International Journal of Minerals, Metallurgy and Materials

V olume 20 , Number 2 , F ebruary 2013 , P age 205

DOI: 10.1007/s12613-013-0714-7

Effect of silica fume on the fresh and hardened properties of fly

ash-based self-compacting geopolymer concrete

Fareed Ahmed Memon1,2) , Muhd Fadhil Nuruddin1) , and Nasir Shafiq1)

1) Department of Civil Engineering, Faculty of Engineering, Universiti Teknologi PETRONAS, Tronoh 31750, Malaysia
2) Department of Civil Engineering, Faculty of Engineering, Mehran University of Engineering and Technology,
Jamshoro 76062, Pakistan
(Received: 2 January 2012; revised: 2 March 2012; accepted: 9 March 2012)

Abstract: The effect of silica fume on the fresh and hardened properties of fly ash-based self-compacting geopolymer
concrete (SCGC) was investigated in this paper. The work focused on the concrete mixes with a fixed water-to-geopolymer
solid (W/Gs) ratio of 0.33 by mass and a constant total binder content of 400 kg/m3 . The mass fractions of silica fume
that replaced fly ash in this research were 0wt%, 5wt%, 10wt%, and 15wt%. The workability-related fresh properties of
SCGC were assessed through slump flow, V-funnel, and L-box test methods. Hardened concrete tests were limited to
compressive, splitting tensile and flexural strengths, all of which were measured at the age of 1, 7, and 28 d after 48-h
oven curing. The results indicate that the addition of silica fume as a partial replacement of fly ash results in the loss
of workability; nevertheless, the mechanical properties of hardened SCGC are significantly improved by incorporating
silica fume, especially up to 10wt%. Applying this percentage of silica fume results in 4.3% reduction in the slump flow;
however, it increases the compressive strength by 6.9%, tensile strength by 12.8% and flexural strength by 11.5%.

Keywords: geopolymer concrete; silica fume; fresh properties; compressive strength; tensile strength; flexural strength

1. Introduction properties to those from Portland cement [1].

Geopolymeric materials have become the focus of in-
The use of cement as a binder in a concrete mixture terest and received the considerable attention because of
is often criticized by circles concerned with environmen- the environmental benefits, such as the reduction in con-
tal conservation. With the increased use of cement in sumption of natural resources and the decrease in pro-
concrete, there have been environmental concerns both in duction of CO2 . Unlike ordinary Portland cement, the
terms of damage caused by the extraction of raw materi- production of raw materials for geopolymers does not re-
als and the emission of carbon dioxide (CO2 ) during ce- quire a high level of energy consumption because the high-
ment manufacture. This has brought pressures to reduce temperature calcining is not required. It is demonstrated
the cement consumption in the construction industry. An that the geopolymeric cement generates 5-6 times less
attempt in this regard is the development of geopolymer CO2 than Portland cement [2-3]. Therefore, the use of
concrete [1]. The development of geopolymer concrete is geopolymer concrete technology not only significantly re-
an important step because of the potential application of duces CO2 emissions but also utilizes the industrial waste
geopolymers from a wide variety of industrial waste ma- and/or byproduct, converting a potentially hazardous ma-
terials to produce the added-value construction materials, terial to a valuable construction material [4].
resulting in the low-cost and environment-friendly material Increasing the productivity and improving the work-
with the similar mechanical performance and appearance ing environment have the high priority in the development

Corresponding author: Fareed Ahmed Memon E-mail: engrfam@gmail.com


c University of Science and Technology Beijing and Springer-Verlag Berlin Heidelberg 2013
206 Int. J. Miner. Metall. Mater., V ol. 20 , No. 2 , F eb. 2013

