Materials and Design 60 (2014) 320–327

Contents lists available at ScienceDirect

Materials and Design

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

A comparison study of the mechanical properties and drying shrinkage

of oil palm shell and expanded clay lightweight aggregate concretes
Payam Shafigh a,⇑, Hafez Ghafari b, Hilmi Bin Mahmud a, Mohd Zamin Jumaat a
a
Department of Civil Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia
b
Sustainable Deliverable, 140 N Wilson Ave, Pasadena, CA 91106, USA

a r t i c l e i n f o a b s t r a c t

Article history: For making artificial lightweight aggregate, selected raw materials are fed into a rotary kiln at high tem-
Received 17 October 2013 perature. Providing such a high temperature is costly and generally, the process of making artificial light-
Accepted 1 April 2014 weight aggregate is not environmentally friendly. The use of natural lightweight aggregate for making
Available online 12 April 2014
lightweight concrete can lead to low-cost construction. The use of a solid waste lightweight aggregate
namely oil palm shell (OPS) as coarse aggregate, is not only environmentally friendly but leads to a
Keywords: low-cost material. This study is a comparison between some engineering properties of OPS lightweight
Oil palm shell
concrete and an artificial lightweight (expanded clay) concrete with low water to cement ratio, along
Expanded clay
Lightweight aggregate
with having good workability and without any segregation. The test results show that OPS concrete
Mechanical properties has better mechanical properties and a higher efficiency factor than expanded clay lightweight concrete.
Efficiency factor The ceiling strength of expanded clay lightweight concrete occurs at an early age; while it happens in OPS
Drying shrinkage concrete at a later age. The crack pattern of the tested specimens shows that OPS is much stronger than
expanded clay. On the other hand, the compressive strength of OPS lightweight concrete is more sensitive
to lack of curing. Although OPS lightweight concrete shows twice the amount of drying shrinkage than
expanded clay lightweight concrete in the short term, this difference reduces significantly at later ages.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction In countries located in tropical regimes such as Malaysia, Indo-

When normal weight concrete is compared with steel, one of the
main disadvantages of concrete is its low strength-weight ratio.
Such a disadvantage of concrete can be resolved proportionally
when lightweight concrete is made, particularly a high strength
lightweight concrete. Structural lightweight concrete is catego-
rized as a special type of concrete [1]. In most cases, structural
lightweight concrete is made with a lightweight aggregate as coarse
aggregate and normal weight sand as the fine aggregate. There are
many different types of natural and artificial lightweight aggregates
with different characteristics. Such a disparity in properties means
that the lightweight concrete made of each type of lightweight
aggregate has special engineering properties. Not all of the available
aggregates are equally suitable for a particular application [2]. For
example, among all the lightweight aggregates only a few can be
used as aggregate for producing high strength lightweight concrete
[3]. Therefore, for each type of lightweight aggregate the engineer-
ing properties should be investigated extensively before any
application in the construction industry.
⇑ Corresponding author. Tel.: +60 172437383; fax: +60 379675318.
E-mail addresses: pshafigh@gmail.com, pshafigh@um.edu.my (P. Shafigh).
nesia and Nigeria, there is a type of lightweight aggregate from the
agricultural industry; namely, oil palm shell (OPS). OPS is a solid
waste from the palm oil industry. Research over two decades has
shown that lightweight concrete incorporating OPS has good
mechanical properties [4] and durable performance [5]. It has been
shown that reinforced concrete beams made of OPS lightweight
concrete with normal strength have satisfactory shear and flexural
performance [6,7]. The experimental bond strength of OPS light-
weight concrete has been shown to be significantly higher than
the design bond strength [8]. In addition, recent studies have dem-
onstrated that OPS can be used as lightweight aggregate for pro-
ducing high strength lightweight concrete [9].
In most cases, artificial lightweight aggregates are manufactured
by a sintering process of raw materials from natural resources. The
sintering process is an expensive method of manufacturing due to
the large energy consumption [10]. For example, for manufacturing
lightweight aggregates, such as expanded clay, shale and slate, se-
lected raw materials are fed into a rotary kiln with a temperature
of about 1200 °C. In the case of expanded perlite lightweight aggre-
gate, a temperature of about 1800 °C is needed. However, with a
new technique the temperature needed for rotary kiln is reduced
to about 860 °C. Providing such a high temperature is very

