Materials and Design 74 (2015) 125–137

Contents lists available at ScienceDirect

Materials and Design

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

An investigation of the microstructure and durability of a fluidized bed

fly ash–metakaolin geopolymer after heat and acid exposure
Ping Duan a,b,c,⇑, Chunjie Yan a,b,c, Wei Zhou a,b, Wenjun Luo a,b, Chunhua Shen d
a
Faculty of Materials Science and Chemistry, China University of Geosciences, Wuhan 430074, China
b
Engineering Research Center of Nano-Geomaterials of Ministry of Education, China University of Geosciences, Wuhan 430074, China
c
Zhejiang Research Institute, China University of Geosciences, Hangzhou 311305, China
d
Center of Materials Research and Analysis, Wuhan University of Technology, Wuhan 430070, China

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

Article history: This study aims to investigate durability and microstructure of fluidized bed fly ash and metakaolin based
Received 28 November 2014 geopolymer exposed to elevated temperatures and acid attack. Geopolymer specimens were prepared by
Revised 5 March 2015 combination of fly ash and metakaolin activated by sodium silicate and sodium hydroxide solutions and
Accepted 7 March 2015
were cured in microwave radiation environment plus a heat curing period. Compressive strength and
Available online 9 March 2015
several key durability parameters for geopolymer and ordinary Portland cement (OPC) were assessed
and compared. Microstructure formation and development was characterized in terms of morphology
Keywords:
and pore structure as well as simulation.
Geopolymer
Fly ash
The experimental results reveal a dense microstructure of geopolymer compared to OPC. In terms of
Metakaolin resistance to the acidic solution and elevated temperatures, geopolymer is superior to OPC as indicated
Durability by the relatively lower strength loss and lower mass change. Compressive strength shows a dramatic
OPC drop in OPC while geopolymer shows a strength increase after 400 °C. The mass loss curves of geopoly-
mer are similar to OPC, but it shows relatively lower mass loss compared to OPC. The result of saturated
water absorption after 28 days curing indicates less water absorption in geopolymer before and after
thermal and acid exposure. Durability of geopolymer is demonstrated by monitoring the pore structure.
Crown Copyright Ó 2015 Published by Elsevier Ltd. All rights reserved.

1. Introduction furnace. It normally contains more than 95% of glass. Generally,

they are ground to fine powder, called ground granulated blast
Cement production is associated with the emission of consider- furnace slag), metakaolin (produced by calcining kaolin at
able amount of greenhouse gas [1]. The amount of carbon dioxide 650–800 °C. The main components are amorphous Al2O3 and
released in the manufacture of cement is about one ton for one ton SiO2 with high pozzolanic activity. Besides the filling effect, meta-
cement clinker. Therefore, it is considered vital to search for alter- kaolin reacts with calcium hydroxide, which is one of the hydration
native low CO2 emission binders for concrete in order to reduce its products of Portland cement, to form calcium silicate hydrate gels)
carbon footprint and the development of alternative binders utiliz- or a combination of them instead of cement and results in less CO2
ing industrial by-products is one of existing strategies. emission to reduce the environmental impact of the cement
Geopolymer is such an emerging alternative binder, which is production.
prepared using by-product materials such as fly ash (by-product During the last decade, increased research efforts [2–9] have
of coal combustion in thermal power plants with an average size been directed to this area due to the wide range of potential appli-
of less than 20 lm and low bulk density (0.54–0.86 g/cm3), high cations of these by-product materials, which has gained increasing
surface area (300–500 m2/kg) and light texture with spherical in attention due to the energy conservation, economic and environ-
shape and consist of solid spheres, cenospheres, irregular-shaped mental considerations.
debris and porous unburnt carbon), blast furnace slag (produced By-products of industry are some of the most complex and
when iron ore is reduced by coke at about 1350–1550 °C in a blast abundant of anthropogenic materials. They cause water and soil
pollution, disrupt ecological cycles and environmental hazards
[10]. The current worldwide annual production of fly ash, one of
⇑ Corresponding author at: Faculty of Materials Science and Chemistry, China such by-products, is estimated around 750 million tones [11]. At
University of Geosciences, Wuhan 430074, China.
present, only a minor part of this material is utilized (20–30%) on
E-mail address: dp19851128@sina.com (P. Duan).

http://dx.doi.org/10.1016/j.matdes.2015.03.009
0261-3069/Crown Copyright Ó 2015 Published by Elsevier Ltd. All rights reserved.
126 P. Duan et al. / Materials and Design 74 (2015) 125–137