of concrete construction over the last two decades. Self- of concrete that can achieve the combined advantages of
compacting concrete (SCC) is another development in con- both geopolymer concrete and SCC [18-19]. Literature
crete technology and is gaining popularity nowadays. SCC, review indicated that, up to date, no research has been
which is different from the conventional concrete especially conducted on SCGC. This research study was therefore
in its fresh state, can flow into places and around obstruc- intended to explore the feasibility and potential of SCGC
tions by its own weight to fill the formwork completely and made with locally available constituent materials by ex-
self-compact without any segregation and blocking [5]. amining its basic physical and mechanical properties. The
SCC offers several advantages in technical, economic, present work investigated the selected properties of fly ash-
and environmental terms over conventional concrete, such based SCGC containing the different percentages of silica
as the improvement of concrete quality, reduction in con- fume, such as workability, compressive strength, splitting
struction time, easier placement in congested reinforcing tensile strength, and flexural strength. The results indi-
bars, homogenization and completion of consolidation, in- cate the significant improvement in strength properties of
crease in bond strength, reduction in noise levels due to the SCGC, and suggest that silica fume can beneficially be
absence of vibration, lower overall costs, and safe working used in the production of SCGC.
environment [6-8]. This matches the reduced availability
of skilled supervision and the reduced tolerance for visual 2. Experimental
defects, excessive noise, and industrial injuries. 2.1. Materials
The constituent materials used for SCC are the same The materials used in this study were fly ash, silica
as those for conventional concrete, except that SCC con- fume, fine and coarse aggregates, alkaline solution, super-
sists of the different proportions and the additional chem- plasticizer, and water.
ical and mineral mixtures. SCC mixes generally have a
2.1.1. Fly ash
higher proportion of ultra-fine materials. SCC has the
Geopolymer concrete is produced by the activating
higher flowability due to its lower coarse aggregate con-
alumino-silicate-based source material with an alkaline so-
tent, lower water-binder ratio, and the use of superplasti-
lution. Fly ash contains high contents of silica and alumina
cizer and viscosity modifying agents [9-10]. Supplementary
and hence is considered as a suitable source material for
cementitious materials and mineral fillers are commonly
making geopolymer concrete. In the present study, low-
utilized to reduce cost, improve workability, and improve
calcium (ASTM C618 Class F) [20] fly ash was used as a
hardened properties of concrete [11-12]. The use of sup-
source material for the synthesis of SCGC. Fly ash was
plementary cementitious materials, such as fly ash, ground
obtained from Manjung Power Station, Perak, Malaysia,
granulated blast-furnace slag, and silica fume, is well estab-
which complied with BS EN 450-1 [21]. Specific gravity
lished because of the improvement in concrete properties
of fly ash was found to be 2.38, and it had the Brunauer-
and also for the environmental and economical reasons [5].
Emmett-Teller (BET) surface area of 4260 m2 /kg, mea-
Silica fume, one of the most popular pozzolanas, is a
sured by the nitrogen-adsorption method. The chemical
byproduct of the silicon smelting process, which is used in
composition of fly ash determined by X-ray fluorescence
concrete to get improved properties. The use of silica fume
(XRF) analysis is presented in Table 1.
in concrete is attributed to its positive effect on the me-
chanical properties of concrete [13]. Silica fume increases
Table 1. Chemical composition of fly ash and silica
the compressive strength, modulus of elasticity, flexural
fume wt%
strength, and durability of hardened concrete [11]. Gener-
ally, silica fume is used in amounts between 5% and 10% Oxide Fly ash Silica fume
Silicon dioxide (SiO2 ) 51.30 90.20
by mass of the total cementitious material [14]. It is used
Aluminum oxide (Al2 O3 ) 30.10 0.82
either as a cement replacement material to reduce the ce- Ferric oxide (Fe2 O3 ) 4.57 1.67
ment content or as an additive material to enhance the Calcium oxide (CaO) 8.73 1.24
performance of concrete [15]. A considerable body of lit- Phosphorus pentoxide (P2 O5 ) 1.60 —
erature is available, which documents the benefits of silica Sulfur trioxide (SO3 ) 1.40 1.40
fume both as a pozzolan and as a filler material [16]. Potassium oxide (K2 O) 1.56 4.02
Titanium dioxide (TiO2 ) 0.70 —
Silica fume, because of its physical properties and
chemical characteristics, has been reported to be more re-
active than fly ash [17]; hence, in this research, it was used 2.1.2. Silica fume
in combination with fly ash to evaluate its effect on the The commercially available silica fume obtained from
fresh and hardened properties of self-compacting geopoly- Elkem materials in dry densified state with Grade 920 was
mer concrete (SCGC). SCGC is relatively a new concept used in this study, complying with the mandatory require-
and can be regarded as the revolutionary development in ments of BS EN 13263-1:2005+A1:2009 [22]. The chemical
the field of concrete technology. It is an innovative type composition of silica fume is given in Table 1. It had the
F.A. Memon et al., Effect of silica fume on the fresh and hardened properties of fly ash-based ... 207

specific gravity of 2.22 and the BET surface area of 17950 The physical properties and sieve analysis results of fine
m2 /kg. aggregate are shown in Tables 2 and 3, respectively.

2.1.3. Fine aggregate Table 2. Physical properties of fine and coarse ag-
gregate
Natural Malaysian sand with the maximum size of Property Fine aggregate Coarse aggregate
4.75 mm was used as fine aggregate. Fine aggregate had Type Uncrushed Crushed
Maximum size / mm 4.75 14
a specific gravity of 2.61 and water absorption of 1.18%.
Specific gravity 2.61 2.66
The sieve analysis of fine aggregate was performed in ac- Fineness modulus 2.76 —
cordance with ASTM C136-01 [23] and compared with the Water absorption / % 1.18 0.96
requirement of BS 882:1992 [24] and ASTM C33-03 [25]. Unit weight / (kg·m–3 ) 1658 1578

Table 3. Sieve analysis of fine aggregate

Sieve size / Retained Retained mass Cumulative retained Passing mass BS 882:1992 ASTM C33-03
mm mass / g fraction / wt% mass fraction / wt% fraction / wt% range / wt% range / wt%
5.00 0 0 0 100 89-100 95-100
2.36 157.13 15.85 15.85 84.15 60-100 80-100
1.18 172.79 17.43 33.28 66.72 30-100 50-85
600 141.46 14.27 47.55 52.45 15-100 25-60
300 325.56 32.84 80.39 19.61 5-70 5-30
150 183.89 18.55 98.94 1.06 0-15 0-10
<150 10.51 1.06 100 — — —

