Construction and Building Materials 184 (2018) 111–127

Contents lists available at ScienceDirect

Construction and Building Materials

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

Magnesium sulfate resistance of geopolymer mortar

Hafez E. Elyamany, Abd Elmoaty M. Abd Elmoaty ⇑, Ahmed M. Elshaboury
Structural Engineering Department, Faculty of Engineering, Alexandria University, Egypt

h i g h l i g h t s

 The performance of geopolymer mortar in magnesium sulfate solution was evaluated.

 Effect of curing temperature, NaOH molarity on sulfate resistance were studied.
 Increasing curing temperature up to 90 °C improves strength and durability properties.
 Addition GGBS and silica fume yields better properties compared to fly ash only.
 Various geopolymer mortars achieved higher sulfate resistance than OPC mortar.

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

Article history: Sulfate attack is one of the reasons which cause deterioration and damages of concrete structures
Received 17 March 2018 throughout the world. Thus, sulfate attack resistance is an important durability and serviceability
Received in revised form 12 June 2018 concern for materials used in construction. Various geopolymer mortars and OPC mortars were manufac-
Accepted 26 June 2018
tured to evaluate magnesium sulfate resistance. This paper studied the effect of curing temperature,
Available online 2 July 2018
sodium hydroxide solution molarity, alkaline solution to binder ratio and binder type on magnesium sul-
fate resistance of geopolymer mortar and make a comparison between OPC mortar and various geopoly-
Keywords:
mer mortars. Specimens were immersed in 10% magnesium sulfate solution up to 48 weeks. The
Geopolymer mortar
Fly ash
evaluated properties in this study were water absorption and voids ratio, visual inspection, microstruc-
Curing temperature ture of specimens, weight change, compressive strength, flexural strength and expansion strain. Results
Magnesium sulfate indicated that increasing curing temperature, sodium hydroxide solution molarity and decreasing
Voids ratio alkaline solution to binder ratio enhanced magnesium sulfate resistance of geopolymer mortar. Mixes
Weight change containing 50% fly ash, 35% ground granulated slag (GGBS) and 15% silica fume achieved the best
Residual compressive strength performance in magnesium sulfate solution. Generally, various geopolymer mortars achieved better per-
SEM formance in magnesium sulfate solution compared to OPC mortars.
Ó 2018 Elsevier Ltd. All rights reserved.

1. Introduction deterioration before their intended service life, especially those

Concrete is considered one of the most important materials
used in construction all over the world. Ordinary portland cement
(OPC) is used as the binder to produce the concrete. The demand of
concrete is increasing continuously for the need of development of
infrastructure facilities [1]. The environmental issues associated
with the production of OPC are well known. Despite portland
cement concrete has many advantages, it has been proven that
portland cement concrete suffers from different problems such as
durability, when exposed to sea water, sulfuric soils or freezing
weather, and carbon dioxide emission during cement manufactur-
ing process. Many concrete structures have shown serious
⇑ Corresponding author.
E-mail address: abduo76@yahoo.com (A.E.M. Abd Elmoaty).
are constructed in a corrosive environment [2].
Portland cement has harmful impacts to the environment due
to its significant contribution to the amount of greenhouse gas,
resulting from the high volume of carbon dioxide emitted during
its production, which represents about 65% of global warming.
The production of ordinary portland cement is responsible for
two ecological problems; it consumes a lot of energy and releases
a large amount of carbon dioxide as previously noted consequently
[3]. Therefore, the need for alternative binders able to achieve a
sustainable and ecologically aware concrete proved to be essential
to reduce the huge emission of CO2 which linked to portland
cement industry and reach good durability characteristics when
exposed to aggressive environment. In 1978, Joseph Davidovits
has led researchers to the discovery of green (eco-friendly)
concrete, commonly named ‘‘geopolymer concrete”.

https://doi.org/10.1016/j.conbuildmat.2018.06.212
0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
112 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127
Davidovits proposed that an alkaline liquid could be used to
react with the silicon (Si) and the aluminum (Al) in a source mate-
rial of geological origin or in by product materials such as fly ash,
blast furnace slag, and rice husk ash to produce binders. He named
the term ‘geopolymer’ because the chemical reaction that occurs in
this case is a polymerization process [4,5]. Unlike ordinary port-
land cements, geopolymers do not form calcium silicate-hydrates
(CSHs) for matrix formation and strength, but use the polyconden-
sation of silica and alumina precursors and a high alkali content to
obtain structural strength. Therefore, in some studies geopolymers
are sometimes referred to as alkali activated aluminosilicate
binders [4,6,7].
Heat-cured low-calcium fly ash-based geopolymer concrete
showed high-early strength gain, excellent resistance to sulfate
attack, good acid resistance, undergoes low creep, and suffers very
little drying shrinkage [8]. Curing temperature plays an important
role in improving the microstructure and mechanical strength of
geopolymer system. Generally, higher temperature accelerates
polymerization process compared to ambient temperature. As fly
ash based geopolymer paste reacts slowly at low ambient temper-
ature when compared to heat cured specimens [9], these mixes are
usually subjected to curing temperatures ranging from 30 °C to 85
°C and a relative humidity of about 95% [6,10]. It is investigated
that the amount of calcium content in geopolymer mix has a signif-
icant effect on the resulting hardened geopolymer because it was
found that calcium oxide form calcium silicate hydrate (CSH) along
with the aluminosilicate geopolymer gel.
Moreover, several previous studies reported availability of mix-
ing fly ash based geopolymers with silica fume, metakaolin and
blast furnace slag [11,12]. Temuujin et al. [13] confirmed that the
addition of calcium compounds CaO and Ca(OH)2 improves
mechanical properties of the fly ash-based geopolymers cured at
ambient temperature. In addition, the geopolymerization process
is also influenced by other factors such as, the type and properties
of aluminosilicate sources and composition of alkaline solution
[14,15].
Sulfate attack is an important durability and serviceability con-
cern for geopolymer materials used in construction. Previous expe-
rience with portland cement and blended cement concretes
showed cases of concrete deterioration when exposed to sulfate
attack in the environment [16,17]. Because of reactions involve
CH, C–S–H and the aluminate component of hardened cement
paste which occur due to sulfate attack on OPC concrete [18,19].
Expansion and cracking are caused, directly or indirectly, by ettrin-
gite and gypsum formation, while softening and disintegration are
caused by destruction of C–S–H [18,20]. Bakharev, Sanjayan and
Chen [21] carried out durability tests on alkali activated slag and
found that they perform better than ordinary portland cements.
The performance of geopolymer concretes in aggressive environ-
ments was studied using tests on absorption and acid resistance
[22].
Wallah and Rangan [23] reported that geopolymers have excel-
lent durability properties as it exhibits extremely small changes in
Table 1
Chemical composition of Fly ash, GGBS, Silica fume and Portland cement.
Chemical composition Fly ash (%)
Silicon dioxide (SiO2) 60
Iron oxide (Fe2O3) 2.50
Aluminum oxide (Al2O3) 28
Calcium oxide (CaO) 2.50
Magnesium oxide (MgO) 1.00
Sulfur trioxide (SO3) 0.40
Na2O 0.50
K2O 1.00
Loss on ignition (LOI) 2.50
length and also showed little increase in mass after one year of
exposure in sulfate solution. In another study, Bakharev [24] used
geopolymer materials prepared by activation using different types
of alkali solutions to be immersed in sulfate solution with various
concentrations.
Studying the durability of geopolymers against magnesium sul-
fate attack especially that contain granulated blast furnace slag is a
topic needing more investigations because the researches in this
field are limited. The main objective of this research is to deter-
mine the effect of curing temperature, sodium hydroxide solution
molarity, alkaline solution to binder ratio and binder type on the
performance of geopolymer mortar exposed to magnesium sulfate
attack up to one year and compare between geopolymer mortar
and portland cement mortar. Magnesium sulfate resistance was
evaluated in the terms of weight change, residual compressive
and flexural strength, expansion strain percent, visual inspection
of specimens and microstructure evaluating using scanning elec-
tron microscope (SEM), X ray diffraction (XRD) and thermogravi-
metric analysis (TGA). Voids ratio and water absorption tests
were also performed.
2. Experimental program
2.1. Materials
In this study, class F fly ash according to ASTM C618, ground granulated blast
furnace slag (GGBFS) and silica fume were used as the main silicon-alumina source
materials for geopolymer mortar. Ordinary portland cement Type I according to
ASTM C 150 was used to compare the sulfate resistance of geopolymer mortar with
portland cement mortar. The chemical compositions of the used fly ash, GGBFS, sil-
ica fume and portland cement are presented in Table 1. Natural siliceous sand with
fineness modulus of 2.76 and specific gravity of 2.62 was used as natural aggregate.
A combination of sodium hydroxide (NaOH) and sodium silicate (Na2SiO3) solu-
tions were used as alkali activators of binders. Sodium hydroxide (NaOH) in pellet
form with 98% purity was dissolved in potable water to make the solution of the
desired concentration. Sodium silicate solution was obtained from a local commer-
cial producer. It was colorless and had a chemical composition of 14.7% Na2O 29.4%
SiO2 and 55.9% H2O. The specific gravity of sodium silicate was 1.52
High range water reducer naphthalene-based admixture (Type F) according to
ASTM C494/C494M with specific gravity of 1.2 was used to improve workability
of the fresh mortar.
2.2. Test parameters
Four parameters were considered in this study. These parameters included cur-
ing temperature, sodium hydroxide solution molarity, alkaline solution to binder
ratio and binder type. Three types of binders were used. These types were fly ash
(ASTM Class F), ground granulated blast furnace slag (GGBS) and silica fume. Three
curing temperature degrees of 30°, 60° and 90 °C were considered. For sodium
hydroxide solution molarity, four concentrations of 10 M, 12 M, 14 M and 16 M
were considered. The considered alkaline solution to binder ratios were 0.35,
0.40, 0.45 and 0.50. For all mixtures, the ratio of sodium silicate solution-to-
sodium hydroxide solution of 2, ratio of sand-to-binder of 3, extra water and admix-
ture of 6% and 3% by weight of binder respectively were kept constant. Portland
cement mix with water cement ratio of 0.35 is used as control mix in order to com-
pare between OPC mortars and geopolymer mortars exposed to magnesium sulfate
attack. To study the previous parameters, twenty-seven geopolymer mixes (9 for
GGBS (%) Silica fume (%) Cement (%)
36.74 96.81 19.6
0.40 0.45 3.42
10.78 0.25 5.30
43.34 0.16 61.6
3.21 0.26 3.40
0.50 0.14 2.30
0.18 0.14 –
0.17 0.28 –
0.60 1.30 2.60
 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127 113

