Construction and Building Materials 38 (2013) 348–355

Contents lists available at SciVerse ScienceDirect

Construction and Building Materials

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

Effect of different sand grading on strength properties of cement grout

Siong Kang Lim a,1, Cher Siang Tan b,⇑, Kah Pin Chen a,2, Min Lee Lee a,3, Wah Peng Lee a,4
a
Faculty of Engineering & Science, Universiti Tunku Abdul Rahman, Malaysia
b
Faculty of Civil Engineering, Universiti Teknologi Malaysia, Malaysia

h i g h l i g h t s

" Three different refined sand grading were used in prepare of cement grout (CG).
" Finer sand filler contributes to higher mechanical strengths of CG.
" CG with finer sand is more durable when exposed to tropical air condition.
" Cubes’ compressive strength was lower compared to that of one-half prisms.

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

Article history: This paper aims to study the consistency and strength properties of cement grouts prepared with three
Received 22 May 2012 different sand grading namely 100% passing through 1.18 mm sieve (P1.18 mm), 0.90 mm sieve
Received in revised form 1 August 2012 (P0.90 mm) and 0.60 mm sieve (P0.60 mm), respectively. The measured flowing time indicated that
Accepted 14 August 2012
the specimens with the finer sand grading had lower flowability than those of the coarser sand grading.
Available online 29 September 2012
As the results, the finer sand grading specimens required a higher water to cement ratio to achieve an
equivalent workability. The specimens with the coarser sand grading obtained higher 7 and 28 days com-
Keywords:
pressive strengths than those of the finer sand grading when the lower water to cement ratio was
Cement grout
Sand grading
adopted (0.61–0.63). However, when the higher water to cement ratio (0.65–0.67) was used, the speci-
Strengths mens with the finer sand grading achieved higher long-term compressive, flexural and splitting tensile
Water to cement ratio strengths than those of the coarser sand grading.
Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction
The application of grout materials and grouting technique are
common for construction sector nowadays. Records abound, ce-
ment-based grout or self-compacting repair mortar have been
widely used since the 1800s and even earlier [1,2]. Grouting is a
process of fluids injection that set into fissures, cracks or voids
[3,4]. Nowadays, several types of grout materials have been used,
including cement, cement and sand, clay–cement, slag cement,
gypsum–cement, epoxy–polymer, clays–asphalt, pulverized fuel
ash and a large number of colloid and low viscosity chemicals
[5]. Cement grout with a high flowability is widely used in the con-
crete remedial works due to its flowable and self-compacting
⇑ Corresponding author. Tel.: +60 12 7587501.
E-mail addresses: sklim@utar.edu.my, siongkang@hotmail.com (S.K. Lim),
tcsiang@utm.my, chersiang@hotmail.com (C.S. Tan), chenkp@utar.edu.my
(K.P. Chen), mllee@utar.edu.my (M.L. Lee), leewp@utar.edu.my (W.P. Lee).
1
Tel.: +60 19 5226141.
2
Tel.: +60 12 4623060.
3 surface area of the fine aggregates. According to Haach et al. [9],
Tel.: +60 12 8808979.
4
Tel.: +60 16 2339393.
behaviors. It can easily flow into fine cracks and fissures attributed
to its fluidity. High fluidity of cement grout is a vital requirement of
high cohesion or segregation resistance during flow to form a
uniform and homogeneous mix. As the fluid cement grout can be
fully compacted without vibration, the application of the self-
compacting cement grout/mortar can therefore reduce labor and
machinery costs, improve compaction and hence enhance durabil-
ity of the critical cover zone of a structural member [6]. Relatively
few detailed studies have been reported on the influence of differ-
ent grading of sand filler on the properties of cement mixes espe-
cially self-compacting cement mortar. De Schutter and Poppe [7]
noticed that sand type has a significant effect on the cement mor-
tar properties. The authors observed that geometrical parameters
of sand based on the grading curve, like fineness modulus, relative
specific surface and apparent weight, can be correlated with the
water demand of the sand in the mortar, and may also influence
the hardened properties of mortars. Westerholm et al. [8] found
that the viscosity of mortar was influenced by the fines content
of fine aggregates, which may increase with the increased total
the cementitious-based mortars (cementitious to sand ratio = 1:3;

