Construction and Building Materials 272 (2021) 121963

Contents lists available at ScienceDirect

Construction and Building Materials

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

Fly-ash-based geopolymer composites incorporating cold-bonded

lightweight fly ash aggregates
Hatice Öznur Öz ⇑, Hasan Erhan Yücel, Muhammet Günesß, Turan Sß evki Köker
Department of Civil Engineering, Niğde Ömer Halisdemir University, Niğde, Turkey

h i g h l i g h t s

A total of 9 GCs with/without cold-bonded LWFA were produced.

Na2SiO3/NaOH was determined as 2.5 by weight in all mixtures.

Replacement of quartz aggregate with LWFA increased the workability of GCs.

LWFA up to 25% developed the physical, mechanical and durability properties of GCs.

The results also were supported by the microstructural analysis of GCs.

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

Article history: In this study, effects of cold-bonded lightweight fine aggregate (LWFA) on the geopolymer composites
Received 11 October 2020 (GCs) were investigated in terms of fresh, physical, mechanical and durability properties. LWFA and dif-
Received in revised form 3 December 2020 ferent diameters quartz aggregates were used in the production of fly-ash-based GCs. However, LWFA
Accepted 4 December 2020
was only replaced with the quartz aggregate of 1.2–2.5 mm. A mixture of 12 M NaOH and Na2SiO3 solu-
Available online 22 December 2020
tion was prepared as 1/2.5. Unit weight, specific gravity, water absorption, apparent porosity, compres-
sive and flexural strengths, ultrasonic pulse velocity and water sorptivity coefficient of GCs were
Keywords:
determined. According to the test results, workability of GCs improved as the usage rate of LWFA
Geopolymer composite
Microstructural analyses
increased. Due to the surface characteristics of LWFA, the mechanical properties of GC were enhanced
Strength and Durability by the geopolymerization reaction conducted between LWFA with alkali activators. Thus, the compres-
Physical properties sive strength of GC with 25% LWFA was 39.6% higher than that of the control GC at 28th day. This con-
LWFAs clusion supported by TGA/DTA and FTIR analysis indicated that N-A-S-H gel of GC incorporating 25%
LWFA was higher compared to control mixture. Also, GCs incorporating 40% and higher LWFA showed
lower performance than control GC.
Ó 2020 Elsevier Ltd. All rights reserved.

1. Introduction cement (OPC) [5,6]. Unlike OPC, GCs are produced with the binders
rich in aluminium and silicon such as fly ash (FA), silica fume, blast
Geopolymer composites (GCs) are inorganic polymer developed furnace slag, metakaoline do not generate CO2 and consume energy
with geopolymerization of aluminosilicate materials in the pres- [1,7,8]. OPC concrete needs to C-S-H gels for structural strength
ence of a strong alkaline solution such as NaOH, Na2SiO3 (water while GCs depend on polycondensation with alkali activators of
glass), KOH or their combination, etc. [1–4]. GCs have three- aluminium and silicon oxides to attain strength [9,10]. According why
dimensional framework structure to an amorphous semi- to previous studies, FA-based GC offered great strength and dura- fly
crystalline by the polymerization of aluminium, silica and oxygen bility performances which are not obtained in OPC concrete ash
with the existence of an exceptionally concentrated alkali as the [5,9,11,12]. Therefore, FA including a sufficient amount of silica
reaction catalyst [5,6]. Considering the sustainable development, and alumina, having a lower water requirement was preferred
the design of GCs that attracted great attention since discovering especially as a suitable material in the production of GC [13,14].
were raised new generation of the friendly environmentally alter- Consequently, FA-based GCs producing a shorter curing period
native compared to concrete developed with ordinary Portland than concrete and having high strength and durability as well as
environmental advantages, can be used instead of conventional
⇑ Corresponding author. concrete [1,7,8]. Moreover, widespread usage of FA mineral in GC
E-mail address: oznuroz@ohu.edu.tr (H.Öznur Öz).

https://doi.org/10.1016/j.conbuildmat.2020.121963
0950-0618/Ó 2020 Elsevier Ltd. All rights reserved.
 Hatice Öznur Öz, Hasan Erhan Yücel, M. Günesß et al. Construction and Building Materials 272 (2021) 121963