each type of binder) in addition to one portland cement mortar mix were used as
presented in Table 2. Initial compressive and flexural strengths before sulfate expo-
sure are also presented in Table 2.
In this study, FA means mixes containing 100% fly ash and FAGS means mixes
containing 50% fly ash and 50% GGBS while FAGSS mixes refers to mixes containing
of 50% fly ash, 35% GGBS and 15% silica fume.
2.3. Mixing, casting and curing
Mixing process of geopolymer consisted of various steps. Alkaline solution was
prepared by mixing sodium silicate solution with sodium hydroxide solution
together before 24 h from adding to dry constituents as suggested by Rangan
[25]. Sand was well mixed with binder in a mechanical mixer. The prepared alkaline
solution was mixed with the admixture and extra water then added to dry con-
stituents and mixed for about 4 min to obtain the fresh mortar. The fresh mortar
was cast into the molds immediately after mixing then specimens were vibrated
using a vibrating table for 2 min for well compaction. The molds were left at ambi-
ent temperature. The specimens were de-molded after 24 h from casting and then
cured in the hot air oven for 48 hrs at various curing temperatures and kept at
ambient temperature (20–23 °C) until testing.
properties. Cubes of 70  70  70 mm were used to evaluate water absorption
and voids ratio according to ASTM C642-06 before immersing in magnesium sulfate
solution. Compressive strength according to BS 1881: Part 3 and flexural strength
according to ASTM C348–02 were obtained at 7 days as initial strength then also
determined after 12,24,36 and 48 weeks in sulfate solution. For weight change,
specimens were initially primed in water for 14 days and their weight in saturated
surface dry condition were recorded as initial weight. Each month, specimens were
removed from the sulfate solution, and wiped clean prior to the weight measure-
ment. Through the exposure period, specimens were weighed and the percentage
of change in weight was calculated as a percent of initial weight. Prisms of
25x25x285 mm were used to evaluate expansion strain according ASTM C157-08
each month till 48 weeks of exposure. Thermogravimetric analysis (TGA) was per-
formed with a thermo-balance Instrument. This device consists of an electronic bal-
ance which is placed inside an oven and coupled to a control microprocessor and a
processing station data. Samples in the form of fine powder of 1 g approximately
are slid and placed inside the analysis equipment for testing. Samples are then cal-
cined to a temperature of 900 °C and that range increases 10 °C per minute.
3. Test results and discussion

3.1. Effect of curing temperature

2.4. Experimental tests

Series of tests were suggested to evaluate magnesium sulfate resistance of
geopolymer mortars and OPC mortar. Specimens were exposed to 10% solution of
magnesium sulfate after 7 days from casting and kept immersed up to 48 weeks.
Specimens were left fully immersed in the sulfate solution having total volume as
four times the volume of immersed specimens and was replaced every month to
maintain the concentration of sulfates. The effect of magnesium sulfate on the
geopolymer mortar specimens were constantly monitored at regular intervals dur-
ing the exposure period by different tests such as weight change, compressive and
flexural strength loss, expansion strain, visual inspection and SEM. X ray diffraction
(XRD) and thermogravimetric analysis (TGA) were also performed to evaluate the
microstructure of mortar specimens before and after sulfate exposure. Water
absorption and voids ratio were determined to study their effect on durability
This section discusses the effect of curing temperature on voids
ratio and sulfate resistance of geopolymer mortar. The molarity of
NaOH and alkaline solution to binder ratio were kept constant at
16 M and 0.35 respectively.
3.1.1. Water absorption and voids ratio
The effect of curing temperature on water absorption and voids
ratio of geopolymer mortar before sulfate exposure is presented in
Table 3. From the results, it is obvious that curing temperature has
a significant effect on geopolymer mortar absorption and voids

Table 2
Considered parameters and mix proportions in addition to the corresponding 7 days compressive and flexural strength.

Type of binder Molarity Curing Alkaline 7 days’ 7 days’ Flexural Note

temperature solution to compressive strength before
binder ratio strength before exposure (MPa)
exposure (MPa)
FA FAGS FAGSS FA FAGS FAGSS
100% 50% Fly ash + 50% Fly ash + 35% GGBS 16 30 0.35 12 36 37 3.5 4.0 5.2 Effect of curing
Fly 50% GGBS + 15% Silica fume 16 60 0.35 31 40 43 5.5 4.9 6.4 temperature
ash (FAGS) (FAGSS) 16 90 0.35 38 43 47 6.0 5.7 7.5
(FA) 10 60 0.35 22 28 31 4.0 3.3 4.5 Effect of NaOH
12 60 0.35 25 30 36 4.7 3.7 5.2 solution molarity
14 60 0.35 27 35 40 5.2 4.5 6.0
16 60 0.40 29 34 36 4.8 4.0 5.7 Effect of alkaline
16 60 0.45 26 30 32 4.5 3.7 4.8 solution to binder
16 60 0.50 18 26 27 3.7 3.0 4.1 ratio
OPC mortar C S/C W/C 7 days’ 7 days’ Flexural –
compressive strength before
strength before exposure (MPa)
exposure (MPa)
1 3 0.35 32 4.8

Table 3
Effect of curing temperature, NaOH molarity and alkaline solution to binder ratio on water absorption and voids ratio of geopolymer mortar and OPC mortar.