http://dx.doi.org/10.1016/j.matdes.2014.04.001
0261-3069/Ó 2014 Elsevier Ltd. All rights reserved.
 P. Shafigh et al. / Materials and Design 60 (2014) 320–327
expensive. Priyadharshini et al. [10] suggest that for producing
greener artificial lightweight aggregate, industrial by-products
such as heavy metal sludge, mining residue, steel slag, bottom ash
and fly ash, should be used instead of natural raw materials. In addi-
tion, Hossain et al. [11] suggest using natural lightweight aggregate
such as pumice, instead of processed artificial lightweight aggre-
gates. They demonstrate that natural lightweight aggregate con-
crete as a construction material can lead to low-cost construction.
Inasmuch as OPS is a waste material that does not need any heating
process, the use of this lightweight aggregate for manufacturing
lightweight concrete not only leads to low-cost construction but
also causes this type of concrete to be a more environmentally
friendly material. Struble and Godfrey [12] stated that environ-
ment, economy and society are three components of sustainability,
which the environment is the most important component nowa-
days and the economy component is given less attention in devel-
oped countries. Therefore, it is clear that OPS lightweight concrete
can meet the requirements of sustainability.
The main aim of this study is to make a comparison between
some of the engineering properties of OPS lightweight concrete
as an environmentally friendly and low-cost material with that
of an artificial lightweight aggregate concrete made of lightweight
expanded clay aggregate. These two types of concrete are made
with low water to cement ratio but with good workability and
without any segregation. Various properties are compared such
as compressive strength in different curing conditions, splitting
tensile and flexural strengths, water absorption and drying
shrinkage.
2. Experimental work
2.1. Materials used
The binder used was ordinary Portland cement (OPC), which
was obtained from a local cement company with a specific gravity,
Blaine surface area, initial and final setting times of 3.14 g/cm3,
3510 cm2/g, 65 min and 140 min, respectively. A superplasticizer
(SP) based on polycarboxylic ether (PCE) was used in all mixes.
Local mining sand with a maximum nominal size of 4.75 mm,
specific gravity of 2.65 and fineness modulus of 2.70 was used as
fine aggregate. Crushed old OPS and expanded clay with a maxi-
mum nominal size of 8 mm as well as crushed granite with a max-
imum nominal size of 12.5 mm were used as coarse aggregate in
the concrete mixtures. The specific gravity and 24-h water absorp-
tion of OPS were 1.2% and 20%, respectively, while for expanded
clay they were 0.66% and 28%, respectively.
2.2. Mix proportions
The mix proportion of lightweight concrete containing OPS (mix
P) was selected based on the method reported by Shafigh et al. [9].
This mix had a water to cement ratio of 0.29% and 1% SP (by mass
of cement). The slump value of mix P was 145 mm. Mix L was
made of expanded clay lightweight aggregate. In this mix the vol-
ume content of lightweight aggregate was the same as OPS. There-
fore, the main difference between the two mixes is the type of
lightweight aggregate used. However, it should be noted that mix
L had a slightly lower SP dosage and water content. This is because
unlike OPS aggregates expanded clay aggregates are round. This
shape helps to improve workability. In addition, the density of ex-
panded clay is significantly lighter than the OPS (approximately
45%). Therefore, if the SP and water used in mix L were similar to
mix P then the expanded clay aggregate would float upwards.
Therefore, mix L was made with a low water to cement ratio of
0.26 and SP dosage of 0.5% with a slump value of 120 mm. The
321
mix proportions of the two mixes are shown in
Table 1.
2.3. Test methods
The concrete specimens were cast in 100-mm cubes, cylinders of
100-mm diameter and 200-mm height, prisms of 100  100 
500 mm3 and prisms of 100  100  300 mm3 steel moulds for
measuring compressive strength, splitting tensile strength, flexural
strength and drying shrinkage, respectively. All specimens were
compacted using a vibrating table. The specimens were demoulded
24 h after casting. At least three specimens were prepared for
obtaining the average value for mechanical properties and two
specimens were used for the drying shrinkage test. The compressive
strength of specimens was determined on the 1st, 3rd, 7th, 28th,
56th, and 90th day in accordance to BS 1881: Part 116 using a com-
pression testing machine of 3000 kN capacity with a rate of loading
controller. The drying shrinkage test, under laboratory environment
condition, was conducted immediately after demoulding. The
shrinkage value for each age is the average of six readings.
2.4. Curing regimes
For determining the effect of the curing regimes on the 28-day
compressive strength of concretes, the specimens were cured in
five curing conditions, as follows:
(a) Continuous moist curing (FW): specimens were immersed in
water at a temperature of 23 ± 3 °C.
(b) Air drying (AC): specimens were kept in the laboratory envi-
ronment with RH of 67–82% and temperature of 29 ± 3 °C
after demoulding.
(c) 3 days (3 W), 5 days (5 W) and 7 days (7 W) partial early
curing: curing in water for 2, 4 and 6 days, respectively, after
demoulding and then air drying in a laboratory environment
until the age of testing.
3. Results and discussion
3.1. Development of compressive strength
The compressive strength development of P and L mixes under
continuous moist curing up to 90 days is shown in Fig. 1. Both of
the mixes have a sharp compressive strength gain until the early
age of 7 days. At this age, P and L mixes have about 95% and 97%,
respectively, of the 28-day compressive strength. It is clear from
Fig. 1 that the strength gain continues in mix P until the age of
90 days and after 7-day age. However, such a trend is not observed
in mix L. In mix P, the 7 days to 56 and 90 days compressive
strength is 91% and 87%, respectively, and 96% and 96%, respec-
tively, for mix L. These results show that mix L reaches its ceiling
strength at the age of 7 days. However, such a ceiling strength
was not observed in mix P. The ceiling strength of concrete de-
pends upon the type of aggregate [13]. In lightweight aggregate
concrete, when the strength reaches the ceiling strength, further
addition of cementitious materials will not significantly raise the
maximum attainable strength [14]. Although mix L had a stronger
mortar due to the lower water to cement ratio than mix P, its ceil-
ing strength occurs at an early age of 7 days. This shows that OPS is
stronger and stiffer than the expanded clay. Okpala [15] reports
that the OPS has 37% porosity, is fibrous in nature, and has a com-
pressive crushing strength of about 12.1 MPa. The Los Angeles
abrasion value of the shell is approximately 76–91% less than gran-
ite aggregate [4,15].
Neville and Brooks [16] reported that generally, concrete made
with expanded shale or clay aggregate has a higher strength than
322
Table 1
Mix proportions of lightweight concrete (kg/m3).
Mix no. Cement Water
P 550 158.4
L 550 141.4
55
Compressive strength
50
45
40
(MPa)
35
30
25
Mix P Mix L
20
0 10 20 30 40 50 60 70
Age (days)
Fig. 1. Development of compressive strength of lightweight concrete.
when any other lightweight aggregate is used. However, it can be
seen that OPS provides significantly greater compressive strength
than this artificial lightweight aggregate. The 28, 56 and 90 days
compressive strength of mix P is about 44%, 49% and 55% greater
than mix L, respectively.
3.2. Effect of curing type on the compressive strength
Table 2 shows the 28-day compressive strength of concretes
under different types of curing regime. Compared to water curing
(FW), both types of lightweight aggregate concrete showed the
lowest compressive strength of about 13–14% under air drying
regimes (AC). Under partial early curing regimes of 3W, OPS and
expanded clay lightweight concretes showed a reduction in com-
pressive strength of about 12.1% and 5.4% compared to standard
curing, respectively. However, under the partial early curing
regimes of 5W and 7W, the OPS concrete showed slightly lower
compressive strength than standard curing, while the expanded
clay lightweight concrete indicated significantly greater compres-
sive strength. In general, expanded clay lightweight concrete
showed better performance in strength gain than OPS concrete un-
der partial early curing regime. To understand the reasons for such
different behaviours, three points should be considered. First, the