worldwide basis, while the rest is still disposed of in landfills, thus high-temperature applications. They estimated that the structure
contributing to the pollution of soil, water and air. was stable enough to resist to high temperatures.
The environmental impact of coal fly ash is well known for its On balance, ash-based geopolymer was prepared based on a
massive generation, large usage of land for disposal and short precursor derived from fly ash, generally consisting of spherical
and long term impact on surrounding areas. The principal environ- particles, glassy or amorphous as well as crystalline phases.
mental concern stems from the possible leaching of heavy metals Recently, increased utilization of fluidized bed technology led to
and organic compounds and their migration into ground water or the production of a great amount of fluidized bed fly ash including
nearby surface water. In addition, fly ash could also affect human little glassy spherical particles, which forms by burning coal
health through direct inhalation or ingestion of airborne or settled gangue in a low temperature. Circulation fluidized bed combustion
ash. The unproductive use of land and its maintenance results in is an advanced, clean and reliable coal firing technology for power
long-term financial burden [12]. generation. There are many unreacted CaO, desulphurized prod-
Aggressive efforts have been undertaken recently to recycle fly ucts CaSO4 and a little CaCO3 remaining in the fluidized bed fly ash.
ash [13–17] in concrete production as mineral admixture, soil Coal fly ash is typically found in the form of coarse bottom ash
amendment, zeolite synthesis, and as filler in polymers. However, and fine fly ash, which represent 5–15 and 85–95 wt% of the total
these applications are not sufficient for complete utilization of fly ash, respectively. Bottom ash refers to the ash that falls down
ash, thereby it is imperative to develop new recycling techniques through the airflow to the bottom of the boiler and is mechanically
for fly ash. removed. The term coal fly ash is often used to refer to fine fly ash,
The concept of sustainable solutions for fly ash is closely linked particles of which are captured from flue gas and collected by elec-
with technologies with aligned vision for environment, economy trostatic or mechanical precipitation.
and societal goals. Fly ash was widely used as the source material Bottom ash is a coarse, granular, incombustible by-product that
to manufacture geopolymer products owing to its aluminosilicate is collected from the bottom of furnaces that burn coal for the gen-
composition, fine size, significant amount of glassy content and eration of steam, the production of electric power, or both. Bottom
availability across the world [14,18–20]. This is one of the existing ash is coarser than fly ash, with grain sizes spanning from fine sand
important strategies and it is believed to be the sustainable solu- to fine gravel.
tion for utilization of fly ash. It will consume large part of fly ash In general, fluidized bed fly ash differs from coal fly ash in terms
and relieve the pollution of soil, water and air. of particle shape, chemical composition and amorphous phase con-
The term geopolymer was introduced in 1970 by Davidovits, tent. Fluidized bed fly ash particles are approximately 1–300 lm in
who made a significant breakthrough in understanding and size, with irregular shape, while coal fly ash particles are normally
development of binders from metakaolin and alkaline metal slightly finer at approximately 1–200 lm in size and the content of
solutions. Geopolymer was originally applied to three-dimensional amorphous phase is usually higher than that of fluidized bed fly
aluminosilicate materials formed by condensation of a solid alumi- ash. Coal fly ash is widely used as pozzolanic material for partial
nosilicate source such as dehydroxylated kaolinite (metakaolin) replacement of Portland cement due to its spherical shape and high
with an alkali silicate solution under highly alkaline conditions [21]. reactivity. Comparing with coal fly ash, fluidized bed fly ash has
This new material was likely to have enormous potential to higher contents of lime (CaO), gypsum (CaSO4) and crystalline
become an alternative to Portland cement and it is receiving phase thus its usage as pozzolanic material is limited. The unique
increased attention due to the need of new binders with enhanced thermal history, featuring low combustion temperatures of
durability performance [22–28]. 800–950 °C, makes fluidized bed fly ash differ greatly in physical
Although different source materials can be used to prepare and chemical characteristics from coal fly ash, whose typical firing
geopolymer binders, fly ash, which provides the greatest opportu- temperatures are 1200–1400 °C.
nity for commercial utilization and has the potential to reduce the It is well acknowledged that compositions and structure of
carbon footprint, has been extensively used and found to be the precursors have significant effects on performance of geopolymer.
most practical source material suitable for concrete applications Previous research efforts about the relevant topic mainly focused
due to the plentiful worldwide raw material supply, which is on cement paste [58–70], or preparation and properties of geopoly-
derived from coal-fired electricity generation [29]. mer but below 1000 °C, and acid attack was not mentioned
Excellent properties of fly ash-based geopolymer concrete have [71–76]. Existing references on the utilization of fluidized bed fly
been reported in the last decades [30–38]. Fly ash-based geopoly- ash to prepare geopolymer and the durability aspects investigation
mer concrete has properties favourable for its potential use as a are scant, and the ash used belonged to high calcium fly ash, there
cementitious material due to excellent durability aspects. Some have been very few published references on fluidized bed fly ash
authors [39–43] have reported good engineering properties of based geopolymer exposed to elevated temperature at 1000 °C
geopolymer concrete that were favourable for its use as a construc- from ambient temperature and acid attack.
tion material. In the present study, geopolymer specimens were prepared by
Geopolymer can be composed from metakaolin or wastes, such combination of fluidized bed fly ash and metakaolin activated by
as fly ash, slag, and tailing [44–48]. Geopolymer can be used as alkaline activator and were cured in microwave radiation environ-
building materials and fillers due to their flameproof characteris- ment. Compressive strength was investigated and several key
tics [49–51]. In addition, geopolymer has several other advantages durability parameters for geopolymer and ordinary Portland
such as high strength, acid/alkaline resistance and heat resistance cement (OPC) were assessed and compared. Microstructure forma-
[44]. tion and development was characterized in terms of morphology,
Having more outstanding mechanical properties and envi- pore structure and simulation.
ronmental friendliness, geopolymer, is considered to be a new
cementing material with widely potential application value in con- 2. Experimental
struction. With the aim of reducing consumption of non-renewable
raw materials whilst increasing the use of industrial by-products 2.1. Materials
(residue), research has recently focused on the alkali activation of
metakaolin and fly ash. Fluidized bed fly ash was provided by Shenhua Junggar Energy
Some authors [52–57] explored some basic aspects of metakao- Corporation in Junggar, Inner Mongolia, China. Metakaolin was
lin or fly ash-based geopolymer activated using sodium alkali for obtained from Yunnan, China. The chemical analysis of those raw
 P. Duan et al. / Materials and Design 74 (2015) 125–137 127

Table 1
Chemical compositions of raw materials by XRF analysis (mass, %).

SiO2 Al2O3 Fe2O3 MgO CaO Na2O K2O MnO TiO2 LOIa
Cement 21.35 7.67 3.31 3.08 62.60 0.35 0.39 0.05 0.25 0.95
Fly ash 29.47 51.72 2.25 0.15 5.21 0.05 0.35 0.03 1.83 8.58
Metakaolin 53.32 42.09 2.33 0.21 0.09 0.49 0.64 0.02 0.63 0.08
a
LOI: mass loss of dried sample after calcined between 1000 and 1100 °C.

Fig. 1. The micrographs (SEM) of fluidized bed fly ash.

materials mentioned above was listed in Table 1 and their micro- All samples were calcined from ambient temperature to 1000 °C
graphs were shown in Figs. 1 and 2, respectively. Alkali activator in an electric muffle furnace with a heating rate of 2 °C/min and
was a combination of sodium silicate and sodium hydroxide then cooled to room temperature. Sample characterization was
(99.2% NaOH) in analytical reagent degree. The liquid portions in conducted on geopolymer and OPC specimens to evaluate the
the mixture were 10 M sodium hydroxide (NaOH) and sodium effects of the elevated temperature exposure on properties of
silicate (Na2SiO3) with 14.51% Na2O, 33.39% SiO2, and 48.53% H2O. them.
Portland cement (CEM I 42.5) (relative density 3100 kg/m3,
specific surface area 369.6 m2/kg) was used as binder and was
compared to geopolymer. The properties of cement were shown 2.2.1. Compressive strength
in Table 2. After casting, the molds containing the specimens were covered
with a plastic sheet and stored in the laboratory environment
(20 °C and 90% RH) for 24 h. The paste specimens with size of
2.2. Test procedure 40 mm  40 mm  40 mm were then removed from the molds.
The compressive strength was determined using a universal
Geopolymer specimens were synthesized by alkaline-activation testing device with a loading capacity of 3000 kN. The loading rates
of combination of fluidized bed fly ash and metakaolin with mass applied during the compression tests was 0.6 MPa/s. For each
ratios of 1.0 in alkali silicate solutions (MR = 1.6, MR means mixture, the compressive strength was measured at the age of
modulus of alkaline activator). The liquid/solid (L/S) mass ratio acquired days.
was kept constant and equaled to 0.4 (the liquid consists of water
in alkaline solutions and the extra water, the solid materials consists
of fly ash, metakaolin or a combination of them), which depended 2.2.2. Drying shrinkage
on an acceptable workability for each paste sample. Fresh geopoly- Paste specimens of the size 20  20  200 mm3 were prepared
mer pastes and cement paste were cast in triplet steel cubes molds for the drying shrinkage test. The specimens were removed from
with size of 40  40  40 mm3 and 20  20  20 mm3, respectively, the molds after being cured for 24 h. The prism specimens were
and vibrated to remove entrained air bubbles. The molds were then installed onto the setup for the length change tests and cured in
sealed with polyethylene film and set into a standard curing box in a room with constant temperature and relative humidity
microwave radiation environment for 5 min. After initial curing at (20 ± 3 °C, RH = 90 ± 5%). Length changes of the prism before and
40 °C for 1 day, the samples were released from the molds and were after high temperature exposure were recorded by reading the dial
subjected to further curing at 40 °C for acquired ages. gauge regularly [77].
128 P. Duan et al. / Materials and Design 74 (2015) 125–137

Fig. 2. The micrographs (SEM) of metakaolin.

Table 2 2.2.6. Micrographs test

The properties of cement. The morphological changes that occurred during thermal expo-
Compressive Flexural Initial setting time Final sure were obtained by means of JSM-5610LV scanning electronic
strength (MPa) strength (MPa) (min) setting microscopy (SEM). The specimens for morphology observation
3 days 28 days 3 days 28 days
time (min) were cut into prisms of about 8  8  2 mm3.