2.1.4. Coarse aggregate condition. Coarse aggregate had a specific gravity of 2.66
and water absorption of 0.96%. The sieve analysis of coarse
A well-graded locally available crushed stone was used aggregate was performed in accordance with ASTM C136-
as coarse aggregate in this study. The maximum size of 01 [23] and compared with the requirement of BS 882:1992
coarse aggregate was selected as 14 mm to avoid any block- [24] and ASTM C33-03 [25]. The physical properties and
ing of SCC. To remove fine silt particles, the coarse aggre- sieve analysis results of coarse aggregate are shown in Ta-
gate was washed and used in saturated surface dry (SSD) bles 2 and 4, respectively.

Table 4. Sieve analysis of coarse aggregate

Sieve size / Retained Retained mass Cumulative retained Passing mass BS 882:1992 ASTM C33-03
mm mass / g fraction / wt% mass fraction / wt% fraction / wt% range / wt% range / wt%
20.0 0 0 0 100 100 90-100
14.0 116.81 5.88 5.88 94.12 90-100 —
10.0 798.55 40.20 46.08 53.92 50-85 20-55
5.00 995.40 50.11 96.19 3.81 0-10 0-10
2.36 20.47 1.03 97.22 2.78 — 0-5
<2.36 55.20 2.78 100 — — —

2.1.5. Alkaline solution 2.1.6. Superplasticizer

In geopolymerization, alkaline solution plays an im- For achieving higher workability and desired flowabil-
portant role. A combination of sodium silicate or potas- ity of fresh concrete, a commercially available superplas-
sium silicate and sodium hydroxide or potassium hydroxide ticizer, named as Sika Viscocrete-3430 supplied by Sika
has been widely used as an alkaline activator. In this study, Kimia Sdn Bhd, Malaysia, and the extra water, different
a combination of sodium silicate and sodium hydroxide was from that used for the preparation of sodium hydroxide so-
used as an alkaline solution. lution, were used. The ordinary drinking water was used
Sodium hydroxide in pellets (99% purity), supplied by available in the concrete laboratory for this purpose.
Quick Lab Sdn Bhd, Malaysia, and sodium silicate solution 2.2. Mix proportions
(Grade A53 with Na2 O = 14.26wt%, SiO2 = 29.43wt%,
and water = 56.31wt%), obtained from Malay-Sino Chem- For this study, four mixtures, one control mix without
ical Industries Sdn Bhd, Malaysia, were used. To prepare silica fume and three other mixes with different propor-
sodium hydroxide solution, sodium hydroxide pellets were tions of silica fume, were prepared. Fly ash was replaced
dissolved in tap water. Both liquid solutions were mixed with silica fume at the contents of 5%, 10%, and 15% by
together and an alkaline solution was prepared. mass. The water to geopolymer solids (W/Gs) ratio by
208 Int. J. Miner. Metall. Mater., V ol. 20 , No. 2 , F eb. 2013

mass for all the mixes was maintained at 0.33 and the for the binder were used. The mixture designation and the
total powder content was fixed at 400 kg/m3 . To obtain quantities of various materials for each designed concrete
the requested workability characteristics of SCGC, a water mix are given in Table 5.
content of 12% and superplasticizer dosage of 6% by mass

Table 5. Mix proportions

Silica Aggregate / Sodium Sodium SP mass Extra water Oven curing

Mix ID
Fly ash /
fume / (kg·m−3 ) silicate / hydroxide W / Gs
fraction / mass frac-
(kg·m−3 ) ratio Time / Temper-
(kg·m−3 ) Fine Coarse −3 3
(kg·m ) kg/m mol/L wt% tion / wt%
h ature / ◦ C
M1 (control mix) 400 0 850 950 143 57 10 0.33 6 12 48 70
M2 (5wt% SF) 380 20 850 950 143 57 10 0.33 6 12 48 70
M3 (10wt% SF) 360 40 850 950 143 57 10 0.33 6 12 48 70
M4 (15wt% SF) 340 60 850 950 143 57 10 0.33 6 12 48 70
Note: SF stands for the silica fume, W/Gs the ratio of water-to-geopolymer solid by mass, and SP the superplasticizer.

