Construction and Building Materials: Wei-Hao Lee, Jhi-Hao Wang, Yung-Chin Ding, Ta-Wui Cheng
Construction and Building Materials: Wei-Hao Lee, Jhi-Hao Wang, Yung-Chin Ding, Ta-Wui Cheng
Construction and Building Materials: Wei-Hao Lee, Jhi-Hao Wang, Yung-Chin Ding, Ta-Wui Cheng
h i g h l i g h t s
a r t i c l e i n f o a b s t r a c t
Article history: The purpose of this study is to focus on the durability of geopolymer concrete after nine months of an
Received 26 October 2018 indoor and outdoor curing period. The geopolymer paste was prepared with fly ash and ground granu-
Received in revised form 23 March 2019 lated blast furnace slag as raw materials and sodium silicate/sodium aluminate as an alkali activator.
Accepted 24 March 2019
The geopolymer concrete was prepared with a 1:2.5:2.4 geopolymer:sand:gravel ratio. The influence of
Available online 28 March 2019
sodium aluminate, wollastonite additions and NaOH concentration on the microstructure, physical and
mechanical properties were evaluated. The compressive strength of the geopolymer concrete can reach
Keywords:
67 MPa and 53 MPa after 180 days of indoor and outdoor curing, respectively. Rapid Chloride Ion
Ground granulated blast furnace slag
Coal fly ash
Permeability Test (RCPT) shows the geopolymer concrete has excellent chloride resistance over pro-
Geopolymer longed curing time. After 180 days of accelerated wetting-drying cycles, the continued growth of com-
Concrete pressive strength indicates the good weathering resistance of geopolymer concrete. According to the
Durability test results obtained during this study, the geopolymer has a high potential to be a key material for civil
RCPT construction.
Ó 2019 Elsevier Ltd. All rights reserved.
https://doi.org/10.1016/j.conbuildmat.2019.03.291
0950-0618/Ó 2019 Elsevier Ltd. All rights reserved.
808 W.-H. Lee et al. / Construction and Building Materials 211 (2019) 807–813
cations are Si4+ and Al3+, and the divalent Ca2+ and Mg2+ act as net- Table 2
work modifiers along with any alkalis present [12]. The geopoly- Experimental design of geopolymer paste.
mer reaction starts with the destruction of the slag bonds (Si-O- Geopolymer FA,% GGBS, % Alkaline Solution S/L
Si, Al-O-Al and Al-O-Si) followed by the formation of hydration paste
SiO2/Na2O SiO2/Al2O3 NaOH, (M)
ratio
products.
GP 70 30 1.28 – 6 0.8
FA is mainly composed of quartz, hematite, mullite, amorphous GPA 50
silicate and aluminate [13]. According to previous research,
geopolymers can be prepared by mixing an alkali solution with
FA. This geopolymer paste has high mechanical strength and
2.2.2. Preparation of geopolymer concrete
durability.
Previous research experiments by our group show that a 1:1 ratio of GGBS:FA
Previous studies have shown the potential of using natural wol- would obtain better compressive strength results for geopolymer concrete. Based
lastonite (Wo) microfibers as a reinforcing material in cementi- on this information GGBS and FA were mixed at 1:1 ratio with an additional 3 wt%
tious materials [14,15]. Vickers discovered fibrous Wo (CaSiO3) of Wo added during blending. After 3 min pre-mixing, the mixture was then acti-
can effectively reduce the shrinkage of fly ash based geopolymer vated with alkali solutions of 1.28 SiO2/Na2O (with controlled NaOH concentra-
tions of 2 M, 3 M, 4 M. 5 M and 6 M) and 50 SiO2/Al2O3 molar ratios at 0.8 S/L
to 1%. This is because Wo can form a denser geopolymer structure
ratio. After 4 min of thorough mixing, sand and gravel were then added into
of low water dissipation and low shrinkage [16]. geopolymer paste for an additional 3 min blending to prepare geopolymer con-
Geopolymers generally only consider the characteristics of Si crete with geopolymer:sand:gravel ratio of 1:2.5:2.4. After the geopolymer con-
and Al in the system, with Als playing the most important role crete was removed from the molds, samples were divided into two separate
sections for indoor curing and outdoor curing until the testing date. For indoor
when considering long term durability. When calcium is added
curing, samples were put in plastic containers and the humidity was set at 90%
to the system, both hydration and polycondensation occur in the in room temperature. For outdoor curing, samples were moved to the roof of a
structures. Therefore, the purpose of this research was to focus five-story building so that they would be exposed to an uncontrolled outdoor
on the change durability, as well as the characteristics and environment. Once cured for 14, 28, 56, 90, 180 and 270 days, the cylindrical con-
microstructure changes of geopolymer concrete after nine months crete specimens (U 10 cm 20 cm) were subjected to physical and mechanical
properties tests. The experimental design of geopolymer concretes are listed in
of an indoor and outdoor curing period when GGBS and sodium
Table 3.