production is very important in terms of waste recycling for sus- two types of green lightweight aggregates with cement,
tainable development. granulated blast furnace slag and coal FA by using 10–20% of
The above-mentioned features of FA for GC design have been the former, 45–40% of latter and 45–40% of the last one.
led to the idea of using artificial lightweight FA aggregates (LWFAs) Mechanical strength of developed LWAs was around 50% of the
developed by the cold bonding method (CBM) instead of natural natural aggregate. Additionally, it was noted that the compres-
why cold aggregates in GC production. Nowadays, the manufacturing of sive strength, splitting tensile strength and ultrasonic pulse
bonded lightweight aggregates (LWAs) by using a CBM had the subject of velocity of geopolymer concrete were higher than Portland
method
scientific studies in the construction industry as it is considered cement concrete. According to test results conducted by Shi
economical and environmentally friendly thanks to minimum et al. [27], lightweight foamed geopolymer concretes exhibited
energy consumption [15–17]. CBM is one of the bonding methods better mechanical performance than the conventional ordinary
identified the ability of pozzolanic minerals such as FA and blast Portland cement foamed concrete having the same density. The
furnace slag to react with Ca(OH)2 at 23 ± 2 °C to obtain as a water usage of artificial/natural LWA in geopolymer design is known
resistant material. Pelletized LWAs subjected to curing a number of to be extremely important because of the improvement of
days to develop an aggregate for desired strength to be utilized in geopolymer characteristics as well the economical and environ-
the production of concrete [15]. In addition, there are significant mental considerations. In the light of these informations, new
advantages to using artificial LWAs such as protection of natural methods should be investigated in order to enhance the usage
raw materials and the decrease of pollution and cost accompanied areas of LWFA in the construction industry.
generally the removal of waste [15]. In this study, it was aimed to obtain the effects of using LWFA in
A various of processes have been thought to take place the production of GC via the fresh, physical, mechanical and dura-
between the cement matrix and the aggregates, associated with bility tests. For this, GC mixtures were designed with quartz sand
why
AA is quite striking features of LWFAs: rough surface leading to forma- in the range of three different diameters. Then, LWFA was replaced
better tion of chemical bond thanks to the pozzolanic characteristic, with quartz aggregate (1.2–2.5 mm). Although the quite light and
than
NA mechanical interlocking and water absorption due to their por- porous structure of LWFA in comparison quartz aggregate, the rea-
ous structures [18]. For these reasons, concretes/mortars con- sons for the strength increase provided by its high alumina and sil-
taining LWFA have been became the subject of scientific ica content were investigated by microstructural analysis methods.
researches. Güneyisi et al. [16] reported that the compressive In addition to SEM/EDX and XRD analysis, TGA/DTA and FTIR anal-
strength of self-compacting mortar incorporating %50 LWFA in yses were performed to observe the N-A-S-H variation in the
comparison the reference mixture reduced in the percentage of geopolymer matrix.
58% at 3 days, but this reduction due to the pozzolanic activity
of LWFA fell to 48% and 44% at 7 and 28 days, respectively. Cori-
naldesi and Moriconi [19] determined a similar behavior that the 2. Experimental study
lower density and higher porosity of LWFA compared to the nat-
ural sand were the reason of decreasing the mechanical proper- 2.1. Materials
ties of concrete. Critical observation of Gesoğlu et al. [20]
reported that the reducing influence of LWFA on compressive GCs were designed by using class F fly ash (FA) and alkaline
strength was determined to be more permanent for 28 and activator solution as binder materials. The chemical compositions
90 days. However, Wasserman and Bentur [18] suggested that of FA were determined by using X-ray Fluorescence (XRF) as shown
the strength of the concrete could not be explained by the in Table 1. Moreover, grain size distribution of FA was presented in
strength and porosity of the aggregates only, and also, it is Fig. 1. The alkali activators consisted of a mixture of Na2SiO3 and
reported that the absorption and pozzolanic nature of the LWFAs 12 M NaOH solution. The ratio of Na2SiO3 solution to NaOH solu-
can show a positive effect on the improved strength. Kayali [21] tion was 2.5 by weight. The composition of the Na2SiO3 solution
indicated that high performance concrete developed with LWFAs consisted of 29.4% SiO2, 14.7% Na2O and 59.9% water by mass.
was approximate 20% stronger and at the same time 22% lighter The alkaline solution is the only liquid component in all mixtures.
than concrete produced by using normal weight aggregate. Two types of aggregates; quartz aggregates of different diame-
Moreover, Gesoğlu et al. [22] revealed that the amount of ters as natural aggregate and cold-bonded LWFAs as artificial
cement had an important influence on the strength of LWFAs aggregate were used together in GC production (Fig. 2). The views
which in turn governed the variation in strength of the light- of quartz aggregates used in three different diameters as 0–
weight concretes. Considering the literature studies of GCs pro- 0.4 mm, 0.6–1.2 mm, 1.2–2.5 were given in Fig. 2. The saturated
duced with LWAs, Wongsa et al. [23] indicated that the surface dry specific weight of these aggregates was found as
mechanical properties of lightweight geopolymer concrete 2.65. LWFAs were produced by the cold bonding method with
(LWGC) incorporating pumice aggregate and crushed clay brick the help of the pelletizing disc seen in Fig. 3 [16]. A dry mixture
were lower than those of geopolymer concrete containing natu- prepared with 10% Portland cement and 90% FA by weight was col-
ral aggregates because of its lower texture that was also indi- lected through moistening in a 45 angle tilted pan and at a rotation
cated by reducing in the density and Los Angeles abrasion of 42 rpm at 23 ± 2 °C. LWFAs were obtained at the first 10 min by
resistance. However, Wongsa et al. [23] suggested that LWGCs sprayer water behaved as a coagulant agent during the pelletiza-
designed with crushed clay brick aggregate was suitable for tion. The next 10 min was significant to achieve fresh pellets, fur-
structural lightweight concrete and the manufacture of concrete ther stiffness and more uniformity of LWFAs. At the end of the
blocks. Moreover, LWGC containing expanded polystyrene pro- pelletization process, the fresh pellets were storaged in the plastic
duced by Colangelo et al. [24] showed higher strength and lower closed bags for the self-curing in a curing room at 70% relative
thermal conductivity than that of cementitious composites. In humidity and 23 ± 2 °C throughout 28 days. Physical and chemical
another study on LWGC, Priyanka et al. [25] reported that the properties of Portland cement and FA was presented in Table 1. At
compressive strength and density of FA-based geopolymer con- the complete of the curing process, the cold-bonded LWFAs were
cretes developed with lightweight expanded clay aggregates sieved into two fractions of 1.2–2.5 mm as lightweight fine aggre-
decreased compared to that of control mixture. Other than this gates to be used in the production of GCs. The water absorption
study, Rehman et al. [26] have been synthetically developed and specific gravity of LWFAs were determined with respect to

