Construction and Building Materials 180 (2018) 12–22

Contents lists available at ScienceDirect

Construction and Building Materials

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

Developing geopolymer concrete by using cold-bonded fly ash

aggregate, nano-silica, and steel fiber
Jumah Musdif Their ⇑, Mustafa Özakça
Department of Civil Engineering, Gaziantep University, 27310 Gaziantep, Turkey

h i g h l i g h t s

GPCs have been synthesized using cold-bonded fly ash aggregate (CFA).

No elevated temperatures have been used in this study.

The dosage of superplasticizer to achieve the target slump value is reduced by the use of CFA.

The pore structure of GPC incorporated permeable aggregate is significantly improved by the utilization of nano-silica (NS).

Better performance of GPC is achieved by the use of 1% steel fiber and 2% NS.

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

Article history: This experimental study presents the effect of nano-silica and steel fiber on the transport properties and
Received 30 January 2018 compressive strength of alkali-activated slag/fly ash concrete incorporated cold-bonded fly ash aggregate.
Received in revised form 11 May 2018 In order to reduce energy consumption and provide environmental impact, the cold-bonded process was
Accepted 30 May 2018
used to manufacture cold-bonded fly ash aggregate. Twenty four geopolymer concrete mixtures incorpo-
Available online 2 June 2018
rated cold-bonded fly ash aggregate, nano-silica, and steel fiber were produced with sodium hydroxide
concentration of 12 M and cured at ambient temperature. Transport properties of geopolymers were
Keywords:
examined through water penetration, water sorptivity, and gas permeability at 28 and 90 days. Results
Cold-bonded process
Compressive strength
indicate that geopolymer concrete incorporated cold-bonded fly ash aggregate can be produced with
Geopolymer compressive strength as high as 28.23 and 36.62 at 28 and 90 days, respectively. However, 2% nano-
Permeability properties silica and 1% steel fiber volume fraction were the most significant parameters that caused remarkable
Nano-silica improvement of investigated properties. Moreover, the incorporation of waste materials in aggregate
Steel fiber and geopolymer concrete production can alleviate environmental problems.
Ó 2018 Published by Elsevier Ltd.

1. Introduction advantages of producing artificial aggregate from waste materials

such as a promising an alternative to recycle waste materials,
As in most of the industrialized countries, Turkey generates reducing the consumption of non-renewable natural resources,
large amounts of waste materials. The annual production of Fly and producing lighter weight aggregate compared to the natural
Ash (FA) and blast furnace slag in Turkey reach about 15 million ones [5,6]. Three common different procedures used for manufac-
tons and 600,000 tons, respectively. The disposal process of waste turing artificial aggregate; cold bonding, autoclaving, and sintering.
materials has been one of the most challenging matters in our Using of these methods form spherical shape aggregates with a rel-
country and worldwide [1]. Employing of solid waste materials atively smooth surface which are preferred because they more
in the construction industry are commonly used nowadays as it readily flow past each other as the low specific surface area needs
introduces usable construction materials. less binder and water in the design of concrete [7,8]. In addition,
Concrete technology can introduce suggestions for recycling comparing to autoclaving or sintering method, the cold-bonding
industrial wastes like FA and Ground Granulated Blast Furnace Slag process method is the lowest energy consumption and has the
(GGBFS). The scientific investigators reengineered some types of least detrimental environmental impact [9,10].
solid wastes in the production of artificial aggregate [2–4]. Several Studies conducted recently by several authors with the use of
Cold-bonded FA aggregate (CFA) in conventional concrete have
⇑ Corresponding author. revealed the performance of them to be satisfactory comparing
E-mail address: [email protected] (J.M. Their). to natural aggregate [9–13]. The studies reported the CFA to have

https://doi.org/10.1016/j.conbuildmat.2018.05.274
0950-0618/Ó 2018 Published by Elsevier Ltd.
 J.M. Their, M. Özakça / Construction and Building Materials 180 (2018) 12–22 13

improved the fresh properties on account of its spherical shape, geopolymer, incorporated with SF and NS. FA is used as a sustain-
and smooth texture [11,14]. Also, the studies reported that the able precursor for manufacturing artificial aggregate through the
incorporation of CFA in conventional concrete led to reduce cold-bonded process. It has long been established that concrete
strength and increase susceptibility to suffer drying shrinkage as durability is significantly influenced by pore structure. Thus, the
compared to natural aggregates [9,11,14]. For example, the present investigation focuses on the compressive strength, water
replacement of natural aggregate by CFA reduced compressive permeability, sorptivity, and gas permeability.
strength and increased drying shrinkage of ordinary Portland
Cement (PC) concrete by about 25% and 50%, respectively [14].
2. Experimental details and methodology
However, water to binder (w/b) ratio, binder content, and Nano-
Silica (NS) were used as effective factors by Güneyisi et al. [11]
2.1. Materials
to avoid the negative side effect of CFA on concrete properties.
Considering the production of PC, the environmental impact
In this investigation, the mineral oxides used as an aluminosil-
associated with its activities includes: discharging of a consider-
icate source to produce GPC were low calcium FA, GGBFS, and NS.
able amount of greenhouse gases, e.g. carbon dioxide (CO2) emis-
Class-F FA was utilized with a specific gravity of 2.25 and specific
sions, depleting of fossil fuels as nearly 40% of the cost of
surface area of 379 m2/kg conforming to ASTM C 618. Locally avail-
production of PC is energy-related, and exploiting of natural
able high calcium precursor GGBFS was provided by a cement pro-
resource [15,16]. Also, the recognition of long-term problems in
duction factory with the specific gravity of 2.88 and specific surface
PC structures has acted as an impetus to develop a novel genera-
area of 575 m2/kg. NS has a specific surface area of 170–230 m2/g.
tion of cement with durability and environmental sustainability
Tests of the particle size distribution and microstructure properties
[15,17].
of dry powdered precursors (FA, GGBFS, NS) were conducted with a
The high performance of GeoPolymer Concrete (GPC) in terms
dry-dispersion laser diffraction (Sympatec RODOS T4.1 Particle
of its high durability, high sustainability, and adequate strength
Size Analyzer) and image analysis using Scanning Electron Micro-
can be considered the key motivation for lessening the use of PC
scopy (SEM). The chemical and physical properties of the raw pow-
in construction projects. In addition, geopolymerization reaction
der materials are presented in Table 1. Fig. 1 depicts the
is an environmentally-friendly process and can significantly
microstructure images of FA, GGBFS, and NS. Fig. 2 shows the par-
account for reducing global CO2 emissions [18–20]. In this reaction,
ticle size distribution curves of FA and GGBFS.
the use of alkali silicate solution to reactivate raw aluminosilicate
The alkaline activator used for the geopolymerization process
materials such as FA, GGBFS, red mud, mine waste, etc., at ambient
was a combination of sodium hydroxide (NaOH) and sodium sili-
or slightly high temperatures, allows the creation of cementitious
cate (Na2SiO3) solutions. In this study, potable water was used to
gel [21]. Xu and Deventer [22] reported that the most reasonable
dissolve pure NaOH pellets to prepare an aqueous compound with
aluminosilicate source is FA due to its low cost, availability, and
the desired concentration. The chemical composition of water
greater potential for producing geopolymers. It was found that cal-
glass as provided by the supplier refers to the presence of 14.7%
cium oxide (CaO) in FA forms calcium silicate hydrate which was
Na2O, 29.4% SiO2, and the rest is water. superplasticizer based
believed to have a significant influence on the hardened properties
on polycarboxylic-ether formulation with a specific gravity of
of FA based GPC [23].
1.08 ± 0.02 and PH-value of 7 ± 1 was used in all mixes to attain
Reports in literature recommended using heat or steam cured
the required slump value for fresh GPC.
geopolymer in an aggressive environment because of their low
In this research, hooked-end SF with a yield strength
effective porosity [24–28]. However, a lesser strength is obtained
1500 N/mm, young modulus E = 200 GPa, density q = 7850 kg/m3,
when FA GPC is cured at an ambient temperature of 25 °C
length L = 30 mm, and aspect ratio l/d = 54.55 was utilized. Fig. 3
[27,28]. In this context, some researchers recently sought to
shows the SF used in the production of GPC.
improve the quality of geopolymer-based FA cured at ambient
Two types of coarse aggregate were used to manufacture GPC,
temperature. Phoo-ngernkham et al. [29] developed high calcium
viz., natural river and CFA. The maximum size, fineness modulus,
FA geopolymer incorporated with 1%, 2%, and 3% NS and nano-
and specific gravity of natural coarse aggregate were calculated
aluminum. Nath et al. [30] partially replaced FA with GGBFS, PC,
as 16 mm, 5.7%, and 2.7, respectively. Regarding the production
and calcium hydroxide. The results of the study revealed that
of CFA, This research based on previous studies concentrating on
workability was optimal with a binder for the FA geopolymer
the use of CFA in conventional concrete [32,33]. For this, the same
blended with these additives.
type of FA as used in geopolymer production was utilized in the
Steel Fiber (SF) reinforced concrete sections are of interest in
pelletizing process. CFA was produced in the construction labora-
some construction applications. For this, some studies are used
tory through the agglomeration of 90% FA and 10% cement by
SF for reinforcing geopolymers. Bernal et al. [15] observed that
the water absorption and porosity in alkali-activated slag concrete
reinforced with SF were reduced by about 20% in comparison to the Table 1
mixes without fibers. Aydın and Baradan [31] reported the worka- Chemical composition and physical properties of fly ash, slag, and nano-silica.
bility and drying shrinkage of alkali-activated slag/silica fume mor- Item, % Fly ash Slag Nano-silica
tars decreased with the inclusion of SF. However, as there are few
CaO 4.24 34.19 –
references concerning the influence of fibers on the transport prop- SiO2 57.2 40.42 >99.8
erties of GGBFS/FA based geopolymer, it is of importance to dis- Al2O3 24.4 10.6 –
cover capabilities of geopolymers, when they are reinforced by Fe2O3 7.1 1.28 –
fibers. Furthermore and to the knowledge of the authors, no inves- MgO 2.4 7.63 –
SO3 0.29 0.68 –
tigation had been made up to date to study the performance of CFA
K2O 3.37 – –
in alkali-activated binder concrete cured at ambient or oven tem- Na2O 0.38 – –
perature. Therefore there is a necessity for further researches to Cl – 0.0128 –
be performed in this regard. Loss of ignition (1000 °C) 1.58 2.74 <1.5
Specific gravity 2.25 2.88 2.2
The main motivation of the paper is to produce and evaluate the
Blaine fineness, m2/kg 379 575 –
transport properties of highly sustainable green concrete through Surface-volume ratio, m2/g – – 170–230
the use of CFA to synthesize low calcium FA-blended GGBFS
14 J.M. Their, M. Özakça / Construction and Building Materials 180 (2018) 12–22