27.3 47.5 6.3 8.7 132 187

3. Results and discussion

2.2.3. Acid attack 3.1. Temperature profile

Cement and geopolymer pastes were placed in a cylinder mold
with size of Ø 100 mm  50 mm. Test pieces were kept in a moist It is believed that geopolymer possesses good thermal stability
room. After 24 ± 2 h curing, the test pieces were removed from the property due to fire resistance stems from ceramic-like properties.
molds. The test pieces were immerged in the sulfuric acid (2%) plus To evaluate thermal resistance of the investigated samples and the
hydrochloric acid (2%) solutions for 28 days. After the immersion influence of high-temperatures on microstructure, dimension sta-
test, compressive strength loss compared to the original strength bility and mechanical properties, all samples were calcined from
was measured and appearance of the exposure surfaces was room temperature to 1000 °C in an electric furnace with a heating
observed to evaluate the resistance to acid attack. rate of 2 °C/min and then cooled naturally to room temperature.
Sample characterization was conducted on untreated and calcined
2.2.4. Water sorptivity samples to evaluate the effects of the elevated temperature expo-
Water sorptivity test was carried out on specimens with size of sure on OPC and geopolymer with size of 40  40  40 mm3 and
Ø100 mm  50 mm for both cement and geopolymer paste. The 20  20  20 mm3, respectively.
specimens were oven dried for 3 days at 50 °C followed by 28 days The temperature of the air inside the furnace was measured by
sealed storage. After this curing stage, the specimens were an in-built thermocouple. The temperature at the centre of the spe-
immerged into deionized water for several days and the mass cimen was measured by a K-type thermocouple inserted in the
changes before and after immerged into water were measured to specimens during casting. The thermocouples were connected to
evaluate the resistance to water permeability. electronic data loggers that recorded the measured temperatures.
The result provided in Fig. 3 reveals the temperature evolution
2.2.5. Mercury intrusion porosimetry at different distances from the heated surface of a cubic OPC and fly
The porosity and pore size distribution of paste were measured ash–metakaolin geopolymer to the center of paste specimens
by mercury intrusion porosimetry (MIP, AutoPore IV 9500 type) during heating process. Maximum thermal gradient between the
with a maximum pressure of 207 MPa. The contact angle was exposure surface and center can be observed to occur between
140° and the measurable pore size ranged from about 6 nm to 180 and 270 min. This clearly shows that the distribution of tem-
360 lm. The samples in the shape of pellets of 5  5  5 mm3 in perature is not uniform and a thermal gradient exists. This thermal
size for pore structure testing were separated from the crushed gradient is likely to have adverse effects on the integrity of the OPC
specimens. The samples were immerged in ethanol to avoid rehy- and geopolymer paste at elevated temperatures.
dration immediately after being crushed and dried at about 105 °C Heat resistance is the ability of materials withstanding a high
for 24 h before pore structure test. temperature treatment while not destroying during the treatment.
 P. Duan et al. / Materials and Design 74 (2015) 125–137 129

On the other hand, the compressive strength of OPC specimens

drops considerably between temperatures of 400 °C and 600 °C.
This strength deterioration of OPC is attributed to the Ca(OH)2
decomposition that occurs at about 400 °C. In geopolymer, alumi-
nosilicate gel is the major binding phase that provides interparticle
bonding, which in turn enhances the macroscopic strength [78–80]
and takes over the strength gain behavior of geopolymer paste.
The compressive strength of geopolymer changes dramatically
at early ages. The compressive strength of OPC and geopolymer
paste specimens at 3 days of curing were also tested and compared
before and after elevated temperatures exposure and the results
has been summarized in Fig. 5.
The results provided in Fig. 5 reveal that the OPC suffers strength
loss after high temperature exposure, however, the fly ash–
metakaolin geopolymer gains strength around 400 °C. Compressive
strength development shows a dramatic drop of nearly 100% in
OPC paste specimen after 600 °C, and a final strength of 25 MPa
for geopolymer can be observed after the 1000 °C temperature
Fig. 3. Temperature evolution of geopolymer and OPC at various distances to the exposure.
surface. As mentioned above, water accumulated inside the geopolymer
structure is the main contributor to heat resistance and it would
Water accumulated inside the geopolymer structure is the main release during the heating to reduce the heat temperature and also
contributor to fire resistance and it would release during the to form a porous microstructure. The water releases during
heating process to reduce the heat temperature and also to form geopolymer formation is expelled from the geopolymer matrix
a porous microstructure. Therefore, geopolymer shows lower tem- during heating process, causing discontinuous nano pores through-
perature at the same distance to the surface compared to OPC. out the matrix, which improves the strength of the geopolymer
[81] before 400 °C. On the other hand, the generation of bubble
3.2. Compressive strength at elevated temperatures and foam-like structure subjects to high temperature makes the
geopolymeric system porous and it reduces the density, which
The compressive strength of OPC and geopolymer paste speci- shows a negative effect on the compressive strength after 400 °C.
mens at 28 days of curing were tested and compared before and
after elevated temperatures exposure and the results has been 3.3. Thermal stability
summarized in Fig. 4.
The results provided in Fig. 4 show that the OPC suffers strength Thermal stability of the selected geopolymer specimens was
loss after high temperature exposure. However, the fly ash– also investigated in terms of dimension evolution after exposed
metakaolin geopolymer gains strength after the same high temper- to elevated temperature (1000 °C) for 2 h. Fig. 6 presents the pho-
ature exposure around 400 °C. Compressive strength development tographs of geopolymer before and after the thermal exposures. It
indicates a dramatic drop of nearly 100% in OPC paste specimen can be observed that geopolymer gets a certain degree of volume
while fly ash–metakaolin geopolymer shows a strength increase shrinkage after the thermal exposures.
of about 3% after 400 °C, attaining a peak strength of 72 MPa at Actually, OPC exhibits a higher rate of shrinkage compared to
400 °C. Subsequently, this strength can be observed to deteriorate geopolymer [82–84], which can cause severe defects when it is
gradually. However, a final strength of 46 MPa is recorded after practically applied.
the 1000 °C temperature exposure, which is about a 25% decrease The increase in strength with increasing temperature up to
over the strength of the reference specimen at ambient temperature. 400 °C may also be attributed to the hardening of geopolymer

80

70
Compressive strength (MPa)

60

50

40

30

Portland cement
20
Geopolymer
10

0
20 100 200 300 400 500 600 700 800 900 1000
Temperature (°C)

Fig. 4. Compressive strength of geopolymer and OPC pastes at 28 days of curing at Fig. 5. Compressive strength of geopolymer and OPC pastes at 3 days of curing at
various temperatures. various temperatures.
130 P. Duan et al. / Materials and Design 74 (2015) 125–137

OPC OPC Geo Geo OPC OPC Geo Geo

Fig. 6. Dimension evolution of OPC and geopolymer after the thermal exposure.