cube samples under the standard and partial early curing regimes
had different conditions in terms of skin humidity at the time of
the compressive strength test. The concrete specimens under
standard curing were in a saturated condition, while the concrete
specimens under partial early curing regimes were in a dry condi-
tion. The saturated skin of a specimen has a negative effect on the
compressive strength [17]. Mehta and Monteiro [18] reported that
air dried specimens have 20–25% higher strength than specimens
in a saturated condition. As Lamond and Pielert [19] reported, this
effect is in the range of 5–20%, while in the case of OPS concrete, it
was in the range of 2–9% [20]. The lower strength of the saturated
Table 2
The 28-day compressive strength of concrete under air drying and partial early curing regimes.
Mix no. 28-day compressive strength under different curing environments (MPa)
FW
P 44.5
L 30.8
P. Shafigh et al. / Materials and Design 60 (2014) 320–327
Super plasticizer Sand Coarse aggregate
Granite Oil palm shell Expanded clay
5.50 720 178 326 0
2.75 720 178 0 180
concrete is attributed to the disjoining pressure within the cement
paste [18]. As can be seen from the reports, there are differences in
strength due to the effect of dry skin which show that dry and sat-
urated skin affect the compressive strength of different types of
concrete. However, even by considering the skin effect, the light-
weight concrete made of expanded clay aggregate shows equiva-
lent compressive strength in the cases of partial early curing of
5W and 7W with the 28-day compressive strength of specimens
80 90 100 under standard moist curing. Zhao et al. [21] have studied the
effect of initial water curing on the compressive strength of
self-compacting concrete. They observed that specimens under
air drying condition had better 28-day compressive strength
(about 3–5%) compared to moist curing condition. In addition, they
concluded that an initial water curing of 7 days and subsequent
room exposure are more beneficial for compressive strength devel-
opment of self compacting concrete than initial water curing for 3,
14 and 28 days.
Secondly, it should be noted that the water absorption of light-
weight aggregates is significantly more than for normal weight
aggregates. When lightweight aggregates are used in concrete in
a pre-saturated condition, the water inside the lightweight aggre-
gate is additional internal water, which is not part of the mixing
water. This water can penetrate from the lightweight aggregate
into the surrounding matrix to a distance of up to several millime-
ters during the first seven days of hydration [22]. Lura et al. [23]
have measured water transport from saturated pumice lightweight
aggregate to hardening cement paste with water to cement ratio of
0.3. They reported that considerable transport of water from light-
weight aggregate to hydrating cement paste took place in the first
days after casting. One of the benefits of internal curing is in-
creased hydration and strength development [24]. In this study,
OPS and expanded clay aggregates were used in the SSD condition.
One reason for the greater compressive strength of expanded clay
lightweight concrete in the 5W and 7W conditions compared to
water curing may be due to the better performance of the ex-
panded clay lightweight aggregates than the OPS for providing bet-
ter cement hydration because of internal curing.
Thirdly, the role of the interfacial transaction zone (ITZ) in con-
crete for different lightweight aggregates should be considered.
Fig. 2 shows a part of two cylindrical specimens of two types of
concrete after the compressive strength test. As can be seen in this
figure, there is no separation between the cement matrix and the
expanded clay. However, it is clear that the OPS were separated
from the matrix. In the case of expanded clay, it has been shown
that the cement paste diffuses into the aggregate shell through
the interface between the lightweight aggregate and cement paste
and aggregate [25]. However, the bond between the OPS and the
cement paste is not strong because of the organic effect and
AC 3W 5W 7W
38.3 39.1 42.5 43.4
26.6 29.2 35.4 34.4
 P. Shafigh et al. / Materials and Design 60 (2014) 320–327
Fig. 2. Crack pattern of (a) OPS and (b) expanded clay lightweight concrete specimens after compression test.
smoothness of the OPS surface [26]. Therefore, it seems that after
initial partial early curing and when specimens were exposed to
drying, micro cracks affect the ITZ in OPS concrete due to drying
shrinkage. This causes lower compressive strength than standard
curing. However, in expanded clay lightweight concrete there were
no losses of strength due to micro cracks.
It has been shown in a previous study [20] that factor, such as
water to cement ratio as well as OPS and cement content influence
the sensitivity of OPS concrete in poor curing environments. How-
ever, initial early water curing can significantly reduce the strength
loss of OPS concrete.
3.3. Efficiency factor
The efficiency factor (strength/weight ratio) of concrete is an
important factor in the behaviour of a concrete structure. Struc-
tural lightweight concrete has a significantly greater strength/
weight ratio than normal weight concrete at the same strength.
This advantage of structural lightweight concrete means that this
type of concrete can be competitive with even high strength nor-
mal weight concrete. Moravia et al. [27] reported in their study
that although a lightweight aggregate concrete made of expanded
clay aggregate demonstrates lower compressive strength than nor-
mal weight concrete, it has significantly greater efficiency factor.
Table 3 shows the 28-day compressive strength, density and effi-
ciency factors for both mixes. It can be seen that although the
dry density of mix P is about 5% more than mix L, the strength of
mix P is significantly more than mix L (about 44%), which contrib-
utes to the greater efficiency factor for mix P. The results show that
the efficiency factor of OPS concrete (mix P) is about 30% more
than for the expanded clay lightweight concrete (mix L). This com-
parison clearly shows another advantage of using OPS lightweight
aggregates in concrete. It should be noted that OPS is a solid waste
while expanded clay is a manufactured aggregate, for which
considerable energy and natural raw materials are needed in its
Table 3
Density and efficiency factor of concrete.
Mix no. 28-day compressive strength (MPa)
P 44.5
L 30.8
323
production. In addition, this artificial lightweight aggregate is an
imported material in the Malaysian market and compared to OPS
aggregate, is expensive.
The efficiency factor for a normal weight concrete with a 28-day
compressive strength similar to mix P and with a dry density of
2350 kg/m3 is 18,936 N M/kg, which is about 14% less than the effi-
ciency factor for the OPS concrete. This shows the better perfor-
mance of OPS concrete compared to normal weight concrete at
the same compressive strength.
3.4. Splitting tensile strength
As specified by ASTM: C330, the minimum splitting tensile
strength of a structural lightweight aggregate concrete should be
2.0 MPa. As can be seen in Table 4, both concretes showed signifi-
cantly more splitting tensile strength (61% for OPS concrete and
38% for expanded clay concrete) than the minimum requirement
of ASTM: C330.
The splitting tensile to compressive strength ratio of OPS con-
crete is 7.2% which is lower than for the expanded clay lightweight
concrete with a ratio of about 9%. This may be due to the compres-
sive strength of OPS concrete being higher than for the expanded
clay lightweight concrete. Neville [28] reported that as the com-
pressive strength increases the tensile strength also increases,
although at a decreasing rate. However, it should be noted that
high strength lightweight aggregate concrete has a splitting tensile
to compressive strength ratio in the range of 6–7% [29]. In the case
of OPS concrete it has been reported that the tensile to compres-
sive strength ratio is comparable with lightweight aggregate con-
crete made of artificial lightweight aggregate at the equivalent
grade [30].
Compared to compressive strength, tensile strength of concrete
is low that generally it is assumed to be zero for designing a rein-
forced concrete member. However, for some purposes such as the
design of highway and airfield slabs, shear strength, concrete
Density (kg/m3) Efficiency factor (N M/kg)
Demould Air dry Oven dry
2046 2011 1898 22,128
1854 1809 1801 17,025
324 P. Shafigh et al. / Materials and Design 60 (2014) 320–327