0950-0618/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.conbuildmat.2012.08.030
 S.K. Lim et al. / Construction and Building Materials 38 (2013) 348–355 349

cementitious = solely Portland cement or Portland cement + lime)

prepared with the finer sand filler (passing through 1.18 mm sieve)
required a higher w/c than that of the coarser sand filler (passing
through 4.75 mm sieve) to achieve an equivalent consistency. Red-
dy and Gupta [10] observed that the cementitious-based mortar
prepared with the finer sand filler required 25–30% more water
to achieve the desired consistency than that of the coarser sand fil-
ler. The compressive strength does not seem to be affected by the
sand grading. However, the sand grading could influence the
dimensional stability and modulus of elasticity in compression of
the cementitious-based mortars [9,10]. From the foregoing, there
are still limited studies on the effect of sand grading on the charac-
teristics especially long-term strengths of self-compacting cement
mortar (CG). Therefore, it is of necessity to further study the effect
of sand grading on the strength properties of CG for a concrete
remedial work. This paper studies the effectiveness of using three
different sand fillers namely P1.18 mm, P0.90 mm and P0.60 mm
Fig. 1. Particle size distribution curves for P1.18 mm, P0.90 mm and P 0.60 mm
on workability and strength properties of CG.
sand samples.

2. Experimental program
2.1. Materials
The production of cement grout (CG) in this study was carried out by using raw
materials namely ordinary Portland cement (OPC), oven dried river sand of different
grading, and clean tap water. The different batches of sand samples were subjected
to varying degrees of tropical natural weathering exposures. As a result, the sam-
ples may contain different initial moisture content. To standardize the preparation
procedures of the specimens, it was necessary to oven dry the sand samples at
105 °C for 24 h to remove the total moisture content. It is easier to control the w/
c ratios used in this study by using the oven-dried sand than the natural sand sam-
ples with inconsistent moisture content. Localized OPC manufactured by YTL Ce-
ment was used as a binder. It complies with the Type I Portland cement in
accordance with ASTM C150 [11]. Table 1 shows the chemical composition and
the physical properties of the YTL branded OPC. Fig. 1 shows the grading curves
of the oven dried refined river sand samples used in this study. Three categories
of refined sand grading were used, namely 100% passing through 1.18 mm sieve
(P1.18 mm), 0.90 mm sieve (P 0.90 mm) and 0.60 mm sieve (P0.60 mm), respec-
tively. The sand fillers used are classified as zone 4 fine sand with the fineness mod-
ulus in the range of 1.48–2.01 (refer to Table 2, the dominant sizes of the sand fillers
ranged from 0.30 to 0.60 mm, specific gravity = 2.60) in accordance with BS
882:1992 [12].
(P1.18 mm, P0.90 mm and P0.60 mm) were carried out in accordance with ASTM C
938 [13]. The laboratory trials aimed to obtain the optimum w/c of respective mix-
tures that corresponded to the optimum strength without compromising its flow-
ability and workability. The cement to sand ratio (c/s) was kept at unity. The trial
w/c ratios for the respective mixtures ranged from 0.61 to 0.67 with an interval
of 0.02. Table 2 (series 1) shows a summary of the trial mixes details.
2.3. Specimens preparation
The cubic, prismatic and cylindrical specimens were prepared by using cubic
mold with a size of 50  50  50 mm, prismatic mold with a size of
40  40  160 mm and cylindrical mold with a size of 100  200 mm, respectively.
The specimens were demolded after 24 h of casting and then subjected to two dif-
ferent curing regimes as described below:
(i) Water: specimens were submerged in water with temperatures in the
range of 25–28 °C until testing ages.
(ii) 7Water + Air: specimens were subjected to 7 days of initial water curing
with temperatures in the range of 25–28 °C, and then were further exposed
to air curing that in a sheltered outdoor environment with temperatures in
the range of 29–33 °C and relative humidity in the range of 55–65% until
testing ages.

2.2. Mix proportions 2.4. Testing methods

Firstly, the OPC and oven-dried sieved sand were blended thoroughly until a
uniform dry mix was obtained. Water was then added into the dry mix, and was
mixed thoroughly in a concrete mixer until a uniform fresh cement grout was ob-
tained. The whole process of mixing took about 5–10 min. The fresh cement grout
was used to cast the different types of specimens. Laboratory trials (series 1) for
three different types of CG mixtures prepared with the predetermined sand grading
Table 1
Chemical composition and physical properties of YTL branded OPC.
OPC
2.4.1. Flowability
The flowability of the fresh mixed cement grout (CG) was determined by using
flow cone method as described in ASTM C 939 [14]. A quick flowing time indicates a
high fluidity/workability of fresh CG.
2.4.2. Compressive strength
Compressive strength was determined by using a Universal compression test
machine with a constant loading rate of 1 kN/s in accordance with ASTM C 942
[15], which is equivalent to ASTM C 109/C [16]. The compressive strength was ob-
tained by applying an axial compressive load on the 50  50  50 mm cubic
specimen.