Fig. 1. SEM of (a) FA, (b) GGBFS, and (c) NS.

Fig. 2. Particle size distribution of (a) FA and (b) GGBFS.

weight, with an available pelletizer disc. After 20 min, the aggre- constant binder content of 600 kg/m3 and alkaline activator solu-
gate fresh pellets formed. They were then cured in sealed plastic tion of about 217 kg/m3 having a Na2SiO3/NaOH ratio of 2.5. In
bags at ambient temperature (without heat) for 28 days in order order to verify the mix design procedure, the design of GPC was
to be ready for use in geopolymer production. Worth mentioning, considered with Activators/Precursors ratio of 0.45, and percentage
after completion of the curing stage, sieving of the CFA was imple- of solids in NaOH and Na2SiO3 are 12.9 and 32.1% by weight of pre-
mented to separate coarse particles 4–16 mm. Fig. 4 depicts the cursors, respectively. Accordingly, the total binder content of each
general steps of the CFA production. mixture is equal to the weight of FA, GGBFS, NaOH, and WG. The
Fine aggregate is a mixture of river sand and crushed limestone natural coarse aggregate was partially replaced with CFA at
sand used in all geopolymer mixes. Table 2 and Fig. 5 illustrate replacement ratios of 0, 50, and 100% by weight in the first, second,
the physical properties and gradation curves of aggregates, and third series, respectively.
respectively. NaOH flake was dissolved in water to prepare a liquid having a
concentration of 12 M. NaOH was selected in order to avoid the
2.2. Production of geopolymer hindrance of polycondensation reaction [34]. The alkali activator
solution was produced by blending NaOH solution (after cooling
The mix proportions of GPC are shown in Table 3. It demon- the NaOH solution down to an ambient temperature of
strates three different series that were prepared based on a (27–30 °C) with water glass. After that, the mixed solution was
 J.M. Their, M. Özakça / Construction and Building Materials 180 (2018) 12–22 15

Table 2
Physical properties of different types of aggregate.

Item Fineness Specific Water

modulus gravity absorption, %
Crushed sand 2.42 2.43 0.95
River sand 2.86 2.68 0.58
Natural coarse aggregate 5.66 2.72 0.43
Coarse CFA 5.54 1.75 21.5

100
Crushed sand

80 River sand

Passing (%)
Natural coarse
60
Artificial
40
Fig. 3. Steel fiber with hook-ended. B1