aluminum were introduced. In this study, geopolymer paste was
prepared using GGBS and FA as raw materials, and sodium hydrox-
2.3. Tests
ide, sodium aluminum and sodium silicates as alkali activators. The
microstructure of the geopolymer paste was analyzed using NMR. The 27Al and 29Si NMR analysis was performed to identify the microstructure of
The geopolymer concretes were prepared and tested for their geopolymer paste after 90 days of curing by using a Bruker 400 MHz Avance III
physical and mechanical properties, weathering resistivity and NMR spectrometer. The compressive strength test is conducted according to ASTM
chlorine permeability for up to 270 days of indoor and outdoor cur- C39/C39M-16b definition. The physical property of geopolymer concrete is mea-
sured according to CNS 619/R3013 definition. Rapid Chloride Ion Permeability test
ing conditions.
(RCPT) is according to ASTM C1202-12definition.
2. Experimental
3. Results and discussions
2.1. Materials
27 29
3.1. Al and Si NMR analysis of geopolymer paste
GGBS and FA were obtained from China Hi-Ment Resources Corporation in Tai-
wan. GGBS has a particle size between 0.6 135.7 lm with D50 at 12.3 lm. The FA
has a particle size between 0.7 201.9 lm with a D50 at 18.14 lm. The particle size The 27Al NMR spectrum results of GP and GPA (7 days curing)
of Wo is between 2.6 323.9 lm with a D50 at 44.0 lm. Particle size distribution are shown in Fig. 1. For GP cement, tetra-coordinated Al(4) atoms
results were determined by using a Microtrac Honeywell Particle Size Analyzer corresponds to the chemical shift of 56 ppm while hexagonal Al(4)
Model X-100. The chemical compositions of the GGBS, FA and Wo were analyzed coordination exhibits a chemical shift of about 4 ppm. The 27Al
according to CNS 12223 definition as listed in Table 1. The river sand used to pre-
pare geopolymer concrete has a particle size range between 150 and 850 lm with
spectrum of the original fly ash usually contains two signals, one
D50 at 250 lm. The gravel has a size range between 2.36 and 19.1 mm. The specific centered at 50 ppm for Al(4) and a second one at 4 ppm assigned
density of the river sand is 2.69 g/cm3 and aggregates is 2.72 g/cm3. to Al(6). These two signals are mainly associated with the presence
Alkali solutions with various SiO2/Na2O molar ratios were prepared by mixing of mullite. According to Davidovits study, the geopolymerization of
sodium silicate solution (9.5 wt% Na2O, 29 wt% SiO2) and sodium hydroxide. SiO2/
fly ash can shift the first resonance to higher values (56 ppm) with
Al2O3 molar ratio was kept 50 and controlled by sodium aluminate. The alkali solu-
tions were prepared and ready to use the day before experiment. Al(4Si) environment [17]. These are the characteristics of well con-
densed sialate networks in geopolymer. For GPA geopolymer, two
chemical shifts were found at 85 ppm and 34 ppm corresponding
2.2. Methods
with tetra-coordinated Al(4) and penta-coordinated Al(5) [18]. This
2.2.1. Preparation of geopolymer paste is probably because the aluminum silicate in the alkali solution
Geopolymer paste was synthesized by adding GGBS and FA mixture with 30/70 shifts the resonance to the higher frequency.