2
Hatice Öznur Öz, Hasan Erhan Yücel, M. Günesß et al. Construction and Building Materials 272 (2021) 121963

Table 1
Chemical composition and physical properties of FA and Portland cement.

Parameters (%) CaO SiO2 Al2O3 Fe2O3 MgO SO3 K2O Na2O TiO2 LOI Specific Gravity
Fly Ash 1.47 61.30 22.20 7.02 1.70 0.06 2.34 0.27 0.90 2.60 2.31
Portland Cement 60.2 20.50 7.78 3.09 2.66 2.33 0.82 0.22 0.30 2.55 3.10

Fig. 1. Grain size distribution of FA.

Fig. 2. (a) LWFA (1.2–2.5 mm), (b) Coarse quartz sand (1.2–2.5 mm), (c) Medium quartz sand (0.6–1.2 mm), and (d) Fine quartz sand (0–0.4 mm).

3
Hatice Öznur Öz, Hasan Erhan Yücel, M. Günesß et al. Construction and Building Materials 272 (2021) 121963

100 °C for 24 h. After curing, the samples were removed from

the molds and kept at room temperature until the specified test
age for each testing method. According to literature studies [29–
31], the variations of compressive strength have been increased
slightly after 7 days. For example, Van Jaarsveld et al. [30] reported
that the strength of geopolymer cured for 4 h can reach 70% of its
ultimate strength. It was observed that approximately 68%-99% of
ultimate strength was obtained at the age of 7th days in this study.
Therefore, most of the tests for this experimental study were con-
ducted at 1st, 7th and 28th days.