Type of binder Molarity Curing Alkaline Water absorption % Voids ratio % Note
Temperature solution to
FA FAGS FAGSS FA FAGS FAGSS
binder ratio
100% 50% Fly ash + 50% Fly ash + 35% 16 30 0.35 4.27 3.10 2.63 8.98 6.90 5.69 Effect of curing
Fly 50% GGBS GGBS + 15% Silica fume 16 60 0.35 3.82 2.97 2.55 7.90 6.60 5.50 temperature
ash (FAGS) (FAGSS) 16 90 0.35 3.70 2.86 2.47 7.72 6.30 5.30
(FA) 10 60 0.35 4.32 3.30 3.20 9.15 7.35 6.10 Effect of NaOH
12 60 0.35 4.10 3.2 2.92 8.63 7.00 5.90 solution molarity
14 60 0.35 3.91 3.05 2.80 8.25 6.80 5.70
16 60 0.40 4.00 3.10 2.65 8.40 6.86 5.71 Effect of alkaline
16 60 0.45 4.26 3.20 2.73 8.97 7.10 5.9 solution to binder
16 60 0.50 4.40 3.35 2.85 9.25 7.48 6.10 ratio
OPC mortar C S/C W/C Water absorption % Voids ratio % –
1 3 0.35 6.95 14.8
114 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127

ratio where water absorption and voids ratio decrease as curing of sulfate exposure may be due to migration of alkalis from the
temperature increases for the various types of binders. For exam- specimens into the solution and also probably because a part of
ple, for FA mixes, the reduction of water absorption is 10.50% geopolymer products were broke down and dissolved [26].
and 13.3% while the reduction in voids ratio is 12% and 14% for
geopolymer mortar cured at 60 °C and 90 °C compared to geopoly-
mer mortar mix cured at 30 °C. This finding is also observed in case 3.1.3. Residual compressive strength
of other types of binders (FAGS and FAGSS mixes). The previous Residual compressive strength for geopolymer mortar speci-
trend for water absorption and voids ratio may be due to the mens immersed in 10% magnesium sulfate solution is shown in
enhancement of polymerization process which associated with Fig. 2 for various types of binders. The residual compressive
curing temperature increasing as two-dimensional polymer chains strength was calculated as a percentage of 7 days’ strength before
are converted to three-dimensional chains [6,9,10]. exposure. From the test results it can be noticed that residual com-
pressive strength increases when curing temperature increases. As
an example, for FA mixes, the final residual compressive strength
3.1.2. Weight change after 48 weeks of exposure is 66.7%, 80.6% and 84.2% at curing tem-
Fig. 1 shows the effect of curing temperature on weight change peratures of 30 °C, 60 °C and 90 °C respectively. This trend is the
as a result of magnesium sulfate attack of geopolymer mortar with same for other types of binders (FAGS and FAGSS mixes). It is obvi-
different types of binders. From the experimental test results, an ous that results show fluctuations in compressive strength up to
increase in weight over the exposure period is recorded compared 24 weeks. These fluctuations may be due to the diffusion of Mg
to initial weight. This increase is observed until around 36 weeks into the matrix which occurred at the same time of migration of
and then a decrease in specimens’ weight is noticed. Results indi- alkali ions from specimens into the solution [24]. After this period,
cate that the increase in curing temperature leads to a decrease a continuous reduction in strength is observed up to 48 weeks. As
in weight gain. In other words, it can be noticed that the least it is shown in the figures, the positive effect of the increase of cur-
weight gain is observed at 90 °C for all types of binders which ing temperature is still found. Residual compressive strength
would be better compared other curing temperatures. This may increases when curing temperature increases. This trend may be
be due to the lower voids ratio of high temperature cured speci- related to the positive effect of curing temperature on voids ratio.
mens. In general, weight gain for specimens is probably due to The optimum residual compressive strength is noticed at 90 °C for
the voids ratio and formation of reaction products such as gypsum all binders. Loss of strength may be attributed to the presence of
and ettringite which found as white deposits on the exposed
surface and within the pores. The drop in weight after 36 weeks
30°C 60°C 90°C
120.0
Residual compressive

30°C 60°C 90°C

strength %

2.50 100.0
Weight change ( % )

2.00
80.0
1.50
60.0
1.00
40.0
0.50
0 12 24 36 48 60
0.00 Exposure period ( weeks )
0 4 8 12 16 20 24 28 32 36 40 44 48 52 (a) FA mixes
Exposure period (weeks)
(a) FA mixes
30°C 60°C 90°C
120.0
Residual compressive

30°C 60°C 90°C

2.50 100.0
strength %
Weight change ( % )

2.00
80.0
1.50

1.00 60.0

0.50 40.0
0.00 0 12 24 36 48 60
0 4 8 12 16 20 24 28 32 36 40 44 48 52 Exposure period ( weeks )
Exposure period (weeks) (b) FAGS mixes
(b) FAGS mixes
30°C 60°C 90°C
120.0
30°C 60°C 90°C
Residual compressive

2.50
100.0
Weight change ( % )

2.00
strength%

1.50 80.0

1.00 60.0
0.50
40.0
0.00 0 12 24 36 48 60
0 4 8 12 16 20 24 28 32 36 40 44 48 52 Exposure period ( weeks )
Exposure period (weeks) (c) FAGSS mixes

Fig. 1. Effect of curing temperature on weight change of various geopolymer Fig. 2. Effect of curing temperature on residual compressive strength of various
mortars immersed in magnesium sulfate solution. geopolymer mortars immersed in magnesium sulfate solution.
 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127 115

micro cracks as a result of gypsum and ettringite formation in the
pores in addition to migration of alkalis from geopolymers into the
solution within pores. This trend is similar to trends reported in
previous researches [24,27].
3.1.4. Residual flexural strength
The effect of curing temperature on the residual flexural
strength due to magnesium sulfate attack is presented in Fig. 3.
30°C
Residual flexural strength %
From the test results, it is clear that the increase of curing temper-
ature from 30 °C to 90 °C enhances the magnesium sulfate resis-
tance (in the term of residual flexural strength). For example, the
final residual flexural strength of FA mixes after 48 weeks is
65.7%, 78.2% and 81.7% at curing temperatures of 30 °C, 60 °C
and 90 °C respectively. The same behaviour was observed in case
of FAGS and FAGSS mixes. The trend of residual flexural strength
is almost the same of residual compressive strength. This may be
also because the increase of curing temperature affects positively
on voids ratio and improves residual flexural strength.
60°C 90°C

120.0
3.1.5. Expansion strain
100.0 The effect of curing temperature on expansion strain of
80.0 geopolymer mortars exposed to magnesium sulfate solution up
to 48 weeks is shown in Table 4. As an example, it can be noticed
60.0 that expansion strain percent of FA mixes is in range of 0.051–
0.059% at curing temperatures from 30 °C to 90 °C which confirms
40.0
0 12 24 36 48 60 that curing temperature hasn’t a significant effect on expansion
Exposure period ( weeks ) strain. Generally, increasing curing temperature decreases the
(a) FA mixes resulting expansion strain. This behaviour may be due to the pos-
itive effect of curing temperature on geopolymerization process
30°C 60°C 90°C
Residual flexural strength %

120.0 and voids ratio.

100.0
3.2. Effect of sodium hydroxide solution molarity
80.0

60.0 In this section, curing temperature and alkaline solution to bin-

der ratio were kept constant at 60 °C and 0.35 respectively.
40.0
0 12 24 36 48 60
Exposure period ( weeks )
3.2.1. Water absorption and voids ratio
(b) FAGS mixes
Table 3 summarizes the effect of NaOH molarity on water
absorption and voids ratio of geopolymer mortar. Results show
Residual flexural strength %

30°C 60°C 90°C

120.0
that water absorption and voids ratio decrease when NaOH molar-
100.0 ity increases for the various types of binders. As an example, for FA
mixes, the reduction in water absorption is 5.1%, 9.5% and11.6 %
80.0 while the reduction in voids ratio is 5.7%, 9.8% and 13.7% for
geopolymer mortar with 12 M, 14 M and 16 M respectively com-
60.0
pared to geopolymer mortar mix with 10 M. This trend is the same
40.0 for other types of binders (FAGS and FAGSS mixes). This finding
0 12 24 36 48 60 may be because the increase in sodium hydroxide solution molar-
Exposure period ( weeks )
ity from 10 M to 16 M leads to enhance the density of mix and
(c) FAGSS mixes
reduce the voids content as the increase in molarity contributed
Fig. 3. Effect of curing temperature on residual flexural strength of various to reduce the water content. This improvement in properties
geopolymer mortars immersed in magnesium sulfate solution. agrees with several previous studies [28,29].

Table 4
Effect of different parameters on 7 days’ strength and durability properties after 48 weeks in 10% magnesium sulfate solution.