paste caused by drying and further hydration, as indicated by Xu According to previous published paper [87], the drying shrink-
et al. [85], who observed an increase in strength when exposure age of cement paste may be due to a higher volume of mesopores,
temperature increased to 250 °C. In our study, the increase in the which causes a higher capillary stress by the water meniscus
strength was recorded up to a temperature of 400 °C. developed in the capillary pores of the paste, resulting in a higher
In the beginning of hydration from the first water–cement (or level of drying shrinkage [88].
fly ash and metakaolin) contact to the initial setting, the capillary
stress can be negligible, and the autogenous shrinkage is mainly 3.5. Mass loss at elevated temperatures
related to the Le Chatelier’s contraction and is thus identical to
the chemical shrinkage [86]. Considering the aforementioned The mass losses of the OPC and geopolymer samples due to
results, it can be stated that the autogenous shrinkage of geopoly- elevated temperature were determined from the mass change
mer may be due to self-desiccation in the hardened state rather before and after heat exposure. The average values of mass loss
than volume contraction by chemical shrinkage in fresh state. after exposure to different temperatures have been plotted in
Fig. 8. An increasing trend of the mass loss in OPC up to 600 °C
3.4. Thermal shrinkage can be observed. However, the exact values of the mass loss in
the OPC could not be determined after 600 °C because of spalling
Thermal shrinkage results for the OPC and geopolymer samples attributed to the Ca(OH)2 decomposition that occurs at this tem-
have been illustrated in Fig. 7. All samples were tested after perature. The geopolymer specimen shows similar mass loss curves
pre-drying at 105 °C for 24 h to remove free water and to avoid as OPC before 500 °C but with relatively lower mass loss. It can be
excessive shrinkage in the initial stage of the measurement. seen from Fig. 8 that most of the mass loss takes place at around
As is shown in Fig. 7, all samples exhibit shrinkage up to 1000 °C 500 °C for geopolymer. The rate of mass loss reduces and become
due to dehydroxylation of chemically bound water. The shrinkage stable in geopolymer after this temperature while OPC continues
is particularly evident for cement sample where dehydration of the at a similar rate until 600 °C. The mass loss of the geopolymer speci-
C–A–S–H phase and C–S–H phase is believed to dominate the mens at 1000 °C is 3.39% in total compared to that at ambient
dilatometry profile, resulting in loss of contact with the pushrod temperature.
at about 1000 °C. The shrinkage for geopolymer may be due to
sintering and further geopolymerization above 600 °C.
3.6. Electron microscopy after thermal exposure
Polynomial trend lines can be used to show the trend of
shrinkage evolution in geopolymer specimens with increasing
The morphological changes of OPC and geopolymer during the
temperature. A polynomial relationship also exists between
thermal exposure were illustrated in the micrographs of Fig. 9.
shrinkage evolution and the elevated temperature for cement
specimens. The fitting coefficients (R2) are all excess 0.99 for
geopolymer and OPC paste. Obviously, cement exhibits evident 5.0
and higher shrinkage values compared to geopolymer pastes.
4.5

4.0

3.5
Mass loss (%)

3.0

2.5

2.0

1.5

1.0 Portland cement

Geopolymer
0.5

0.0
0 200 400 600 800 1000
Temperature (°C)
Fig. 7. Thermal shrinkage of Portland cement and geopolymer paste with elevated
temperatures. Fig. 8. Changes in mass loss of the OPC and geopolymer at elevated temperatures.
 P. Duan et al. / Materials and Design 74 (2015) 125–137 131

OPC 20°C OPC 100°C

OPC 200 °C OPC 400 °C

GEO 20 °C GEO 100 °C

GEO 200 °C GEO 400 °C

OPC 600 °C GEO 600 °C

Fig. 9. The morphological changes of geopolymer and OPC at various temperatures.

The geopolymer microstructures become denser with the corresponding microstructure at 600 °C becomes looser with
increase of temperature up to 400 °C. This change has occurred in microcracking. However, the microstructure of geopolymer
the microstructure because of sintering and further geopolymeriza- remains to be stable after exposure to high temperatures and the
tion of fly ash and metakaolin with the increasing temperature. rate of mass loss reduces and becomes stable finally.
As mentioned above, an increasing trend of the mass loss in the Geopolymer possesses notable denser microstructure than OPC.
OPC up to 600 °C can be observed because of spalling, and the The Si/Na atomic ratio in an alkaline silicate solution affects the
132 P. Duan et al. / Materials and Design 74 (2015) 125–137

polymerization degree of the dissolved species [89–90]. The

alkaline aluminosilicate system is symbolized by N–A–S–H.
The formation process of these systems is a polycondensation. The
amorphous N–A–S–H gel in geopolymer becomes more compact.

3.7. Resistance to acid attack

Geopolymer and OPC specimens were immerged in the sulfuric

acid (2%) plus hydrochloric acid (2%) for 28 days to assess the resis-
tance to acid attack. Fig. 10 shows the photographs of geopolymer
before and after the acid exposure.
Geopolymer specimens show very small change in appearance
after 28 days of immersion in the acidic solutions. Some softening
of the surface cover and insignificant change of the color can be
noticed in the geopolymer specimens after exposure to the acid solu- Fig. 11. Compressiver strength evolution of the geopolymer and OPC specimens
exposed to sulfuric acid (2%) plus hydrochloric acid (2%) solutions.
tion, and there is only 0.7% mass change in geopolymer specimens.
Surfaces of OPC specimens exposed to the acidic solutions indi-
cate severe deterioration compared to geopolymer. After 10 days, Significant fluctuations of strength measured in OPC specimens
the surface layer of the samples is converted to some reaction are possibly connected to the decomposition of hydration products
products and is corroded to a depth of 4 mm, however, the surface and migration of alkalis from the specimens into solution imposed
layer of geopolymer samples deteriorates to a depth of 2 mm after by the acid attack. Changes of strength measured in geopolymer
28 days. After 28 days, OPC samples are severely deteriorated and specimens are possibly connected to the breakdown of some
exhibit severely deteriorated surface layer with 3.3% mass change. geopolymer alkali components and migration of alkalis from the
specimens into acidic solutions.
3.8. Compressive strength exposed to acid
3.9. Water absorption
The results provided in Fig. 11 uncover the evolution of the
compressive strength of the samples exposed to the acidic solu- Sorptivity tests were conducted to determine mass change in the
tions. The geopolymer samples activated by sodium hydroxide water absorption process of cement paste and geopolymer. The
and sodium silicate solution perform well with about 10.4% results of saturated water absorption on OPC, metakaolin and fly
strength decline in the 28-day exposure, which continues over ash geopolymer specimens after several days curing have been pre-
the next 28 days, reaching 22.2% after the test. Exponent trend sented in Fig. 12. It indicates less water absorption in geopolymer
lines can be used to show the trend of strength evolution in compared to cement paste. Minor difference between metakaolin
geopolymer specimens because significant fluctuations of strength and fly ash geoplymer can be observed. Based on the pore structure
are not observed for geopolymer in the acidic solutions. results mentioned below, the comparison among metakaolin
OPC paste sample shows 34.4% strength loss after 28-day expo- geopolymer, fly ash geopolymer and OPC paste shows that the total
sure, and about 57.8% after the next 28 days. The OPC samples are porosity (cumulative intrusion volume) is decreasing for geopoly-
severely deteriorated in the first 28-day exposure. An exponent mer pastes, which indicates that geopolymer possesses denser
relationship is also found between compressive strength and the microstructure. It also indicates that geopolymer could withstand
exposure time of OPC. outside medium ingress including water and acid, which relates

Fig. 10. Photographs of geopolymer and OPC before and after the acid exposures: (a) geopolymer before acid exposures; (b) geopolymer after acid exposures; (c) OPC before
acid exposures; (d) OPC after acid exposures.
 P. Duan et al. / Materials and Design 74 (2015) 125–137 133

Fig. 14. Change of mass in water absorption test of geopolymer and cement
Fig. 12. Mass change of geopolymer and cement samples in water absorption test. samples after acid exposure.