Chemical constituents
Silicon dioxide (SiO2) (%) 21.1 2.4.3. Flexural strength and jig-section compressive strength
Aluminium oxide (Al2O3) (%) 5.2 The prismatic specimens with a size of 40  40  160 mm (width 
Ferric oxide (Fe2O3) (%) 3.1 depth  length) were subjected to the center-point loading flexural test in accor-
Calcium oxide (CaO) (%) 64.4 dance with ASTM C 348 [17]. The test was conducted at a fixed rate of 0.1 mm/
Magnesium oxide (MgO) (%) 1.1 min by using Instron Testing Machine. The prismatic specimens were tested until
Sulfur oxide (SO3) (%) 2.5 they were broken into two halves and their flexural strength was computed. In
Sodium oxide (Na2O) (%) 0.2 addition, both portions from each broken prismatic specimen were used for com-
Potassium oxide (K2O) (%) 0.6 pressive strength testing in accordance with ASTM C 349 [18]. The one-half pris-
Titanium oxide (TiO2) (%) 0.2 matic specimen shall have a length of not less than 65 mm and free of crack,
Phosphorous oxide (P2O3) (%) <0.9 chipped surface or other obvious defects. The compressive strength of jig-section
Carbon content (C) (%) – of the one-half-prismatic specimens was computed by dividing maximum imposed
load with 1600 mm2 of testing area (jig-section area).
Physical properties
Loss on ignition (LOI) 2.4
Specific gravity 3.15 2.4.4. Splitting tensile strength
Fineness in blaine (cm2/g) 3170 Splitting tensile strength of cylindrical specimens with a size of 100  200 mm
Fineness (% passing 45 lm) 93.0 (diameter  height) was determined by using a Universal compression test ma-
chine with a constant loading rate of 0.5 kN/s in accordance with ASTM C 496 [19].
350 S.K. Lim et al. / Construction and Building Materials 38 (2013) 348–355

Table 2
A summary of experimental works for trial mixes (series 1).

Mix details Sand gradation c/s w/c Curing condition Investigated properties
LT-P1.18 – 100% Passing through Cement to sand ratio was 0.61–0.67, Increased by Totally immersed 1. Fresh properties (flowing
0.61–0.67 1.18 mm sieve; fineness kept constant at 1 for interval of 0.02 in water with time)
modulus = 2.01 comparison purpose temperature in
range of 25–28 °C
LT-P0.90 – 100% Passing through 0.61–0.67, Increased by 2. 7-Day and 28-day
0.61–0.67 0.90 mm sieve; fineness interval of 0.02 compressive strengths of
modulus = 1.85 50  50  50 mm cube
specimens (Average value of
three specimens for each mix
proportion)
LT-P0.60 – 100% Passing through 0.61–0.67, Increased by
0.61–0.67 0.60 mm sieve; fineness interval of 0.02
modulus = 1.48

Note: LT = laboratory trial.

All the tests for flowability, compressive strength, flexural strength, jig-section that when the w/c ratios of each category of mixture was increased
compressive strength, and splitting tensile strength were done in triplicate, but only
to the range of 0.65–0.67 (the higher workability), the CG prepared
the average values were reported in this study.
with the P0.60 sand filler obtained the highest 28-day compressive
strength compared to other equivalent mixtures prepared with
coarser sand filler. The CG prepared with the P1.18 sand filler ob-
3. Results and discussion
tained the highest 28-day compressive strength among all the
equivalent mixtures when the water to cement ratio was equal to
3.1. Laboratory trials (series 1)
0.61 (the lower workability). It is believed that the finer sand grad-
ing can contribute to a more consistent and uniform CG mixture
The optimal trial mixes (series 1) to be selected for further
than those of the cement grouts with coarser sand grading when
investigations (series 2) shall possess a normal concrete strength
the w/c ratio was high (0.67). When an even higher w/c ratio was
(around 25 MPa) without compromising its workability and flow-
used, the cement grout with the coarser sand grading has a higher
ability [5].
flowability than the finer sand grading as indicated by the flowing
Table 3 shows the effects of water to cement ratios on flowabil-
time shown in Table 3. The higher free water content in the mix
ity, and the results of 7 and 28 days compressive strengths. The
caused the higher flowability. This was attributed to the smaller sur-
flowing time in Table 3 clearly indicated that the workability and
face area of the coarser sand grading compared to that of the finer
flowability of the CGs were dependant on the water content in the
sand grading. Reddy and Gupta [10] provided the similar explana-
mixes. When the w/c of each category of trial mix was increased
tions on their results. The higher flowability may lead to a higher
from 0.61 to 0.67, the flowing time decreased. The results in the
bleedings, as well as a separation of light and heavy particles, and
Table 3 clearly show that the w/c of the selected mixes (in bold) in
consequently produced a non-uniform mix. Practically and accord-
each category increased when a finer grading of sand filler was used
ing to ASTM C 938 [13], the flowing time of cement grout shall be
in order to obtain equivalent workability and flowability. The re-
controlled within the range of 19–23 s to produce a consistent and
sults showed good agreements with the findings of previous studies
uniform mix. This is important to eliminate segregation and mini-
[7–10]. In addition, Table 3 shows that a higher w/c yielded a lower
mize bleeding phenomena, and eventually create more solid and
strength performance for each category of mixture. Reddy and
less porous microstructures of cement-based specimens [21].
Gupta [10] also observed a similar finding. They claimed that the
water–cement ratio increased with the increased fineness of sand
to maintain a 100% flow of the studied mortars. Besides, they also 3.2. Results of cement grout (CG) prepared with optimal w/c (series 2)
noted that an increase in water to cement ratio when a fine sand
was used would lead to a decrease in dry density of the mortars, Based on the initial laboratory trials results, the water to ce-
and hence caused a reduction in strength. In this study, it was noted ment ratio of 0.67 was selected for productions of the CGs, namely