Binder type Effect of curing temperature

FA FAGS FAGSS OPC
Curing temperature 30 °C 60 °C 90 °C 30 °C 60 °C 90 °C 30 °C 60 °C 90 °C
Expansion strain% 0.059 0.053 0.051 0.041 0.038 0.035 0.026 0.024 0.021 0.09

Binder Type Effect of sodium hydroxide solution molarity

FA FAGS FAGSS
NaOH Molarity 10M 12M 14M 16M 10M 12M 14M 16M 10M 12M 14M 16M
Weight change % +1.35 +1.29 +1.25 +1.20 +0.66 +0.58 +0.53 +0.5 +0.27 +0.25 +0.20 +0.18
Residual flexural strength after exposure % 72.50 74.50 76.90 78.20 75.80 78.40 80 83.70 77.80 80.80 83.30 85.90
Expansion strain % 0.058 0.057 0.055 0.053 0.043 0.041 0.04 0.038 0.029 0.027 0.026 0.024
Binder Type Effect of alkaline solution to binder ratio
FA FAGS FAGSS
Alkaline solution to binder ratio (S/b) 0.35 0.40 0.45 0.50 0.35 0.40 0.45 0.50 0.35 0.40 0.45 0.50
Weight change % +1.20 +1.13 +1.25 +1.49 +0.50 +0.60 +0.64 +0.79 +0.18 +0.27 +0.30 +0.41
Residual flexural strength after exposure % 83.70 80 75.70 73.30 85.90 82.50 79.20 75.60 85.90 82.50 79.20 75.60
Expansion strain% 0.053 0.055 0.058 0.062 0.038 0.041 0.043 0.046 0.024 0.026 0.029 0.031
116 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127

3.2.2. Weight change
The effect of sodium hydroxide solution molarity on weight
change of geopolymer mortar after 48 weeks of magnesium sulfate
solution exposure for different types of binders is presented in
Table 4. All mixes recorded an increase in weight over the exposure
period compared to initial weight. it can be noticed from the
results that sodium hydroxide solution molarity has insignificant
effect on weight change for the same binder. In general, the least
weight gain is associated with the highest molarity. This trend
may be because the increase of sodium hydroxide solution molar-
ity reduces voids ratio as mentioned which decreases weight gain.
3.2.3. Residual compressive strength
The effect of sodium hydroxide solution molarity on residual
compressive strength with time for various types of binders is
shown in Fig. 4. From figures, it’s clear that residual compressive
strength increases when NaOH molarity increases. For example,
the final residual compressive strength of FA mixes after 48 weeks
of exposure is 72.7%, 76%, 77.8% and 80.6% for NaOH molarities of
10 M, 12 M, 14 M and16 M respectively. This behaviour is the same
in case of other binders (FAGS and FAGSS mixes). This trend may be
because increasing molarity affects positively on reducing voids
ratio and water content.
3.2.4. Residual flexural strength
Table 4 presents the final residual flexural strength for geopoly-
mer mixes after 48 weeks of exposure to magnesium sulfate con-
sidering the effect of sodium hydroxide solution molarity.
Residual flexural strength was calculated as a percentage of 7 days’
strength before exposure. From the results it can be observed that
increasing NaOH molarity affected positively on residual flexural
strength. As an example, the final residual flexural strength of FA
mixes after 48 weeks is 72.50%, 74.50%, 76.90% and 78.20% at
NaOH molarities of 10 M, 12 M, 14 M and16 M respectively which
is similar to the trend in case of other binders (FAGS and FAGSS
mixes). The behaviour of residual flexural strength is almost the
same of residual compressive strength behaviour. The optimum
residual flexural strength is observed at 16 M for all types of
binders.
3.2.5. Expansion strain
The effect of sodium hydroxide solution molarity on expansion
strain of geopolymer mortars after 48 weeks in 10% magnesium
sulfate solution is indicated in Table 4. From table, it can be
observed that expansion strain percent of FA mixes is in the range
of 0.053–0.058% and for FAGS mixes, it ranges from 0.038% to
0.043% while for FAGSS mixes, it ranges from 0.024% to 0.029%
for NaOH molarities ranging from 10 M to 16 M. From results it
can be observed that NaOH molarity doesn’t affect greatly on
expansion strain of geopolymer mortar for the same type of binder.
However, it’s clear that expansion strain decreases by NaOH molar-
ity increasing. This may be because increasing molarity enhances
the density of the mix which reduces permeability and expansion
consequently.
3.3. Effect of alkaline solution to binder ratio (S/b)
Curing temperature and sodium hydroxide solution molarity
were kept constant at 60 °C and 16 M respectively in this section.

10 M 12 M 14 M 16 M 3.3.1. Water absorption and voids ratio

120
The effect of alkaline solution to binder ratio on water absorp-
Residual compressive

tion and voids ratio of geopolymer mortar is presented in Table 3.

strength %

100
Results show that water absorption and voids ratio increase when
alkaline solution to binder ratio increases for the various types of
80
binders. As an example, for FA mixes, the increase in water absorp-
60
tion is 4.7%, 11.5% and 15.2% while the increase in voids ratio is
6.3%, 13.5% and 17.1% for geopolymer mortar with 0.40, 0.45 and
40 0.50 alkaline solution to binder ratio respectively compared to
0 12 24 36 48 60 geopolymer mortar mix with ratio of 0.35. This trend is also
Exposure period ( weeks ) observed for other types of binders (FAGS and FAGSS mixes).
(a) FA mixes According to Hardjito and Rangan [25], water in a geopolymer mix-
10 M 12 M 14 M 16 M ture don’t play any role in the chemical reaction that takes place.
120
This water expelled from the geopolymer mixture during the cur-
Residual compressive

100 ing and drying periods, leaves some discontinuous nano-pores in

strength %

the matrix which improve the workability of the mixture during

80 handling and casting. In addition to the increased water content,
60 an amount of alkalis remains unreacted with binder when alkaline
solution content exceeds the appropriate limit which leaves
40 behind more pores and increases water absorption and voids ratio.
0 12 24 36 48 60
Exposure period ( weeks ) Results confirm that the determination of appropriate amount of
(b)FAGS mixes alkaline liquid is essential to achieve effective reaction with binder
10 M 12 M 14 M 16 M to obtain the desired properties. The previous trend is in agree-
120 ment with various literatures. [30,31].
Residual compressive

100
strength %

3.3.2. Weight change

80 Weight change of geopolymer mortar exposed to 10% magne-
sium sulfate solution for 48 weeks considering the effect of differ-
60 ent ratios of alkaline solution to binder for various types of binders
40
is presented in Table 4. As shown in the table, weight gain over the
0 12 24 36 48 60 exposure period is recorded for all mixes compared to initial
Exposure period ( weeks ) weight. Results reveal that the least weight gain for all mixes is
(C) FAGSS mixes
observed at solution to binder ratio of 0.35 and maximum weight
Fig. 4. Effect of NaOH molarity on residual compressive strength of various gain is noticed at the ratio of 0.50. It’s clear that the weight gain of
geopolymer mortars immersed in magnesium sulfate solution. specimens decreases when alkaline solution to binder ratio
 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127 117

deceases which confirms the positive effect on decreasing voids
ratio by decreasing alkaline solution to binder ratio.
3.3.3. Residual compressive strength
Fig. 5 show the residual compressive strength for different
geopolymer mortars after exposing to magnesium sulfate attack
at various ages up to 48 weeks. From figures, the negative effect
of increasing solution to binder ratio on residual compressive
strength is observed. Maximum residual compressive strength is
noticed at solution to binder ratio of 0.35 while the least residual
compressive strength is observed at the ratio of 0.50 for all binders.
As an example, the residual compressive strength of FA mixes is
80.6%, 75.9%, 73.1% and 72.2% for solution to binder ratios of
0.35, .0.40, 0.45 and 0.50 respectively. The same trend is noticed
for the other types of binder (FAGS and FAGSS mixes).
3.3.5. Expansion strain
Table 4 presents the effect of alkaline solution to binder ratio on
expansion strain of geopolymer mortars with different types of
binder after 48 weeks in 10% magnesium sulfate solution. As
shown in the table, expansion strain for all geopolymer mixes
increases by the increase of alkaline solution to binder ratio. For
example, the expansion strain of FA mixes ranges from 0.053% to
0.062% for alkaline solution to binder ratios ranging from 0.35 to
0.50. The same trend is observed in case of the other types of bin-
ders. This may be because of water content increasing which is
associated with the increase of solution to binder ratio where this
water may create voids in the specimen which facilitate magne-
sium sulfate penetration and causes expansion.
3.4. Effect of binder type