ratio into an alkali solution at 0.55 S/L ratio. As shown in Table 2, geopolymers of Fig. 2 shows 29Si NMR spectrum of GP and GPA. For GP activated
(GP) and (GPA) were prepared by pre-mixing GGBS and FA for 3 min and then
mixed with the designated alkali solution for another 4 min. After thorough mixing,
with 1.28 SiO2/N2O solution, a broad featureless 29Si signal cen-
geopolymer paste was cast in a U10 cm 20 cm cylinder mold and left in ambient tered at 88 ppm was identified. The width of the 29Si signal spans
temperature for 3 days. The cast geopolymer paste was then removed from the between 80 ppm and 100 ppm indicating a full range of com-
mold and further cured for 90 days before NMR analysis. pletely condensed siloxane units [18]. The Q4(0Al) was also found
to correspond with a chemical shift of 104 ppm. For the GPA
specimen activated with 1.28 SiO2/N2O and 50 SiO2/Al2O3 solution,
Table 1
Chemical composition of GGBS, FA and Wo. a sharp 29Si signal of Q4(4Al) centered at 83 ppm and Q4(0Al) signal
corresponds to the chemical shift of 106 ppm. Because the 29Si
Chemical compositions SiO2 Al2O3 Fe2O3 CaO K2O Others
signals are broadened and overlap each other, the individual signal
GGBS 26.6 11.4 0.4 58.6 0.6 2.4 cannot be clearly identified, the Gaussian curve fitting of 29Si spec-
FA 69.3 13.3 7.9 5.1 2.2 3.2
trum using Peak fit software developed by Jandel Scientific was
Wo 37.5 2.3 0.8 58.1 – 1.3
used to calculate the relative deconvoluted peak areas of each
W.-H. Lee et al. / Construction and Building Materials 211 (2019) 807–813 809
Table 3
Experimental design of geopolymer concrete.
45
40 1.28 SiO2/N2O, 50
SiO2/Al2O3
35
25
20
15
10
0
Q401 Q41 2 Q42 3 Q43 4 Q44 5 Q1 6 Q2 7
Fig. 3. Fraction area of silicon sites presented as Q44, Q43, Q42, Q41, Q40, Q1 and Q2.
29
Fig. 2. Si NMR spectrum of GP and GPA.
individual 29Si unit. As shown in Fig. 3, the fraction area of Q44 for
GP and GPA are 40% and 15.9%, respectively. This can be attributed
to the addition of aluminum silicate into alkali solution to promote
the formation of Q44 structure. The higher fraction area of Q44 indi-
cates a denser geopolymer structure. For GP specimen, fraction
areas of Q4(0-4Al) are all less than 20%. Davidovits indicated the
more Q34、Q24 and Q04 signifies a higher degree of alkali activated
reactions and may result in geopolymer of lower strength and Fig. 4. Effect of curing day and curing environment on the compressive strength of
durability [17]. geopolymer concretes.
810 W.-H. Lee et al. / Construction and Building Materials 211 (2019) 807–813
27
Fig. 5. Al NMR spectrum of GCW6. Fig. 7. Chloride ion penetration of geopolymer concrete.
W.-H. Lee et al. / Construction and Building Materials 211 (2019) 807–813 811
of GC6 and GCW6 concrete at different curing stages. According to 55.4 MPa as test cycles increase from 14 to 180 days. This result
ASTM standard, the charge passed (coulombs) over 4000 is consid- shows GCW6 has very high weathering resistance while still con-
ered to have very high penetrability; 2000–4000 is moderate; tinues its geopolymer reaction even under severe weathering con-
1000–2000 is low and 100–1000 is very low. After 28 days of cur- dition. Fernandez-Jimenez et al. [33] indicated the increase of
ing, all specimens have a total charge lower than 4000 coulombs. strength could be attributed to the crystallization of aluminosili-
The total charge decreases as the curing days increase for all spec- cates that contain excellent anti-weathering characteristics. The
imens. This is because longer curing periods result in denser XRD spectrum of GCW6 shows gismondine phase (CaAl2SiO2O8H2-
geopolymer matrixs, and thus, reduce the chloride ion penetrabil- O) after 90 days of curing as shown in Fig. 9. The gismondine phase
ity. Miranda et al. [26] also indicated the more dense sodium sili- toward the C-A-S-H structure is the probable cause for strength
cate activated fly ash geopolymer correspond to greater improvement [33]. Olivia and Nikraz [34] also indicated the
impermeability and may retard the diffusion-driven penetration repeated drying at 80 °C for up to 100 cycles may cause structural
of chloride ions. Chindaprasirt and Chalee [27] found that increas- and phase composition change that increase the strength of
ing NaOH concentrations in geopolymer concrete decreased the geopolymer concrete.