2.3. Preparation, casting, and curing of ist Construction Chemicals and Concrete Systems (EF-
specimens NARC) [26] guidelines.
3.1.1. Slump flow test
For the production of fresh SCGC, fine powdered ma-
Slump flow test is the most commonly used and gives
terials (i.e., fly ash, silica fume, and fine aggregate) were
a good assessment of filling ability. This test is used to
firstly placed in a 100-L pan mixer and blended manu-
evaluate the horizontal free flow (deformability) of con-
ally. Afterwards, the coarse aggregate in SSD condition
crete in the absence of obstructions. To perform the test,
was added to the mixer and mixed mechanically for about
a slump cone is placed on a rigid and non-absorbent lev-
2.5 min. At the end of this dry mixing, a well-shacked pre-
eled plate and filled with concrete without tamping. After
mixed liquid mixture, containing alkaline solution, super-
filling the slump cone, it is raised vertically and concrete
plasticizer, and extra water, was added in the mixer. This
is allowed to flow out freely. The diameter of concrete in
duration was not less than 3 min. The freshly prepared
two perpendicular directions is measured, and the average
concrete mix was then assessed for the essential workabil-
of the two measured diameters is recorded. The higher the
ity tests required for characterizing SCC. Slump flow, V-
slump flow, the better the filling ability of concrete to fill
funnel, and L-box tests were performed for this purpose.
formwork is. SCC is assumed to have a good filling ability
After testing SCGC in fresh state, the test of hardened and consistency if the diameter of spread is in the range of
properties for specimens was prepared. The fresh concrete 650-800 mm [7].
was thoroughly mixed by hand and poured into moulds of During the slump flow test, the time from lifting the
cubes, cylinders, and prisms without compaction, filling all
cone until the flow spread up to a 500 mm diameter circle,
the spaces of moulds by its own weight. The top surface is also recorded. This flow time is termed as T50cm slump
of specimens was scraped to remove the excess material
flow. A lower time indicates a greater flow ability.
and achieve the smooth finish. From each mix, nine cu-
3.1.2. V-funnel test
bical, nine cylindrical, and six prism specimens were cast.
This test is primarily used to measure the filling ability
Cubes were used for compressive strength test, cylinders
(flowability) of SCC and can also be used to evaluate the
were used for splitting tensile strength, and prism spec-
ability of concrete to flow through the restricted spacing
imens were used for the test of rupture modulus. After
without segregation and blocking. To perform this test,
casting, without any delay, the specimens along with steel
the funnel is completely filled with concrete without com-
moulds were placed in the oven at a temperature of 70◦ C
paction, and the time taken for the concrete to flow out
for 48 h. At the end of this oven curing period, test spec-
completely through the orifice is recorded as V-funnel flow
imens were removed from the moulds and placed in room
time. The funnel flow time between 6 and 12 s is generally
temperature conditions for the air curing until the test day.
desired for SCC [7].
3. Specimen test 3.1.3. L-box test
L-box test characterizes the filling and passing abil-
3.1. Fresh properties
ity of SCC. L-box test is more sensitive to blocking. To
Fresh properties of SCGC mixes were evaluated based perform this test, the vertical section of box is filled with
on three key characteristics of SCC: filling ability, pass- concrete, and the concrete is allowed to flow through the
ing ability, and resistance to segregation. These charac- closely spaced reinforcing bars at the bottom into the hori-
teristics were measured using slump flow, V-funnel, and zontal section of box. When the concrete stops flowing, the
L-box tests following the European Federation of Special- heights of concrete at the end of horizontal section (H2 )
F.A. Memon et al., Effect of silica fume on the fresh and hardened properties of fly ash-based ... 209

and in the vertical section (H1 ) are measured to compute been found to be somewhat 15% higher than those deter-
the blocking ratio (H2 /H1 ). mined by third point loading (two-point loading). Center-
3.2. Hardened properties point loading also gives more variable values; as a result,
this test method is very rarely used.
In the hardened state, compressive, splitting tensile,
In this study, flexural strength test was performed on
and flexural strengths of various SCGC mixes were inves-
the prism specimens (100 mm×100 mm×500 mm) in ac-
tigated. All of these tests were performed at the ages of 1,
cordance with BS EN 12390-5 [31] and ASTM C78-02 [32].
7, and 28 d after oven curing for 48 h.
The specimens were placed on the support blocks of the
3.2.1. Compressive strength testing machine, and the load was applied constantly with-
Compressive strength is one of the most common mea- out any shock until the specimen failed.
sures used to evaluate the quality of hardened concrete and
is considered as the characteristic material value for the 4. Results and discussion
classification of concrete. Compressive strength test for all 4.1. Fresh properties
the variants of concrete mixes with different silica fume
The results of fresh properties of various SCGC mixes
contents was carried out on three cubical specimens (100
with the different contents of silica fume are presented in
mm×100 mm×100 mm) in accordance with BS EN 12390-
Table 6 and Figs. 1-4. As reported in literatures, because
3 [27] using a 2000-kN Digital Compressive and Flexural
of the higher surface area and extremely fine particle size,
Testing Machine. During the test, the compressive load
silica fume increased the water requirement of concrete.
was applied at the rate of 3.0 kN/s until the specimen
As a result, the workability of fresh concrete decreased.
failed. The reported compressive strength was the average
This hypothesis became true in this investigation. The
strength of three specimens.
addition of silica fume as a partial replacement of fly ash
3.2.2. Tensile strength
in SCGC resulted in the loss of workability. This might
Tensile strength of concrete, although is not as im- be explained by the increased surface area of silica fume
portant as compressive strength and not considered in the particles. Generally, the mixtures containing silica fume
structural design commonly [15], however, the determina- exhibited worse performance than the control mix in re-
tion of concrete tensile strength is necessary to control the gards to the fresh properties. Silica fume, because of its
cracks and improve the ductility [28]. higher surface area than fly ash, absorbed the excessive
Tensile strength of various SCGC mixes was measured water in the geopolymer system [33]. It was observed that
by cylinder splitting test on 100 mm×200 mm concrete the concrete mixes containing higher percentages of silica
cylinders, following the procedure prescribed by BS EN fume were more cohesive and appeared to be sticky; fluid-
12390-6 [29] and ASTM C496/C 496M-04 [30]. The cylin- ity and flowability of various SCGC mixtures were reduced
der was loaded uniformly on two diametrically opposite when the proportion of silica fume increased. These effects
lines at a rate of 0.94 kN/s without any sudden shock loads. of silica fume on fresh properties of SCGC were somewhat
3.2.3. Flexural strength similar to results reported by Andri [33], which indicated
Flexural strength, also known as the modulus of rup- that the workability of freshly prepared low-calcium fly
ture, is a measure of a plain concrete beam to resist the fail- ash-based geopolymer concrete was decreased with the in-
ure in bending. Flexural strength is determined either by crease in the quantity of silica fume. The effect of silica
third point loading or center-point loading test methods. fume on the individual fresh property test was discussed
Flexural strength determined by center-point loading has in the subsequent paragraphs.