Table 3
Screening of trial mixes results.

Mix notation Hardened density (kg/m3) Flow time (s) 7-Day compressive strengtha (MPa) 28-Day compressive strengtha (MPa)
LT-P1.18 – 0.61 1928 29 24.6 42.0
LT-P1.18 – 0.63 1910 25 20.2 35.4
LT-P1.18 – 0.65 1920 20 16.9 29.0
LT-P1.18 – 0.67 1881 18 17.6 24.2
LT-P0.90 – 0.61 1925 29 21.7 35.6
LT-P0.90 – 0.63 1920 25 18.5 29.8
LT-P0.90 – 0.65 1931 21 17.0 28.5
LT-P0.90 – 0.67 1904 19 18.8 24.2
LT-P0.60 – 0.61 1928 30 21.2 34.3
LT-P0.60 – 0.63 1931 26 17.0 31.8
LT-P0.60 – 0.65 1905 22 18.7 31.0
LT-P0.60 – 0.67 1905 20 22.6 27.5

Note: LT: laboratory trial; P1.18 : sand 100% passing 1.18 mm sieve; P0.90: sand 100% passing 0.90 mm sieve; P0.60: sand 100% passing 0.60 mm sieve; 0.61–0.67: w/c ratios.
a
Each compressive strength value is based on average of three readings.
 S.K. Lim et al. / Construction and Building Materials 38 (2013) 348–355
Table 4
A summary of experimental works for selected mixes (series 2).
Mix notation Sand grading
CG-1.18-W (ILT-P1.18 – 0.67) 100% passing 1.18 mm sieve size
CG-1.18-A (ILT-P1.18 – 0.67)
CG-0.90-W (ILT-P0.90 – 0.67) 100% passing 0.90 mm sieve size
CG-0.90-A (ILT-P0.90 – 0.67)
CG-0.60-W (ILT-P0.60 – 0.67) 100% passing 0.60 mm sieve size
CG-0.60-A (ILT-P0.60 – 0.67)
Note: c/s = cement to sand ratio, w/c = water to cement ratio, ILT = initial laboratory trial, CG = cement grout. W: Water = full immersion in water from day-1 until testing age;
A: 7Water + air = 7 days initial water curing and further air curing until testing age.
Fig. 2. Compressive strength development of CG cubic specimens under water
curing condition.
CG-P1.18, CG-P0.90 and CG-P0.60, respectively (Table 4). The fol-
lowing investigations focused predominantly on the strength prop-
erties of the respective CG specimens, i.e. compressive, flexural and
splitting tensile strengths. Two curing conditions were adopted as
described in the Section 2.3.
3.2.1. Compressive strength
Figs. 2 and 3 show that the compressive strength of CG cubic
specimens at all ages increased when finer grading of sand fillers
were used. The CG-P0.60 specimen obtained the highest compres-
sive strength at all testing ages among the three studied mixtures
under the two different curing conditions. This could be explained
by the more uniform and well graded P0.60 mm sand used in the
CG compared to the other two sand grading. It is believed that
the P0.60 mm sand filler had contributed to the internal voids
reduction between the sand particles, as well as tightened the
bonding between hydrated cement pastes and sand fillers. Neville
[21] stated that the volume of all voids in concrete would influence
its strength. When the volume of voids in a cement-based mix was
reduced, its strength will be increased. The results proved that the
CG specimens prepared with the finer sand filler could perform
better, in terms of compressive strength, at the higher w/c (0.65–
0.67) compared to the equivalent CG specimens prepared with
the coarser sand filler.
Fig. 3 shows that the 7water + air-cured specimens developed
their initial strength more rapidly than those of the equivalent
water-cured specimens up to 56 days of age. However, the com-
pressive strength development of the water-cured specimens had
overtaken the later strengths of the air-cured specimens after
56 days of age. In fact, the air-cured specimens faced a decrease
in compressive strength after the mentioned age, as shown in
Fig. 3. In addition, the other strength properties of the air-cured
specimens also exhibited the same decreasing trend, as shown in
Figs. 7, 9 and 12, respectively. This was probably caused by the ra-
pid rate of initial hydration of the air-cured specimens in view of
the tropical air weather has a higher temperature than that of
water. Eventually, this could produce porous hydrated cement
351
c/s w/c Curing condition Investigated properties
1 0.67 Water 1. 28, 56, 90 and 120 days compressive strengths.
7Water + air 2. 28, 56, 90 and 120 days flexural strengths.
0.67 Water 3. 28, 56, 90 and 120 days one-half prismatic specimens’
jig-section compressive strengths.
7Water + air 4. 28, 56, 90 and 120 days splitting tensile strengths.
0.67 Water
7Water + air
Fig. 3. Compressive strength development of CG cubic specimens under 7water + -
air curing condition.
paste structures. Based on the strength decreasing trends at later
ages as shown in the above-mentioned figures, it is questionable
that these poor calcium silicate hydrate formations can be dis-
solved by the dry and warm tropical climates, such as those in
Malaysia. Consequently, the compressive strength was reduced.
However, the real mechanism is still unclear, and merit further