0.09
(a)
0.08 Cement paste

0.07 Metakaolin

Cumulative intrusion (mL/g)

Geopolymer
0.06
Fly ash
Geopolymer
0.05

0.04

0.03

0.02
Fig. 13. Change of mass in water absorption test of geopolymer and cement
samples after thermal exposure. 0.01

0.00
to better permeability resistance, and the results are consistent 1 10 100 1000 10000 100000 1000000
with the water absorption test result. It can be concluded that alkali Pore diameter (nm)
activated condition is optimum for better pore structure modifica-
tion. The strength is also higher at this optimal condition. 0.18
As the compressive strength of OPC specimens drops consider- (b)
ably after 500 °C (see Figs. 4 and 5) because of spalling, water 0.16
absorption test after thermal exposure was carried out for geopoly- Cement paste
mer and OPC samples exposed to 500 °C. 0.14

The results of saturated water absorption on OPC, metakaolin Metakaolin

0.12
and fly ash geopolymer specimens exposed to 500 °C have been
dV/dlogD (mL/g)

Geopolymer
presented in Fig. 13. It indicates markedly less water absorption 0.10 Fly ash
in geopolymer compared to cement paste. Minor difference Geopolymer
between metakaolin and fly ash geoplymer can be observed, and 0.08
water absorption of samples is relatively higher when exposed to
elevated temperatures. 0.06
Water absorption test after acid exposure was also made for
0.04
geopolymer and OPC samples exposed to acidic solutions for
28 days. The results of saturated water absorption are presented 0.02
in Fig. 14. Similarly, it shows markedly less water absorption in
geopolymer compared to cement paste. Minor difference between 0.00
metakaolin and fly ash geoplymer can be observed, and water 1 10 100 1000 10000 100000 1000000
absorption is relatively higher than samples not exposed to Pore diameter (nm)
acidic solutions but lower than samples exposed to elevated tem-
Fig. 15. The porosity and pore size distribution of geopolymer and OPC paste.
peratures. These results are consistent with the compressive
strength development results (see Figs. 4, 5 and 11), which uncover
that better permeability resistance is closely related to the compres- materials. As pointed out by Diamond [91], MIP is the only
sive strength. available procedure that can cover nearly the whole range of sizes
that must be tallied.
3.10. Pore structure after acid exposure The porosity and pore size distribution measurements per-
formed by MIP of metakaolin and fly ash geopolymer samples and
Mercury intrusion porosimetry (MIP) has been widely used to OPC paste are shown in Fig. 15. The results at 28 days are plotted
determine pore structure and pore size distribution of porous to identify how pore size distribution varies with ages.
134 P. Duan et al. / Materials and Design 74 (2015) 125–137

Fig. 16. The porosity and pore size distribution of geopolymer and OPC paste after Fig. 17. The porosity and pore size distribution of geopolymer and OPC paste after
thermal exposure. acid exposure.

The comparison among metakaolin geopolymer, fly ash geopoly- microstructure. It indicates that geopolymer could withstand out-
mer and OPC paste shows that the total porosity (cumulative intru- side medium ingress including water and acid, which relates to
sion volume) (Fig. 15(a)) is decreasing for geopolymer pastes, and better permeability resistance, and the result of MIP is consistent
fly ash geopolymer shows a significant change. The differential with the water absorption and acid exposure test. It can be
curves of pore size distribution are almost identical between concluded that alkali activated condition is optimum for better
different samples. The critical pore diameters, defined as the peaks pore structure modification.
in the differential curves (Fig. 15(b)), giving the rate of mercury
intrusion per change in pressure (differential curves) [92], do not 3.11. Formation of microstructure and simulation
show a significant change. However, the total pore volume in OPC
is higher than that in geopolymer. It is evident that exposure of MIP tests have been verified from the formation of microstruc-
cement to acidic solutions shifts the pore size peak to a higher value ture aspect, which was simulated with a numerical model. For the
and also leads to the development of a macropore system with pore simulation of the reaction processes and of the formation of the
size exceeding 100 nm. microstructure in cement and geopolymer systems, the numerical
Unlike cement pastes which hydrate with time, the strength of model HYMOSTRUC was developed [93].
the geopolymer samples remain high after exposure to sulfuric acid In this model, the degree of hydration was simulated as a func-
(2%) plus hydrochloric acid (2%) solutions. Geopolymer possesses a tion of the particle size distribution, the chemical composition of
relatively lower strength loss when compared to conventional geopolymer materials, the water/solid ratio and the reaction tem-
cement paste. This is also confirmed by MIP tests with almost iden- perature. In a computer digitized format of HYMOSTRUC [93], the
tical results, which show the total porosities of the OPC samples are raw solid particles were modelled as digitized spheres randomly
significantly higher than the porosities of the fly ash and metakaolin distributed in a three-dimensional body and the hydrating
based geopolymer. This is also extremely true for metakaolin and fly geopolymer grains were simulated as growing spheres.
ash geopolymer samples and OPC paste exposed to high temperature In this model, the geopolymer particles are modelled as spheres
(Fig. 16) and acid (Fig. 17). Geopolymer has a relatively lower of different diameters. Metakaolin and fly ash are not regular
strength loss when compared to conventional cement paste after spheres, and this may be the limitation of this simulation model.
thermal exposure and acid exposure. This is also confirmed by MIP However, the hydration proceeded at the particle/water interface
test, which shows the total porosities of the OPC samples are sig- and the hydration products formed around the original particles
nificantly higher than the porosities of the fly ash and metakaolin in this model are true when geopolymerization occurs.
based geopolymer. Additionally, major difference between OPC and The microstructure formation and development in geopolymer
geopolymer can be observed due to marked strength loss of OPC after is schematically presented in Fig. 18, which can be described via
thermal exposure and acid exposure (see Figs. 4, 5 and 11). simplified mechanisms, geopolymer samples show a homogenous
Based on the pore structure results by MIP, the total porosity microstructure with reaction products evenly distributed in the
(cumulative intrusion volume) is lower for geopolymer pastes matrix. Some large cavities form and are seemed as ‘‘isolated’’ by
compared to OPC, which means that geopolymer possesses denser aluminosilicate gels.
 P. Duan et al. / Materials and Design 74 (2015) 125–137 135

respectively. Only a little amount of products could be observed

at early ages. However, with the reaction time increases, plenty of
hydration products fill the voids in the matrix and the microstruc-
ture becomes denser at 6 h (Fig. 18(e)) and 24 h (Fig. 18(f)), respec-
tively. Obviously, large amount of outer hydration products form at
longer ages and the microstructure at 3 days (Fig. 18(g)), 7 days
(Fig. 18(h)) and 28 days (Fig. 18(i)) can be observed to be much
denser and only a few voids exist in the matrix.
As mentioned in Fig. 9, geopolymer presents denser micro-
structure compared to OPC paste. Additionally, the comparison
among metakaolin geopolymer, fly ash geopolymer and OPC paste
shows that the total porosity (cumulative intrusion volume) in
Figs. 15–17 is decreasing for geopolymer pastes. The total pore vol-
ume in OPC is higher than that in geopolymer. The differential
curves of pore size distribution also indicate that geopolymer has
optimized pore structure. All these simulation results are demon-
strated by the obtained experimental results.

4. Conclusions

This study intends to broaden application of fluidized bed fly

ash and metakaolin in cement or concrete. Durability and
microstructure of fluidized bed fly ash and metakaolin geopolymer
exposed to elevated temperatures and acid attack were investigat-
ed. Compressive strength and several key durability parameters for
geopolymer and ordinary Portland cement (OPC) were assessed
and compared. Microstructure formation and development was
characterized in terms of morphology by SEM and pore structures
by simulation. The results are summarized below:

(1) The distribution of temperature is not uniform and a thermal

gradient exists after exposure to high temperature. Water
accumulated inside the geopolymer structure is the main
contributor to heat resistance, geopolymer presents lower
temperature at the same distance to the surface compared
with OPC. OPC suffers strength loss, however, the fly ash
and metakaolin geopolymer gains strength after the same
high temperature exposure. OPC shows a higher rate of
shrinkage compared to geopolymer after the thermal expo-
sures. Polynomial trend lines can be used to show the trend
of shrinkage evolution in geopolymer specimens and OPC
with increasing temperature.
(2) The geopolymer microstructures become denser with the
increase of temperature up to 400 °C. This change in the
microstructure may be due to sintering and further geopoly-
merization of fly ash and metakaolin with the increasing
temperature.
(3) Geopolymer specimens have very small change in appear-
ance after 28 days of immersion in the acidic solutions, and
there is only 0.7% mass change in geopolymer specimens.
Surfaces of OPC specimens exposed to the acidic solutions
indicate severe deterioration compared to geoplymer and
there is 3.3% mass change in the OPC specimens. An exponent
relationship can be used to show the trend of strength evolu-
tion in geopolymer specimens and OPC after exposure to
acidic solutions.
(4) Sorptivity tests on OPC, metakaolin and fly ash geopolymer
specimens after several days curing show the lesser water
absorption in geopolymer compared to cement paste. The
Fig. 18. Formation of the microstructure in geopolymer at various curing times
comparison among metakaolin geopolymer, fly ash geopoly-
( un-reacted particles; inner products; outer
products; pores). mer and OPC paste shows that the total porosity is decreas-
ing for geopolymer pastes, which is closely related to better
permeability resistance.
The original initial geopolymer particles (Fig. 18(a)) are random- (5) The total pore volume in traditional OPC is higher than that
ly distributed in a three-dimensional body. Some hydration prod- in geopolymer. It is evident that exposure of cement to
ucts form at 1 h (Fig. 18(b)), 2 h (Fig. 18(c)) and 3 h (Fig. 18(d)), acidic solutions shifts the pore size peak to a higher value
136 P. Duan et al. / Materials and Design 74 (2015) 125–137