Table 4
Measured and predicted splitting tensile strength.

Mix no. P L
Experiment
Compressive strength (MPa) 44.5 30.8
Splitting tensile strength (MPa) 3.22 2.77
Splitting tensile strength to compressive strength ratio (%) 7.2 9.0
Predicted splitting tensile strength (MPa)
Equation No. Description Ref.
pffiffiffiffiffiffi
ft ¼ 0:59 fcy 1 Concrete strength ranging from 21 to 83 MPa [32] 3.52 2.93
(9.3%") (5.8%")
0:67 2 CEB-PIP [33] 3.30 2.58
ft ¼ 0:301fcy
(2.5%") (6.9%;)
ft = fcy/(0.1fcy + 7.11) 3 Normal weight concrete with compressive strength ranging from 35 to 115 MPa [34] 3.34 3.22
(3.7%") (16.2%")
pffiffiffiffiffiffi
ft ¼ 0:54 fcy ðw=bÞ
0:07 4 Normal weight concrete with compressive strength ranging from 35 to 115 MPa and w/b ratio in the [34] 3.52 2.95
range of 0.22 to 0.55 (9.3%") (6.5%")
pffiffiffiffiffiffi
ft ¼ 0:4887 fcu 5 Crushed OPS concretes with cube compressive strength ranging from 35 to 53 MPa [20] 3.26 2.71
(1.2%") (2.2%;)
pffiffiffiffiffiffi
2
ft ¼ 0:20 3 fcu 6 OPS concretes containing original OPS aggregates with cube compressive strength ranging from 17 to [35] 2.51 1.97
37 MPa (22%;) (28.9%;)
pffiffiffiffiffiffi
ft ¼ 0:46 fcy 7 Natural Tuff LWAC with high strength [36] 2.74 2.28
(14.9%;) (17.7%;)
pffiffiffiffiffiffi
2
ft ¼ 0:23 3 fcu 8 Pelletized blast furnace slag LWAC with cube compressive strength ranging from 10 to 65 MPa [28] 2.89 2.26
(10.2%;) (18.4%;)
pffiffiffiffiffiffi
ft ¼ 0:51 fcy 9 High strength lightweight concrete [37] 3.04 2.53
(5.6%;) (8.7%;)
pffiffiffiffiffiffi
2
ft ¼ 0:27 3 fcu 10 Cold-bonded fly ash LWAC with a cube compressive strength ranging from 20 to 47 MPa [38] 3.39 2.65
(5.3%") (4.3%;)
0:675
ft ¼ 0:358fcu 11 Lightweight expanded polystyrene aggregate concretes [39] 4.64 3.62
(44%") (30.7%")
0:7
ft ¼ 0:20fcy 12 Normal weight concrete [40] 2.85 2.20
(11.5%;) (20.6%;)
0:85
ft ¼ 0:14fcu 13 Normal weight concrete with cube compressive strength ranging from 25 to 115 MPa [41] 3.53 2.58
(9.6%") (6.9%;)

