ACI STRUCTURAL JOURNAL
Title No. 114-S60
Long-Term Mechanical Properties of Different Fly Ash
Geopolymers
by Chamila Gunasekara, Sujeeva Setunge, and David W. Law
Geopolymer concrete is a sustainable construction material with
the potential to act as a replacement for portland-cement (PC)
concretes. A detailed investigation of the mechanical properties of
four different fly ash geopolymer concretes was carried out up to
1 year of age. Compressive, flexural, and splitting tensile strengths,
elastic modulus, and Poisson’s ratio of four geopolymer concretes
at 1 year ranged between 28 and 88 MPa (4.06 and 12.76 ksi),
3.92 and 6.3 MPa (0.568 and 0.914 ksi), 1.86 and 4.72 MPa (0.27
and 0.684 ksi), 10.3 and 29 GPa (1493.5 and 4205 ksi), and 0.16
and 0.28, respectively. The results show an increase in performance
observed between 90 and 365 days for all concretes depending on
the fly ash properties. Tarong displayed the highest increase while
Gladstone had the least, although Gladstone did display the best
performance throughout. The nature of the gel matrix formed, in
terms of uniformity and compactness, was observed to determine
the mechanical properties. The nature of the interfacial transition
zone formed between coarse aggregate and mortar and its density
was observed to govern the tensile strength. An increase in porosity
and microcracks was seen to negatively affect the compactness of
the gel matrix, which in turn affected the elastic modulus.
Keywords: compressive strength; elastic modulus; fly ash; geopolymer
concrete; microstructure; tensile strength.
INTRODUCTION
Geopolymer concrete is an emerging construction mate-
rial with significant benefits in reducing environmental
impacts of concrete. Compared to portland-cement (PC)
concrete, geopolymer concrete can reduce CO2 emission by
26 to 45% with no adverse economic effects.1-3 Geopolymer
concrete can be produced from different types of fly ash:
high-calcium fly ash,4-7 low-calcium fly ash8-13 and bottom
ash.14-17 However, geopolymer concrete produced using
100% low-calcium Class F fly ash has been the focus of the
largest area of research in recent years due to its high level
of availability and cost effectiveness.
A major challenge faced by the construction industry in
adopting fly ash geopolymer concrete is the variability of
fly ash from different sources and the effect this can have
on the range of mechanical properties produced. Hardjito
and Rangan18 used low-calcium Class F fly ash from an
Australian power station, but in three different batches to
produce geopolymer concrete, and investigated mechanical
properties at 90 days. The splitting tensile strength, elastic
modulus, and Poisson’s ratio of geopolymer concrete with
compressive strength in the range of 40 to 90 MPa (5.8 to
13.05 ksi) gave results between 4.4 and 7.4 MPa (0.638
and 1.073 ksi), 23.0 and 30.8 GPa (3335 and 4466 ksi), and
0.12 and 0.16, respectively. While splitting tensile strength
was found to be higher than the values recommended by the
Australian standard,19 the measured elastic modulus were at
ACI Structural Journal/May-June 2017
TECHNICAL PAPER
the lower end of the values calculated according to the stan-
dard. The stress-strain relationship of geopolymer concrete in
compression fits well with the expression developed for PC
concrete, with the strain at peak stress in the range of 0.0024
to 0.0026. Fernandez-Jimenez et al.12 produced geopolymer
concrete using a low-calcium Class F Spanish fly ash, and
while a 45 MPa (6.525 ksi) compressive strength was achieved
in the first 24 hours, it only increased to 60 MPa (8.7 ksi)
after 28 days. The geopolymer concrete had a higher flex-
ural strength, which increased from 4.62 to 8.8 MPa (0.67
to 1.276 ksi) between 1 and 90 days. However, it showed a
lower elastic modulus compared to PC concrete—18.4 and
30.3 GPa (2.668 and 4.394 ksi), respectively—at 28 days.
According to the authors, it might be possible that fly-ash-
based geopolymer concrete exhibits similar behavior to
high-performance PC concrete. In addition, the values of
elastic modulus based on formula proposed by different
codes proved to be unreliable. This was attributed to the
existence of variables that had not been taken into consider-
ation in the formula, but nevertheless influenced the proper-
ties of the geopolymer.
Ryu et al.10 investigated compressive and splitting tensile
strength of geopolymer concrete up to 28 days using a
low-calcium Class F fly ash obtained from a Korean power
plant. The splitting tensile strength of the geopolymer
concrete, with compressive strength in the range of 25.8
to 30.3 MPa (3.741 to 4.394 ksi), was in the range 2.0 to
2.6  MPa (0.29 to 0.377 ksi) in the period between 3 and
28 days. The data also showed a splitting tensile strength
to compressive strength ratio ranged between 7.8 and 8.2%,
similar to that of PC concrete. Neupane et al.20 also inves-
tigated the 28-day mechanical properties of geopolymer
concrete using a low-calcium Class F Australian fly ash,
but with two different concrete grades: 50 and 65 MPa
(7.25 and 9.425 ksi). The flexural strength, splitting tensile
strength, and elastic modulus of geopolymer concrete, with
compressive strength in the range of 59 to 82.5 MPa (8.555
to 11.963 ksi), gave results in the range of 5.6 to 8.7 MPa
(0.812 to 1.262 ksi), 4.4 to 6.2 MPa (0.638 to 0.899 ksi), and
28 to 39 GPa (4060 to 5655 ksi), respectively. The author
observed that the flexural and splitting tensile strengths
attained by geopolymer concrete are significantly higher
than the prescribed value in the Australian standards19 for
the same grade of concrete. However, the elastic modulus
ACI Structural Journal, V. 114, No. 3, May-June 2017.
MS No. S-2016-113.R2, doi: 10.14359/51689454, received August 16, 2016, and
reviewed under Institute publication policies. Copyright © 2017, American Concrete
Institute. All rights reserved, including the making of copies unless permission is
obtained from the copyright proprietors. Pertinent discussion including author’s
closure, if any, will be published ten months from this journal’s date if the discussion
is received within four months of the paper’s print publication.
743
Table 1—Chemical composition of fly ash
By weight, %

Fly ash SiO2 Al2O3 SiO2/Al2O3 Fe2O3 CaO P2O5 TiO2 MgO K2O SO3 MnO Na2O LOI*
Gladstone 47.87 28.0 1.71 14.09 3.81 1.81 1.99 0.93 0.62 0.27 0.21 0.41 0.43
Pt. Augusta 49.37 31.25 1.58 4.47 4.80 1.65 2.94 1.28 2.21 0.24 0.04 1.30 0.51
Collie 53.82 29.95 1.80 9.24 1.03 1.28 2.19 0.58 0.79 0.34 0.04 0.75 0.63
Tarong 75.66 19.0 3.98 1.38 0.30 1.0 1.83 0.0 0.63 0.03 0.02 0.15 1.16
*
Loss on ignition (unburnt carbon content).

