F, Yigit I, Sahin Y. Combined effect of silica fume and steel ber
on the mechanical properties of high strength concretes. Constr Build Mater
2008;22(8):187480.
[18] Sahin Y, Koksal F. The inuences of matrix and steel bre tensile strengths on
the fracture energy of high-strength concrete. Constr Build Mater 2011;25(4):
18016.
[19] Gao J, Sun W, Morino K. Mechanical properties of steel ber-reinforced, highstrength, lightweight concrete. Cem Concr Compos 1997;19(4):30713.
[20] Nataraja MC, Dhang N, Gupta AP. Toughness characterization of steel berreinforced concrete by JSCE approach. Cem Concr Res 2000;30(4):5937.
[21] Mohammadi Y, Singh SP, Kaushik SK. Properties of steel brous concrete
containing mixed bres in fresh and hardened state. Constr Build Mater
2008;22(5):95665.
[22] Barros JAO, Figueiras JA. Flexural behavior of SFRC: testing and modeling. J
Mater Civ Eng 1999;11(4):3319.
[23] Barros JAO, Cruz JS. Fracture energy of steel ber-reinforced concrete. Mech
Compos Mater St 2001;8(1):2945.
[24] Chen B, Liu J. Effect of aggregate on the fracture behavior of high strength
concrete. Constr Build Mater 2004;18(8):58590.
[25] Alengaram UJ, Al Muhit BA, Jumaat MZ, Liu MYJ. A comparison of the thermal
conductivity of oil palm shell foamed concrete with conventional materials.
Mater Des 2013;51:5229.
[26] Teo DCL, Mannan MA, Kurian VJ. Structural concrete using oil palm shell (OPS)
as lightweight concrete. Turk J Eng Env Sci 2006;30(4):2517.
[27] Alengaram UJ, Jumaat MZ, Mahmud H. Ductility behavior of reinforced palm
kernel shell concrete beams. Eur J Sci Res 2008;23(3):40620.
[28] Mo KH, Yap SP, Alengaram UJ, Jumaat MZ, Bu CH. Impact resistance of
hybrid bre-reinforced oil palm shell concrete. Constr Build Mater 2014;50:
499507.

[29] Shah AA, Ribakov Y. Recent trends in steel bered high-strength concrete.
Mater Des 2011;32(89):412251.
[30] Barragan BE, Gettu R, Martin MA, Zerbino RL. Uniaxial tension test for steel
bre reinforced concrete a parametric study. Cem Concr Compos
2003;25(7):76777.
[31] Marar K, Eren O, Celik T. Relationship between impact energy and compression
toughness energy of high-strength ber-reinforced concrete. Mater Lett
2001;47(45):297304.
[32] Gencel O, Brostow W, Datashvili T, Thedford M. Workability and mechanical
performance of steel ber-reinforced self-compacting concrete with y ash.
Compos Interface 2011;18(2):16984.
[33] Neves RD, de Almeida JCOF. Compressive behaviour of steel bre reinforced
concrete. Struct Concr 2005;6(1):18.
[34] Shagh P, Mahmud H, Jumaat MZ. Effect of steel ber on the mechanical
properties of oil palm shell lightweight concrete. Mater Des 2011;32(7):
392632.
[35] Campione G, Miraglia N, Papie M. Mechanical properties of steel bre
reinforced lightweight concrete with pumice stone or expanded clay
aggregates. Mater Struct 2001;34(4):20110.
[36] Lee HK, Song SY. Inuence of ber volume fraction and ber type on
mechanical properties of FRLACC. J Reinf Plast Comp 2010;29(7):108998.
[37] Kayali O, Haque MN, Zhu B. Some characteristics of high strength ber
reinforced lightweight aggregate concrete. Cem Concr Compos 2003;25(2):
20713.
[38] Akcaozoglu S, Atis CD. Effect of granulated blast furnace slag and y ash
addition on the strength properties of lightweight mortars containing waste
PET aggregates. Constr Build Mater 2011;25(10):40528.
[39] Sivakumar A, Santhanam M. Mechanical properties of high strength concrete
reinforced with metallic and non-metallic bres. Cem Concr Compos 2007;
29(8):6038.
[40] Kumar V, Sinha AK, Prasad MM. Static modulus of elasticity of steel ber
reinforced concrete. In: Proceedings of the international conference on cement
combinations for durable concrete, Dundee Scotland, 2005. p. 52736.
[41] Kurugol S, Tanacan L, Ersoy HY. Youngs modulus of ber-reinforced and
polymer-modied lightweight concrete composites. Constr Build Mater 2008;
22(6):101928.
[42] Kuder K, Lehman D, Berman J, Hannesson G, Shogren R. Mechanical properties
of self consolidating concrete blended with high volumes of y ash and slag.
Constr Build Mater 2012;34:28595.
[43] Shariq M, Prasad J, Abbas H. Effect of GGBS on age dependent static modulus of
elasticity of concrete. Constr Build Mater 2013;41:4118.
[44] Banthia N. Fiber reinforced concrete. ACI SP-142ACI, Detroit, MI; 1994. p. 91
119.
[45] Kou S, Poon C, Agrela F. Comparisons of natural and recycled aggregate
concretes prepared with the addition of different mineral admixtures. Cem
Concr Compos 2011;33(8):78895.
[46] Atis CD, Karahan O. Properties of steel ber reinforced y ash concrete. Constr
Build Mater 2009;23(1):3929.
[47] Neville AM, Brooks JJ. Concrete technology. ELBS, 1987.
[48] Tadepalli PR, YL Mo, Hsu TTC. Mechanical properties of steel bre concrete.
Mag Concr Res 2013;65(8):46274.
[49] Cui HZ, Lo TY, Memon SA, Xu W. Effect of lightweight aggregates on the
mechanical properties and brittleness of lightweight aggregate concrete.
Constr Build Mater 2012;35:14958.
[50] Eskandari H, Muralidhara S, Raghuprasad BK, Reddy BV. Size effect in self
consolidating concrete beams with and without notches. Sadhana 2010;35(3):
30317.
[51] Lin Z, Li VC. Crack bridging in ber reinforced cementitious composites with
slip-hardening interfaces. J Mech Phys Solids 1997;45(5):76387.