ft is the splitting tensile strength (MPa), fcy is the cylinder compressive strength (MPa), fcu is the cube compressive strength (MPa), ("): indicating overestimate, (;): indicating
underestimate, w/b is water to binder ratio.

pavements and other slabs, the tensile strength must be known
[28]. Bhanja and Sengupta [31] have stated that due to low tensile
strength of concrete, it cracks in tension very easy and the cracks
can propagate easily. They demonstrated that cracks may cause
serviceability and durability problems. Therefore, due to the
importance of knowing tensile strength many equation prediction
models were developed to predict tensile strengths from the com-
pressive strength. Table 4 shows several existing equations for pre-
dicting the splitting tensile strength from the compressive
strength, as well as the predicted splitting tensile strength for both
types of lightweight aggregate concrete. In the equations where
the cylinder compressive strength was used, a coefficient of 0.8
[42] was used for converting the cube to cylinder compressive
strength. It can be seen from this table that if just OPS concrete
is considered, Eqs. (2), (3) and (5) show an acceptable estimation
(an error below 5%), and that if expanded clay lightweight concrete
is considered, Eqs. (5) and (10) showed statistically insignificant
error. However, if both types of concrete are considered the use
of Eq. (5) is recommended.
3.5. Flexural strength
Table 5 shows the test results for the flexural strength for both
types of concrete. It can be seen that similar to compressive and
splitting tensile strengths, the flexural strength of OPS concrete is
more than the expanded clay lightweight concrete. In sequence,
it shows that the OPS concrete (mix P) has a compressive strength,
flexural strength and splitting tensile strength of about 44%, 30%
and 16% greater than the expanded clay lightweight concrete,
respectively. However, the flexural to compressive strength and
splitting tensile to flexural strength ratios of OPS concrete are less
than for expanded clay lightweight concrete. Nevertheless, the dif-
ferences are not significant. It should be noted that, typically, the
flexural strength to compressive strength ratio of concrete is of
the order of 15% [18]. Compared to normal weight concrete, light-
weight concrete has a lower flexural to compressive strength ratio
[43]. Generally, high strength lightweight concrete has a flexural to
compressive strength ratio in the range of 9–11% [29]. It can be
seen that mix P has a marginally greater flexural to compressive
strength ratio than the value mentioned for high strength light-
weight aggregate concrete.
The splitting to flexural strength ratio of OPS concrete is about
65%, which is lower than that of the expanded clay lightweight
concrete. For a wide range of compressive strength of OPS con-
cretes (up to 53 MPa), reports [20,44] showed that the splitting
tensile to flexural strength ratio varies between 51% and 72%. This
ratio is affected by the amount of OPS in the mixture [20].
Table 5 also shows the existing prediction equation for flexural
strength from compressive strength as well as the predicted flex-
ural strength with the percentage error prediction. As can be seen
in this table, Eqs. (19) and (20) give the lowest error in the case of
OPS concrete. In the case of expanded clay concrete, Eqs. (16) and
(22) show the lowest error. However, there is no equation for esti-
mating the flexural strength from the compressive strength with
low error (an error below 5%) at the same time for both types of
concrete. If it be assumed that a prediction error of up to 10% is
acceptable then a general equation for both types of concrete can
be developed as follows:
pffiffiffiffiffi
fr ¼ k fcu ð23Þ
where fr is flexural strength in MPa, fcu is cube compressive strength
in MPa, and k is a constant coefficient in the range of 0.69–0.73.
 P. Shafigh et al. / Materials and Design 60 (2014) 320–327 325