Table 2—Physical and mineralogical properties of

fly ash
Properties investigated Gladstone Pt. Augusta Tarong Collie
Specific gravity 2.26 2.23 2.08 2.42
BET surface area, m2/kg 2363 1228 1876 1095
At 10
43.1 46.7 43.0 40.9
microns
At 20
Fineness, % 61.9 62.1 63.0 54.6
microns
At 45
82.7 80.2 81.8 70.0
microns
Amorphous content, % 71.8 59.5 66.3 72.5
Quartz 6.8 29.2 14.8 18.2
Crystalline, % Fig. 1—Particle size distribution.
Mullite 17.9 7.5 18.9 8.7
Note: 1 kg/m2 = 0.2048 lb/ft2. type of fly ash, and in each study using a mixing process
unique to that study. This research, dealing with the perfor-
is found to be almost equal with the calculated value from mance of geopolymer concrete up to 1 year using a range
Australian standards19 for a similar grade of PC concrete. of low-calcium Class F fly ashes with same mixing process,
Diaz-Loya et al.11 produced geopolymer concrete using provides a systematic long-term study of the mechanical
13 different low-calcium Class F fly ashes collected from properties of a range of geopolymer concretes. Hence, the
power stations located across the United States, and inves- research data presented herein will be extremely useful to
tigated the mechanical properties at 28 days. The flexural understand the long-term behavior of geopolymer concretes
strength, elastic modulus, and Poisson’s ratio of geopolymer made with the wide range of low-calcium Class F fly ashes
concrete, with compressive strength in the range 10.34 to that are available across the world.
49.24 MPa (1.499 to 7.14 ksi), gave values between 2.74 to
4.66 MPa (0.397 to 0.676 ksi), 7.46 to 19.28 GPa (1081.7 to EXPERIMENTAL PROCEDURE
2795.6 ksi), and 0.1 to 0.15, respectively. Materials
Given the range of values reported for the mechanical Low-calcium fly ash (class F) conforming to Australian
properties, a clear understanding of the effect of fly ash char- standard AS 3582.121 were obtained from Gladstone, Port
acteristics on the long-term performance of the geopolymer (Pt.) Augusta, Collie, and Tarong power stations in Australia.
concrete is essential to be able to construct real geopolymer The chemical composition, particle size distribution, and
structures with confidence. The present study investigated mineralogical composition of each fly ash, determined by
the key mechanical properties of geopolymer concretes X-ray fluorescence (XRF), Malvern particle size analyzer
made with four different low-calcium fly ashes with a instruments, and X-ray diffraction (XRD), respectively, are
range of chemical, physical, and mineralogical properties; shown in Table 1, Fig. 1, and Table 2. The Brunauer Emmett
the same mixing process was used throughout to enable Teller (BET) method by N2 absorption was used to deter-
a direct comparison of the properties to be undertaken. A mine fly ash surface area. Alkaline solution used in geopoly-
detailed microscopic analysis was conducted to investigate mers consisted of a mixture of commercially available
the changes of microstructure, especially at interfacial tran- sodium silicate (specific gravity = 1.53, Na2O = 14.7%, and
sition zone (ITZ) and the aluminosilicate gel structure. The SiO2 = 29.4% by mass), and sodium hydroxide (15M). Both
observed test results up to 1 year of age are reported, and coarse and fine aggregate were prepared in accordance with
the variations in properties observed are then explained in AS 1141.5.22 The fine aggregate was river sand in uncrushed
conjunction with the microstructure development. form with a specific gravity of 2.5 and a fineness modulus
of 3.0. The coarse aggregate was crushed basalt aggregate
RESEARCH SIGNIFICANCE of two grain sizes: 7 mm (0.2758 in.), 2.58 specific gravity,
Published research to date on geopolymer concrete has and 1.60% water absorption; and 10 mm (0.394 in.), 2.62%
studied the mechanical properties up to 90 days using a single specific gravity, and 0.74% water absorption.

744 ACI Structural Journal/May-June 2017

Table 3—Optimum mixture design details (kg/m3)
Aggregates, kg Activator, kg
Geopolymer Fly ash, kg Sand 7 mm 10 mm Na2SiO3 (liquid) NaOH (15 M) Added water, kg
Gladstone 416 699 309 618 292 65 8
Pt. Augusta 416 699 309 618 292 65 8
Collie 420 706 312 624 241 92 15
Tarong 412 693 306 612 342 39 0
Notes: 1 kg/m = 0.0624 lb/ft ; 1 kg = 2.2046 lb; 1 mm = 0.0394 in.
3 3

Mixture designs
The mixture design is an optimized concrete mixture
developed in previous research for each of the four fly
ashes.23 In this study, the Na2O dosage is fixed at 15%
while the total aggregate in the concrete was kept to 64%
of the entire mixture by volume for all mixtures. The ratio
of mixture components in Table 3 was calculated based on
the absolute volume method24; as a result, the total weight of
binder and water was varied to keep the volume of material
and water/solid ratio (0.37) constant.
The mass of water in the mixture was taken as the sum
of mass of water contained in the sodium silicate, sodium
hydroxide, and added water. The mass of solid is taken as the
sum of fly ash, the solids in the sodium silicate solution, and
the sodium oxide pellets.
Mixing, casting, and curing
The mixing of geopolymer concrete was carried out using
a 120 L (4.238 ft3) concrete mixer. The dry materials (fly ash,
fine and coarse aggregates) were mixed first for 4 minutes.
Then activator and water were added to the dry mix and
mixed continuously for another 8 minutes. The mixture
was then poured into molds and vibrated using a vibration
table for 1 minute to remove air bubbles. After vibration, the
molds were kept at room temperature for 1 day and then heat
cured in an oven for 24 hours at 80°C (176°F) temperature
with 95% relative humidity. Molds were removed from the
oven and left to cool to room temperature before demolding,
and then kept at room temperature until being tested.
Testing
The compressive strength test was performed by MTS
machine with a loading rate of 20 MPa/min (2.9 ksi/min)
according to AS 1012.9.25 The flexural and splitting tensile
strength tests were conducted in accordance with AS 1012.1126
and AS 1012.10,27 respectively. The flexural tensile strength
test was carried out on a MTS machine with additional testing
apparatus under a four-point bending test with a loading rate
of 1 MPa/min (0.145 ksi/min). The splitting tensile strength
test was performed on an MTS machine equipped with split-
ting tensile strength test equipment under a loading rate of
1.5 MPa/min (0.218 ksi/min). The elastic modulus and Pois-
son’s ratio were determined using a concrete testing machine
coupled with the compressometer/extensometer with a
loading rate of 0.25 MPa/s (0.036 ksi/s) in accordance with
AS 1012.17.28 Slump and dry density tests were conducted
in accordance with AS 1012.3.129 and AS 1012.12.2.30 All
tests were conducted at 28, 90, and 365 days of casting. The
Fig. 2—Long-term density development and workability.