20 B2

stored in plastic containers for about 1 day (24 h) at room temper- B3

ature prior to usage. FA and GGBFS were sourced locally. They were 0
0.25 0.5 1 2 4 8 16
incorporated at the same weight ratio, a mix of 50% FA and 50%
Sieve size (mm)
GGBFS. Table 3 shows that a total of four geopolymer series (0.0,
0.5, 1.0, and 1.5% SF) were produced. In addition, each series con- Fig. 5. Gradation curves of aggregate. (B1, B2, and B3 aggregate mixture contained
sisted of two groups with and without the incorporation of 2% 0%, 50%, 100% coarse CFA).
NS. As a result, in the current research, the experimental program
involved the design and preparation of a total 24 GPC mixtures.
compacted with a vibrator table. In terms of mold dimensions,
The suggested designation of geopolymer mixtures was based on
two different types of cylinders were utilized, (Ø 150  300) mm
the investigated parameters (i.e. CFA replacement level, NS con-
and (Ø 100  200) mm; and, two cubes were used, (100  100 
tent, and SF volume fraction). For example, CFA50-NS2-F100 indi-
100) mm and (150  150  150) mm.
cates the replacement level of coarse CFA is 50%, 2% NS, and 1%
The surfaces of cast samples were covered with a polyethylene
volume fraction of SF.
film to minimize the carbonation effect and prevent water evapo-
The mixing process of geopolymer for casting began with the
ration. They were then allowed to harden at room temperature for
preparation of CFA by immersing it in water for about 25–30 min
24 h. Subsequently, the specimens were demolded and stored
as about 93% of the 24 h water absorption occurred within first
before testing in an ambient temperature curing room with a con-
30 min [5]. Thereafter, dry towels were used to achieve a saturated
trolled temperature of 27–30 °C.
surface dry condition. This procedure was adopted in order to pre-
vent or to lessen the water absorption property of CFA which if
occurs, reduces the early values of the slump test. Such an occur- 2.3. Test procedures
rence would also negatively influence the polycondensation/crys-
tallization reaction. Next, a concrete mixer was used for about At 28 and 90 days of curing, a compressive strength test was
1 min to mix the coarse and fine aggregate together. Thereafter, applied on the cubes, all of which had dimensions of 100  100
the blended powder raw materials and alkali activator solution,  100 mm, in accordance with BS EN 12390-3 [35], using a
including superplasticizer, were gradually added to the concrete 3000 KN universal testing machine. The test was replicated three
mixer one-by-one. They were mixed for approximately 6 min, with times and the average value of these tests was calculated and
a 2-minute rest between 3-minute intervals of mixing. The GPC reported for each type of GPC mixture.
was designed to achieve a slump value in the range of 9 ± 2 cm Tests on water depth penetration under pressure of geopolymer
and hence, the superplasticizer dosage was modified for each mix- samples were performed in accordance with code provisions (DIN
ture to attain the specified workability (slump specified value). 1048) [36]. Cubes of 150 mm edge were oven-dried at 65–70 °C
Once the mixing process was completed, the resulting fresh mix- immediately after attaining the ages of 28 and 90 days. Next, the
ture was poured into pre-oiled molds in three layers. It was well- dried cubes were subjected to 0.5 N/mm2 of pressurized tap water

Fig. 4. production of CFA. (a) Dry Powder, (b) Fresh pellets, and (c) Pellets in bag for curing.
16 J.M. Their, M. Özakça / Construction and Building Materials 180 (2018) 12–22

Table 3
Details of geopolymer mix proportions in kg/m3.

Code of mixture Precursors Activators Aggregate Steel Reducing

fiber admixture
Fly ash Slag Nano-silica Sodium Sodium Natural Cold-bonded River Crushed
hydroxide silicate (coarse) (coarse) (sand) (sand)
CFA0-NS0-F0 206.9 206.9 0.0 53.2 133 945.5 0.0 465.8 422.3 0.0 13.4
CFA50-NS0-F0 206.9 206.9 0.0 53.2 133 472.7 304.1 465.8 422.3 0.0 10.5
CFA100-NS0-F0 206.9 206.9 0.0 53.2 133 0.0 608.3 465.8 422.3 0.0 8.7
CFA0-NS2-F0 202.8 202.8 8.3 53.2 133 944.9 0.0 465.5 422.1 0.0 15.2
CFA50-NS2-F0 202.8 202.8 8.3 53.2 133 472.4 304.0 465.5 422.1 0.0 12.0
CFA100-NS2-F0 202.8 202.8 8.3 53.2 133 0.0 607.9 465.5 422.1 0.0 9.6
CFA0-NS0-F50 206.9 206.9 0.0 53.2 133 938.7 0.0 462.4 419.3 39.3 15.8
CFA50-NS0-F50 206.9 206.9 0.0 53.2 133 469.3 302.0 462.4 419.3 39.3 12.9
CFA100-NS0-F50 206.9 206.9 0.0 53.2 133 0.0 603.9 462.4 419.3 39.3 11.3
CFA0-NS2-F50 202.8 202.8 8.3 53.2 133 938.1 0.0 462.1 419.0 39.3 17.2
CFA50-NS2-F50 202.8 202.8 8.3 53.2 133 469.0 301.8 462.1 419.0 39.3 13.5
CFA100-NS2-F50 202.8 202.8 8.3 53.2 133 0.0 603.5 462.1 419.0 39.3 12.2
CFA0-NS0-F100 206.9 206.9 0.0 53.2 133 931.9 0.0 459.1 416.3 78.5 17.9
CFA50-NS0-F100 206.9 206.9 0.0 53.2 133 465.9 299.8 459.1 416.3 78.5 13.7
CFA100-NS0-F100 206.9 206.9 0.0 53.2 133 0.0 599.5 459.1 416.3 78.5 12.5
CFA0-NS2-F100 202.8 202.8 8.3 53.2 133 931.3 0.0 458.8 416.0 78.5 18.4
CFA50-NS2-F100 202.8 202.8 8.3 53.2 133 465.6 299.6 458.8 416.0 78.5 14.1
CFA100-NS2-F100 202.8 202.8 8.3 53.2 133 0.0 599.2 458.8 416.0 78.5 12.7
CFA0-NS0-F150 206.9 206.9 0.0 53.2 133 925.1 0.0 455.7 413.2 117.8 20.5
CFA50-NS0-F150 206.9 206.9 0.0 53.2 133 462.5 297.6 455.7 413.2 117.8 17.4
CFA100-NS0-F150 206.9 206.9 0.0 53.2 133 0.0 595.2 455.7 413.2 117.8 16.2
CFA0-NS2-F150 202.8 202.8 8.3 53.2 133 924.5 0.0 455.4 412.9 117.8 22.2
CFA50-NS2-F150 202.8 202.8 8.3 53.2 133 462.2 297.4 455.4 412.9 117.8 19.9
CFA100-NS2-F150 202.8 202.8 8.3 53.2 133 0.0 594.8 455.4 412.9 117.8 17.6

at the base of the cubes for 3 days. Then, after splitting the speci- then placed on glass rods on a tray such that the head of water
mens into two symmetric parts, the maximum depth of penetrated was 5 mm above the bottom surface of each specimen, so as to
water was measured. It should be noted that the samples that were absorb water vertically. During the test, the weight of the samples
incorporated with SF were split with a jackhammer after removal was measured at 1, 4, 9, 16, 25, 36, 49, and 64 min, so as to find the
from the press machine. The test setup from the current investiga- volume of water absorption. The sorptivity coefficient at 28 and 90
tion is presented in Fig. 6. days was obtained from the slope of a line drawn between the vol-
In order to specify the capillary suction performance of the ume of absorbed water versus the square root of the selected time
geopolymer specimens, a sorptivity test was conducted on three interval. The representation of the sorptivity test is presented in
samples as per to ASTM C1585-13 [37]. Each mixture sample had Fig. 7.
dimensions of 100  100  100 mm and testing was done at the Typically, the gas permeability test for concrete is evaluated
ages of 28 and 90 days of self-curing. The specimens were first based on Darcy’s Law. In this study, and as reported by RILEM
dried by the oven at 65–70 °C to improve the capillary suction. [38], the CEMBUREAU procedure was applied to study oxygen per-
Then, the specimens were stored in sealed containers to cool to meability of GPC. Each (Ø 150  300) mm cylinder was cut with a
room temperature. Afterward, the lateral sides and surface face diamond saw into three-disc specimens with a size of (Ø 150 
of each specimen were completely insulated with paraffin wax to 50) mm each. It is necessary to have the same moisture condition
prevent any penetration of water and evaporation that might occur for all samples. Therefore, disc specimens were oven-dried using
through the bottom surface during the test procedure. Next, the the same procedure adopted in the water permeability and sorp-
weight of each specimen was measured and recorded. Each was tivity tests, until the weight variation value did not exceed 1%.
The specimens were then wrapped with cling film and stored in
plastic sealed containers, one for every three samples, before test-
ing the samples at 28 and 90 days. The configuration and layout of
the gas permeability test and the current testing apparatus are pre-
sented in Fig. 8.
The apparent oxygen permeability coefficient K beneath steady
state laminar flow conditions is evaluated using a modified Darcy’s
equation as follows:

Fig. 6. Test setup for the penetration depth of water under pressure studies. Fig. 7. Water sorptivity test configuration.
 J.M. Their, M. Özakça / Construction and Building Materials 180 (2018) 12–22 17

(a) 60
0% SFA

Compressive strength (MPa)

NS 0% NS 2%
50% SFA
100% SFA
50

40

30

20

10
0 0.5 1 1.5 0 0.5 1 1.5
Steel fiber volume fraction (%)
(b) 70
0% SFA
NS 0% NS 2%

Compressive strength (MPa)

50% SFA
60 100% SFA

50

40

30

20

Fig. 8. Configuration view of gas permeability apparatus. 10

0 0.5 1 1.5 0 0.5 1 1.5
Steel fiber volume fraction (%)
2P2 QLl
K¼   Fig. 9. Variation of compressive strength for SF volume fraction of geopolymer
A P21  P 22 concrete: (a) 28-day, and (b) 90-day.

where K is the coefficient of gas permeability (m2), L is sample

height (m), l is oxygen viscosity (2.02  105N s/m2), A is sample in Fig. 9, increasing the curing age from 28 to 90 days slightly
cross-sectional area (m2), P1 is inlet gas pressure (N/m2), P2 is outlet decreased the efficiency of SF on the compressive strength of con-
gas pressure (N/m2), and Q is volume flow rate (m3/s). crete. For example, with the incorporation of 1% SF in the mixtures
free of NS, and made with 0, 50, and 100% CFA, at 28 days the com-
pressive strength increased by 11, 14, and 20% related to those free
3. Test results and discussions of SF respectively.
As Fig. 9 presents, the geopolymer mixes incorporating nano-
3.1. Compressive strength particles achieved higher compressive strength values. The average
of enhancement ratios resulting from the use of NS was found to be
The variation trend of 28 and 90 days compressive strength for 13.5 and 19.8% for concretes aged 28 and 90 days respectively. In
different GPC mixes are graphically presented in Fig. 9. It can be light of the previous studies, such trends were expected as the
seen that the compressive strength varied from 28.2 to 46.7 MPa addition of nanoparticles developed harder binder materials
and from 36.6 to 56.5 MPa at 28 and 90 days, respectively. These [29,43]. The use of NS seemed to slightly reduce the harmful effect
results are comparable to those of previous studies adopted by on compressive strength caused by replacing natural coarse aggre-
Adak et al. [39], Pacheco-Torgal et al. [40], and Kouamo et al. gate with CFA. For instance, the compressive strength at 28 days of
[41]. However, it can be clearly detected that a decrease in com- CFA100-NS0-F150 and CFA100-NS2-F150 was lower by 19 and
pressive strength at 28 and 90 days can be clarified with an inferior 12%, respectively, when compared to that of CFA0-NS0-F150. The
performance of aggregate which corresponds to the incorporation improvement in compressive strength is due to the efficiency of
of CFA in geopolymer. For instance, when compared to geopoly- NS in increasing the geopolymerization reaction products in the
mer, concrete having an age of 28 days made without the incorpo- matrix [44].
ration of CFA, the compressive strength, irrespective of NS content,
decreased from 8% to 15% when coarse CFA had a replacement 3.2. Water penetration
level of 50%. In those with 100% CFA, a higher reduction of 12–
24% can be observed. Kastiukas et al. [42] reported that the porous Resistance to water penetration can be a reliable tool to assess
nature of aggregate (i.e. lightweight aggregate) weakens the com- concrete durability. Typically, a higher depth value of water pene-
pressive strength of GPC. However, similar trends have also been tration refers to lower durability and vice versa. In general, types of
observed in normal concrete incorporated with CFA [33,42]. nanoparticles [11,45] were found to diminish concrete permeabil-
Fig. 9 depicts that the incorporation of SF in GPC slightly ity. However, permeable concrete was produced with the use of
increased compression-resistant capacities. It is estimated that this high absorption aggregate, as reported by several researchers
improvement reached highest values at a volume fraction of 1% for [14,33,46]. Öz [14] reported that the depth of water penetration
all types of concretes produced in this study. Based on the compar- of concrete made with natural river aggregate was 15.5 mm
ison of compressive strength results of geopolymer mortar made whereas it was increased to 33.5 mm when lightweight aggregate
with the different volume fraction of SF, Aydın and Baradan [31] had been used. Therefore, resistance to water penetration in GPC is
found that the optimal SF content is 2%. Conversely, as indicated expected to be considerably affected by NS and CFA.
18 J.M. Their, M. Özakça / Construction and Building Materials 180 (2018) 12–22

The depths of water penetration in GPC samples are illustrated (a) 40

NS 0% NS 2%
in Table 4. Further, Fig. 10 depicts the relative variation of water

Change of water penetration (%)

30
depth calculated with respect to the CFA0-NS0-F0 mixtures at 28
and 90 days. It is evident from Table 4 and Fig. 10 that the inclusion 20
of CFA plays a remarkable role in increasing water penetration 10
regardless of type and curing age of tested samples. As indicated
in the graphed results, the maximum depth obtained for the 0
CFA100-NS0-F0 mixture was greater than the allowable limit that -10
was reached by TS EN 12390-8 [47] (see Fig. 10).
The water absorption test of CFA and natural aggregate, clearly -20
revealed that CFA is more porous than natural river aggregate. Fur- -30
thermore, porous aggregate leads to a greater pore area at the 0% SFA 50% SFA 100% SFA
interfacial transition zone. That has been recognized to signifi- -40
0 0.5 1 1.5 0 0.5 1 1.5
cantly affect a fluid’s permeation [48]. In general, the results of Steel fibre volume fraction (%)
water permeability for geopolymer mixes free of SF and NS appear (b) 40
to trend as those results measured by Kockal and Ozturan [10]. NS 0% NS 2%

Change of water penetration (%)

They observed water penetration at 28 days for normal concrete 30
and CFA concrete as 21 and 36 mm, respectively. 20
As presented in Fig. 10, when increasing the NS content from 0%
to 2%, the depth of water penetration reduces considerably. A 10
higher reduction in water penetration was observed with the 0
CFA0-NS2-F100 mixture of about 34% and 33% at 28 and 90 days,
respectively. All GPC mixtures incorporated with NS showed lower -10

water penetration at 28 days than the control mixture except the -20
CFA100-NS2-F0 (see Fig. 10). A possible explanation for lower pen-
-30
etration is that as the geopolymerization proceeded, the pore
0% SFA 50% SFA 100% SFA
refinement effect of NS caused enhancement of the water penetra- -40
tion resistance. Also, like such trends suggest, the production of 0 0.5 1 1.5 0 0.5 1 1.5
Steel fibre volume fraction (%)
GPC with NS significantly improves pore structure even with the
incorporation of high permeable aggregate. Adak et al. [39] pro- Fig. 10. Normalized water penetration of geopolymer concrete with respect to
duced low calcium FA based GPC, and more amount of crystalline control mixture: (a) 28-day, and (b) 90-day.
compound in the geopolymer matrix was observed due to the use
of NS. At the same time, the use of NS as an additive to geopolymer
paste increases the rate of calcium silicate hydrate formation [29]. from 28 to 90 days became lower when NS was added. For
Mo et al. [48] reported that a higher hydration rate causes a lower instance, the reduction in water permeability of CFA50-NS0-F100
number of pores in the concrete microstructure. Following that, was 29% whilst it was 22% for the same mix made with 2% NS.
the comparison between the water penetration measurements The results presented in Fig. 10 revealed that in this research,
clearly revealed that the average reduction of water permeability there was an improvement in water penetration resistance

Table 4
Test results for water penetration, water sorptivity, and gas permeability of various geopolymer concrete.