Table 5 twice that of expanded clay lightweight concrete. However, the dif-
Measured and predicted flexural strength. ference between the shrinkage of the two concretes reduces over
Mix no. P L time. Fig 4 shows the ratio of drying shrinkage of OPS and ex-
Experiment panded clay lightweight concretes with time. The variation of
Compressive strength (MPa) 44.5 30.8 shrinkage ratios are shown by highlighted areas. The areas are di-
Flexural strength (MPa) 4.95 3.81 vided into two parts of A and B. Part A shows that the ratio de-
Flexural strength to compressive strength ratio (%) 11.1 12.4 creases significantly over time up to about 90 days. The mean of
Splitting tensile strength to flexural strength ratio (%) 65 73
area A is a line with the following function:
Predicted flexural strength (MPa)
Equation No. Description Ref. y ¼ 0:0101x þ 2:16 ð24Þ
fr ¼ 1:03
0:12fcu 14 Crushed OPS concretes [20] 5.98 4.10
with cube compressive (20.8%") (13.3%")
where y is the ratio of drying shrinkage of OPS to expanded clay
strength ranging from 35 lightweight concretes and x is the age (days).
to 53 MPa The B area shows that after three months the shrinkage ratio
pffiffiffiffiffiffi
2
fr ¼ 0:30 3 fcu 15 OPS concretes with [44] 3.77 2.95 varies between 1.24 and 1.46 with a mean value of 1.35. Generally
compressive strength (23.8%;) (28%;)
it can be said that the shrinkage value of OPS concrete is about
ranging from 15 to
37 MPa twice that of expanded clay concrete at the early ages. This differ-
pffiffiffiffiffiffi
fr ¼ 0:69 fcu 16 Expanded clay [42] 4.60 3.83 ence in shrinkage value reduces to about 0.35 times at the age of
lightweight aggregate (7.1%;) (0.5%") 90 days. To explain such a difference in the amount of drying
concrete with cube shrinkage for the two types of lightweight concrete the following
compressive strength
ranging from 29 to
points should be considered.
43 MPa Previous studies [28,50] show that for both normal and light-
pffiffiffiffiffiffi
2
fr ¼ 0:46 3 fcu 17 Lightweight concrete [45] 5.78 4.52 weight concretes there is a correlation between shrinkage and
made with expanded (16.8%") (18.6%") the modulus of elasticity of concrete. A concrete with a higher
shale and clay aggregates
modulus of elasticity shows lower drying shrinkage. It has been re-
with cube compressive
strength ranging from 20 ported [51] that an expanded clay lightweight concrete with a
to 60 MPa compressive strength of 30% less than OPS concrete has a higher
pffiffiffiffiffiffi
fr ¼ 0:58 fcy 18 Lightweight concrete [36] 3.46 2.88 modulus of elasticity of about 40%. Therefore, it is expected that
made with tuff (30.1%;) (24.4%;) the OPS concrete in this study will show greater drying shrinkage.
lightweight aggregate
pffiffiffiffiffiffi
19 Lightweight aggregate [46] 5.07 4.22
The reason for the two existing areas of A and B in Fig. 4 is prob-
fr ¼ 0:76 fcu
concrete (2.4%") (10.8%") ably due to the main difference in the properties between the two
pffiffiffiffiffiffi
fr ¼ 0:73 fcu 20 High strength lightweight [47] 4.87 4.05 concretes; namely, SP dosage and water to cement ratio.
concrete (1.6%;) (6.3%") It should be noted that the main reason for the shrinkage
pffiffiffiffiffiffi
fr ¼ 0:53 fcy 21 ACI equation for [48] 3.16 2.63
behaviour of a concrete originates from the cement paste. In drying
lightweight concrete with (36.1%;) (31.0%;)
normal sand condition, the water inside the C–S–H is removed. The loss of water
pffiffiffiffiffiffi from the small capillary pores (<50 nm) significantly affects the
fr ¼ 0:7 fcu 22 Indian standard [49] 4.67 3.88
(5.7%;) (1.8%") volume change of concrete [52]. The degree of hydration affects
fr is the flexural strength (MPa), fcy is the cylinder compressive strength (MPa), fcu is
the shrinkage of concrete [53]. The water loss from smaller pores
the cube compressive strength (MPa), ("): indicating overestimate, (;): indicating causes a higher volume change in the concrete. However, using a
underestimate. superplasticizer in concrete causes better dispersion of cement
particles, which causes more hydration products within a short
time. Mix P has 0.5% and 12% more SP and water content respec-
3.6. Drying shrinkage tively than mix L and both of these concretes have a low water
to cement ratio. It seems that mix P, with the better dispersion
Fig. 3 shows the development of the drying shrinkage of two of cement particles and higher existing water content, has more
types of lightweight aggregate concrete for up to one year. As hydration products. Therefore, it is expected that mix P will exhibit
can be seen in this figure, the drying shrinkage of OPS lightweight more drying shrinkage than mix L, especially at an early age. In
concrete is more than for the expanded clay lightweight concrete addition, due to the cement content of both concretes being high
at all ages. The increasing rate of drying shrinkage of both con- (550 kg/m3) and mix L having a lower water to cement ratio than
cretes is very sharp in the first month. The amount of drying mix P, it is expected that mix L contains a higher magnitude of
shrinkage of OPS concrete within the first week was more than