(Note: 1 kg/m3 = 0.0624 lb/ft3; 1 mm = 0.0394 in.)
reported test results in each specific test are an average of
three samples.
The microstructure development was observed using scan-
ning electron microscopy (SEM) imaging employing back-
scatter electron detector with 15 eV of energy. Energy-dis-
persive X-ray spectroscopy (EDS) analysis was performed
using nano-analysis software (AZtec 2.1) to determine the
chemical composition (Si/Al atomic ratio) of the reacted
geopolymer concrete. Specimens were cut using a diamond
saw to a size of 2 to 4 mm (0.0788 to 0.1576 in.) in height
and 5 to 10 mm (0.197 to 0.394 in.) in diameter. The samples
were subsequently carbon coated and then mounted on the
SEM sample stage with conductive, double-sided carbon
tape. A total of three samples were investigated for each
geopolymer concrete.
EXPERIMENTAL RESULTS
Density and workability
The dry density of four geopolymer concretes ranged
between 2074 and 2205 kg/m3 (129.42 and 137.59 lb/ft3)
(Fig. 2), all of which increased with age. The range of densi-
ties observed is in agreement with the literature11,31 and is
lower than PC concrete, which is characteristically cited
as 2400 kg/m3 (149.76 lb/ft3).19 In addition, all concretes
displayed a very high, collapsed slump. This is attributed
to the spherical shape of fly ash particles combined with the
lubricating effect of sodium silicate solution. Thus, work-
ability is represented by the slump flow diameter (ϕ) and
ranged between 350 and 735 mm (13.79 and 28.96 in.) for
the four materials.

ACI Structural Journal/May-June 2017 745

Fig. 3—Long-term compressive strength development.
(Note: 1 MPa = 0.145 ksi.)
Fig. 4—Long-term flexural strength development. (Note:
1 MPa = 0.145 ksi.)
Compressive strength
Figure 3 shows long-term compressive strength develop-
ment of the four different fly ash geopolymer concretes. It is
observed that the compressive strength of all geopolymers
tended to increase with time, but in different increment ratios,
depending on the fly ash properties. Gladstone geopolymer
displayed the highest strength while Collie the lowest at all
ages, with the strengths of Pt. Augusta and Tarong geopoly-
mers falling between the two. The compressive strength
of Gladstone ranged from 82.5 to 87.4 MPa (11.963 to
12.673 ksi) while the Collie varied from 24.9 to 28.7 MPa
(3.611 to 4.162 ksi) between 28 and 365 days. It was noted
that Gladstone obtained 95% of its ultimate strength in first
28 days, showing a lowest percentage increase in strength
between 28 and 365 days. Collie geopolymer demonstrated
a higher strength gain: 15.3% between 28 and 365 days.
However, the actual increase in strength is only by 3.8 MPa
(0.551 ksi), which is lower than 4.9 MPa (0.711 ksi) increase
observed for Gladstone. Tarong and Pt. Augusta geopolymers
displayed a high compressive strength development: 44.6%
(13.2 MPa [1.914 ksi]) and 27.4% (10.1 MPa [1.465 ksi]),
respectively, between 28 and 365 days. Pt. Augusta had
significant strength development: 8.7 MPa (1.262 ksi) by 90
days, but little further increase (1.4 MPa [0.203 ksi]) in the
period between 90 and 365 days. Tarong, however, had a
similar increase both from 28 to 90 days (6.9 MPa [1.0 ksi])
746
Fig. 5—Long-term splitting tensile strength development.
(Note: 1 MPa = 0.145 ksi.)
and from 90 to 365 days (6.3 MPa [0.914 ksi]). These results
demonstrate that there is a strength development over time,
even though heat curing is observed to enhance the initial
geopolymerization process.32,33
Flexural strength
Figure 4 shows long-term flexural strength development
of four fly ash geopolymer concretes. It ranged from 3.36
to 6.30 MPa (0.487 to 0.914 ksi) between 28 and 365 days.
Similar to compressive strength development, the flexural
strength of all concretes tended to increase with time. The
highest and lowest flexural strengths were displayed by
Gladstone and Collie geopolymers, while the Pt. Augusta and
Tarong again fell in between the Gladstone and Collie. While
Tarong showed a lower compressive strength, it achieved a
slightly higher flexural strength than Pt. Augusta. Tarong
had the highest increase both from 28 to 90 days (0.77 MPa
[0.112 ksi]) and from 90 to 365 days (1.59 MPa [0.231 ksi]),
while Collie had the lowest increase in both periods. Overall,
the flexural strength of the four geopolymers ranged from 7
to 14% of the compressive strength, compared to a range of
9 to 12% typically cited for PC concrete.11,24 Diaz-Loya et
al.11 reported a similar trend but over a wider range, noting
that the flexural strength of the geopolymers are between 9
and 26% of the compressive strength.
Splitting tensile strength
Figure 5 shows long-term splitting tensile strength devel-
opment of four fly ash geopolymer concretes. The splitting
tensile strength ranged from 1.15 to 4.72 MPa (0.167 to
0.684 ksi) between 28 and 365 days, and ranged from 4 to
8% of the compressive strength. The splitting tensile strength
of all geopolymer concretes increased with time. Gladstone
and Collie achieved the highest and lowest splitting tensile
strength, respectively, with Pt. Augusta and Tarong geopoly-
mers once more falling between the two. It is interesting to
note that Pt. Augusta showed higher splitting tensile strength
than Tarong, though it had lower flexural strength of the two.
Elastic modulus
The elastic modulus of concrete is defined as the stress
required to induce a unit strain within the elastic limit. The
ACI Structural Journal/May-June 2017

Fig. 6—(a) Stress-strain curve; and (b) elastic modulus
development. (Note: 1 MPa = 0.145 ksi.)
stress-strain curves shown in Fig. 6(a) presents the range of
elastic modulus measured in this study. As the geopolymer
concretes demonstrated brittle failure,11 the descending
branch of the stress-strain curve could not be determined in
any of the tests. Each curve represents the tangent modulus
of elasticity and ranged from 8.2 to 22.7 GPa (1160 to
3291.5 ksi) between Collie and Gladstone geopolymers at
28 days. Figure 6(b) further shows elastic modulus devel-
opment at 90 and 365 days. Similar to the strength proper-
ties discussed, the elastic modulus also tended to increase
with time in all concretes with different increment ratios.