and also leads to the development of a macropore system. [22] P. Duxson, J. Provis, G. Lukey, J.S.J. van Deventer, The role of inorganic polymer
technology in the development of green concrete, Cem. Concr. Res. 37 (2007)
The compressive strength shows agreement with the results
1590–1597.
of pore structure. [23] F. Pacheco-Torgal, J.P. Gomes, S. Jalali, Alkali-activated binders: a review. Part
(6) Geopolymer samples show homogenous microstructure 2: About materials and binders manufacture, Constr. Build. Mater. 22 (2008)
with reaction products evenly distributed in the matrix. 1315–1322.
[24] F. Pacheco-Torgal, J.P. Gomes, S. Jalali, Alkali-activated binders: a review. Part
Only a little amount of products can be observed at early 1: Historical background, terminology, reaction mechanisms and hydration
ages. However, with the reaction time increases, plenty of products, Constr. Build. Mater. 22 (2008) 1305–1314.
hydration products fill the voids in the matrix and the [25] M. Shojaei, K. Behfarnia, R. Mohebi, Application of alkali-activated slag
concrete in railway sleepers, Mater. Des. 69 (2015) 89–95.
microstructure becomes denser after 6 h curing ages. [26] C. Bilim, O. Karahan, C.D. Atis, S. Ilkentapar, Influence of admixtures on the
properties of alkali-activated slag mortars subjected to different curing
conditions, Mater. Des. 44 (2013) 540–547.
Acknowledgements [27] A. Nazari, Compressive strength of geopolymers produced by ordinary
Portland cement: application of genetic programming for design, Mater. Des.
43 (2013) 356–366.
This work was supported by the Fundamental Research Funds [28] P. Posi, C. Teerachanwit, C. Tanutong, S. Limkamoltip, S. Lertnimoolchai, V.
for the Central Universities (CUGL150806), China University of Sata, et al., Lightweight geopolymer concrete containing aggregate from
recycle lightweight block, Mater. Des. 52 (2013) 580–586.
Geosciences, Wuhan, Public Service Project of the Chinese
[29] P. Sukmak, S. Horpibulsuk, S.L. Shen, Strength development in clay-fly ash
Ministry of Land and Resources (201311024) and the National geopolymer, Constr. Build. Mater. 40 (3) (2013) 566–574.
Natural Science Foundation of China (NNSF-51202172). [30] D. Hardjito, S.E. Wallah, D.M.J. Sumajouw, On the development of fly ash-based
geopolymer concrete, ACI Mater. J. 101 (6) (2004) 467–472.
[31] D. Hardjito, S.E. Wallah, D.M.J. Sumajouw, Fly ash-based geopolymer concrete,
References Aust. J. Struct. Eng. 6 (1) (2005) 77–86.
[32] B.V. Rangan, Studies on low-calcium fly ash-based geopolymer concrete,
Indian Concr. Inst. J. 4 (2006) 9–17.
[1] K.H. Yang, J.K. Song, K.I. Song, Assessment of CO2 reduction of alkali-activated
[33] D.M.J. Sumajouw, D. Hardjito, S.E. Wallah, Fly ash-based geopolymer concrete:
concrete, J. Clean Prod. 39 (2013) 265–272.
study of slender columns, J. Mater. Sci. 42 (9) (2007) 3124–3130.
[2] R.I. Malek, Z.H. Khalil, S.S. Imbaby, D.M. Roy, The contribution of class-F fly ash
[34] P.K. Sarker, Analysis of geopolymer concrete columns, Mater. Struct. 42 (2009)
to the strength of cementitious mixtures, Cem. Concr. Res. 35 (2005) 1152–
715–724.
1154.
[35] A. Bagheri, A. Nazari, Compressive strength of high strength class C fly ash-
[3] X.Y. Wang, Effect of fly ash on properties evolution of cement based materials,
based geopolymers with reactive granulated blast furnace slag aggregates
Constr. Build. Mater. 69 (10) (2014) 32–40.
designed by Taguchi method, Mater. Des. 54 (2014) 483–490.
[4] P.K. Sarker, S. Kelly, Z.T. Yao, Effect of fire exposure on cracking, spalling and
[36] E. Papa, V. Medri, E. Landi, B. Ballarin, F. Miccio, Production and
residual strength of fly ash geopolymer concrete, Mater. Des. 63 (11) (2014)
characterization of geopolymers based on mixed compositions of metakaolin
584–592.
and coal ashed, Mater. Des. 56 (2014) 409–415.
[5] P.S. Deb, P. Nath, P.K. Sarker, The effects of ground granulated blast-furnace
[37] S. Pangdaeng, T. Phoo-ngernkham, V. Sata, P. Chindaprasirt, Influence of curing
slag blending with fly ash and activator content on the workability and
conditions on properties of high calcium fly ash geopolymer containing
strength properties of geopolymer concrete cured at ambient temperature,
Portland cement as additive, Mater. Des. 53 (2014) 269–274.
Mater. Des. 62 (10) (2014) 32–39.
[38] J.J. Feng, R.F. Zhang, L.L. Gong, Y. Li, W. Cao, X.D. Cheng, Development of porous
[6] P. Nath, P.K. Sarker, Effect of GGBFS on setting, workability and early strength
fly ash-based geopolymer with low thermal conductivity, Mater. Des. 65
properties of fly ash geopolymer concrete cured in ambient condition, Constr.
(2015) 529–533.
Build. Mater. 66 (9) (2014) 163–171.
[39] A.M. Fernandez-Jimenez, A. Palomo, C. Lopez-Hombrados, Engineering properties
[7] Z.H. Zhang, H. Wang, Y.C. Zhu, A. Reid, J.L. Provis, F. Bullen, Using fly ash to
of alkali-activated fly ash concrete, ACI Mater. J. 103 (2) (2006) 106–112.
partially substitute metakaolin in geopolymer synthesis, Appl. Clay Sci. 88–89
[40] M. Sofi, J.S.J. van Deventer, P.A. Mendis, Engineering properties of inorganic
(2014) 194–201.
polymer concretes (IPCs), Cem. Concr. Res. 37 (2007) 251–257.
[8] T. Sun, J.Y. Chen, X.R. Lei, C.Y. Zhou, Detoxification and immobilization of
[41] A.M. Rashad, A comprehensive overview about the influence of different
chromite ore processing residue with metakaolin-based geopolymer, J.
admixtures and additives on the properties of alkali-activated fly ash, Mater.
Environ. Chem. Eng. 2 (1) (2014) 304–309.
Des. 53 (2014) 1005–1025.
[9] T.W. Cheng, M.L. Lee, M.S. Ko, T.H. Ueng, S.F. Yang, The heavy metal adsorption
[42] M.O. Yusuf, M.A.M. Johari, Z.A. Ahmad, M. Maslehuddin, Evolution of alkali
characteristics on metakaolin-based geopolymer, Appl. Clay Sci. 56 (2) (2012)
activated ground blast furnace slag-ultrafine palm oil fuel ash based concrete,
90–96.
Mater. Des. 55 (2014) 387–393.
[10] Z.T. Yao, X.S. Ji, P.K. Sarker, J.H. Tang, L.Q. Ge, M.S. Xia, et al., A compre-
[43] M.O. Yusuf, M.A.M. Johari, Z.A. Ahmad, M. Maslehuddin, Effects of H2O/Na2O
hensive review on the applications of coal fly ash, Earth-Sci. Rev. 141 (2015)
molar ratio on the strength of alkali activated ground blast furnace slag-
105–121.
ultrafine palm oil fuel ash based concrete, Mater. Des. 56 (2014) 158–164.
[11] R.S. Blissett, N.A. Rowson, A review of the multi-component utilization of coal
[44] P. Duxson, A. Fernández-Jiménez, J.L. Provis, G.C. Lukey, A. Palomo, J.S.J. van
fly ash, Fuel 97 (2012) 1–23.
Deventer, Geopolymer technology: the current state of the art, J. Mater. Sci. 42
[12] R. Kumar, S. Kumar, S.P. Mehrotra, Towards sustainable solutions for fly ash
(9) (2007) 2917–2933.
through mechanical activation, Resour. Conserv. Recy. 52 (2) (2007) 157–179.
[45] K.C. Goretta, F. Gutierrez-Mora, D. Singh, J.L. Routbort, G.C. Lukey, J.S.J. van
[13] M. Saridemir, Effect of specimen size and shape on compressive strength of
Deventer, Erosion of geopolymers made from industrial waste, J. Mater. Sci. 42
concrete containing fly ash: application of genetic programming for design,
(9) (2007) 3066–3072.
Mater. Des. 56 (2014) 297–304.
[46] D.L.Y. Kong, J.G. Sanjayan, K. Sagoe-Crentsil, Comparative performance of
[14] B. Nematollahi, J. Sanjayan, Effect of different superplasticizers and activator
geopolymers made with metakaolin and fly ash after exposure to elevated
combinations on workability and strength of fly ash based geopolymer, Mater.
temperatures, Cem. Concr. Res. 37 (12) (2007) 1583–1589.
Des. 57 (2014) 667–672.
[47] Sindhunata, J.S.J. van Deventer, G.C. Lukey, H. Xu, Effect of curing temperature
[15] W. Wongkeo, P. Thongsanitgam, A. Ngamjarurojana, A. Chaipanich,
and silicate concentration on fly-ash-based geopolymerization, Ind. Eng.
Compressive strength and chloride resistance of self-compacting concrete
Chem. Res. 45 (10) (2006) 3559–3568.
containing high level fly ash and silica fume, Mater. Des. 64 (2014) 261–269.
[48] C.K. Yip, G.C. Lukey, J.S.J. van Deventer, Effect of blast furnace slag addition on
[16] W. Chalee, P. Ausapanit, C. Jaturapitakkul, Utilization of fly ash concrete in
microstructure and properties of metakaolinite geopolymeric materials,
marine environment for long term design life analysis, Mater. Des. 31 (3)
Ceram. Trans. 153 (2004) 187–209.
(2010) 1242–1249.
[49] T.W. Cheng, J.P. Chiu, Fire-resistant geopolymer produce by granulated blast
[17] P. Lorca, R. Calabuig, J. Benlloch, L. Soriano, J. Paya, Microconcrete with partial
furnace slag, Miner. Eng. 16 (3) (2003) 205–210.
replacement of Portland cement by fly ash and hydrated lime addition, Mater.
[50] J.W. Ginacaspro, P.N. Balaguru, R.E. Lyon, Use of inorganic polymer to improve
Des. 64 (2014) 535–541.
the fire response of balsa sandwich structures, J. Mater. Civ. Eng. 18 (3) (2006)
[18] R.F. Li, G.H. Wu, L.T. Jiang, D.L. Sun, Interface microstructure and compressive
390–397.
behavior of fly ash/phosphate geopolymer hollow sphere structures, Mater.
[51] R.E. Lyon, P.N. Balaguru, A. Foden, U. Sorathia, J. Davidovits, M. Davidovics,
Des. 65 (2015) 585–590.
Fire-resistant aluminosilicate composites, Fire Mater. 21 (2) (1997) 67–73.
[19] A. Islam, U.J. Alengaram, M.Z. Jumaat, I.I. Bashar, The development of
[52] P.G. He, D.C. Jia, T.S. Lin, M.R. Wang, Y. Zhou, Effects of high-temperature heat
compressive strength of ground granulated blast furnace slag-palm oil fuel
treatment on the mechanical properties of unidirectional carbon fiber
ash-fly ash based geopolymer, Mater. Des. 56 (2014) 833–841.
reinforced geopolymer composites, Ceram. Int. 36 (4) (2010) 1447–1453.
[20] T. Phoo-ngernkham, P. Chindaprasirt, V. Sata, S. Hanjitsuwan, The effect of
[53] N. Ranjbar, M. Mehrali, U. Johnson Alengaram, H.S.C. Metselaar, M.Z. Jumaat,
adding nano-SiO2 and nano Al2O3 on the properties of high calcium fly ash
Compressive strength and microstructural analysis of fly ash/palm oil fuel ash
geopolymer cured at ambient temperature, Mater. Des. 55 (2014) 58–65.
based geopolymer mortar under elevated temperatures, Constr. Build. Mater.
[21] J. Davidovits, Geopolymers-inorganic polymeric new materials, J. Ther. Anal.
65 (8) (2014) 114–121.
37 (8) (1991) 1633–1656.
 P. Duan et al. / Materials and Design 74 (2015) 125–137 137