600

500
Drying shrinkage
(microstrain)

400

300

200

100
Mix P Mix L
0
0 50 100 150 200 250 300 350 400
Age (days)
Fig. 4. Relationship between drying shrinkage of OPS and expanded clay light-
Fig. 3. Drying shrinkage of P and L mixes. weight concretes ratio and age.
326 P. Shafigh et al. / Materials and Design 60 (2014) 320–327
un-hydrated cement. The un-hydrated cement in the concrete mix-
ture may be considered as part of the aggregate [18]. This means
that the aggregate volume in mix L is higher than for mix P. Hooton
et al. [54] reported that the only parameter of the mixture design
that has a significant influence on the drying shrinkage is the total
aggregate volume. The significant influence of aggregate volume
on the drying shrinkage of concrete is clearly shown in the follow-
ing equation [18]:
Sc ¼ ð1  gÞn Sp
where Sc is the shrinkage of concrete, Sp is the shrinkage of the
cement paste, g is the volume fraction of aggregate and n is a
constant that varies between 1.2 and 1.7 depending on the elastic
modulus of aggregate. Therefore, in general, it can be concluded
that for both types of lightweight concrete, when they are con-
structed with the lowest possible water to cement ratio along with
good workability, OPS concrete shows higher drying shrinkage,
particularly for the early ages.
4. Conclusions
Oil palm shell (OPS) is a solid waste from the palm oil industry.
It is lightweight and can be used as aggregate for producing struc-
tural lightweight aggregate concrete. To show the potential appli-
cations and advantages of using of OPS in concrete, some
engineering properties of OPS and a conventional artificial light-
weight aggregate (expanded clay) concretes were compared. Both
of these lightweight aggregate concretes were made with a low
water to cement ratio along with good workability and without
any segregation. From the test results of this study it can be con-
cluded that although OPS lightweight concrete has a higher dry
density than the expanded clay lightweight concrete (about 5%),
its 28-day compressive, flexural and splitting tensile strengths
are greater than the expanded clay lightweight concrete by about
44%, 30% and 16%, respectively. There is an equation for predicting
splitting tensile strength from the compressive strength for both
concretes with low error. The ceiling strength of expanded clay
lightweight concrete occurred an early age of 7 days, but for OPS
lightweight concrete it occurred at a later age. The efficiency factor
of OPS lightweight concrete is more than for the expanded clay and
normal weight concrete. OPS concrete has greater drying shrinkage
than expanded clay lightweight concrete of about 100% at the early
ages; however, it reduces to 35% at 90-day age and beyond.
Acknowledgement
The authors gratefully acknowledge the financial support from
University of Malaya under the University of Malaya Research
Grant (UMRG), Grant No. RP018/2012A.
References
[1] Kostmatka SH, Kerkhoff B, Panarese WC. Design and control of concrete
mixtures. 14th ed. USA: Portland Cement Association; 2002.
[2] Duggal SK. Building materials. 3rd ed. India: New Age International (P)
Limited; 2008.
[3] Zhang MH, Gjorv OE. Development of high-strength lightweight concrete. ACI
Spe Pub 1990;121:667–82.
[4] Teo DCL, Mannan MA, Kurian VJ. Production of lightweight concrete using oil
palm shell (OPS) aggregates. 4th International conference on construction
materials: performance, innovations and structural implications, Nagoya,
Japan; 2009.
[5] Teo DCL, Mannan MA, Kurian VJ. Durability of lightweight OPS concrete under
different curing conditions. Mater Struct 2010;43:1–13.
[6] Ahmed E, Sobuz HR. Flexural and time-dependent performance of palm shell
aggregate concrete beam. KSCE J Civ Eng 2011;15(5):859–65.
[7] Alengaram UJ, Jumaat MZ, Mahmud H, Fayyadh MM. Shear behaviour of
reinforced palm kernel shell concrete beams. Construct Build Mater
2011;25(6):2918–27.
[8] Teo DCL, Mannan MA, Kurian VJ, Ganapathy C. Lightweight concrete made
from oil palm shell (OPS): structural bond and durability properties. Build
Environ 2007;42:2614–21.
[9] Shafigh P, Jumaat MZ, Mahmud H, Alengaram UJ. A new method of producing
high strength oil palm shell lightweight concrete. Mater Des
2011;32:4839–43.
[10] Priyadharshini P, Ganesh GH, Santhi AS. A review on artificial aggregates. Int J
Earth Sci Eng 2012;5(3):540–6.
[11] Hossain KMA, Ahmed S, Lachemi M. Lightweight concrete incorporating
pumice based blended cement and aggregate: mechanical and durability
characteristics. Construct Build Mater 2011;25:1186–95.
[12] Struble L, Godfrey J. How sustainable is concrete? Proc Int Workshop Sust Dev
ð25Þ Concr Technol, Beijing 2004;20–21:201–10.
[13] Clarke JL, editor. Structural lightweight concrete. Blackie Academic &
Professional; 1993.
[14] Lamond JF, Pielert JH. Significance of tests and properties of concrete and
concrete-making materials. PA: ASTM International Press; 2006.
[15] Okpala DC. Palm kernel shell as a lightweight aggregate in concrete. Build
Environ 1990;25:291–6.
[16] Neville AM, Brooks JJ. Concrete technology. London: Longman Group Ltd.;
1987.
[17] Haque MN. Strength development and drying shrinkage of high-strength
concretes. Cem Concr Compos 1996;18:333–42.
[18] Mehta PK, Monteiro PJM. Concrete: microstructure, properties, and materials.