3.3.4. Residual flexural strength This section discusses the effect of binder type at various curing
The effect of alkaline solution to binder ratio on residual flexu- temperatures, NaOH molarities, solution to binder ratios durability
ral strength of geopolymer mortar with different types of binder
after 48 weeks in magnesium sulfate solution is presented in
Table 4. Experimental test results indicate that residual flexural FA FAGS FAGSS OPC
2.50
strength is affected by solution to binder ratio where the residual
2.00
Weight change ( % )
strength decreases by the increase of alkaline solution to binder 1.50
ratio for all types of binders. For example, the residual flexural 1.00
strength of FA mixes is 83.7%, 80%, 75.7% and 73.3% for solution 0.50
to binder ratios of 0.35, 0.40, 0.45 and 0.50 respectively. This beha- 0.00
-0.50 0 4 8 12 16 20 24 28 32 36 40 44 48 52
viour is almost the same that of residual compressive strength.
-1.00
This trend is related to the increase of voids ratio. -1.50
-2.00
-2.50
0.35 0.4 0.45 0.5 -3.00
120 Exposure period (weeks)
Residual compressive

100 Fig. 6. Effect of binder type on weight change of various geopolymer mortar cured
at 30 °C and OPC mortar immersed in 10% magnesium sulfate solution.
strength %

80

60

FA FAGS FAGSS OPC

40 100
88.4 88.4 91.5
0 12 24 36 48 60 90 83.3 86.5 80.6
85 84.2
Residual Compressive

Exposure period ( weeks ) 80

66.7
(a) FA mixes 70
strength %

59.4
60
50
0.35 0.4 0.45 0.5 40
120 30
20
Residual compressive

100 10
0
strength %

30 60 90
80
Curing temperature °C

60
Fig. 7. Effect of binder type on residual compressive strength of geopolymer and
OPC mortars cured at various curing temperatures after 48 weeks in magnesium
40 sulfate solution.
0 12 24 36 48 60
Exposure period ( weeks )
(b)FAGS mixes

0.35 0.4 0.45 0.5 FA FAGS FAGSS

120 100
85 88.4
residual Compressive

82.9 85
Residual compressive

90 78.6 80.6 80 83.3 77.8 80.6

76
100 80 72.7
strength %

strength %

70
80 60
50
40
60 30
20
40 10
0 12 24 36 48 60 0
Exposure period ( weeks ) 10 12 14 16
(c) FAGSS mixes Sodium hydroxide solution molarity

Fig. 5. Effect of alkaline solution to binder ratio on residual compressive strength of Fig. 8. Effect of binder type on residual compressive strength of geopolymer mortar
different geopolymer mortars immersed in magnesium sulfate solution. with various NaOH molarities after 48 weeks in magnesium sulfate solution.
118 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127

properties of geopolymer mortars after magnesium sulfate On the other hand, Table 3 shows that water absorption and
exposure. voids ratio of OPC mortar specimens are 6.95% and 14.8% respec-
tively which are higher than those of various geopolymer mortars.
3.4.1. Water absorption and voids ratio
Results shown in Table 3 indicate that, at the same curing tem-
perature, NaOH molarity and alkaline solution to binder ratio, FA FAGS FAGSS OPC
100
FAGSS mixes give lowest voids ratio and water absorption com-

residual flexural strength %

86 88
90 83.7 85.9 81.7
78.8
pared to FAGS and FA mixes while FA mixes have the highest voids 80 75 78.2
65.7
ratio and water absorption. As an example, at NaOH molarity of 16 70 62
60
M, curing temperature of 60 °C and alkaline solution to binder ratio
50
of 0.35, using FAGSS mixes decrease water absorption by 33.2% and 40
14.1% and voids ratio by 30.3% and 16.7% when compared to FA 30
and FAGS mixes respectively. 20
10
Also, it can be noticed that the effect of these previous param- 0
eters is highly noticed for FA mixes compared to FAGS and FAGSS 30 60 90
mixes. The presence of GGBS which is the main resource of Ca Curing temperature °C
(Table 1) makes the mixtures more compacted and yields higher
Fig. 11. Effect of binder type on residual flexural strength of geopolymer and OPC
density due to the formation of additional geopolymerization mortars cured at various curing temperatures after 48 weeks in magnesium sulfate
products because geopolymers consisting of both fly ash and slag solution.
contain multiple gels (e.g., N–A–S–H, C–A–S–H) which reduce
voids ratio. This explanation agrees with several previous studies
[32,33,34]. Moreover, silica fume is widely used to enhance the FA FAGS FAGSS

residual flexural strength %

100
durability of concrete because it decreases the permeability of 90 83.3 83.7 85.9
77.8 78.4 80.8 76.9 80 78.2
concrete as a micro filler due to its high fineness and high silica 80 72.5 75.8 74.5

content [35]. 70
60
When slag is replaced with silica fume in a geopolymer mix 50
based on fly ash and slag (FAGSS), the Si and Al in the slag 40
30
contributed to the formation of both the aluminosilicate and the 20
C-S-H gel in the specimens and this process increases the reactivity 10
of the slag which improves density and reduces voids ratio. It was 0
10 12 14 16
also investigated that the alkali activator reacted more favorably Sodium hydroxide solution molarity
with the silica fume than with the fly ash [36]. This trend agrees
with previous researches [37–39]. Fig. 12. Effect of binder type on residual flexural strength of geopolymer mortar
with various NaOH molarities after 48 weeks in magnesium sulfate solution.

FA FAGS FAGSS
100
residual Compressive strength %

88.4 86.1
90 85 82.4 84.4
81.5 FA FAGS FAGSS
80.6 80
75.9 76.9 100
Residual flexural strength %

80 73.1 72.2
90 83.7 85.9
70 78.2 80 82.5 79.2
75.7 76.6
80 75
71.1 73.3 75.6
60
70
50 60
40 50
30 40
20 30
20
10
10
0
0
0.35 0.4 0.45 0.5 0.35 0.4 0.45 0.5
Alkaline solution to binder ratio (S/b) Alkaline solution to binder ratio (S/b)

Fig. 9. Effect of binder type on residual compressive strength of geopolymer mortar Fig. 13. Effect of binder type on residual flexural strength of geopolymer mortar
with various solution to binder ratios after 48 weeks in magnesium sulfate solution. with various (S/b) ratios after 48 weeks in magnesium sulfate solution.

FA FAGS FAGSS OPC FA FAGS FAGSS OPC

Residual compressive strength %

120.0 120.0
Residual flexural strength %

100.0 100.0

80.0 80.0

60.0 60.0

40.0 40.0
0 12 24 36 48 60 0 12 24 36 48 60
Exposure period ( weeks ) Exposure period ( weeks )

Fig. 10. Residual compressive strength test results of geopolymer mortars cured at Fig. 14. Residual flexural strength test results of geopolymer mortars cured at 30 °C
30 °C and OPC mortar exposed to 10% magnesium sulfate solution. and OPC mortar exposed to 10% magnesium sulfate solution.
 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127 119

3.4.2. Weight change

FA FAGS FAGSS OPC The effect of binder type at various curing temperatures, NaOH
0.100
molarities, solution to binder ratios on weight gain of specimens
Expansion strain ( % )

0.080 after 48 weeks in magnesium sulfate solution can be observed in

0.060
Table 4. At the same curing temperature, NaOH molarity and alka-
line solution to binder ratio, results indicate that FAGSS mixes have
0.040 the least weight gain then FAGS mixes while FA mixes have max-
0.020 imum weight gain. For example, at NaOH molarity of 16 M, curing
temperature of 60 °C and alkaline solution to binder ratio of 0.35,
0.000
0 4 8 12 16 20 24 28 32 36 40 44 48 52
FAGSS mixes reduce weight gain by 85% and 64% when compared
Exposure period (weeks) to FA and FAGS mixes respectively. Previous values are based on
results presented in Table 4. These results may be due to the pres-
Fig. 15. Effect of binder type on expansion strain of various geopolymer mortar ence of GGBS and silica fume in FAGSS mixes which enhances the
cured at 30 °C and OPC mortar exposed to 10% magnesium sulfate solution.
density of the mixture and reduces voids ratio compared to FA

Fig. 16. Geopolymer and OPC mortar specimens (a) FA, (b) FAGS, (c) FAGSS, (d) OPC after 48 weeks in 10% magnesium sulfate solution.