Code of mixture Water penetration Water sorptivity Gas permeability

(mm) (mm/min1/2) (106  m2)
28 days 90 days 28 days 90 days 28 days 90 days
CFA0-NS0-F0 23.5 18 0.088 0.071 2.650 2.176
CFA50-NS0-F0 27.0 20.5 0.095 0.074 3.020 2.546
CFA100-NS0-F0 30.5 22.5 0.102 0.077 3.329 2.797
CFA0-NS2-F0 18.5 15.0 0.076 0.066 2.104 1.791
CFA50-NS2-F0 22.5 17.5 0.081 0.068 2.325 1.955
CFA100-NS2-F0 25.0 19.5 0.082 0.070 2.748 2.308
CFA0-NS0-F50 21.5 16.5 0.081 0.066 2.500 2.05
CFA50-NS0-F50 25.5 19.5 0.093 0.073 2.939 2.351
CFA100-NS0-F50 29.0 22.0 0.101 0.077 3.199 2.625
CFA0-NS2-F50 18.0 14.0 0.073 0.063 2.030 1.724
CFA50-NS2-F50 20.5 16.0 0.076 0.065 2.291 1.961
CFA100-NS2-F50 22.5 18.0 0.081 0.066 2.387 2.013
CFA0-NS0-F100 20.0 14.5 0.077 0.065 2.353 1.937
CFA50-NS0-F100 24.0 17.0 0.090 0.071 2.802 2.215
CFA100-NS0-F100 27.5 20.0 0.096 0.072 3.051 2.548
CFA0-NS2-F100 15.5 12.0 0.067 0.059 1.774 1.566
CFA50-NS2-F100 17.5 13.5 0.074 0.063 2.063 1.857
CFA100-NS2-F100 20.0 15.5 0.075 0.063 2.163 1.918
CFA0-NS0-F150 19.5 16.0 0.084 0.066 2.415 2.251
CFA50-NS0-F150 23.5 18.5 0.093 0.072 2.957 2.557
CFA100-NS0-F150 26.5 21.0 0.099 0.076 3.170 2.772
CFA0-NS2-F150 16.5 14.0 0.070 0.061 2.036 1.736
CFA50-NS2-F150 17.5 15.0 0.077 0.065 2.404 2.006
CFA100-NS2-F150 20.0 16.5 0.079 0.066 2.457 2.105
 J.M. Their, M. Özakça / Construction and Building Materials 180 (2018) 12–22 19

induced by the utilization of SF. As can be seen from Table 4, the high porosity caused an increase in the concrete capillarity. Türk-
addition of SF slightly decreased the depth of penetrated water men and Kantarcı [46] partially substituted natural aggregate with
in mixes made with CFA and 0% NS. Whereas, when NS and SF four volume fractions of expanded perlite aggregate. They found
are used together in geopolymer mixes, the measured values of that apparent porosity depends on the replacement ratio of aggre-
water penetration depths are significantly reduced even for those gate. Moreover, they reported that capillarity decreased with the
made with CFA. For instance, at 90 days, the water penetration longest curing time. The same behavior can be clearly seen in
depth values of CFA50-NS0-F50 and CFA50-NS2-F50 were found Table 4 as the 28-day sorptivity diminished after 90 days. More-
to be lower than that of CFA50-NS0-F0 by about 5% and 21%, over, moderate water sorptivity values were achieved in this study
respectively. Furthermore, such trends suggest that the use of NS with respect to those reported in previous studies. Liu et al. [50]
in SF reinforced GPC incorporated with CFA will remarkably raise utilized lightweight coarse aggregate to develop lightweight-
the resistance of water penetration under pressure [49]. Moreover, foamed concrete with geopolymer technology. They reported
the SF volume fraction of 1% could be considered optimum in terms water sorptivity indexes between 0.775 and 2.711 mm/min1/2. In
of improving water permeability (see Fig. 10). This also appears to another study carried out by Shaikh [51], the water sorptivity
be in agreement with the discussion on compressive strength as index varied from 0.023 to 0.033 mm/min1/2 for GPC containing
presented earlier. recycled coarse aggregates.
Water sorptivity of concrete is significantly influenced by the
3.3. Water sorptivity use of nanoparticles, as shown in Fig. 11 A decrease in water
sorptivity was observed with 2% NS, regardless of SF volume frac-
Sorptivity introduces a reliable indicator for concrete durability tion, CFA replacement level, and curing age. For example, with NS,
as it relates to the mechanism of moisture transport into unsatu- the 0.091 mm/min1/2 and 0.071 mm/min1/2 water sorptivity
rated concrete specimens. In general, during the sorptivity test, indexes of CFA50-NS0-F100 reduced to 0.074 mm/min1/2 and
the driving force that allows water to ingress into concrete is cap- 0.059 mm/min1/2 at 28 and 90 days, respectively. However, simi-
illary suction. Table 4 illustrates the water sorptivity calculated lar to the behavior seen during water penetration, NS seemed to be
values of different GPC. The normalized sorptivity of GPC of the more effective at 28 days compared to 90 days as lower improve-
control mixture at 28 and 90 days are presented in Fig. 11, respec- ment in sorptivity coefficients were observed in the latter. On
tively. By means of their 28-day sorptivity indexes of 0.102 and the other hand, the comparison between the water sorptivity of
0.067 mm/min1/2, CFA100-NS0-F150 was considered as the most mixtures incorporating natural and/or CFA and NS indicates that
permeable concrete and CFA0-NS2-F100 as the least, respectively. the use of NS reduces the negative side effect on water sorptivity
The water sorptivity results of concrete containing CFA were as seen with the use of a CFA. This is especially true between con-
higher than those of mixtures made without CFA. In addition, it crete made with 50% CFA and those produced with 100% CFA. For
was observed that the higher the CFA volume, the higher water instance, at 28 days, the water sorptivity of CFA50-NS0-F0 and
sorptivity. For example, the 28-day water sorptivity of CFA0-NS0- CFA100-NS0-F0 were found to be 0.095 and 0.102, respectively,
F0 increased by 8% and 15% due to the use of 50% and 100% CFA, whilst they were 0.081 and 0.082 for those made with NS (see
respectively. This trend revealed that the use of aggregates having Table 4). This behavior may be related to the surface texture of nat-
ural aggregate because CFA has porous and rough outer layers
making it more effective to build a harder interfacial zone with
(a) 30 hydrated products, as in the case of using NS [52,53].
In the current investigation, various effects of the addition of SF
Change of water sorptivity (%)

NS 0% NS 2%
20 were observed in the water sorptivity of GPC by applying four dif-
ferent dosages of fiber (0%, 0.5%, 1.0%, and 1.5% by volume). The
10 addition of SF when NS was absent varied the water sorptivity
coefficients into two routes (Fig. 11). The first route is represented
0 by the enhancement in concrete capillarity due to the addition of
SF in mixtures made with natural coarse aggregate only. The sec-
-10
ond route showed the use of SF and CFA in GPC production slightly
enhanced water sorptivity. These keep into account that the water
-20 0% SFA
50% SFA sorptivity of these concretes were lower than the control mixture.
100% SFA However, in the case of using both NS and SF, all concretes attained
-30
0 0.5 1 1.5 0 0.5 1 1.5 water sorptivity coefficients higher than the control mixture as
Steel fibre volume fraction (%) presented in Fig. 11. The slight reduction in concrete capillarity
(b) 30 induced by the addition of SF may be related to the ability of SF
NS 0% NS 2% to block possible water pathways inside concrete structures [54].
Change of water sorptivity (%)