chloride diffusion coefficient and chloride penetration in geopoly-
mer concrete. Aravindan et al. [28] pointed out the reduction of
3.3. Effect of NaOH concentration on the compressive strength of
chloride ion penetration over a longer curing period. Generally,
geopolymer concrete
the GCW6 specimens have lower coulombs values than GC6 spec-
imens. This is also because the effect of Wo can reduce the shrink-
In the process of geopolymer reaction, NaOH plays an important
age and micro-cracks of geopolymer concrete. The total charge
role on both the compressive strength and structure of geopoly-
passed of GCW6 dropped to as low as 759 and 1279 coulombs after
mers. The NaOH concentration in the aqueous phase of the
cured for 270 days under indoor and outdoor environments,
geopolymer system acts on the dissolution process, as well as on
respectively. However, different results were found elsewhere.
the bonding of solid particles in the final structure [35]. Under a
Zhu et al. [29] observed that alkali-activated fly ash paste and mor-
strong alkali solution, Si and Al ions can be dissolved from alumi-
tar had higher chloride penetration rates than that of Portland
nosilcate materials and form SiO4 and AlO4 tetrahedral units [17].
cement. The porosity and tortuosity are the two most significant
Fig. 10 shows the compressive strength of geopolymer concretes
factors affecting the chloride penetration. Pasupathy et al. [30]
prepared with alkali solution of different NaOH concentrations,
reported that chloride penetration in the geopolymer concrete
as listed in Table 3. The GCW4 (4 M NaOH) have higher strengths
was high in saline environments. Law et al. [31] reported that
at each curing phase and curing environment within those pre-
the RCPT may not be a suitable method for assessing the chloride
pared with 3 M and 5 M NaOH alkali solution. As shown in
resistance of geopolymer concretes due to rapid heating that was
Fig. 10(a), GCW4 strength increases from 51.2 MPa to 63.7 MPa
found to reach 60 °C before the end of the test. A similar phe-
as curing time increases from 14 to 56 days. From 56 to 180 days,
nomenon occurred in this study. After RCPT started, the tempera-
GCW4 only gains another extra 3.3 MPa. The same observation was
ture increased sharply. This may be because of the high level of
also found on GCW3 and GCW5. These results indicate the com-
sodium ions in the geopolymer system affecting the conductivity.
pressive strength development of geopolymer concrete is practi-
This brings into question whether or not the RCPT method is suit-
cally completed within the first 56 days of curing. GCW3 has the
able for measuring geopolymer durability. Further research works
lowest compressive strength in all four specimens tested. This is
and discussion are still needed.
because the low NaOH concentration solution dissolved insuffi-
In order to evaluate the durability of concrete exposed to sea-
cient Al and Si ions to form SiO4 and AlO4 tetrahedral units and
water conditions, geopolymer concretes were subjected to acceler-
results in lower cementing concrete structure [36]. Similar results
ated wetting–drying cycles, developed by Kasaiand and Nakamura
also found elsewhere [37-39]. However, GCW5 was prepared with
[32], in Na2SO4 solution followed by compressive strength tests.
higher NaOH concentration solution, but has lower compressive
GCW6 was immersed in 5% Na2SO4 solution for 24 h followed by
strength than GCW4. Zuhua et al. [40] suggested when OH con-
24 h drying at 80 °C. Fig. 8 shows the compressive strength of
centration was high enough, dissolution of Si and Al ions was
GCW6 after subjecting to 14 to 180 days wetting-drying cycles.
accelerated, but polycondensation was hindered. Lee and Deventer
The compressive strength of GCW6 increases from 42.3 MPa to
[41] also indicated that excess hydroxide ion concentration can
4. Conclusions
29
Si spectra of geopolymer cement prepared with 1.28 SiO2/Na2-
(A) indoor curing O, 50 SiO2/Al2O3 alkali solution has higher fractions of Q44 due to
more Si and Al ions leached from GGBS and FA and formed a denser
C-A-S-H structure.
The compressive strength of GCW6 concretes steadily increases
from 20.4 MPa to 58.6 MPa as the curing period increases from 14
to 270 days. This is because the Wo microfibers can bridge micro-
cracks and improve the bonding strength of geopolymer concrete.