Fig. 17. SEM micrographs of FA specimen before exposure (a) and after 48 weeks of exposure (b), (c).
120 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127

Fig. 18. SEM micrographs of FAGS specimen before exposure (a) and after 48 weeks of exposure (b), (c).

Fig. 19. SEM micrographs of FAGSS specimens before exposure (a) and after 48 weeks of exposure (b), (c).

Fig. 20. X-ray diffractogram of FAGS mix before exposure.

 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127 121

mixes. Fig. 6 shows a comparison of weight change between vari-
ous geopolymer mortars cured at 30 °C and OPC mortar. From the
figure, it can be observed that all geopolymer mortars
achieve weight gain while OPC mortar shows initial increase in
weight then there is a continuous decrease in weight where the
weight loss of OPC mortar specimen after 48 weeks of exposure
is 2.5%.
3.4.3. Residual compressive strength
Figs. 7–9 show the effect of binder type on various geopolymer
mortar with different curing temperatures, NaOH molarities and
alkaline solution to binder ratios. From figures, it can be noticed
that FAGSS mixes achieve the highest residual compressive
strength while FA mixes have the least residual compressive
strength at the same curing temperature, NaOH molarity and

(a) FA mix

(b) FAGS mix

Fig. 21. X-ray diffractogram of geopolymer mortar after 48 weeks of 10% magnesium sulfate exposure.
122 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127
(c) FAGSS mix
Fig. 21 (continued)
alkaline solution to binder ratio after 48 weeks from immersing in
magnesium sulfate solution. It’s clear that lower voids ratio for
FAGSS mixes affects positively on durability properties which
enhances residual compressive strength compared to those of
FAGS and FA mixes. It can also be noticed that OPC mortar has min-
imum residual compressive strength compared to those of various
geopolymer mortars as shown in Fig. 7. This may be because OPC
mix has the highest voids ratio compared to those of FA, FAGS
and FAGSS mixes, thus magnesium sulfate diffusion would be
easier in OPC specimen and cause the maximum strength reduc-
tion. Fig. 10 shows the residual compressive strength of test results
for FA, FAGS, FAGSS mixes cured at 30 °C and OPC mix exposed to
10% magnesium sulfate solution.
3.4.4. Residual flexural strength
The effect of binder type on residual flexural strength of
geopolymer mortar at various curing temperatures, NaOH molari-
ties, alkaline solution to binder ratios after 48 weeks from immers-
ing in magnesium sulfate solution is shown in Fig. 11. From these
figures, it can be observed that at the same curing temperature,
NaOH molarity and alkaline solution to binder ratio, FAGSS mixes
yield the highest residual flexural strength compared to those of
other types of binder (FAGS and FAGSS mixes). It can be noticed
from Fig. 11 that OPC mix has the least residual flexural strength
compared to different types of geopolymer mortars. This may also
be related to the highest voids ratio of OPC specimens compared to
geopolymer mortar specimens. Residual flexural strength test
results of FA, FAGS, FAGSS mixes cured at 30 °C and OPC mix
exposed to 10% magnesium sulfate solution is shown in Fig. 14.
From this figure, it can be observed that residual flexural strength
of OPC specimen is the highest at 12 weeks of exposure, but it
recorded a continuous decrease up to 48 weeks and became the
lowest residual strength.
3.4.5. Expansion strain
From Table 4, it can be observed that, at the same curing tem-
perature, NaOH molarity and alkaline solution to binder ratio,
FAGSS mixes record the least expansion strain while FA mixes have
the highest expansion strain. This may be because FA mixes have
the highest voids ratio compared to those of other geopolymer
mortars which enable magnesium sulfate solution to access into
pores of the specimen. In addition, this could be because some of
fly ash particles don’t react completely with the solution, so the
unreacted part may work as a pozzolanic material not geopolymer
which enable to form gypsum phases. According to previous
researches, another possible reason is due to migration of Mg
and S from magnesium sulfate solution into the matrix and the
migration of Ca and alkalis from inside of the specimen to the sur-
face area which causes the destruction of silicon-oxygen-silicon
bonds in xAl2O3
ySiO2 gel structure. As a result, sodium aluminum
silicate hydrate (N–A–S–H) gels reacted with MgSO4, creating low
strength magnesium aluminum silicate hydrate (M–A–S–H) gels.
[40,41].
Although FAGS and FAGSS mixes have high content of CaO,
these mixes yield lower expansion percent compared to those of
FA mixes. This may be because CaO in GGBS was consumed in reac-
tions of hydration and geopolymerization process which resulted
in very dense gel and strong cross-linked networks leading to
decrease the voids and block the pores in gel. Gypsum crystals
were also formed in FAGS and FAGSS mixes which cause expansion
but lower than FA mixes which have the highest voids ratio.
On the other hand, OPC mortar specimen records an expan-
sion strain percent of 0.09% which is the highest value compared
to those of various geopolymer mortars. Expansion strain is gen-
erated by the formation of high volume gypsum and ettringite
crystals along the walls of capillary and gel pores [42] which
resulted from reacting of sulfate ions with calcium hydroxide.
Fig. 15 shows the expansion strain test results of FA, FAGS,
FAGSS mixes cured at 30 °C and OPC mix exposed to 10% mag-
nesium sulfate solution. The figure shows that OPC mortar gives
the highest values of expansion strain compared to those of
geopolymer mortars. For geopolymer mortars, the rate of expan-
sion increases obviously after 24 weeks of exposure. Generally,
the increase of voids ratio increases the diffusion coefficient of
sulfate ions into the specimens which leads to the reduction of
sulfate [43]. ACI 201 indicated that the concrete permeability
has a significant effect not only for sulfate resistance but also
for all chemical attack [44].
 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127 123

Fig. 22. TGA and DTG results of FA samples before exposure (a) and after 48 weeks of exposure (b).

3.5. Visual inspection and scanning electron microscope (SEM)
In this section, only the effect of binder type on visual inspec-
tion and SEM of geopolymer and OPC mortars after 48 weeks of
exposure to 10% magnesium sulfate is studied. Curing temperature,
NaOH molarity and alkaline solution to binder ratio of 60 °C, 16 M
and 0.35 respectively were kept constant. Portland cement mortar
(OPC) with water cement ratio of 0.35 was immersed in 10% mag-
nesium sulfate solution up to 48 weeks to compare its appearance
with geopolymer mortars. Fig. 16 show the appearance of speci-
mens after 48 weeks of magnesium sulfate solution exposure for
FA, FAGS, FAGSS and OPC mixes. From the figure, it can be noticed
that various geopolymer mortars did not show any change in shape
and remained without visible cracks but surfaces received white
deposits due to sulfate exposure which soft and powdery became
harder with time. In addition, few needles were observed in the
surface of specimens. Thus, it can be expected that geopolymer
specimens weight will increase. On the contrary, OPC mortar spec-
imen showed a surface erosion and spalling of edges which cause
weight loss of specimen.
On the other hand, scanning electron microscope (SEM) is per-
formed on unexposed specimens as well as on those exposed to
124 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127

Fig. 23. TGA and DTG results of FAGS samples before exposure (a) and after 48 weeks of exposure (b).