20 According to the water sorptivity variation presented in Fig. 11,

it can be asserted that a 1% volume fraction of SF provided the opti-
10
mum concrete capillarity regardless of the age of curing.
0
3.4. Gas permeability
-10
Concrete permeability involves the liquid and/or gas movement
-20 0% SFA through a hardened concrete mass under a definite pressure gradi-
50% SFA
ent. It is an inherent property of concrete that essentially relates to
100% SFA
-30 the characteristics and geometric arrangement of the ingredients
0 0.5 1 1.5 0 0.5 1 1.5
of concrete [55]. At initial pressure value, capillary pores dominate
Steel fibre volume fraction (%)
the ingress of gas while the gel pores prevent ingress. But at higher
Fig. 11. Normalized water sorptivity of geopolymer concrete with respect to pressures, such small pores work to resist gas penetration to a cer-
control mixture: (a) 28-day, and (b) 90-day. tain extent. In this context, the introduction of nano-particles and
20 J.M. Their, M. Özakça / Construction and Building Materials 180 (2018) 12–22

varying replacement levels of CFA are expected to take a role in the NS2-F0 mixture. This improvement can be attributed to the poz-
gas transport property of GPC. Jalal et al. [56] believed that con- zolanic reaction which increased with the utilization of NS. In addi-
crete permeability is principally controlled by the porosity and tion, the improvement in the pozzolanic reaction led to the
solidity of hydrated paste present in the bulk paste matrix and improvement in the quality of the bulk matrix and aggregate-
interfacial transition zone of concrete. matrix interfacial zone. The employment of NS in concrete produc-
The gas permeability coefficients of GPC are shown in Table 4. In tion to enhance the transport properties of concrete was also
addition, similar to previous representations, Fig. 12 depict the rel- reported by Supit and Shaikh [45]. They found that the durability
ative coefficient of gas permeability variations at 28 and 90 days, of mixtures incorporating 2% NS performed better than the control
respectively. Table 4 shows that the permeability coefficient of concrete.
GPC at 28 and 90 days varied from (1.774–3.329)  1016 to The same behavior as observed in this study in two previous
(1.57–2.8)  1016 m2, respectively. This results appear to be sim- tests, namely, water sorptivity and water penetration, was
ilar to the gas permeability of K = (1.6–5.3)  1016 m2 for an detected again as shown in Fig. 12 regarding the influence of SF
alkali-activated slag concretes, reported by Balcikanli and Ozbay on the gas permeability coefficient of the CFA concrete produced.
[57]. Mangat and Lambert [58] reported that the aggregate- It was observed that the use of SF without CFA directly contributed
matrix interfacial zone in conventional matrices tends to be more to the achievement of lower permeability coefficients. On the other
permeable than the bulk matrix remote from aggregate surfaces. hand, the use of SF had no appreciable effect when CFA was
However, that case study does not invalidate GPC because the replaced. For example, in the first situation, the coefficients of
alkali activator in geopolymer may interact chemically with the GPC free of NS tested at 28 days with volume fractions of SF at
particles of aggregate in addition to the binder. In relation to our 0%, 0.5%, 1.0%, and 1.5% were found to be (2.650, 2.500, 2.353
study, the aforementioned statement of Mangat and Lambert and 2.415)  1016 m2. In the second instance and at the same vol-
seems to be valid with NS-incorporated mixtures while the appli- ume fractions, those incorporated with 100% CFA were (3.329,
cation of CFA only in GPC caused an increase in the coefficient of 3.199, 3.015 and 3.170)  1016 m2. This insignificant effect of SF
gas permeability (see Fig. 12). on the improvement of geopolymer permeability possibly corre-
As presented in Fig. 12, the improvement resulting from the use lates to the presence of cracks that were located in the interfacial
of CFA were observed in CFA100-NS0-F150 and CFA50-NS2-F100. zone between the fibers and hardened matrix [15,59].
The reduction in gas permeability coefficients appear to be related However, the use of NS corrected that negative behavior as was
to the relationship between CFA employed as the CFA and the detected in mixtures that integrated NS with SF and CFA, regard-
interfacial transition zone quality. Therefore, it can be concluded less of SF volume fraction and replacement level of CFA (see
that the presence of NS with CFA triggered an improved and stron- Fig. 12).
ger pozzolanic reaction, and hence, an improved interfacial transi-
tion zone quality was created.
Fig. 12 shows that the gas permeability coefficients of varying 4. Correlation between transport properties and compressive
GPC mixtures were reduced by employing NS. The lower coeffi- strength
cient was found to be 1.57  1016 m2 at 90 days for the CFA0-
Always researchers widely use the experimental results correla-
tion for assessment of the findings reported. In this study, the
(a) 40 transport properties were evaluated as they correlate with the
NS 0% NS 2% compressive strength of GPC. The correlation between compressive
Change of gas permeability (%)

30
strength and other transport characteristics were studied at 28 and
20 90 days are graphically illustrated in Fig. 13, there was a strong
10 correlation between water penetration and compressive strength.
This confirms the general relationship between porosity and
0
strength. Therefore, it can be deduced that water depth decreases
-10 with incremental compressive strength. Fig. 14 shows two correla-
-20
tions between water sorptivity index and the compressive strength
of produced GPC at curing ages of 28 and 90 days. Evidently, the
-30
adopted tests provided a definitive correlation making the test suf-
0% SFA 50% SFA 100% SFA
-40 ficient to utilize in predicting compressive strength.
0 0.5 1 1.5 0 0.5 1 1.5
Regarding the correlation between gas permeability and com-
Steel fibre volume fraction (%)
pressive strength, the same trend as those discussed above
40
(b) NS 0% NS 2%
30 60
Change of gas permeability (%)

y = 91.331e-0.04x 28 days 90 days

Compressive strength (MPa)

20 55 R² = 0.90

50
10
45
0
40
-10
35
-20
30
-30 y = 71.813e-0.03x
0% SFA 50% SFA 100% SFA 25 R² = 0.9501
-40
0 0.5 1 1.5 0 0.5 1 1.5 20
Steel fibre volume fraction (%) 10 15 20 25 30 35
Water penetration (mm)
Fig. 12. Normalized gas permeability of geopolymer concrete with respect to
control mixture: (a) 28-day, and (b) 90-day. Fig. 13. Correlation between compressive strength and water penetration.
 J.M. Their, M. Özakça / Construction and Building Materials 180 (2018) 12–22 21

60 close to that of the same concrete mixture made with 50%

CFA. The increase in the SF volume fraction of 1.5% led to an
Compressive strength (MPa)