3rd ed. New York: McGraw-Hill; 2006.
[19] Lamond JF, Pielert JH. Significance of tests and properties of concrete and
concrete-making materials. New York: ASTM International; 2006.
[20] Shafigh P, Jumaat MZ, Mahmud H, Hamid NAA. Lightweight concrete made
from crushed oil palm shell: tensile strength and effect of initial curing on
compressive strength. Construct Build Mater 2012;27:252–8.
[21] Zhao H, Sun W, Wu X, Gao B. Effect of initial water-curing period and curing
condition on the properties of self-compacting concrete. Mater Des
2012;35:194–200.
[22] Zhutovsky S, Kovler K, Bentur A. Assessment of water migration distance in
internal curing of high-strength concrete, vol. 220. ACI Special Publication;
2004. p. 181–200.
[23] Lura P, Bentz DP, Lange DA, Kovler K, Bentur A, van Breugel K. Measurement of
water transport from saturated pumice aggregates to hardening cement paste.
Mater Struct 2006;39(9):861–8.
[24] Bentz DP, Lura P, Roberts JW. Mixture proportioning for internal curing. Concr
Int 2005;2:35–40.
[25] Lo T, Cui HZ. Spectrum analysis of the interfacial zone of lightweight aggregate
concrete. Mater Lett 2004;58:3089–95.
[26] Mannan MA, Ganapathy C. Concrete from an agricultural waste-oil palm shell
(OPS). Build Environ 2004;39:441–8.
[27] Moravia WG, Gumieri AG, Vasconcelos WL. Efficiency factor and modulus of
elasticity of lightweight concrete with expanded clay aggregate. IBRACON
Struct Mater 2010;3(2):195–204.
[28] Neville AM. Properties of concrete. 14th ed. Malaysia: CTP-VVP; 2008.
[29] Holm TA, Bremner TW. State of the art report on high strength, high durability
structural low-density concrete for applications in severe marine
environments. US Army Corps of Engineers, Engineering Research and
Development Center, ERDC/SL TR-00-3; 2000.
[30] Shafigh P. Mechanical properties of high strength lightweight concrete
containing oil palm shell. PhD thesis, University of Malaya, Malaysia; 2013.
[31] Bhanja S, Sengupta B. Influence of silica fume on the tensile strength of
concrete. Cem Concr Res 2005;35:743–7.
[32] ACI 363R-92. State-of-the-art report on high-strength concrete. ACI Committee
Report 363, American Concrete Institute, Detroit, 363R1-363R55; 1992.
[33] Committee Euro-International du Beton (CEB-PIP), CEB-PIP Model Code 1990,
Thomas Telford, London; 1993.
[34] Zain MFM, Mahmud HB, Ilham A, Faizal M. Prediction of splitting tensile
strength of high-performance concrete. Cem Concr Res 2002;32:1251–8.
[35] Shafigh 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):2127–34.
[36] Smadi M, Migdady E. Properties of high strength tuff lightweight aggregate
concrete. Cem Concr Compos 1991;13:129–35.
[37] Slate FO, Nilson AH, Martinez S. Mechanical properties of high strength
lightweight concrete. ACI J Proc 1986;83(7):606–13.
[38] Gesoglu M, Ozturan T, Guneyisi E. Shrinkage cracking of lightweight concrete
made with cold-bonded fly ash aggregates. Cem Concr Res 2004;34:1121–30.
[39] Babu DS, Babu KG, Wee TW. Properties of lightweight expanded polystyrene
aggregate concretes containing fly ash. Cem Concr Res 2005;35:1218–23.
[40] Oluokun FA. Prediction of concrete tensile strength from compressive
strength: evaluation of existing relations for normal weight concrete. ACI
Mater J 1991;88(3):302–9.
[41] Khan MI, Lynsdale CJ. Strength, permeability, and carbonation of high-
performance concrete. Cem Concr Res 2002;32:123–31.
[42] Lo TY, Cui HZ, Li ZG. Influence of aggregate pre-wetting and fly ash on
mechanical properties of lightweight concrete. Waste Manage 2004;24:333–8.
[43] Domagała L. Modification of properties of structural lightweight concrete with
steel fibres. J Civ Eng Manage 2011;17(1):36–44.
[44] Alengaram UJ, Jumaat MZ, Mahmud H. Influence of sand content and silica
fume on mechanical properties of palm kernel shell concrete. Int Conf
Construct Build Technol ICCBT 2008:251–62.
 P. Shafigh et al. / Materials and Design 60 (2014) 320–327
[45] CEB/FIP Manual of Design and Technology. Lightweight Aggregate Concrete.
First pub 1977, Great Britain.
[46] Short A, Kinnburgh W. Lightweight concrete. London: Applied Science
Publishers Ltd.; 1978.
[47] Zhang MH, Gjorv OE. Mechanical properties of high-strength lightweight
concrete. ACI Mater J 1991;88(3):240–7.
[48] ACI Committee 318-05. Building code requirements for structural concrete
(ACI 318-05) and commentary (ACI 318R-05). American Concrete Institute;
2005.
[49] Shetty MS. Concrete technology – theory and practice. New Delhi, India: S.
Chand and Company Limited; 2004.
[50] Alfes CH. Modulus of elasticity and drying shrinkage of high-strength concrete
containing silica fume, vol. 132. ACI Special Publication; 1992. p. 1651–72.
327
[51] Shafigh P, Mahmud H, Jumaat MZ. Oil palm shell lightweight concrete as a
ductile material. Mater Des 2012;36:650–4.
[52] Wongkeo W, Thongsanitgarn P, Chaipanich A. Compressive strength and
drying shrinkage of fly ash-bottom ash–silica fume multi-blended cement
mortars. Mater Des 2012;36:655–62.
[53] Basma AA, Jawad YA. Probability model for the drying shrinkage of concrete.
ACI Mater J 1995;92(3):246–51.
[54] Hooton RD, Stanish K, Angel JP, Prusinski J. The effect of ground granulated
blast furnace slag (slag cement) on the drying shrinkage of concrete – a critical
review of the literature, vol. 263. ACI Special Publication; 2009. p. 79–94.