Table 6. Fresh properties of various SCGC mixes with the different contents of silica fume
Slump flow
Mix ID V-funnel flow time / s L-box blocking ratio (H2 /H1 )
Diameter / mm T50cm / s
EFNARC range 650-800 2-5 6-12 0.8-1.0
M1 (Control mix) 695 3.5 9.0 0.96
M2 (5wt% SF) 680 4.0 9.5 0.96
M3 (10wt% SF) 665 4.5 11.0 0.94
M4 (15wt% SF) 640 6.5 15.0 0.90

4.1.1. Slump flow test of 650-800 mm. A maximum slump flow value of 695 mm
The results of slump flow test are shown in Fig. 1. is achieved for the control mix M1 (0wt% SF). With the
Test results indicate that, except for Mix M4 (15wt% SF), increase in the quantity of silica fume, the flow of various
slump flow for all the mixes is within the EFNARC range SCGC mixes decrease. This may be attributed to the very
210 Int. J. Miner. Metall. Mater., V ol. 20 , No. 2 , F eb. 2013

fine particle size and the increased surface area of silica

fume.

Fig. 3. Effect of SF content on V-funnel flow time.

4.1.4. L-box test

Fig. 1. Effect of SF content on slump flow. The blocking ratio (H2 /H1 ) of various SCGC mixes is
shown in Fig. 4. The blocking ratio should be between 0.8
4.1.2. T50cm slump flow and 1.0 [7]. While assessing the fresh concrete for passing
The slump flow time for concrete to reach 50 cm diam- ability, it is observed that all the four mixes pass through
eter (T50cm slump flow) was also recorded during the test. the bars of L-box very easily and no blockage is seen in
Fig. 2 illustrates the results of T50cm slump flow. It can any of the mixes. The results of L-box test show that,
be seen that, except for Mix M4 (15wt% SF), all the other although the blocking ratio (H2 /H1 ) gradually decreases
three mixes qualify the permissible limits (2-5 s) given by with the increase in the quantity of silica fume, the ratio
EFNARC [7]. T50cm slump flow varies between 3.5 and (H2 /H1 ) for all the mixes is above 0.8, which is as per
6.5 s. The lowest slump flow time of 3.5 s is recorded for EFNARC standards.
the control mix containing 0wt% silica fume. An increase
in the quantity of silica fume leads to the increase in T50cm
time. This may be due to the increasing paste volume with
silica fume.

Fig. 4. Effect of SF content on L-box blocking ratio

(H2 /H1 ).

4.2. Hardened properties

Fig. 2. Effect of SF content on T 50cm slump flow.
4.2.1. Compressive strength
4.1.3. V-funnel test Silica fume is generally known to improve the mechani-
To assess the flowability and stability of freshly pre- cal properties of concrete [34]. The inclusion of silica fume
pared concrete, all the four mixes with different contents in concrete causes significant changes in the structure of
of silica fume were tested by V-funnel test. The V-funnel the matrix by both a physical action and a pozzolanic reac-
flow time values obtained from the test are shown in Fig. tion. The silica fume particles with small size and spherical
3. According to the results of test, the V-funnel flow time shape can fill the voids created by free water in the ma-
varies between 9 and 15 s. It can be seen that, except for trix. This particle packing effect refines the microstructure
Mix M4 (15wt% SF), all the other three concrete mixtures of concrete, creates a much denser pore structure, and re-
meet the requirements of allowable flow time. The mini- sults in an increase in the mechanical properties of concrete
mum flow time of 9 s is recorded for the control mix con- [35]. The inclusion of silica fume as a partial replacement
taining 0wt% silica fume. With the increase in silica fume of fly ash in SCGC was proved to be very effective in im-
content, the fluidity of concrete decreases. Consequently, proving the hardened properties of concrete. Table 7 and
the V-funnel flow time increases. The same reasons and Fig. 5 illustrate the compressive strength test results of
mechanism mentioned for slump flow test results are also various SCGC mixes with and without silica fume. From
commanding the explanations for the results of V-funnel Fig. 5, it can be seen that the compressive strengths of
test. fly ash-based SCGC with 5wt%, 10wt%, and 15wt% silica
F.A. Memon et al., Effect of silica fume on the fresh and hardened properties of fly ash-based ... 211