investigation. Lim et al. [20] observed similar results. Neville [21]
stated that although a higher temperature promoted the develop-
ment of the early strength, it may adversely affect the later
strength. Neville explained that a rapid initial hydration tended
to form products of poor physical structures. A proportion of the
pores in the porous structures will always remain unfilled. In addi-
tion, Verbeck and Helmuth [22] claimed that the rapid initial rate
of hydration at higher temperatures in their study retarded the
subsequent hydration, and eventually produced a non-uniform dis-
tribution of the products of hydration within the paste. Kjellsen
et al. [23] confirmed the presence of porous C–S–H in between
the cement particles at higher temperatures in their study by using
backscattered electron imaging. Figs. 4 and 5 show the microstruc-
tures of the 120-day water-cured and air-cured CG-P0.60 speci-
mens, respectively. Although both the specimens consisted of a
small portion of ettringite and/or calcium sulfate (gypsum) forma-
tions, the water-cured specimen had a denser microstructure than
that of the air-cured specimen.
3.2.2. Flexural strength
Figs. 6 and 7 show the flexural strength development of the CG
prismatic specimens at different Ae3 ges under water and
7water + air curing conditions, respectively. The flexural strains
of all the prism specimens increased gradually with the increased
imposed load at their center-part. The bending of the specimens
was not obvious. However, it was observed that the cracks were
propagated at the center of bottom part of the specimens, and
eventually caused the specimens to be broken into two-halves at
the center-part. In conjunction with the compressive strength re-
sults of cubic specimens, the CG-P0.60 prismatic specimens ob-
tained the highest flexural strength at all testing ages among the
three studied mixtures under the two different curing conditions.
352 S.K. Lim et al. / Construction and Building Materials 38 (2013) 348–355
Fig. 4. Microstructure of air-cured CG-P0.60 specimen at 120 days of age (5000
magnification).
Fig. 5. Microstructure of water-cured CG-P0.60 specimen at 120 days of age (5000
magnification).
Fig. 6. Flexural strength development of CG prismatic specimens under water
curing condition.
These observations were opposing the findings of Haach et al. [9].
Haach et al. [9] observed that the cement mortars manufactured
with the coarser sand in their study, except for the mortar mix
1:1:6 (Portland cement:lime:sand), exhibited higher flexural
strengths than those of the fine sand. They claimed that coarse
sand probably promoted a better interlocking of particles due to
the large sizes of grains thus increased the flexural strength. In fact,
the strength properties of cementitious mortar are dependent on
several factors such as cementitious to sand ratios, water to
cementitious ratios, raw materials used, sand grading, and sand
particles’ shape [7–10]. In this research, the CG-P0.60 prismatic
specimens obtained the highest flexural strength. A plausible
Fig. 7. Flexural strength development of CG prismatic specimens under 7water + air
curing condition.
Fig. 8. Jig-section compressive strength development of CG one-half-prismatic
specimens under water curing condition.
Fig. 9. Jig-section compressive strength development of CG one-half-prismatic
specimens under 7water + air curing condition.
explanation to this observation is that the fine sand particles are
capable of forming a relatively stronger bonding with the hydrated
cement pastes than the coarse sand particles. It is believed that a
high cement to sand ratio (1:1) and an appropriate water to ce-
ment ratio could contributed to a strong bonding attributed to
the high production amount of the hydrated cement pastes.
3.2.3. Compressive strength of jig-section of one-half prism
Figs. 8 and 9 show the jig-section compressive strength of the
one-half prisms at different ages under the water and 7water + air
curing conditions, respectively. The results in Fig. 8 show reason-
ably good comparisons with the trend of compressive strength
development of cubic specimens. The water-cured CG-P0.60 one-
half-prismatic specimen possessed the highest compressive
strength at 120 days of age among the studied mixtures. In addi-
tion, Fig. 9 shows that the air-cured specimens faced a decrease
in compressive strength after 90 days of age. This could be ex-
plained by the retardants of cement hydration [21,22] and possible
degradation of the hydrated cement pastes caused by the inade-
quate internal moisture. The dry condition of the specimens could
 S.K. Lim et al. / Construction and Building Materials 38 (2013) 348–355 353