Gladstone showed the highest value while Collie gave the
lowest at all ages. A substantial increase of elastic modulus
was noted between 90 and 365 days compared to the initial
28- to 90-day period for all geopolymers, except Collie,
which displayed the opposite trend. Moreover, in the first
90 days, Tarong had a very low elastic modulus, similar to
Collie and lower than Pt. Augusta. However, by 365 days, it
had achieved the second-highest elastic modulus, having a
twofold increase between 90 and 365 days.
Poisson’s ratio
Figure 7 shows long-term Poison’s ratio changes of the
four geopolymer concretes. The Poisson ratio varied from
ACI Structural Journal/May-June 2017
Fig. 7—Long-term Poisson’s ratio development.
0.12 to 0.28 between 28 and 365 days, and increased with
the age. Warner et al.34 reported that Poisson’s ratio values
of PC concrete can range between 0.11 and 0.21, but usually
fall in the range from 0.15 to 0.2. Values obtained by four
geopolymer concretes tended to populate the lower end of
this range at 28 days, but three of them fall into the standard
range (0.15 to 0.2) at 90 days. Tarong geopolymer, however,
exceeded this upper end of PC concrete and displayed a
significantly high Poisson’s ratio at 90 and 365 days.
DISCUSSION
In the geopolymerization process, alumina and silica
species in fly ash rapidly react with highly alkaline activator
solution and produce a three-dimensional polymeric chain
and ring structure consisting of Si-O-Al-O bonds (Eq. (1)),33
where z is the type of bond presence (1, 2, or 3), and n is
the degree of polymerization. The schematic formation of
the resultant geopolymer product is generally represented by
Eq. (2) and (3).
The Si/Al (atomic) ratio determines the structure of
the geopolymer backbone. In this study, Si/Al ratio of
geopolymer concrete ranged between 2.69 and 2.84, 2.55
and 2.64, and 2.48 and 2.58 for Gladstone, Pt. Augusta,
and Collie at 28, 90, and 365 days, respectively. As such,
the geopolymer structure was inferred to be polysialat-si-
loxo (Si-O-Al-O-Si-O). In Tarong geopolymer, the Si/Al
ratio ranged 4.05 to 3.65, and the corresponding gel struc-
ture is identified as polysialate-disiloxo (Si-O-Al-O-Si-O-
Si-O). In all geopolymers, the Si/Al ratio decreased with age
(Fig. 8 and 9). This indicates an ongoing geopolymerization
747
Fig. 8—SEM images of gel-microstructure in: (a) to (c) Gladstone; (d) to (f) Collie geopolymer concretes; and (g) different
aluminosilicate gel formation in Gladstone.
Fig. 9—SEM images of gel-microstructure in: (a) to (c) Pt. Augusta; and (d) to (f) Tarong geopolymer concretes.
process, with continuous gel formation along with incor-
poration of alumina into the silicate backbone. The Si/Al
ratio of all geopolymer concretes, other than Tarong, varied
over a small range, but their mechanical properties differed
significantly. Hence, although the Si/Al ratio indicated that
the geopolymer bond structure should be similar, the nature
of the geopolymer gel matrix formed, in terms of uniformity
and compactness, is significantly different, which is hypoth-
esized as the reason for the mechanical properties observed.
The matrix formed and its development with the time in
Gladstone and Collie geopolymer concretes are displayed in
Fig. 8. Gladstone produced a uniformly distributed geopoly-
meric gel-microstructure at all ages. The degree of reac-
tivity of fly ash governs the dissolution, coagulation, and
gel formation, which is itself dependent on the properties
of source material. Gunasekara et al.23 reported that a higher
quantity of fly ash particles with the range of 10 and 20 µm
748
(3.94 × 10–4 and 7.88 × 10–4 in.) size, coupled with the larger
surface area, increased the fly ash reactivity. Gladstone fly
ash has the highest surface area with a large amount of finer
particles, especially a higher percentage passing at 10, 20,
and 45 micron sieves. It also consists of a high amount of
reactive alumina-silica amorphous phase. The very fine
particles of amorphous phase in Gladstone rapidly react with
the alkali activator and precipitates geopolymeric gel on the
surface of the fly ash particles. This is because alumina-silica
species are primarily leached by dissolution of the amor-
phous phase, not the crystalline phases of fly ash,4 with the
significant part of the alkali-fly ash reaction occurring at the
particle-liquid interface.35 This gel then diffuses through the
surface, covering and coalescing unreacted fly ash spheres
together. Continuous gel precipitation further fills the inte-
rior spaces and forms a uniform, strongly compacted, and
well-condensed gel-microstructure.
ACI Structural Journal/May-June 2017
 The Gladstone fly ash also contains a high CaO percentage.
The CaO reacts with the alkali activator and produces calci-
um-aluminosilicate gel (C-A-S-H) (Fig. 8(g)). The C-A-S-H
gel coupled with the sodium-aluminosilicate (N-A-S-H) gel
provides additional rigidity to the geopolymer backbone. The
combination of the C-A-S-H and N-A-S-H gels result in the
dense microstructure observed for the Gladstone geopolymer.
This dense microstructure in turn results in the high compres-
sive strength at 28 days and beyond. Due to high reactivity
of the material, little fly ash remains available for subsequent
geopolymeric reaction and, hence, the minimal strength
development observed between 28 and 365 days.
Collie geopolymer displayed a significantly different
microstructure compared to that of Gladstone. A large
number of unreacted fly ash particles were observed in the
microstructure at 28 days (Fig. 8(d)). These comprised a
number of unreacted fly ash particles that were separated
from the geopolymeric binder, indicating weak adhesion
between the gel and the particles. Other unreacted spheres
were partially embedded in the precipitated gel. Collie fly ash
has the lowest surface area and a higher fraction of coarser
particles—that is, very low level of particles passing at 10,
20, and 45 micron sieves. Although it contains the highest
percentage of reactive amorphous phase, the reactivity and
dissolution of coarser fly ash particles in alkali activator and
subsequent geopolymeric gel formation is observed to be
much lower, as evidenced by the large number of unreacted
particles. This resulted in the low compressive strength for
geopolymer concrete produced using Collie fly ash. On the
other hand, the microstructure does show a decrease in the
number of unreacted particles at 365 days, indicating that
there has been some additional geopolymerization and gel
formation, as reflected by the increase in strength observed.
This excess gel, however, is distributed unevenly over the
matrix. This resulted in forming nonhomogeneous small gel
units as displayed in Fig. 8(f), rather than forming an inter-
connected uniform gel matrix. Hence, this would indicate
that the compressive strength of geopolymer concrete does
not only rely on the quantity of gel formation, but also on
uniform distribution of gel and the compactness, the combi-
nation of which leads to the dense microstructure observed.
This is hypothesized as the reason that the Collie geopolymer
had a low compressive strength even after 365 days.