fume replacement are higher than Mix M1 (0wt% SF) at was realized earlier in the fresh concrete tests, could have
every age of testing. At 28 d, Mix M1 (0wt% SF) achieves a probably led to a decrease in concrete strength. Similar
compressive strength of 51.43 MPa, whereas mixtures M2 observations were also made by Dutta et al. [37] who re-
(5wt% SF), M3 (10wt% SF), and M4 (15wt% SF) achieve ported that the mortar specimens incorporated with sil-
the compressive strength of 53.38, 55.02, and 53.96 MPa, ica fume showed a better microstructure and exhibited a
respectively, increasing by 3.8%, 6.9%, and 4.9% in com- lesser porosity. They have demonstrated that the addition
parison with the strength of the control mix. Test results of silica fume up to 5% by mass enhanced the compressive
indicate that, until the 10wt% of silica fume, the higher the strength of fly ash-based geopolymer mortars. However,
percentage of silica fume, the higher the values of compres- the further increase of silica fume caused a decrease in
sive strength; after that, the increase in the percentage of compressive strength.
silica fume leads to the decrease in compressive strength. 4.2.2. Splitting tensile strength
The increase in compressive strength with the inclusion of The relationships between tensile, flexural, and com-
silica fume may probably be due to the fact that the sil- pressive strengths in fly ash-based SCGC containing silica
ica fume is finer than fly ash, which results in the dense fume were found to be almost similar as in case of ordinary
particle packing, pore size refinement, and denser concrete Portland cement concrete. An increase in the compressive
matrix. When the fine silica fume is added, it offers the ac- strength generally resulted in a similar relative increase
tive SiO2 , which is advantageous to form the siloxo bridges in the tensile and flexural strengths. The splitting tensile
(–Si–O–Si–O–) during the geopolymerization processing. strength test results of various SCGC mixes with and with-
These bridge chains bond the particles firmly, then a much out silica fume are given in Table 8 and Fig. 6. From Fig.
denser and more compact matrix structure is formed; con- 6, there is an increase in splitting tensile strength with the
sequently, both compressive and bending strengths are en- increase in silica fume contents; however, the maximum
hanced [36]. Andri [33] also reported that the dissolution of strength at all ages occurs at 10wt% fly ash replacement.
silicate monomer and small oligomers from silica fume par- With increasing the amount of silica fume content over
ticles altered the composition of geopolymer framework. 10wt%, the splitting tensile strength decreases. The trend
The additional dissolution and polycondensation process of in splitting tensile strength with silica fume content is sim-
aluminate precursors from the fly ash particles with silicate ilar to that in the case of compressive strength. At 28 d,
monomer and oligomer supplied by silica fume particles re- the splitting-tensile strength of M1 (0wt% SF) achieves
sulted in an improved geopolymer matrix with denser gel 4.14 MPa, whereas mixtures M2 (5wt% SF), M3 (10wt%
structure, which led to the higher bending and compressive SF), and M4 (15wt% SF) achieve the strength of 4.31, 4.67,
strengths. and 4.24 MPa, respectively, increasing by 4.1%, 12.8%, and
2.4% in comparison with the strength of the control mix.
Table 7. Compressive strength test results
Compressive strength / MPa Table 8. Splitting tensile strength test results
Mix ID
1d 7d 28 d
Splitting tensile strength / MPa
M1 (Control mix) 46.98 49.25 51.43 Mix ID
1d 7d 28 d
M2 (5wt% SF) 48.22 51.62 53.38
M1 (Control mix) 3.84 4.07 4.14
M3 (10wt% SF) 50.25 53.76 55.02 M2 (5wt% SF) 3.92 4.18 4.31
M4 (15wt% SF) 48.84 51.55 53.96 M3 (10wt% SF) 4.12 4.40 4.67
M4 (15wt% SF) 3.89 4.09 4.24

Fig. 5. Effect of SF on compressive strength.

Fig. 6. Effect of SF on splitting tensile strength.