Fig. 10. Relationship between jig-section compressive strength of CG one-half-prismatic specimens and compressive strength of CG cubic specimens.

Fig. 11. Splitting tensile strength development of CG cylindrical specimens under

water curing condition. Fig. 12. Splitting tensile strength development of CG cylindrical specimens under
7water + air curing condition.

be caused by evaporation of the internal moisture. Haach et al. [9]

water-cured CG-P0.60 cylindrical specimens obtained the highest
stated that the specimens manufactured with a low w/c ratio (w/
splitting tensile strength at 120 days of age among the studied
c = 0.40) had insufficient water to promote an adequate hydration
specimens under the same curing conditions. All the water-cured
of cement particles. The results from the present study and those
specimens showed an increasing trend of splitting tensile strength
reported by Haach et al. [9] have both revealed that the dry condi-
towards the studied ages due to the proper curing environment
tion of specimens has retarded cement hydration. Nevertheless, it
that had been provided. On the contrary, the air-cured specimens
was observed that the decrease of compressive strength from 90
suffered a decrement of splitting tensile strength after 90 days of
to 120 days of period faced by the specimens prepared with
age as shown in Fig. 12. Fig. 13 shows the drying of specimens after
P1.18 mm sand filler was the highest among the three different
exposed to the tropical air condition (temperatures in the range of
types of mixtures. Fig. 10 shows the relationship between the com-
29–33 °C, and relative humidity in the range of 55–65%) for
pressive strength of cubic specimens and the jig-section compres-
113 days (120 days deduct 7 days of initial water curing). The dry-
sive strength of one-half-prismatic specimens. A linear fitting was
ing gradually degraded the strength after 90 days of age, as shown
used to the correlate the relationship. The results indicated that
in Fig. 11. It was observed that the CG-P0.60 cylindrical specimens
jig-section compressive strength of one-half-prismatic specimens
possessed the highest splitting tensile strength at 120 days of age
was higher than that of cubic specimens. The percentage of differ-
among the studied mixtures that exposed to air condition upon
ence was in the range of 10–30%. The restraint parts at two oppo-
the degradation of strength, as shown in Fig. 11. It is believed that
site sides of the jig-section increased the shear capacity of the
the CG specimens prepared with the finer sand fillers were able to
tested one-half prismatic specimens. Thus, a higher compressive
hold more internal moisture than those specimens prepared with
load was required to crush the one-half prismatic specimens than
the coarser sand fillers. Since the finer sand fillers have larger total
the cubic specimens. The equation and R-squared value in Fig. 10
surface area that required more water to wet the surface, the free
show that the jig-section compressive strengths of the CG-P0.60
water held by the fillers could be used to hydrate the remaining
specimen can be reasonably approximated (difference of about
non-hydrated cement particles. The late hydration could compen-
10%) to the cube compressive strengths. However, the approxima-
sate the strength loss caused by the tropical weathering, and hence
tions are relatively weaker for the specimens of CG-P1.18 and CG-
defer the strength degrading process when they were exposed to
P0.90.
tropical air condition.