[54] O.A. Abdulkareem, A.M. Mustafa Al Bakri, H. Kamarudin, I. Khairul Nizar, A.A. [73] F.U.A. Shaikh, V. Vimonsatit, Compressive strength of fly-ash-based
Saif, Effects of elevated temperatures on the thermal behavior and mechanical geopolymer concrete at elevated temperatures, Fire Mater. (2014), http://
performance of fly ash geopolymer paste, mortar and lightweight concrete, dx.doi.org/10.1002/fam.2240.
Constr. Build. Mater. 50 (11) (2014) 377–387. [74] Y.C. Choi, J.H. Kim, S. Choi, Mechanical performance of fly-ash-based green
[55] L. Zeng, D.Y. Cao, Y. Xu, C.W. Fan, X.Q. Peng, Novel method for preparation of geopolymer composites subjected to elevated temperatures, Asian J. Chem. 26
calcined kaolin intercalation compound-based geopolymer, Appl. Clay Sci. 101 (17) (2014) 5517–5521.
(11) (2014) 637–642. [75] J.Z. Kuang, X.C. Zhao, F. Shi, H.Y. Cao, Study on influence factors of fly ash and
[56] Z. Pan, J.G. Sanjayan, Stress–strain behaviour and abrupt loss of stiffness of metakaolin based geopolymer, Adv. Mater. Res. 538–541 (2012) 2322–2328.
geopolymer at elevated temperatures, Cem. Concr. Compos. 32 (9) (2010) [76] C. Prinya, R. Ubolluk, Utilization of blended fluidized bed combustion (FBC) ash
657–664. and pulverized coal combustion (PCC) fly ash in geopolymer, Waste Manage.
[57] Z. Pan, J.G. Sanjayan, F. Collins, Effect of transient creep on compressive 30 (4) (2010) 667–672.
strength of geopolymer concrete for elevated temperature exposure, Cem. [77] X.H. Yuan, W. Chen, Z.A. Lu, H.G. Chen, Shrinkage compensation of alkali-
Concr. Res. 56 (2) (2014) 182–189. activated slag concrete and microstructural analysis, Constr. Build. Mater. 66
[58] C.C. Mao, H. Ran, H.W. Te, C.F. Toun, Utilization of circulating fluidized bed (15) (2014) 422–428.
combustion (CFBC) fly ash and coal-fired fly ash in Portland cement, Key Eng. [78] A. Palomo, M.W. Grutzeck, M.T. Blanco, Alkali-activated fly ashes: a cement for
Mater. 629–630 (2014) 306–313. the future, Cem. Concr. Res. 29 (8) (1999) 1323–1329.
[59] M. Shaswata, M. Saroj, U.B. Adhikari, Study on the physical and mechanical [79] H. Rahier, B. van Mele, J. Wastiels, Low-temperature synthesized alumino-
property of ordinary Portland cement and fly ash paste, Int. J. Civ. Struct. Eng. 2 silicate glasses. Part II: Rheological transformations during low-temperature
(3) (2012) 731–736. cure and high temperature properties of a model compound, J. Mater. Sci. 31
[60] Ö. Eren, Strength development of concretes with ordinary Portland cement, (1) (1996) 80–85.
slag or fly ash cured at different temperatures, Mater. Struct. 35 (9) (2002) [80] P. Duxson, G.C. Lukey, J.S.J. van Deventer, Physical evolution of Na-geopolymer
536–540. derived from metakaolin up to 1000 °C, J. Mater. Sci. 42 (9) (2007) 3044–3054.
[61] W. Sha, G.B. Pereira, Differential scanning calorimetry study of ordinary [81] R.D. Recommendation, P.D.R. De La Rilem, 129-MHT: test methods for
Portland cement paste containing metakaolin and theoretical approach of mechanical properties of concrete at high temperatures, Mater. Struct. 28
metakaolin activity, Cem. Concr. Compos. 23 (6) (2001) 455–461. (1995) 410–414.
[62] A.M. Mustafa Al Bakri, H. Kamarudin, M. Binhussain, N.I. Khairul, A.R. Rafiza, Y. [82] F. Collins, J.G. Sanjayan, Microcracking and strength development of alkali
Zarina, Comparison of geopolymer fly ash and ordinary Portland cement to the activated slag concrete, Cem. Concr. Res. 23 (4) (2001) 345–352.
strength of concrete, Adv. Sci. Lett. 19 (12) (2013) 3592–3595. [83] E. Douglas, A. Bilodeau, V.M. Malhotra, Properties and durability of alkali-
[63] V. Kirk, K. Aditya, S. Gaurav, N. Narayanan, The rheological properties of activated slag concrete, ACI Mater. J. 89 (5) (1992) 509–516.
ternary binders containing Portland cement, limestone, and metakaolin or fly [84] S.D. Wang, X.C. Pu, K.L. Scrivener, P.L. Pratt, Alkali-activated slag cement and
ash, Cem. Concr. Res. 52 (2013) 196–207. concrete: a review of properties and problems, Adv. Cem. Res. 7 (27) (1995)
[64] G.S. David, W. Stan, O.F. Martin, Heat of hydration of Portland cement– 93–102.
metakaolin–fly ash (PC–MK–PFA) blends, Cem. Concr. Res. 38 (6) (2008) 832– [85] Y. Xu, Y.L. Wong, C.S. Poon, M. Anson, Impact of high temperature on PFA
840. concrete, Cem. Concr. Res. 31 (2001) 1065–1073.
[65] V. Kirk, A. Matthew, O. Tandre, S. Gaurav, N. Narayanan, Hydration and [86] M. Bouasker, P. Mounanga, P. Turcry, A. Loukili, A. Khelidj, Chemical shrinkage
strength development in ternary Portland cement blends containing limestone of cement pastes and mortars at very early age: effect of limestone filler and
and fly ash or metakaolin, Cem. Concr. Compos. 39 (2013) 93–103. granular inclusions, Cem. Concr. Compos. 30 (1) (2008) 13–22.
[66] R.D. Hooton, W. Ma, D. Sample, R. Martin, P.W. Brown, Calorimetric study of [87] F.G. Collins, J.G. Sanjayan, Effect of pore size distribution on drying shrinkage
cement blends containing fly ash, silica fume, and slag at elevated of alkali-activated slag concrete, Cem. Concr. Res. 30 (9) (2000) 1401–1406.
temperatures, Cem. Concr. Aggr. 16 (2) (1994) 93–99. [88] Z. Jiang, Z. Sun, P. Wang, Autogenous relative humidity change an autogenous
[67] N.I. Fattuhi, B.P. Hughes, Ordinary Portland cement mixes with selected shrinkage of high-performance cement pastes, Cem. Concr. Res. 35 (8) (2005)
admixtures subjected to sulfuric acid attack, ACI Mater. J. 85 (6) (1988) 512– 1539–1545.
518. [89] J.L. Provis, J.S.J. van Deventer, Direct measurement of the kinetics of geopoly-
[68] S. Wild, B.B. Sabir, J. Bai, J.M. Kinuthia, Self-compensating autogenous merisation by in-situ energy dispersive X-ray diffractometry, J. Mater. Sci. 42
shrinkage in Portland cement–metakaolin–fly ash pastes, Adv. Cem. Res. 12 (2007) 2974–2981.
(1) (2000) 35–43. [90] J. Davidovits, Geopolymers and geopolymeric materials, J. Ther. Anal. 35
[69] R. Mateusz, O. Jan, Effects of curing conditions on the properties of ternary (1989) 429–441.
(ordinary Portland cement/fly ash/silica fume) concrete, ACI Mater. J. (2014), [91] S. Diamond, Mercury porosimetry: an inappropriate method for the
http://dx.doi.org/10.14359/51687307. measurement of pore size distributions in cement-based materials, Cem.
[70] P. Sandor, Portland cement–fly ash–silica fume systems in concrete, Adv. Cem. Concr. Res. 30 (10) (2000) 1517–1525.
Based Mater. 1 (2) (1993) 83–91. [92] Y. Akkaya, K. van Bui, S.P. Shah, Rheological model for self-consolidating
[71] M.S. Sudarshan, R.V. Ranganath, Properties of fly ash based geopolymer concrete, ACI Mater. J. 99 (6) (2002) 549–559.
concrete exposed to sustained elevated temperatures, Adv. Mater. Res. 250– [93] G. Ye, P. Lura, K. van Breugel, A.L.A. Fraaij, Study on the development of the
253 (2011) 962–968. microstructure in cement-based materials by means of numerical simulation
[72] D.L.Y. Kong, J.G. Sanjayan, S.C. Kwesi, Comparative performance of and ultrasonic pulse velocity measurement, Cem. Concr. Compos. 26 (5) (2004)
geopolymers made with metakaolin and fly ash after exposure to elevated 491–497.
temperatures, Cem. Concr. Res. 37 (12) (2007) 1583–1589.