55 28 days 90 days
increase in the volume of absorbed water.
50
4. Similar to water penetration and sorptivity, the gas permeabil-
45 ity of GPC was influenced by CFA and NS. There was no signifi-
40 cant effect of SF on this property. The refining of pore structure
and discontinuity of pores resulting from NS activity caused a
35 y = 208.78e-22.12x
R² = 0.9105
considerable decrease in the gas permeability coefficient. The
30 y = 2.9667x-1.012 reduction in the gas permeability coefficient of concrete incor-
R² = 0.9105
25 porated with 100% CFA from (0.102 to 0.063)  106 m2 at 90
days can be attributed to the effect of 2% NS. The SF volume
20
0.04 0.05 0.06 0.07 0.08 0.09 0.1 0.11 fraction of 1% in terms of reduction in the gas permeability
Water sorptivity index (mm) for GPC is recommended as optimum.
5. The differences in transport properties become lighter when
Fig. 14. Correlation between compressive strength and water sorptivity index. compressive strength is higher. In addition, the correlations
between compressive strength and transport properties of
GPC revealed that concrete has regression coefficient values
60
higher than 0.90 regardless of aggregate type, the presence of
Compressive strength (MPa)

55 28 days
y = 76.361e-0.287x NS, the SF volume fraction, and the age of concrete.
R² = 0.9417 90 days
50

45
6. Recommendations for future work

40 y = 79.28x-0.712
R² = 0.9212
In general, future work needs to be done to understand the
35 microstructures of GPC made with CFA, NS, and SF. For this several
30 tests can be suggested such as X-ray diffraction (XRD) to determine
the amorphous phases of geopolymer paste, Electronic Microscopy
25
SEM to show the surface morphology of GPC, Mercury Intrusion
20 Porosimetry (MIP) to assess the porosity and the pore size distribu-
1 1.5 2 2.5 3 3.5
tion of aged GPC, etc.
Gas permeability coefficient, K×10-16 (m2)

Fig. 15. Correlation between compressive strength and gas permeability Conflict of interest
coefficient.
None.

between compressive strength and water penetration was

References
observed (see Fig. 15).
Furthermore, all relationships represented in Figs. 13–15 intro- [1] V.S. Ramachandran, Waste and by-products as concrete aggregates, Can. Build.
duce strong correlation coefficients (R2) of 0.90–0.95. The high val- Dig. (1981).
ues of R2 were achieved and the comparison of all test results [2] G. Baykal, A.G. Döven, Utilization of fly ash by pelletization process; theory,
application areas and research results, Resour. Conserv. Recycl. 30 (2000) 59–
showed minimum scatter. In addition, the higher R2 of 0.95 in 77.
water penetration and compressive strength within confidence [3] C.R. Cheeseman, A. Makinde, S. Bethanis, Properties of lightweight aggregate
limits was reached for the for mixes tested at 28 days. produced by rapid sintering of incinerator bottom ash, Resour. Conserv. Recycl.
43 (2005) 147–162.
[4] G. Joseph, K. Ramamurthy, Influence of fly ash on strength and sorption
5. Conclusion characteristics of cold-bonded fly ash aggregate concrete, Constr. Build. Mater.
23 (2009) 1862–1870.
[5] K. Sakai, T. Noguchi, The Sustainable Use of Concrete, CRC Press, 2012.
In this study, the examination of employing CFA in the produc- [6] O. Kayali, Fly ash lightweight aggregates in high performance concrete, Constr.
tion of GPC is studied. The general conclusions of this experimental Build. Mater. 22 (2008) 2393–2399.
study are drawn as follows: [7] Z. Wu, Y. Zhang, J. Zheng, Y. Ding, An experimental study on the workability of
self-compacting lightweight concrete, Constr. Build. Mater. 23 (2009) 2087–
2092.
1. At 28 and 90 days, a reduction in the compressive strength of [8] C. BIBM, E. ERMCO, EFNARC, The European guidelines for self-compacting
GPC was identified with the replacement of natural coarse concrete, Specif. Prod. Use (2005). n.d..
[9] A. Terzić, L. Pezo, V. Mitić, Z. Radojević, Artificial fly ash based aggregates
aggregate with those made from FA through a cold-bonded pro- properties influence on lightweight concrete performances, Ceram. Int. 41
cess. Incorporation of SF did not noticeably enhance compres- (2015) 2714–2726.
sive strength except when used with NS. The highest [10] N.U. Kockal, T. Ozturan, Effects of lightweight fly ash aggregate properties on
the behavior of lightweight concretes, J. Hazard. Mater. 179 (2010) 954–965.
compressive strength of 40.28–56.47 was achieved due to the [11] E. Güneyisi, M. Gesoglu, O.A. Azez, H.Ö. Öz, Effect of nano silica on the
incorporation of 1% SF and 2% NS. workability of self-compacting concretes having untreated and surface treated
2. The introduction of CFA into GPC presented higher water pene- lightweight aggregates, Constr. Build. Mater. 115 (2016) 371–380.
[12] A. Sarabèr, R. Overhof, T. Green, J. Pels, Artificial lightweight aggregates as
tration than the control GPC. An improvement in water penetra-
utilization for future ashes–a case study, Waste Manage. 32 (2012) 144–152.
tion resistance due to the NS pozzolanic effect was [13] R. Manikandan, K. Ramamurthy, Effect of curing method on characteristics of
considerable. No adverse effect of SF on water permeability cold bonded fly ash aggregates, Cem. Concr. Compos. 30 (2008) 848–853.
was identified. Curing at ambient temperature developed a [14] H.Ö. Öz, M. Gesoglu, E. Güneyisi, N.H. Sor, Self-Consolidating concretes made
with cold-bonded fly ash lightweight aggregates, ACI Mater. J. 114 (2017) 385.
gradual geopolymerization reaction which led to the produc- [15] S. Bernal, R. De Gutierrez, S. Delvasto, E. Rodriguez, Performance of an alkali-
tion of concrete at 28 days with a water penetration depth of activated slag concrete reinforced with steel fibers, Constr. Build. Mater. 24
30 mm. (2010) 208–214.
[16] A.R. Marei, Energy and its impacts on cement industry with special reference
3. The 90-day water sorptivity index of about 0.072 mm/min1/2 for to conversion and modernization of cement plants in Arab and developing
the mixtures made with 100% CFA, 2% NS, and 1% SF was very countries, in: Cem. Ind. OIC Memb. Countries, Tech. Pap. 1, 1990.
22 J.M. Their, M. Özakça / Construction and Building Materials 180 (2018) 12–22