When the addition of silica fume passes over 10% by 4.2.3. Flexural strength
mass of fly ash, the instability of concrete samples, which The flexural strength test results for SCGC mixes with
212 Int. J. Miner. Metall. Mater., V ol. 20 , No. 2 , F eb. 2013

different contents of silica fume are given in Table 9 and (5) Based on the assessment of fresh and hardened
Fig. 7. Like compressive and splitting tensile strengths, properties of SCGC, it can be concluded that silica fume
silica fume addition also improved the flexural strength of can beneficially be used in the production of SCGC up to
SCGC mixes. This was believed to be due to the improved 10% by mass of fly ash.
interfacial bond between paste and aggregates. From Fig.
7, it can be seen that the flexural strength of concrete Acknowledgement
increases with the increase in silica fume content up to
The authors would like to acknowledge Universiti
10wt% replacement of fly ash and then tends to decrease.
Teknologi PETRONAS and the Ministry of Science, Tech-
At 28 d, the flexural strength of M1 (0wt% SF) achieves
nology, and Innovation, Malaysia (Research grant 06-02-
4.09 MPa, whereas mixtures M2 (5wt% SF), M3 (10wt%
02-SF0052) for providing the financial support and re-
SF), and M4 (15wt% SF) achieve the strength of 4.18, 4.56,
search facilities.
and 4.21 MPa, respectively, increasing by 2.2%, 11.5%, and
2.9% in comparison with the strength of the control mix. References

Table 9. Flexural strength test results [1] B.V. Rangan, Fly Ash-based Geopolymer Concrete, Curtin
Flexural strength / MPa University of Technology, Perth, 2010, p. 68.
Mix ID
1d 7d 28 d [2] J. Davidovits, Geopolymer cements to minimise carbon-
M1 (Control mix) 3.62 3.76 4.09 dioxide greenhouse-warming, Ceram. Trans., 37(1993), p.
M2 (5wt% SF) 3.80 4.03 4.18 165.
M3 (10wt% SF) 3.98 4.29 4.56
[3] J. Davidovits, Geopolymer Chemistry and Applications,
M4 (15wt% SF) 3.78 4.10 4.21
2nd Ed., Institut Géopolymère, Saint-Quentin, 2008, p.
276.
[4] D. Hardjito, S.E. Wallah, D.M.J. Sumajouw, and B.V.
Rangan, Factors influencing the compressive strength of
fly ash-based geopolymer concrete, Civ. Eng. Dimens.,
6(2004), No. 2, p. 88.
s

[5] O. Boukendakdji, S. Kenai, E.H. Kadri, and F. Rouis, Ef-

fect of slag on the rheology of fresh self-compacted concrete,
Constr. Build. Mater., 23(2009), No. 7, p. 2593.
c [6] C. Druta, Tensile Strength and Bonding Characteristics of
Self-Compacting Concrete [Dissertation], Polytechnic Uni-
Fig. 7. Effect of SF on flexural strength. versity of Bucharest, Bucharest, 2003, p. 5.
[7] European Federation of Specialist Construction Chemi-
5. Conclusions
cals and Concrete Systems (EFNARC), Specification and
(1) Addition of silica fume by substitution to fly ash Guidelines for Self-Compacting Concrete, EFNARC Asso-
is proved to be very effective in improving the hardened ciation House, Surrey, 2002, p. 4.
properties of SCGC. [8] M. Liu, Self-compacting concrete with different levels of
pulverized fuel ash, Constr. Build. Mater., 24(2010), No.
(2) Fluidity and flowability of various SCGC mixes de-
7, p. 1245.
crease when the content of silica fume increases. Concrete
[9] H. Okamura and M. Ouchi, Self-compacting concrete, J.
mixes still meet the requirements of flowability, viscosity,
Adv. Concr. Technol., 1(2003), No. 1, p. 5.
and passing ability of SCC with the silica fume addition
up to 10% by mass of fly ash. [10] P. Nanthagopalan and M. Santhanam, A new empirical
test method for the optimisation of viscosity modifying
(3) With the silica fume addition up to 10% by mass agent dosage in self-compacting concrete, Mater. Struct.,
of fly ash, the compressive, splitting tensile, and flexural 43(2010), No. 1-2, p. 203.
strengths of SCGC are significantly improved, compared [11] E.P. Koehler, Aggregates in Self-Consolidating Concrete
with the control concrete at all ages. The replacement of [Dissertation], The University of Texas at Austin, Austin,
fly ash beyond 10wt% of silica fume causes a reduction in 2007, p. 38.
the hardened properties of concrete.
[12] H.A.F. Dehwah, Mechanical properties of self-compacting
(4) At 28 d, in comparison with the strength of concrete incorporating quarry dust powder, silica fume or
the control mix, the compressive strength, splitting ten- fly ash, Constr. Build. Mater., 26(2012), No. 1, p. 547.
sile, strength and flexural strength are increased by 6.9%, [13] K.C. Biswal1 and S.C. Sadangi, Effect of superplasticizer
12.8%, and 11.5%, respectively. and silica fume on properties of concrete, [in] Proceedings
 F.A. Memon et al., Effect of silica fume on the fresh and hardened properties of fly ash-based ... 213