3.2.4. Splitting tensile strength

Figs. 11 and 12 show the splitting tensile strength at different 3.2.5. Strengths comparison
ages of cylindrical specimens under the water and 7water + air cur- Table 5 shows the strengths comparison for the CGs of various
ing conditions, respectively. At a high fluidity (w/c = 0.67), the ages and curing conditions expressed as percentage to their
354 S.K. Lim et al. / Construction and Building Materials 38 (2013) 348–355

Fig. 13. Moisture content in cylindrical specimens at different ages subjected to two different curing conditions (top: water curing; bottom: air curing).

Table 5
Influence of sand grading on strength properties of different types cement grouts expressed as percentage of respective strength property of P1.18 cement grout (reference mix).

Age Mix notation Cubes’ compressive strength, %; Flexural strength, %; (MPa) Splitting tensile strength, %; Jig-section compressive strength, %;
(MPa) (MPa) (MPa)
Water curing Air curing Water curing Air curing Water curing Air curing Water curing Air curing
Strengths comparison of the studied CGs expressed as percentage of respective strength of P 1.18 CG (reference mix) at respective age
7 days CG-P1.18 100; (17.6) NA 100; (6.6) NA 100; (2.30) NA 100; (19.1) NA
CG-P0.90 107; (18.8) NA 89; (5.9) NA 78; (1.30) NA 138; (26.4) NA
CG-P0.60 128; (22.6) NA 97; (6.4) NA 102; (2.34) NA 119; (22.8) NA
28 days CG-P1.18 100; (24.2) 100; (25.2) 100; (7.7) 100; (7.3) 100; (3.26) 100; (3.16) 100; (25.0) 100; (31.2)
CG-P0.90 100; (24.2) 106; (26.6) 82; (6.3) 100; (7.3) 81; (2.64) 73; (2.32) 123; (30.7) 98; (30.6)
CG-P0.60 114; (27.5) 136; (34.2) 87; (6.7) 108; (7.9) 96; (3.14) 106; (3.35) 129; (32.2) 103; (32.0)
56 days CG-P1.18 100; (27.5) 100; (26.9) 100; (8.5) 100; (8.6) 100; (3.32) 100; (3.64) 100; (33.0) 100; (33.6)
CG-P0.90 108; (29.8) 112; (30.1) 76; (6.5) 90; (7.7) 93; (3.10) 86; (3.13) 101; (33.3) 96; (32.2)
CG-P0.60 111; (30.4) 121; (32.5) 104; (8.8) 100; (8.6) 98; (3.24) 98; (3.56) 106; (35.1) 103; (34.7)
90 days CG-P1.18 100; (27.7) 100; (26.9) 100; (8.6) 100; (7.3) 100; (3.34) 100; (3.44) 100; (38.9) 100; (36.6)
CG-P0.90 120; (33.2) 103; (27.7) 94; (8.1) 95; (6.9) 108; (3.59) 102; (3.52) 90; (34.9) 103; (37.6)
CG-P0.60 128; (35.4) 116; (31.2) 106; (9.1) 111; (8.1) 108; (3.62) 94; (3.22) 102; (39.6) 94; (34.4)
120 days CG-P1.18 100; (28.0) 100; (25.6) 100; (9.1) 100; (6.9) 100; (3.65) 100; (2.32) 100; (42.9) 100; (33.2)
CG-P0.90 123; (34.5) 98; (25.2) 102; (9.3) 94; (6.5) 99; (3.62) 105; (2.43) 84; (36.2) 106; (35.3)
CG-P0.60 131; (36.8) 117; (30.0) 103; (9.4) 106; (7.3) 105; (3.85) 111; (2.57) 103; (44.2) 101; (33.5)

The bold values are to highlight the reference values for the other mix within the same category.
Note: CG: Cement Grout; NA = not applicable (7 days initial water curing). P1.18: Sand 100% passing 1.18 mm sieve; P0.90: sand 100% passing 0.90 mm sieve; P0.60: sand
100% passing 0.60 mm sieve. aEach strength value is based on average readings of three laboratory test specimens; w/c ratio for all mixtures is 0.67.