The microstructure of Pt. Augusta and Tarong geopoly-
mers are shown in Fig. 9. The gel matrixes of both are
heterogeneous, having a number of unreacted/partially
reacted fly ash particles. Both geopolymers had microcracks
on their surface, but with Tarong displaying more microc-
racks and a greater crack width than the Pt. Augusta. Similar
to Gladstone, both Pt. Augusta and Tarong contained a large
amount of fine particles, but they had a lower surface area
and amorphous content. These material differences are
hypothesized as negatively affecting their degree of alkali
reactivity, resulting in the non-uniform, heterogeneous
geopolymer matrix observed for both concretes. The unre-
acted/partially reacted fly ash particles behave as compos-
ites. These composites and the interface between them and
geopolymer matrix is hypothesized as an area of weakness
and, thus, has a significant bearing on the overall strength of
ACI Structural Journal/May-June 2017
the concrete,36 resulting in the lower compressive strength
achieved compared to the Gladstone geopolymer concrete.
Pt. Augusta fly ash has the highest CaO content while
Tarong has the lowest. The C-A-S-H gel produced by
this additional CaO in the Pt. Augusta is identified as one
reason for the higher strength observed compared to the
Tarong, although it has a lower surface area. Tarong fly ash
contains more unburnt carbon content. This acts as an inert
particulate in the gel matrix and thus may cause a spread of
microcracks. Both geopolymers, however, showed a signif-
icant strength development between 28 and 365 days. It is
hypothesized that the relatively low surface area of the two
fly ashes led to a lower initial dissolution of alumina-silicate
species. This dissolution continued over time and produced
more geopolymer gel, which led to the increase in the homo-
geneity observed and resulted in the strength development
observed for both Pt. Augusta and Tarong.
In addition to the nature of the geopolymer matrix formed,
the interfacial transition zone (ITZ) of geopolymer concrete
takes a crucial role in the concrete performance. In PC
concrete, coarse aggregate has an ITZ of approximately
100 µm (0.00394 in.) depth with two layers; one is a duplex
film (1 µm [3.94 × 10–5 in.]) at the surface of the aggregate
and other one is the porous transition zone of 20 to 100 µm
(0.000788 to 0.00394 in.) deep.37 However, Demie et al.37
noted that ITZ in fly ash geopolymer concrete is formed
by combining these two layers, and is defined as the inter-
face between aggregate and geopolymer paste. The flexural
strength and splitting tensile strength of fly ash geopolymer
concrete are strongly dependent on this geopolymer gel-ag-
gregate bond strength. Figure 10 shows a comparison
of ITZ with the bulk geopolymer gel matrix in four
different concretes.
The ITZ is critical because it is known to have a different
microstructure from the bulk of hardened geopolymer paste
and the interface is also considered as the specific location
of early cracking. Scrivener et al.38 reported that ITZ in
PC concrete is caused by the disorder of packing the anhy-
drous cement grains in the transition zone. In geopolymer
concrete, ITZ is caused by incomplete packing of unreacted
fly ash microsphere particles in the transition between the
geopolymer paste and coarse aggregates, as the aluminos-
ilicate gel formation is strongly dependent on the degree of
reactivity of fly ash. This phenomenon is due to the incom-
plete dissolution of a large proportion of fly ash and the
existence of differing composition and size unreacted fly ash
particles. The ITZ was a region of transition that is highly
heterogeneous, non-uniform, and varying from point to
point along each aggregate particle. In this study, Gladstone
showed a denser ITZ between aggregates and geopolymer
matrix (Fig. 10(a) and 10(b)). There is no obvious difference
noted between the microstructure of the ITZ and the bulk
geopolymer matrix. Sarker et al.39 showed that the fracture
plane in the PC concrete passed predominantly around the
aggregates, and those in the geopolymer concrete generally
passed predominantly through the aggregates. This behavior
was observed in Gladstone geopolymer in the tensile testing.
The high bond strength at the interface of the geopolymer
binder and the aggregates lead to the fracture plane passing
749
Fig. 10—SEM images of ITZ in different fly ash geopolymer concrete.
predominantly through the aggregates instead of following
the interface. The strong ITZ observed in Gladstone
geopolymer is identified as the reason for the high flexural
and splitting tensile strengths observed at all ages.
Collie showed a highly heterogeneous and non-uniform ITZ
(Fig. 10(c) and 10(d). The unreacted, partially reacted fly
ash spheres in Collie in the aggregate interface produced a
loose and porous ITZ. The propagated microcracks further
took away the interconnection between aggregate and
geopolymer binder and, hence, produced a weak aggre-
gate-gel bond strength. Moreover, fracture plane of the
Collie geopolymer concrete specimen passed predominantly
around the aggregate, rather than through the aggregates.
This explains the low flexural and tensile strength reported
in Collie geopolymer concrete at 28 days. As the geopo-
lymerization progresses, the space between the aggregate
and the geopolymer paste was gradually replaced by some
additional geopolymer gel as the polymerization continued.
However, due to the low degree of reactivity of Collie fly
ash, the gel formation is expected to be relatively small. The
SEM image shows a relatively porous and loose interface at
365 days, confirming this expectation. This resulted in the
minimal tensile strength development observed in Collie
geopolymer concrete.
Pt. Augusta does not show any discernible difference
between the microstructure of ITZ and the bulk geopolymer
matrix. Narrow microcracks are observed in the aggre-
gate-gel interface at 28 days, but not at 365 days, which
appears as a strong aggregate-gel interface. The continuous
geopolymerization process and gel formation thus densify
(refine) the ITZ filling spaces between aggregate-gel inter-
face. This resulted in the high splitting tensile strength devel-
opment achieved by Pt. Augusta between 28 and 365 days.
SEM showed that Tarong had a weak aggregate-gel inter-
face compared to Pt. Augusta. Tarong, however, did display
progressive geopolymer gel formation with age (Fig. 10(d) to
10(f), but this gel refined the bulk gel-microstructure rather
than the ITZ. This resulted in a weak aggregate-gel inter-
750
face even at 365 days. The high unburnt carbon in Tarong is
understood to weaken the bonding with aggregates. It further
contains high amorphous SiO2 content, which is expected
to result in more Si+4 being leached into the activator. The
optimum Tarong geopolymer mixture also required a higher
sodium silicate concentration than other mixtures. Fernan-
dez-Jimenez et al.12 reported that the presence of excess Si+4
in the alkaline activator solution has a substantial negative
effect on the gel-aggregate bond. The combination of these
factors appeared to have negatively affected the gel-aggre-
gate bond in Tarong geopolymer concrete. This is supported
by the lower splitting tensile strength observed compared to
the Pt. Augusta.