of International Confence on Advances in Civil Engineer- ing and Materials, Philadelphia, 2004, p. 10.
ing, Trivandrum, 2010, p. 94. [26] European Federation of Specialist Construction Chemicals
[14] M.A. Caldarone, High-Strength Concrete: a Practical and Concrete Systems (EFNARC), The European Guide-
Guide, Taylor & Francis, New York, 2008, p. 43. lines for Self-Compacting Concrete: Specification, Produc-
tion and Use, The Self-Compacting Concrete European
[15] A.M. Neville, Properties of Concrete, 4th Ed. Wiley Pub-
Project Group, Surrey, 2005, p. 47.
lishers, New York, 1995, p. 844.
[27] BS EN 12390-3, Testing Hardened Concrete: Part 3. Com-
[16] C. Meyer, The greening of the concrete industry, Cem.
pressive Strength of Test Specimens, British Standards In-
Concr. Compos., 31(2009), No. 8, p. 601.
stitute, London, 2002, p. 5.
[17] S.A. Barbhuiya, J.K. Gbagbo, M.I. Russell, and P.A.M. [28] S. Mindess, J.F. Young, and D. Darwin, Concrete, Pren-
Basheer, Properties of fly ash concrete modified with hy- ticeHall Inc., New Jersey, 2002, p. 315.
drated lime and silica fume, Constr. Build. Mater., [29] SS EN 12390-6, Testing Hardened Concrete: Part 6. Ten-
23(2009), No. 10, p. 3233. sile Splitting Strength of Test Specimens, Swedish Stan-
[18] F.A. Memon, F. Nuruddin, and N. Shafiq, Compressive dards Institute, Stockholm, 2004, p. 6.
strength and workability characteristics of low-calcium fly [30] ASTM C496/C496M-11, Standard Test Method for Split-
ash-based self-compacting geopolymer concrete, Int. J. ting Tensile Strength of Cylindrical Concrete Specimens,
Civ. Environ. Eng., 3(2011), No. 2, p. 72. Annual Book of ASTM Standards, American Society for
[19] F.A. Memon, F. Nuruddin, S. Demie, and N. Shafiq, Ef- Testing and Materials, Philadelphia, 2004, p. 287.
fect of curing conditions on strength of fly ash-based self- [31] BS EN 12390-5, Testing Hardened Concrete: Part 5. Flex-
compacting geopolymer concrete, Int. J. Civ. Environ. ural Strength of Test Specimens, British Standards Insti-
Eng., 3(2011), No. 3, p. 183. tute, London, 2009, p. 5.
[20] ASTM C618, Standard Specification for Coal Fly Ash and [32] ASTM C78/C78M-10, Standard Test Method for Flexural
Raw or Calcined Natural Pozzolan for Use as a Mineral Strength of Concrete, Annual Book of ASTM Standards,
Admixture in Concrete, Annual Book of ASTM Standard, American Society for Testing and Materials, Philadelphia,
American Society for Testing and Materials, West Con- 2004, p. 39.
shohocken, 2004, p. 2. [33] K. Andri, Alkali-activation of Fly Ash/MIRHA Blend in
Geopolymer Concrete for In-situ Application [Disserta-
[21] SS EN 450-1, Fly Ash for Concrete: Part 1. Definition,
tion], Universiti Teknologi PETRONAS, Kuala Lumpur,
Specifications and Conformity Criteria, Swedish Standards
2012, p. 148.
Institute, Stockholm, 2012, p. 40.
[34] S. Bhanja and B. Sengupta, Influence of silica fume on the
[22] BS EN 13263-1, Silica Fume for Concrete: Part 1. Def-
tensile strength of concrete, Cem. Concr. Res., 35(2005),
initions, Requirements and Conformity Criteria, British
No. 4, p. 743.
Standards Institute, London, 2005, p. 10.
[35] J. Newman and B.S. Choo, Advanced Concrete Technol-
[23] ASTM C136, Standard Test Method for Sieve Analysis of ogy: Constituent Materials, Butterworth-Heinemann, Jor-
Fine Aggregates, Annual Book of ASTM Standards, Amer- dan Hill, Burlington, 2003, p. 48.
ican Society for Testing and Materials, Philadelphia, 2004, [36] Z.X. Yang, N.R. Ha, M.S. Jang, K.H. Hwang, and J.K. Lee,
p. 84. The effect of SiO2 on the performance of inorganic sludge-
[24] BS 882: 1992, Specification for Aggregates from Natural based structural concretes, J. Ceram. Process. Res.,
Sources for Concrete, British Standards Institute, London, 10(2009), No. 3, p. 266.
1992, p. 14. [37] D. Dutta, S. Thokchom, P. Ghosh, and S. Ghosh, Effect of
[25] ASTM C33, Specifications for Concrete Aggregates, An- silica fume additions on porosity of fly ash geopolymers, J.
nual Book of ASTM Standards, American Society for Test- Eng. Appl. Sci., 5(2010), No. 10, p. 74.