corresponding P1.18 cement grout (reference mix) strength. The
water cement ratio for all the CG mixes was fixed at 0.67.
Through the strengths comparison, the influence of different sand
grading on strength properties of CGs could be clearly shown. De-
spite some inconsistencies, the overall strength results, especially
those of the 120-day of age, suggested that the P0.60 sand grad-
ing contributed to the optimum strengths of CG among the three
studied sand grading under both the water and air-curing
conditions.
4. Conclusions
Several conclusions can be drawn from the experimental
investigations:
(1) At the lower workability (w/c in the range of 0.61–0.63) and
a high cement to sand ratio (c/s = 1), the studied coarser
sand filler contributed to the higher 7-day and 28-day com-
pressive strengths of CGs than those of the studied finer sand
filler.
(2) At the higher workability (w/c in the range of 0.65–0.67) and
a high cement to sand ratio (c/s = 1), the studied finer sand
filler contributed to the higher compressive, flexural and
splitting tensile strengths of CGs than those of the studied
coarser sand filler.
(3) The sand with the finer grading has a better ability to hold
the internal moisture content of CGs. This could help in
deferring the strength degradation of the specimens when
they were exposed to the tropical air condition. In other
words, the finer grading of sand fillers is capable of produc-
ing a CG of higher durability than that of the coarse grading.
(4) Water curing condition assisted in developing CGs of more
solid and less porous microstructures under the tropical
weather. On the other hand, tropical air curing condition
retarded the long-term strengths development as well as
degraded the long-term strengths of the CG specimens
(5) The jig-section compressive strengths of the CG one-half-
prismatic specimens were higher than those of the CG cubic
specimens. The percentages of differences were in the range
of 10–30%.
 S.K. Lim et al. / Construction and Building Materials 38 (2013) 348–355 355

References [13] ASTM C938. Standard practice for proportioning grout mixtures for preplaced-
aggregate concrete. American Society of Testing and Materials; 2005.
[14] ASTM C939. Standard test method for flow of grout for preplaced-aggregate
[1] Bowen R. Grouting in engineering practice. 2nd ed. London: Apply Science Pub;
concrete (flow cone method). American Society of Testing and Materials; 2005.
1981.
[15] ASTM C 942. Standard test method for compressive strength of grouts for
[2] Bungey JH, Millard SG. Absorption and permeability tests. 3rd ed. In: Testing of
preplaced-aggregate concrete in the laboratory. American Society of Testing
concrete in structures. London: Chapman & Hall; 1996.
and Materials; 2005.
[3] Nonveiller E. Grouting theory and practice. Amsterdam: Elsevier; 1989.
[16] ASTM C 109/C. Test method for compressive strength of hydraulic cement
[4] Houlsby AC. Construction and design of cement grouting. In: A guide to
mortars (using 2 in or 50 mm cube specimens). American Society of Testing
grouting in rock foundations. New York: John Wiley and Sons; 1990.
and Materials; 2005.
[5] Hussin MW, Lim SK, Zakaria F. Engineering properties of high volume slag
[17] ASTM C 348. Standard test method for flexural strength of hydraulic cement
cement grout in tropical climate. Malaysian J Civil Eng 2007;19(1):42–54.
mortars. American Society of Testing and Materials; 2008.
[6] Bartos PJM. Fresh concrete: properties and tests. Amsterdam: Elsevier; 1992.
[18] ASTM C 349. Standard test method for compressive strength of hydraulic
[7] De Schutter G, Poppe AM. Quantification of the water demand of sand in
cement mortars (using portions of broken prisms in flexure). American Society
mortar. Constr Build Mater 2004;18(7):517–21.
of Testing and Materials; 2008.
[8] Westerholm M, Lagerblad B, Silfwerbrand J, Forssberg E. Influence of fine
[19] ASTM C 496. Standard test method for splitting tensile strength of cylindrical
aggregate characteristics on the rheological properties of mortars. Cement
concrete specimens. American Society of Testing and Materials; 1996.
Concr Compos 2008;30(4):274–82.
[20] Lim SK, Ling TC, Hussin MW. Strength properties of self-compacting mortar
[9] Haach VG, Vasconcelos G, Lourenco PB. Influence of aggregates grading and
mixed with GGBFS. Constr Mater 2012;165(2):87–98.
water/cement ratio in workability and hardened properties of mortars. Constr
[21] Neville AM. Properties of concrete. 4th ed. Harlow, England: Pearson Prentice
Build Mater 2011;25:2980–7.
Hall; 2009.
[10] Venkatarama Reddy BV, Gupta A. Influence of sand grading on the
[22] Verbeck GJ, Helmuth RA. Structures and physical properties of cement paste,
characteristics of mortars and soil-cement block masonry. Constr Build
vol. 3. In: Proceeding of 5th international symposium on the chemistry of
Mater 2008;22:1614–23.
cement, Tokyo; 1968. p. 1–32.
[11] ASTM C150. Standard specification for Portland cement. American Society of
[23] Kjellsen KO, Detwiler RJ, Gjorv OE. Backscattered electron imaging of cement
Testing and Materials; 2007.
pastes hydrated at different temperatures. Cem Concr Res 1990;20(2):308–11.
[12] BS 882. Specification for aggregates from natural sources for concrete. British
Standard Institute; 1992.