According to RCPT test, the total charge passed through
geopolymer concrete dropped to 759 coulombs after cured for
270 days which is equivalent to latex-modified concrete and inter-
nally sealed concrete. This is because the longer curing period can
result in denser geopolymer matrix and the reduced penetrability
of the chloride ion.
The wetting–drying cycle tests show GCW6 has very high
weathering resistance and can still continue its geopolymer reac-
tion even under severe weathering conditions. The compressive
strength of GCW6 specimens increases from 42.3 MPa to
55.4 MPa as test cycles increase from 14 to 180 days.
By varying the NaOH concentration of alkali solution, GCW4
(4 M NaOH) has the highest compressive strength (67 MPa) after
(B) outdoor curing 180 days of curing. This is because higher NaOH concentrations
can result in higher dissolution of Al and Si ions and the formation
Fig. 10. Influence of NaOH concentration on the compressive strength of geopoly-
mer concrete after 180 days (A) indoor curing and (B) outdoor curing.
of alumino-silicate leading to an increase in strength. According to
the test results obtained in this study shows geopolymer concrete
has great potential for further development in civil construction
cause aluminosilicate gel precipitation at very early stage and hin- application.
der subsequent geopolymerization.
Fig. 10(b) shows the compressive strength of the same concrete
Conflict of interest
specimens cured under an uncontrolled outdoor environment.
None.
References
[7] I. Lecomte, C. Henrist, M. Liégeoisa, F. Maserib, A. Rulmonta, R. Clootsa, [25] D. Khale, R. Chaudhary, Mechanism of geopolymerization and factors
(Micro)-structural comparison between geopolymers, alkali-activated slag influencing its development: a review, J. Mater. Sci. 42 (2007) 729–746.
cement and Portland cement, J. Eur. Ceram. Soc. 26 (2006) 3789–3797. [26] J.M. Miranda, A. Fernandez-Jimenez, J.A. Gonxales, A. Palomo, Corrosion
[8] G.S. Ryu, Y.B. Lee, K.T. Koh, Y.S. Chung, The mechanical properties of fly ash- resistance in activated FA mortars, Cem. Concr. Res. 35 (6) (2005) 1210–1217.
based geopolymer concrete with alkaline activators, Constr. Build. Mater. 47 [27] P. Chindaprasirt, W. Chalee, Effect of sodium hydroxide concentration on
(2013) 409–418. chloride penetration and steel corrosion of fly ash-based geopolymer concrete
[9] M.A. Criado, A. Fernández-Jiménez, A.G. dl Torre, A. Palomo, An XRD study of under marine site, Constr. Build. Mater. 63 (2014) 303–310.
the effect of the SiO2/Na2O ratio on the alkali activation of fly ash, Cem. Concr. [28] S. Aravindan, N. Jagadish, A. Peter Guspher, Experimental investigation of
Res. 37 (2007) 671–679. alkali-activated slag and fly ash based geopolymer concrete, J. Eng. Appl. Sci.
[10] M. Komljenovi, Z. Bacarevi, V. Bradic, Mechanical and microstructural 10 (10) (2015) 4701–4705.
properties of alkali-activated fly ash geopolymers, J. Hazard. Mater. 181 [29] H. Zhu, Z. Zhang, Y. Zhu, L. Tian, Durability of alkali-activated fly ash concrete:
(2010) 35–42. chloride penetration in pastes and mortars, Constr. Build. Mater. 65 (2014) 51–
[11] P. Duxson, A. Fernández-Jiménez, J.L. Provis, G.C. Lukey, A. Palomo, J.S.J. van 59.
Deventer, Geopolymer technology: the current state of the art, J. Mater. Sci. 42 [30] K. Pasupathy, M. Berndt, J. Sanjayan, P. Rajeev, D.S. Cheema, Durability of low
(9) (2006) 2917–2933. calcium fly ash based geopolymer concrete culvert in a saline environment,
[12] C. Li, H. Sun, L. Li, A review: The comparison between alkali-activated slag (Si Cem. Concr. Res. 100 (2017) 297–310.
+Ca) and metakaolin (Si+Al) cements, Cem. Concr. Res. 40 (2010) 1341–1349. [31] D.W. Law, A.A. Adam, T.K. Molyneaux, I. Patnaikuni, A. Wardhono, Long term
[13] J. Temuujin, A. van Riessen, R. Williams, Influence of calcium compounds on durability properties of class F fly ash geopolymer concrete, Mater. Struct. 48
the mechanical properties of fly ash geopolymer pastes, J. Hazard. Mater. 167 (3) (2014) 721–731.