10% magnesium sulfate solution for 48 weeks. Fig. 17 show SEM
micrographs of FA, FAGS and FAGSS mixes respectively for unex-
posed specimens (a) and specimens exposed to magnesium sulfate
attack (b, c). Unexposed specimens show the formed gel with some
pores which may exist due to un reacting of some binder particles.
Bakharev [24] reported diffusion of magnesium and sulfur into the
surface of specimens from the exposure solution. At the same time
migration of calcium from within the mortar matrix to the surface
area occurred. Formation of gypsum crystals encrusted in the
geopolymer phase (CaSO4
2H2O) were noticed in the graphs. Gyp-
sum and ettringite are known to cause expansion in cement con-
crete when exposed in sulfate solution [18,20], thus, micrographs
show localized cracks which could be attributed to formation of
gypsum and ettringite. These cracks lead to compressive and flex-
ural strength reduction. From graphs after sulfate exposure, it can
be noticed that FAGSS mixes have the least micro cracks and pores
 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127 125

Fig. 24. TGA and DTG results of FAGSS samples before exposure (a) and after 48 weeks of exposure (b).

and dense microstructure compared to FAGS and FA mixes which
confirms the best mechanical and sulfate resistance of FAGSS
mixes as obtained in this study.
3.6. X-ray diffraction analysis (XRD) and thermogravimetric analysis
(TGA/DTG)
Fig. 20 shows X-ray diffractogram of geopolymer mortar
sample (FAGS) before exposure to magnesium sulfate. The
diffractogram shows peaks located around 30° 2H which corre-
sponded to the main phase C-(A)-S-H that overlapped the calcite
phase due to a weathering effect [45]. This pattern is also
observed in case of FA and FAGS samples. After 48 weeks of
10% magnesium sulfate exposure, the XRD patterns presented in
Fig. 21 for FA, FAGS and FAGSS samples, show a new phase with
high intensity (G: gypsum), which can be produced by decompo-
sition of the C-(A)-S-H phase, as illustrated by the lower intensity
of its corresponding diffraction peak [46]. It can be noticed that
126 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127
the intensity of gypsum phase is the least in FAGSS mixes com-
pared to that of FA and FAGS mixes.
On the other hand, Figs. 22–24 show the TGA and DTG data for
the samples before exposure (a) and after 48 weeks of 10% mag-
nesium sulfate exposure (b) for different types of binders. From
these figures, it can be observed that mass loss is much higher
in all samples after exposure to MgSO4, when compared with
unexposed samples. Differential thermograms (DTG) of the sam-
ples show noticeable peaks occurred at various temperatures
which indicate decomposition of various phases. It’s clear that
peaks around a temperature of 100 °C are related to the loss of
free evaporable water which is present in the pores of the
geopolymer gel products, either N–A–S–H or C–(A)–S–H gels
[47]. The other significant degradation occurs between 120 and
200 °C which causes a weight loss ranges from 0.5% to 1.5%
approximately and is associated with the presence of NASH and
CSH gels in the samples before exposure [48]. At the range
between 650 °C and 750 °C approximately, weight loss is
observed in all samples and it is attributed to the decomposition
of the carbonate minerals. After sulfate exposure, peaks are
observed between 110 °C and 200 °C and confirm the presence
of gypsum which starts to decompose between 110 °C and
150 °C [49,50]. According to the XRD results, the C–(A)–S–H
phase has been consumed from the samples through exposure
to MgSO4. Gypsum is known to dehydrate through an intermedi-
ate calcium sulfate hemihydrate (CaSO4
0.5H2O) phase, to
anhydrite (CaSO4) between 150 °C and 200 °C [51].
Moreover, the maximum weight loss in gypsum zone is
observed in FA mixes by 2.67% while FAGSS mixes have the least
weight loss of gypsum by 0.76% which indicate that the amount
of formed gypsum in FAGSS mixes is the least compared to that
of FAGS and FA mixes. These results confirmed the best magne-
sium sulfate resistance of FAGSS mixes.
4. Conclusion
From the present study, the following conclusions may be
drawn.
- Increasing curing temperature from 30 °C to 90 °C leads to a
decrease of water absorption and voids ratio and enhances
magnesium sulfate resistance of various geopolymer mortars.
- Increasing sodium hydroxide solution molarity from 10 M to
16 M decreases water absorption and voids ratio and improves
magnesium sulfate resistance of various geopolymer mortars.
- Increasing alkaline solution to binder ratio from 0.35 to 0.50
increases water absorption and voids ratio and reduces magne-
sium sulfate resistance of various geopolymer mortars.
- Type of binder affect magnesium sulfate resistance. FAGSS mor-
tars containing 50% fly ash, 35% GGBS and 15% silica fume have
the best performance in magnesium sulfate resistance com-
pared to FAGS mortars containing 50% fly ash and 50% GGBS
while FA mortars containing 100% fly ash show the least
performance.
- Various geopolymer mortars didn’t show any change in shape
and remained without visible cracks. White deposits and few
needles due to sulfate exposure were observed which cause
weight gain. On the contrary, OPC mortar specimen showed a
surface erosion and spalling of edges which cause weight loss
of specimen.
- SEM micrographs show the formations of gypsum crystals after
48 weeks of 10% magnesium sulfate exposure for various
geopolymer mortars. FAGSS mixes achieve very dense
microstructure and narrow micro cracks compared to FA and
FAGS mixes.
- XRD and TGA analysis confirm the presence of gypsum phase in
geopolymer mortar after magnesium sulfate exposure and its
intensity is the least in case of FAGSS mixes compared to FA
and FAGSS mixes which prove the best sulfate resistance of
FAGSS mixes.
- Voids ratio increases the diffusion of sulfate ions into the spec-
imens which reduces sulfate resistance and leads to strength
reduction. Thus, attempt of decreasing voids ratio by using
appropriate binder type, curing temperature, NaOH molarity
and alkaline solution to binder ratio is essential.
Generally, various geopolymer mortars achieved better magne-
sium sulfate resistance than OPC mortars.
5. Conflict of interest
None
References
[1] V.M. Malhotra, Introduction: sustainable Development and Concrete
technology, ACI Concr. J. (2000) 1147–1165.
[2] Kumar, P. Mehta, Concrete in the Marine Environment, American Concrete
Institute, Farmington Hills, Michigan, 1991.
[3] R. McCaffrey, Climate Change and the Cement Industry, Global Cement and
Lime Magazine, (Environmental Special Issue), 2002, pp. 15–19.
[4] J. Davidovits, High-Alkali Cements for 21st Century Concretes, Special
Publication, 1994, pp. 383–398.
[5] J. Davidovits, Soft mineralogy and geopolymers. in: Proceedings of the of
Geopolymer 88 International Conference, 1988.
[6] A. Palomo, M.W. Grutzeck, M.T. Blanco, Alkali-activated fly ashes: a cement for
the future, Cem. Concr. Res. 29 (8) (1999) 1323–1329.
[7] D.M. Roy, Alkali-activated cements opportunities and challenges, Cem. Concr.
Res. 29 (2) (1999) 249–254.
[8] B. Vijaya Rangan, Fly ash-based geopolymer concrete, in: Proceedings of the
International Workshop on Geopolymer Cement and Concrete, Allied
Publishers Private Limited, Mumbai, India, December 2010, pp. 68–106.
[9] K. Vijai, R. Kumutha, B.G. Vishnuram, Effect of types of curing on strength of
geopolymer concrete, Int. J. Phys. Sci. 5 (9) (2010) 1419–1423.