Tarong had a higher flexural strength than Pt. Augusta
at all ages. Sarker et al.39 stated that the higher tensile and
bond strengths of geopolymer concrete increased its critical
stress intensity factor, which resulted in less tortuosity of the
fracture plane and relatively more brittle type of failure than
in the PC concrete. Both Tarong and Pt. Augusta followed
a similar trend, but Tarong showed a significantly higher
failure strain than Pt. Augusta (Fig. 6(a)), thus improving the
ductility of the material. The maximum fiber stress reached
in the flexural strength test is higher than the splitting
strength because the propagation of a crack is blocked by
less stressed material near to the neutral axis. The high strain
capacity of Tarong assists in redistribution of stresses. This
is hypothesized as the reason for higher resistance to the
flexural failure in Tarong than Pt. Augusta concrete. Overall,
other than tensile strength improvement, the development of
ITZ in geopolymer concrete with age would further result
into the aforementioned compressive strength improvement
of concretes between 28 and 365 days.
All geopolymer concretes in this study showed a much
lower elastic modulus than PC concrete for similar compres-
sive strengths.19 Liu et al.40 reported that the pore volume and
elastic modulus have a linear relationship: as the density of
the geopolymer concrete increases, the elastic modulus also
increases. Puertas et al.41 reported that elastic modulus is
ACI Structural Journal/May-June 2017
dependent on the porosity and packing efficiency—that is, the
higher the packing density, the higher the modulus of elas-
ticity. This study showed all fly ash geopolymer concretes
had a lower density than PC concrete. The water content in
geopolymer mixture does not participate in the geopolymer
reaction (Eq. (3)) and is expected to evaporate during the
high-temperature curing. Hence, there were more air voids
in the bulk gel matrix along with increased porosity of the
geopolymer concrete compared with PC concrete. Gladstone
and Collie geopolymers have lowest and highest macropo-
rosity, respectively, and Tarong has higher macroporosity than
Pt. Augusta.42 In addition, the microcracks induced on the
surface of geopolymer matrix due to high-temperature curing
can easily spread through these air voids into the geopolymer
matrix. Increase of porosity and crack propagation would
negatively influence packing density of geopolymer micro-
structure. This is hypothesized as the reason for reported
lower elastic modulus compared to PC concrete.
Similar to the aforementioned mechanical properties
discussed, the Gladstone geopolymer showed the highest
elastic modulus while Collie had the lowest. The elastic
modulus of geopolymer concrete is dependent on the degree
of geopolymerization.43,44 Duxson et al.45 further noted
that the elastic modulus is influenced by the alkali-acti-
vated metakaolin microstructure rather than by the chem-
ical composition. The Gladstone had a densely packed,
uniform gel-microstructure with the least amount of
partially dissolved fly ash grains and surface microcracks.
It further showed lower macroporosity42 and a high level of
interconnectivity between geopolymer gel and aggregates,
especially at the aggregate interface. This would account for
the high elastic modulus achieved by Gladstone geopolymer.
In contrast, Collie had a very heterogeneous and non-uni-
form microstructure with many unreacted particles and an
interconnected microcrack network with higher macropo-
rosity,42 showing a higher discontinuity and looser micro-
structure, resulting in the low elastic modulus observed.
The differences in the elastic modulus of Pt. Augusta and
Tarong geopolymers also correlate well with their respec-
tive geopolymer matrixes. Pt. Augusta had a less dense gel
matrix with partially dissolved fly ash grains and voids.
There is no clear difference of microstructure observed with
the age. Hence, this would explain the low elastic modulus
of Pt. Augusta compared with Gladstone at all ages. Tarong
had a less-compacted heterogeneous microstructure up to
90 days. However, beyond 90 days, the microstructure was
significantly denser, consistent with continuing dissolution
and gel formation. The improvement in compactness and
packing density of the gel-microstructure is hypothesized as
the reason for the observed improved elastic modulus devel-
opment in Tarong after 90 days.
SUMMARY AND CONCLUSIONS
Based on the results of the experimental study, the
following conclusions can be made:
1. The 365-day compressive strength of four different fly ash
based geopolymer concretes ranged between 28 and 88 MPa
(4.06 and 12.76 ksi). The order of 5 to 45% strength increase
is observed between 28 and 365 days in four concretes. Tarong
ACI Structural Journal/May-June 2017
had a more significant strength increase from 90 to 365 days
than other geopolymer concretes. The rate of increase was
dependent on the properties of the precursor fly ash.
2. The Gladstone ash was observed to provide the best
performance for all the mechanical properties at all ages,
while Collie ash gave the poorest at all ages.
3. The 365-day flexural and splitting tensile strengths,
of the four different fly ash geopolymer concretes ranged
between 3.92 to 6.3 MPa (0.568 to 0.914 ksi) and 1.86 to
4.72 MPa (0.27 to 0.684 ksi), respectively. The nature of
the ITZ formed between aggregates and aluminosilicate gel
matrix were shown to cause the tensile strength variation
among the four geopolymers.
4. Geopolymer produced using Gladstone fly ash had
a well-compacted dense ITZ, and no obvious difference
was observed compared to the bulk geopolymer matrix. In
contrast, Collie geopolymer had a porous ITZ. The unre-
acted/partially reacted fly ash particles were loosely packed
in this zone. The density of the ITZ was the key factor in
determining the tensile strength of the geopolymer concrete.
5. The 365-day elastic modulus and Poisson’s ratio of
four geopolymers ranged from 10.3 to 29 GPa (1493.5
to 4205 ksi) and 0.16 to 0.28. The significant increase of
elastic modulus has been observed between 90 and 365 days,
compared to the initial 28- to 90-day period for all concretes,
other than Collie. Tarong displayed a twofold increase of
elastic modulus on this period.
6. The packing density of the aluminosilicate gel matrix influ-
ences elastic modulus development. An increase in porosity and
microcracks was observed to negatively affect the compactness
of the gel matrix, which in turn affects the elastic modulus.
AUTHOR BIOS
Chamila Gunasekara is a PhD Student in the Civil, Environmental and
Chemical Engineering Department at RMIT University, Melbourne,
Australia. He received his BSc Eng degree from the University of Pera-
deniya, Sri Lanka, and his MEng (Struct.) degree from Saitama Univer-
sity, Saitama, Japan. His research interests include concrete materials and
geopolymer concretes.
Sujeeva Setunge is a Professor and Deputy Head of Civil Engineering
at RMIT University. She received her BSc Eng degree at the Univer-
sity of Moratuwa, Sri Lanka, and her PhD from Monash University,
Melbourne, Australia. Her research interests include mechanical
behavior, strengthening, creep, and shrinkage of reinforced concrete
and geopolymer concrete.
David W. Law is a Senior Lecturer at RMIT University. His research inter-
ests include the use of sustainable materials, and durability and material
properties of geopolymer concrete.