[17] D. Hardjito, B.V. Rangan, Development and properties of low-calcium fly ash- [41] H.T. Kouamo, A. Elimbi, J.A. Mbey, C.J.N. Sabouang, D. Njopwouo, The effect of
based geopolymer concrete, 2005. adding alumina-oxide to metakaolin and volcanic ash on geopolymer
[18] D. Khale, R. Chaudhary, Mechanism of geopolymerization and factors products: a comparative study, Constr. Build. Mater. 35 (2012) 960–969.
influencing its development: a review, J. Mater. Sci. 42 (2007) 729–746. [42] G. Kastiukas, X. Zhou, J. Castro-Gomes, Development and optimisation of phase
[19] M.S. Reddy, P. Dinakar, B.H. Rao, A review of the influence of source materials change material-impregnated lightweight aggregates for geopolymer
oxide composition on the compressive strength of geopolymer concrete, composites made from aluminosilicate rich mud and milled glass powder,
Microporous Mesoporous Mater. 234 (2016) 12–23. Constr. Build. Mater. 110 (2016) 201–210.
[20] B. Singh, G. Ishwarya, M. Gupta, S.K. Bhattacharyya, Geopolymer concrete: a [43] M. Stefanidou, I. Papayianni, Influence of nano-SiO2 on the Portland cement
review of some recent developments, Constr. Build. Mater. 85 (2015) 78–90. pastes, Composites Part B 43 (2012) 2706–2710.
[21] M. Zhang, M. Zhao, G. Zhang, D. Mann, K. Lumsden, M. Tao, Durability of red [44] K. Dombrowski, A. Buchwald, M. Weil, The influence of calcium content on the
mud-fly ash based geopolymer and leaching behavior of heavy metals in structure and thermal performance of fly ash based geopolymers, J. Mater. Sci.
sulfuric acid solutions and deionized water, Constr. Build. Mater. 124 (2016) 42 (2007) 3033–3043.
373–382. [45] S.W.M. Supit, F.U.A. Shaikh, Durability properties of high volume fly ash
[22] H. Xu, J.S.J. Van Deventer, The geopolymerisation of alumino-silicate minerals, concrete containing nano-silica, Mater. Struct. 48 (2015) 2431–2445.
Int. J. Miner. Process. 59 (2000) 247–266. [46] Ibrahim Türkmen, A. Kantarci, Effects of expanded perlite aggregate and
[23] C.K. Yip, G.C. Lukey, J.L. Provis, J.S.J. van Deventer, Effect of calcium silicate different curing conditions on the physical and mechanical properties of self-
sources on geopolymerisation, Cem. Concr. Res. 38 (2008) 554–564. compacting concrete, Build. Environ. 42 (2007) 2378–2383.
[24] T. Bakharev, Durability of geopolymer materials in sodium and magnesium [47] T.S. EN, others, Testing Hardened Concrete-Part 8: Depth of Penetration of
sulfate solutions, Cem. Concr. Res. 35 (2005) 1233–1246. Water under Pressure, 2002.
[25] T. Ramlochan, P. Zacarias, M.D.A. Thomas, R.D. Hooton, The effect of pozzolans [48] K.H. Mo, U.J. Alengaram, M.Z. Jumaat, M.Y.J. Liu, J. Lim, Assessing some
and slag on the expansion of mortars cured at elevated temperature: Part I: durability properties of sustainable lightweight oil palm shell concrete
expansive behaviour, Cem. Concr. Res. 33 (2003) 807–814. incorporating slag and manufactured sand, J. Cleaner Prod. 112 (2016) 763–
[26] S. Wallah, B.V. Rangan, Low-calcium fly ash-based geopolymer concrete: Long- 770.
term properties, 2006. [49] J. Gao, W. Sun, K. Morino, Mechanical properties of steel fiber-reinforced, high-
[27] X. Guo, H. Shi, L. Chen, W.A. Dick, Alkali-activated complex binders from class strength, lightweight concrete, Cem. Concr. Compos. 19 (1997) 307–313.
C fly ash and Ca-containing admixtures, J. Hazard. Mater. 173 (2010) 480–486. [50] M.Y.J. Liu, U.J. Alengaram, M. Santhanam, M.Z. Jumaat, K.H. Mo,
[28] K. Vijai, R. Kumutha, B.G. Vishnuram, Effect of types of curing on strength of Microstructural investigations of palm oil fuel ash and fly ash based binders
geopolymer concrete, Int. J. Phys. Sci. 5 (2010) 1419–1423. in lightweight aggregate foamed geopolymer concrete, Constr. Build. Mater.
[29] T. Phoo-ngernkham, P. Chindaprasirt, V. Sata, S. Hanjitsuwan, S. Hatanaka, The 120 (2016) 112–122.
effect of adding nano-SiO2 and nano-Al2O3 on properties of high calcium fly [51] F.U.A. Shaikh, Mechanical and durability properties of fly ash geopolymer
ash geopolymer cured at ambient temperature, Mater. Des. 55 (2014) 58–65. concrete containing recycled coarse aggregates, Int. J. Sustain. Built Environ. 5
[30] P. Nath, P.K. Sarker, V.B. Rangan, Early age properties of low-calcium fly ash (2016) 277–287.
geopolymer concrete suitable for ambient curing, Procedia Eng. 125 (2015) [52] U. Schneider, S. Chen, The interface zone around expanded shale grain in
601–607. hardened cement paste, in: Int. Union Test. Res. Lab. Mater. Constr. Int. Conf.,
[31] S. Aydin, B. Baradan, The effect of fiber properties on high performance alkali- 1992, pp. 149–156.
activated slag/silica fume mortars, Composites Part B 45 (2013) 63–69. [53] M.-H. Zhang, O.E. Gjørv, Microstructure of the interfacial zone between
[32] K.I. Harikrishnan, K. Ramamurthy, Influence of pelletization process on the lightweight aggregate and cement paste, Cem. Concr. Res. 20 (1990) 610–618.
properties of fly ash aggregates, Waste Manage. 26 (2006) 846–852. ß imsßek, M. Çullu, G. Durmus, Ö. Can, The effects of macro synthetic
[54] H. Bolat, O. S
[33] A.G. Döven, Lightweight fly ash aggregate production using cold bonding fiber reinforcement use on physical and mechanical properties of concrete,
agglomeration process PhD thesis, Bogaziçi University, Istanbul, 1998. Composites Part B 61 (2014) 191–198.
[34] K.-L. Lin, H.-S. Shiu, C.-L. Hwang, A. Cheng, T.-W. Cheng, Effects of SiO2/Na2O [55] A. Nazari, S. Riahi, The role of SiO2 nanoparticles and ground granulated blast
molar ratio on properties of TFT-LCD waste glass-metakaolin-based furnace slag admixtures on physical, thermal and mechanical properties of self
geopolymers, Environ. Prog. Sustainable Energy 33 (2014) 205–212. compacting concrete, Mater. Sci. Eng. A. 528 (2011) 2149–2157.
[35] B.S. EN, 12390-3: 2002, Testing Hardened Concrete. Compressive Strength of [56] M. Jalal, A.A. Ramezanianpour, M.K. Pool, Split tensile strength of binary
Test Specimens, Br. Stand. Inst. (n.d.), 2002. blended self compacting concrete containing low volume fly ash and TiO2
[36] G. Standard, DIN 1048-test methods of concrete impermeability to water: Part nanoparticles, Composites Part B 55 (2013) 324–337.
2, Dtsch. Inst. Fur Normung, Ger., 1991. [57] M. Balcikanli, E. Ozbay, Optimum design of alkali activated slag concretes for
[37] A. Standard, C1585-13, Stand. Test Method Meas. Rate Absorpt. Water by the low oxygen/chloride ion permeability and thermal conductivity,
Hydraul. Concr. ASTM Int., 2013. Composites Part B 91 (2016) 243–256.
[38] T.C. Rilem, 116-PCD. Permeability of concrete as a criterion of its durability, [58] P. Mangat, P. Lambert, Sustainability of alkali-activated cementitious materials
Mater. Struct. 32 (1999) 174–179. and geopolymers, in: Sustain. Constr. Mater., Second ed., Elsevier, 2016, pp.
[39] D. Adak, M. Sarkar, S. Mandal, Effect of nano-silica on strength and durability 459–476.
of fly ash based geopolymer mortar, Constr. Build. Mater. 70 (2014) 453–459. [59] T. Häkkinen, The influence of slag content on the microstructure, permeability
[40] F. Pacheco-Torgal, D. Moura, Y. Ding, S. Jalali, Composition, strength and and mechanical properties of concrete Part 1 Microstructural studies and basic
workability of alkali-activated metakaolin based mortars, Constr. Build. Mater. mechanical properties, Cem. Concr. Res. 23 (1993) 407–421.
25 (2011) 3732–3745.