(1–3) (2009) 82–88. [32] Y. Kasai, N. Nakamura, Accelerated Test Method for Durability of Cement
[14] N.M.P. Low, J.J. Beaudoin, Mechanical properties of high performance cement Mortars in Sea Water, Proceedings of international conference on performance
binders reinforced with wollastonite microfibers, Cem. Concr. Res. 22 (5) of concrete in marine environment, St Andrews, 1980, pp. 379–396.
(1992) 981–989. [33] A. Fernandez-Jimenez, I. García-Lodeiro, A. Palomo, Durability of alkali-
[15] N.M.P. Low, J.J. Beaudoin, The flexural toughness and ductility of Portland activated fly ash cementitious materials, J. Mater. Sci. 42 (9) (2007) 3055–
cement-based binders reinforced with wollastonite microfibers, Cem. Concr. 3065.
Res. 24 (2) (1994) 250–258. [34] M. Olivia, H. Nikraz, Properties of fly ash geopolymer concrete designed by
[16] L. Vickers, W. Rickard, A. van Riessen, Strategies to control the high Taguchi method, Mater. Des. 36 (2012) 191–198.
temperature shrinkage of fly ash based geopolymers, Thermochim. Acta 580 [35] D. Panias, I.P. Giannopoulou, T. Perraki, Effect of synthesis parameters on the
(2014) 20–27. mechanical properties of fly ash-based geopolymers, Colloids Surf. A:
[17] J. Davidovits, in: Geopolymer Chemistry and Applications, Geopolymer Physiochem. Eng. Aspects 301 (2007) 246–254.
Institute, France, 2008, pp. 61–76. [36] F. Puertas, S. Martinez-Ramirez, S. Alonso, T. Vazquez, Alkali-activated fly ash/
[18] G.A. Webb, Annual Reports NMR Spectroscopy, 80 2016, 90 92. slag cement. Strength behaviour and hydration products, Cem. Concr. Res. 30
[19] M. Mathur, A.K. Misra, P. Goel, Influence of wollastonite on mechanical (2000) 1625–1632.
properties of concrete, J. Sci. Ind. Res. 66 (2007) 1029–1034. [37] P. Chindaprasirt, C. Jaturapitakkul, W. Chalee, U. Rattanasak, Comparative
[20] P.S. Deb, P. Nath, P.K. Sarker, Drying shrinkage of slag blended fly ash study on the characteristics of fly ash and bottom ash geopolymers, Waste
geopolymer concrete cured at room temperature, Procedia Eng. 125 (2015) Manage. 29 (2009) 539–543.
594–600. [38] U. Rattanasak, P. Chindaprasirt, Influence of NaOH solution on the synthesis of
[21] N.M.P. Low, J.J. Beaudoin, Flexural strength and microstructure of cement fly ash geopolymer, Miner. Eng. 22 (12) (2009) 1073–1078.
binders reinforced with wollastonite microfibers, Cem. Concr. Res. 23 (4) [39] X. Guo, H. Shi, W.A. Dick, Compressive strength and microstructural
(1993) 905–916. characteristics of class C fly ash geopolymer, Cem. Concr. Compos. 32 (2)
[22] R. Hameed, A. Turatsinze, F. Duprat, A. Sellier, Metallic fiber reinforced (2010) 142–147.
concrete: effect of fiber aspect ratio on the flexural properties, J. Eng. Appl. Sci. [40] Z. Zuhua, Y. Xiao, Z. Huajun, C. Yue, Role of water in the synthesis of calcined
4 (5) (2009) 67–72. kaolin-based geopolymer, Appl. Clay Sci. 43 (2) (2009) 218–223.
[23] N. Banthia, J. Sheng, Fracture toughness of micro-fiber reinforced cement [41] W.K. Lee, J.S.J. van Deventer, The effects of inorganic salt contamination on the
composites, Cem. Concr. Compos. 18 (4) (1996) 251–269. strength and durability of geopolymer, Colloids Surf. A 211 (2–3) (2002) 115–
[24] D. Hardjito, S.E. Wallah, D.M.J. Sumajouw, B.V. Rangan, Fly ash-based 126.
geopolymer concrete, Aust. J. Struct. Eng. 6 (1) (2005) 77–86.