ACKNOWLEDGMENTS
The authors wish to express their thanks to Cement Australia Pty Ltd. and
Flyash Australia Pty Ltd for the supply of fly ash. The authors also wish to
acknowledge the X-ray facility and Microscopy & Microanalysis facility
provided by RMIT University and the scientific and technical assistance.
Scholarship provided by the School of Civil, Environmental and Chemical
Engineering to the first author is gratefully acknowledged.
REFERENCES
1. McLellan, B. C.; Williams, R. P.; Lay, J.; Van Riessen, A.; and Corder,
G. D., “Costs and Carbon Emissions for Geopolymer Pastes in Comparison
to Ordinary Portland Cement,” Journal of Cleaner Production, V. 19, No. 9,
2011, pp. 1080-1090. doi: 10.1016/j.jclepro.2011.02.010
2. Habert, G.; d’Espinose de Lacaillerie, J. B.; and Roussel, N., “An Envi-
ronmental Evaluation of Geopolymer Based Concrete Production: Reviewing
751
Current Research Trends,” Journal of Cleaner Production, V. 19, No. 11, 2011,
pp. 1229-1238. doi: 10.1016/j.jclepro.2011.03.012
3. Stengel, T.; Reger, J.; and Heinz, D., “LCA of Geopolymer Concrete—
What is the Environmental Benefit?” Proceedings of the 24th Biennial
Conference of the Concrete Institute of Australia, Sydney, NSW, Australia,
2009, pp. 54-62.
4. Hanjitsuwan, S.; Hunpratub, S.; Thongbai, P.; Maensiri, S.; Sata, V.;
and Chindaprasirt, P., “Effects of NaOH Concentrations on Physical and
Electrical Properties of High Calcium Fly Ash Geopolymer Paste,” Cement
and Concrete Composites, V. 45, No. 1, 2014, pp. 9-14. doi: 10.1016/j.
cemconcomp.2013.09.012
5. Chindaprasirt, P.; Rattanasak, U.; and Taebuanhuad, S., “Resistance to
Acid and Sulfate Solutions of Microwave-Assisted High Calcium Fly Ash
Geopolymer,” Materials and Structures, V. 46, No. 3, 2013, pp. 375-381.
doi: 10.1617/s11527-012-9907-1
6. Chindaprasirt, P.; Chareerat, T.; Hatanaka, S.; and Cao, T., “High-
Strength Geopolymer Using Fine High-Calcium Fly Ash,” Journal of Mate-
rials in Civil Engineering, ASCE, V. 23, No. 3, 2011, pp. 264-270. doi:
10.1061/(ASCE)MT.1943-5533.0000161
7. Chindaprasirt, P.; Chareerat, T.; Hatanaka, S.; and Cao, T., “High-
Strength Geopolymer Using Fine High-Calcium Fly Ash,” Journal of Mate-
rials in Civil Engineering, ASCE, V. 23, No. 3, 2011, pp. 264-270. doi:
10.1061/(ASCE)MT.1943-5533.0000161
8. Nematollahi, B., and Sanjayan, J. G., “Effect of Different Superplasti-
cizers and Activator Combinations on Workability and Strength of Fly Ash
Based Geopolymer,” Materials & Design, V. 57, No. 0, 2014, pp. 667-672.
doi: 10.1016/j.matdes.2014.01.064
9. Law, D. W.; Adam, A. A.; Molyneaux, T. K.; Patnaikuni, I.; and Ward-
hono, A., “Long Term Durability Properties of Class F Fly Ash Geopolymer
Concrete,” Materials and Structures, V. 48, No. 3, 2014, pp. 1-11.
10. Ryu, G. S.; Lee, Y. B.; Koh, K. T.; and Chung, Y. S., “The Mechan-
ical Properties of Fly Ash-Based Geopolymer Concrete with Alkaline
Activators,” Construction and Building Materials, V. 47, No. 1, 2013, pp.
409-418. doi: 10.1016/j.conbuildmat.2013.05.069
11. Diaz-Loya, E. I.; Allouche, E. N.; and Vaidya, S., “Mechanical
Properties of Fly-Ash-Based Geopolymer Concrete,” ACI Materials
Journal, V. 108, No. 3, May-June 2011, pp. 300-306.
12. Fernandez-Jimenez, A. M.; Palomo, A.; and Lopez-Hombrados, C.,
“Engineering Properties of Alkali-Activated Fly Ash Concrete,” ACI Mate-
rials Journal, V. 103, No. 2, Mar.-Apr. 2006, pp. 106-112.
13. Morrison, A. G. P., and Nelson, P., “Fly Ash Availability—Potential
Consequences of Transformation of Australia’s Energy Generation Port-
folio to 2050,” Proceedings of World of Coal Ash, Apr. 2005, pp. 56-63.
14. Bondar, D.; Lynsdale, C. J.; Milestone, N. B.; Hassani, N.; and
Ramezanianpour, A. A., “Engineering Properties of Alkali-Activated
Natural Pozzolan Concrete,” ACI Materials Journal, V. 108, No. 1,
Jan.-Feb. 2011, pp. 64-72.
15. Sathonsaowaphak, A.; Chindaprasirt, P.; and Pimraksa, K., “Work-
ability and Strength of Lignite Bottom Ash Geopolymer Mortar,” Journal
of Hazardous Materials, V. 168, No. 1, 2009, pp. 44-50. doi: 10.1016/j.
jhazmat.2009.01.120
16. Chindaprasirt, P.; Chareerat, T.; and Sirivivatnanon, V., “Workability
and Strength of Coarse High Calcium Fly Ash Geopolymer,” Cement and
Concrete Composites, V. 29, No. 3, 2007, pp. 224-229. doi: 10.1016/j.
cemconcomp.2006.11.002
17. Wang, H.; Li, H.; and Yan, F., “Synthesis and Mechanical Proper-
ties of Metakaolinite-Based Geopolymer,” Colloids and Surfaces. A, Phys-
icochemical and Engineering Aspects, V. 268, No. 1-3, 2005, pp. 1-6. doi:
10.1016/j.colsurfa.2005.01.016
18. Hardjito, D., and Rangan, B. V., “Development and Properties of
Low-Calcium Fly Ash-Based Geopolymer Concrete,” Curtin University of
Technology, Perth, Australia, 2005, pp. 1-103.
19. AS 3600, “Concrete Structures,” Standards Australia, 2009,
pp. 1-208.
20. Neupane, K.; Baweja, D.; Shrestha, R.; Chalmers, D.; and Sleep, P.,
“Mechanical Properties of Geopolymer Concrete: Applicability of Rela-
tionships Defined by AS 3600,” Concrete in Australia, V. 40, No. 1, 2014,
pp. 50-56.