[10] A.A. Aliabdo, A.E.M. Abd Elmoaty, M.F. Mohamed, Permeability indices and
corrosion resistance of geopolymer and Portland cement concretes, Mag.
Concr. Res. 70 (12) (2018) 595–609.
[11] M.L. Granizo, S. Alonso, M.T. Blanco-Varela, A. Palomo, Alkaline activation of
metakaolin: effect of calcium hydroxide in the products of reaction, J. Am.
Ceram. Soc. 85 (1) (2002) 225–231.
[12] C.K. Yip, G.C. Lukey, J.L. Provis, J.S. van Deventer, Effect of calcium silicate
sources on geopolymerisation, Cem. Concr. Res. 38 (4) (2008) 554–564.
[13] J.V. Temuujin, A. Van Riessen, R. Williams, Influence of calcium compounds on
the mechanical properties of fly ash geopolymer pastes, J. Hazard. Mater. 167
(1) (2009) 82–88.
[14] P. Duxson, A. Fernández-Jiménez, J.L. Provis, G.C. Lukey, A. Palomo, J.S.J. Van
Deventer, Geopolymer technology: the current state of the art, J. Mater. Sci. 42
(9) (2007) 2917–2933.
[15] E.I. Diaz, E.N. Allouche, S. Eklund, Factors affecting the suitability of fly ash as
source material for geopolymers, Fuel 89 (5) (2010) 992–996.
[16] P.K. Mehta, Sulfate attack on concrete: a critical review, in: R.R. Villarreal (Ed.),
Concrete Durability, Univ. Autonoma de Nuevo Leon, 1993, pp. 107–132.
[17] E.F. Irassar, A. Di Maio, O.R. Batic, Sulfate attack on concrete with mineral
admixtures, Cem. Concr. Res. 26 (1) (1996) 113–123.
[18] C.F. Ferraris Jr, P.S. Clifton, E. Garboczi, 22 Mechanisms of degradation of
Portland cement-based systems by sulfate attack. in: Mechanisms of chemical
degradation of cement-based systems, 1997, pp.185–192.
[19] M. Santhanam, M.D. Cohen, J. Olek, Mechanism of sulfate attack: a fresh look:
Part 2. Proposed mechanisms, Cem. Concr. Res. 33 (3) (2003) 341–346.
[20] H.F.W. Taylor, R.S. Gollop, Some chemical and microstructural aspects of
concrete durability, in: K.L. Scrivener, J.F. Young (Eds.), Mechanisms of
Chemical Degradation of Cement-Based Systems, E & FN Spon, London, 1997,
pp. 177–184.
[21] T. Bakharev, J.G. Sanjayan, Y.B. Cheng, Resistance of alkali-activated slag
concrete to acid attack, Cem. Concr. Res. 33 (10) (2003) 1607–1611.
[22] Manu Santhanam et al., Durability Study of low calcium fly ash geopolymer
concrete, in: Proceedings of third ACF International conference-ACF/VCA,
2008, pp. 1153–1159.
[23] S.E. Wallah, B.V. Rangan, Low-Calcium Fly Ash-Based Geopolymer Concrete:
Long-Term Properties, Research Report GC2 Faculty of Engineering, Curtin
University of Technology, Perth, 2006.
[24] T. Bakharev, Durability of geopolymer materials in sodium and magnesium
sulfate solutions, Cem. Concr. Res. 35 (6) (2005) 1233–1246.
 H.E. Elyamany et al. / Construction and Building Materials 184 (2018) 111–127
[25] D. Hardjito, B.V. Rangan, Development and Properties of Low-Calcium Fly Ash-
based Geopolymer Concrete, Research Report GC1 Faculty of Engineering,
Curtin University of Technology, Perth, 2005.
[26] S. Thokchom, D. Dutta, S. Ghosh, Effect of incorporating silica fume in fly ash
geopolymers, World Acad. Sci. Eng. Technol. 60 (2011) 243–247.
[27] S. Thokchom, P. Ghosh, S. Ghosh, Performance of fly ash based geopolymer
mortars in sulphate solution, J. Eng. Sci. Technol. Rev. 3 (1) (2010) 36–40.
[28] D. Hardjito, S.E. Wallah, D.M. Sumajouw, B.V. Rangan, On the development of
fly ash-based geopolymer concrete, ACI Mater. J. Am. Concr. Inst. 101 (6)
(2004) 467–472.
[29] A.A. Aliabdo, A.E.M.A. Elmoaty, H.A. Salem, Effect of water addition, plasticizer
and alkaline solution constitution on fly ash based geopolymer concrete
performance, Constr. Build. Mater. 121 (2016) 694–703.
[30] P. Nath, P.K. Sarker, V.B. Rangan, Early age properties of low-calcium fly ash
geopolymer concrete suitable for ambient curing, Procedia Eng. 125 (2015)
601–607.
[31] G.F. Huseien, J.A.H.A.N.G.I.R. Mirza, M.O.H.A.M.M.A.D. Ismail, M.W. Hussin, M.
A.M. Arrifin, A.A. Hussein, The effect of sodium hydroxide molarity and other
parameters on water absorption of geopolymer mortars, Indian J. Sci. Technol.
9 (48) (2016).
[32] I. Garcia-Lodeiro, A. Palomo, A. Fernández-Jiménez, D.E. Macphee,
Compatibility studies between NASH and CASH gels. Study in the ternary
diagram Na 2 O–CaO–Al 2 O 3–SiO 2–H 2 O, Cem. Concr. Res. 41 (9) (2011)
923–931.
[33] P.S. Deb, P. Nath, P.K. Sarker, The effects of ground granulated blast-furnace
slag blending with fly ash and activator content on the workability and
strength properties of geopolymer concrete cured at ambient temperature,
Mater. Des. 1980–2015 (62) (2014) 32–39.
[34] D. Dutta, S. Ghosh, Durability Study of Geopolymer Paste Blended with Blast
Furnace Slag.
[35] D. Dutta, S. Thokchom, P. Ghosh, S. Ghosh, Effect of silica fume additions on
voids ratio of fly ash geopolymers, J. Eng. Appl. Sci 5 (10) (2010) 74–79.
[36] N.K. Lee, G.H. An, K.T. Koh, G.S. Ryu, Improved reactivity of fly ash-slag
geopolymer by the addition of silica fume, Adv. Mater. Sci. Eng. (2016).
[37] S.D. Partha, N. Pradip, K.S. Prabir, Strength and permeation properties of slag
blended fly ash based geopolymer concrete, in: Advanced Materials Research,
Trans Tech Publications, 2013, pp. 168–173.
127
[38] P. Nath, P.K. Sarker, Effect of GGBFS on setting, workability and early strength
properties of fly ash geopolymer concrete cured in ambient condition, Constr.
Build. Mater. 66 (2014) 163–171.
[39] M.H. Al-Majidi, A. Lampropoulos, A. Cundy, Effect of alkaline activator, water,
superplasticiser and slag contents on the compressive strength and
workability of slag-fly ash based geopolymer mortar cured under ambient
temperature, Int. J. Civ. Environ. Struct. Constr. Archit. Eng. 10 (3) (2016) 308–
312.
[40] D. Bonen, M.D. Cohen, Magnesium sulfate attack on portland cement paste—II.
Chemical and mineralogical analyses, Cem. Concr. Res. 22 (4) (1992) 707–718.
[41] N.P. Rajamane, M.C. Nataraja, J.K. Dattatreya, N. Lakshmanan, D. Sabitha,
Sulphate resistance and eco-friendliness of geopolymer concretes, Indian
Concr. J. 86 (1) (2012) 13.
[42] P. Kumar Mehta, Paulo J.M. Monteiro, Concrete, McGraw-Hill, New York, 2006.
[43] E. Samson, J. Marchand, K.A. Snyder, Calculation of ionic diffusion coefficients
on the basis of migration test results, Mater. Struct. 36 (3) (2003) 156–165.
[44] American Concrete Institute, Guide to Durable Concrete, ACI 201.2R-92, ACI
Committee 201, Detroit, MI, 1992.
[45] N.K. Lee, H.K. Lee, Reactivity and reaction products of alkali-activated, fly ash/
slag paste, Constr. Build. Mater. 81 (2015) 303–312.
[46] I. Ismail, S.A. Bernal, J.L. Provis, S. Hamdan, J.S. van Deventer, Microstructural
changes in alkali activated fly ash/slag geopolymers with sulfate exposure,
Mater. Struct. 46 (3) (2013) 361–373.
[47] T. Nochaiya, W. Wongkeo, K. Pimraksa, A. Chaipanich, Microstructural,
physical, and thermal analyses of Portland cement–fly ash–calcium
hydroxide blended pastes, J. Therm. Anal. Calorim. 100 (1) (2010) 101–108.
[48] L. Alarcon-Ruiz, G. Platret, E. Massieu, A. Ehrlacher, The use of thermal analysis
in assessing the effect of temperature on a cement paste, Cem. Concr. Res. 35
(3) (2005) 609–613.
[49] M. Borrachero, J. Paya, M. Bonilla, J. Monzó, The use of thermogravimetric
analysis technique for the characterization of construction materials: the
gypsum case, J. Therm. Anal. Calorim. 91 (2) (2008) 503–509.
[50] F. Paulik, J. Paulik, M. Arnold, Thermal decomposition of gypsum, Thermochim.
Acta 200 (1992) 195–204.
[51] Q.L. Yu, H.J.H. Brouwers, Thermal properties and microstructure of gypsum
board and its dehydration products: a theoretical and experimental
investigation, Fire Mater 36 (7) (2012) 575–589.