21. AS 3582.1, “Supplementary Cementitious Materials for Use with
Portland and Blended Cement, Part 1: Fly Ash,” Standards Australia, 1998,
pp. 1-16.
22. AS 1141.5, “Methods for Sampling and Testing Aggregates,
Method 5: Particle Density and Water Absorption of Fine Aggregate,” Stan-
dards Australia, 2000, pp. 1-9.
23. Gunasekara, C.; Law, D. W.; Setunge, S.; and Sanjayan, J. G., “Zeta
Potential, Gel Formation and Compressive Strength of Low Calcium Fly
752
Ash Geopolymers,” Construction and Building Materials, V. 95, No. 1,
2015, pp. 592-599. doi: 10.1016/j.conbuildmat.2015.07.175
24. Neville, A. M., Properties of Concrete, Pearson Education Ltd.,
1996, 844 pp.
25. AS 1012.9, “Method of Testing Concrete, Method 9: Determination of
the Compressive Strength of Concrete Specimens,” Standards Australia, 1999,
pp. 1-13.
26. AS 1012.11, “Methods of Testing Concrete—Determination of the
Modulus of Rupture,” Standards Australia, 2000, pp. 1-5.
27. AS 1012.10, “Methods of Testing Concrete—Determination of Indi-
rect Tensile Strength of Concrete Cylinders (Brasil or Splitting Test),” Stan-
dards Australia, 2000, pp. 1-6.
28. AS 1012.17, “Methods of Testing Concrete—Determination of the
Static Chord Modulus of Elasticity and Poisson’s Ratio of Concrete Speci-
mens,” Standards Australia, 1997, pp. 1-14.
29. AS 1012.3.1, “Determination of Properties Related to the Consis-
tency of Concrete—Slump Test,” Standards Australia, 2014, pp. 1-8.
30. AS 1012.12.2, “Methods of Testing Concrete—Determination
of Mass per Unit Volume of Hardened Concrete—Water Displacement
Method,” Standards Australia, 1998, pp. 1-3.
31. Wardhono, A., “The Durability of Fly Ash Geopolymer and Alkali-
Activated Slag Concretes,” PhD thesis, RMIT University, School of Civil,
Environmental and Chemical Engineering, Melbourne, Australia, pp. 1-326.
32. Bakharev, T., “Geopolymeric Materials Prepared Using Class F Fly
Ash and Elevated Temperature Curing,” Cement and Concrete Research,
V. 35, No. 6, 2005, pp. 1224-1232. doi: 10.1016/j.cemconres.2004.06.031
33. Palomo, A.; Grutzeck, M. W.; and Blanco, M. T., “Alkali-Activated
Fly Ashes: A Cement for the Future,” Cement and Concrete Research,
V. 29, No. 8, 1999, pp. 1323-1329. doi: 10.1016/S0008-8846(98)00243-9
34. Warner, R. F.; Rangan, B. V.; Hall, A. S.; and Faulkes, K. A., Rein-
forced Concrete, Addison Wesley Longman, 1998, pp. 1-495.
35. Diaz, E. I.; Allouche, E. N.; and Eklund, S., “Factors Affecting the
Suitability of Fly Ash as Source Material for Geopolymers,” Fuel, V. 89,
No. 5, 2010, pp. 992-996. doi: 10.1016/j.fuel.2009.09.012
36. Steveson, M., and Sagoe-Crentsil, K., “Relationships between
Composition, Structure and Strength of Inorganic Polymers,” Journal
of Materials Science, V. 40, No. 16, 2005, pp. 4247-4259. doi: 10.1007/
s10853-005-2794-x
37. Demie, S.; Nuruddin, M. F.; and Shafiq, N., “Effects of Micro-struc-
ture Characteristics of Interfacial Transition Zone on the Compressive
Strength of Self-Compacting Geopolymer Concrete,” Construction
and Building Materials, V. 41, No. 1, 2013, pp. 91-98. doi: 10.1016/j.
conbuildmat.2012.11.067
38. Scrivener, K. L.; Crumbie, A. K.; and Laugesen, P., “The Interfacial
Transition Zone (ITZ) between Cement Paste and Aggregate in Concrete,”
Interface Science, V. 12, No. 4, 2004, pp. 411-421. doi: 10.1023/B:INTS.0
000042339.92990.4c
39. Sarker, P. K.; Haque, R.; and Ramgolam, K. V., “Fracture Behaviour
of Heat Cured Fly Ash Based Geopolymer Concrete,” Materials & Design,
V. 44, Feb., 2013, pp. 580-586. doi: 10.1016/j.matdes.2012.08.005
40. Liu, M. Y. J.; Alengaram, U. J.; Jumaat, M. Z.; and Mo, K. H., “Eval-
uation of Thermal Conductivity, Mechanical and Transport Properties of
Lightweight Aggregate Foamed Geopolymer Concrete,” Energy and Build-
ings, V. 72, No. 2014, pp. 238-245.
41. Puertas, F.; Palacios, M.; Manzano, H.; Dolado, J.; Rico, A.; and
Rodríguez, J., “A Model for the C-A-S-H Gel Formed in Alkali-Activated
Slag Cements,” Journal of the European Ceramic Society, V. 31, No. 12,
2011, pp. 2043-2056. doi: 10.1016/j.jeurceramsoc.2011.04.036
42. Gunasekara, C.; Law, D. W.; and Setunge, S., “Long Term Perme-
ation Properties of Different Fly Ash Geopolymer Concretes,” Construction
and Building Materials, V. 124, Oct. 2016, pp. 352-362. doi: 10.1016/j.
conbuildmat.2016.07.121
43. Williams, R. P.; Hart, R. D.; and Van Riessen, A., “Quantification of
the Extent of Reaction of Metakaolin-Based Geopolymers Using X-Ray
Diffraction, Scanning Electron Microscopy, and Energy-Dispersive Spec-
troscopy,” Journal of the American Ceramic Society, V. 94, No. 8, 2011,
pp. 2663-2670. doi: 10.1111/j.1551-2916.2011.04410.x
44. Kirschner, A., and Harmuth, H., “Investigation of Geopolymer
Binders with Respect to Their Application for Building Materials,” Ceram-
ics-Silikáty, V. 48, No. 11, 2004, pp. 7-20.
45. Duxson, P.; Mallicoat, S.; Lukey, G.; Kriven, W.; and Van Deventer, J.,
“The Effect of Alkali and Si/Al Ratio on the Development of Mechanical
Properties of Metakaolin-Based Geopolymers,” Colloids and Surfaces. A,
Physicochemical and Engineering Aspects, V. 292, No. 1, 2007, pp. 8-20.
doi: 10.1016/j.colsurfa.2006.05.044
ACI Structural Journal/May-June 2017
Reproduced with permission of copyright owner.
Further reproduction prohibited without
permission.