2.3. Test procedures

2.3.1. Flow diameters

According to ASTM C1437 [32], the flow diameter values of
Fig. 3. The view of the pelletization device.
fresh GCs were measured by conducting the flow table test. The
minimum flow diameter for the GC can be explained
ASTM C127 [28]. Dry, apparent and saturated surface dry specific 150 ± 10 mm as the flow required to easily insert into molds
gravity of LWFAs with a water absorption value of 21.2% were cal- [33]. Depending on the flow of the mortar, the workability level
culated as 1.46, 2.10 and 1.76, respectively. proposed by Ghosh and Ghosh [34] is classified in Table 3. The bot-
tom diameter of the funnel was 100 mm, the top diameter was
70 mm, and the height was 60 mm. The funnel was placed in the
2.2. Mix proportions and casting middle of the flow tray and filled in 2 stages with fresh GC. Fresh
GCs were compressed 20 times at each stage to provide uniformity.
A total of 9 GC mixtures was designed with constant FA content After the top surface of the funnel was leveled, the funnel was
such that Na2SiO3/NaOH = 2.5 by weight. Mix proportions of GCs instantly lifted vertically. Then, the flow table was dropped 25
were shown in Table 2. The selected binder content (FA + alkaline times in 15 s. The flow diameter was found by reading the 4 diam-
solution) for GC design was 1000 kg/m3. The ratio of the alkaline eters measured symmetrically and their average.
solution (Na2SiO3 + NaOH) to the binder was determined as 0.33.
Two types of aggregates, quartz aggregates (0–0.4; 0.6–1.2; 1.2–
2.3.2. Microstructural analysis
2.5 mm) and LWFA (1.2–2.5 mm), were used in the production of
The microstructural examinations of the GC0, GC25, and GC100
GCs. Quartz aggregates in dimensions of 0–0.4 mm, 0.6–1.2 mm
mixes were performed with SEM/EDX, XRD, TGA/DTA and FTIR
and 1.2–2.5 mm in GC mixtures were utilized at the rates of 45%,
analyzes at 28 days. EDX analysis was obtained from the area
35% and 20% respectively. Firstly, the control mixture (GC0) con-
which the SEM image was taken. XRD analysis performed in steps
taining only quartz aggregates were produced. Then, LWFA in the
of 0.013° was applied at a range 10–90°. TGA analysis was applied
range of 1.2–2.5 mm was used in the ratios of 10%, 20%, 25%,
under nitrogen atmosphere condition with 10 °C/min (scanning
30%, 40%, 50%, 75% and 100% by volume instead of quartz aggregate
rate) in the range of 25–1000 °C. FTIR analysis was performed at
in the range of 1.2–2.5 mm. Thus, GC10, GC20, GC25, GC30, GC40,
20 °C temperature and 60% relative humidity.
GC50, GC75 and GC100 mixtures were obtained, respectively.
First of all, the LWFAs were kept in water for 24 h in order not to
affect the amount of water in the mixture. At the end of 24 h, the 2.3.3. Physical properties
LWFAs were removed from the water and their surfaces were dried Samples in dimensions of 40x40x160 mm3 were used to deter-
with the help of a blow dryer to obtain the saturated surface dry mine the physical properties of GCs. After the air dry weight of the
condition [15,16]. LWFAs in saturated surface dry condition were samples was weighed (W1), the samples were kept in water for
added to the mixture instead of coarse quartz aggregate with a 24 h, their surfaces were cleaned with a towel and weighed
diameter range of 1.2–2.5 mm. In the production phase, dry mixing (W4). Then, the weight in water of the samples was measured with
(FA, LWFA and quartz aggregates) process was completed in 1 min. the help of Archimedes scale (W2) and finally the samples were
Then, alkaline solutions, which were mixed in the beaker for 1 min, kept in the oven at 100 ± 5 °C for 24 h and the dryness weights
were added to the mixture and mixing was continued throughout (W1) were determined. Unit volume weight, dry specific gravity,
4 min. Flow diameter test was performed on the fresh GC. Then, apparent specific gravity and saturated surface dry specific gravity
the GC was molded using the appropriate vibration process. Next, values of GCs were determined at 28th day, and water absorption
the GCs were kept in the oven under the curing temperature of and apparent porosity values of GCs were calculated at 1st, 7th and

Table 2
Mix proportions of GCs (kg/m3).

Kod NaOH Na2SiO3 FA Quartz Aggregates LWFA (1.2–2.5)

0–0.4 0.6–1.2 1.2–2.5
GC0 95.24 238.10 666.67 571.16 444.23 253.85 –
GC10 95.24 238.10 666.67 571.16 444.23 228.47 16.86
GC20 95.24 238.10 666.67 571.16 444.23 203.08 33.72
GC25 95.24 238.10 666.67 571.16 444.23 190.39 42.15
GC30 95.24 238.10 666.67 571.16 444.23 177.70 50.58
GC40 95.24 238.10 666.67 571.16 444.23 152.31 67.44
GC50 95.24 238.10 666.67 571.16 444.23 126.93 84.30
GC75 95.24 238.10 666.67 571.16 444.23 63.46 126.45
GC100 95.24 238.10 666.67 571.16 444.23 – 168.60

4
Hatice Öznur Öz, Hasan Erhan Yücel, M. Günesß et al. Construction and Building Materials 272 (2021) 121963

Table 3
Workability criteria of GC (Ghosh & Ghosh, 2012).

Classification 1 2 3 4 5
Flow Diameter >250 mm 180–250 mm 150–180 mm 120–150 mm <120 mm
Workability Very High High Normal Low Very Low

28th days using Equations (1), 2, 3, 4, 5 and 6. 3 samples were used 3. Results and discussion
for each test age and the results were averaged.
W1 3.1. Flow diameters
Unit volume weigth ¼ ð1Þ
V
Fig. 4 presents the flow diameter results for GC mixtures. The
W1
Dry specific gravity ¼ ð2Þ minimum flow diameter of 150 ± 10 mm for GC can be considered
W4  W2
as the flow required to easily insert into the molds [33]. In this
W1 study, flow diameters of GCs changed in the range of 17.9–
Apparent specific gravity ¼ ð3Þ 19.2 cm. Based on these results, all GC’s workability can be deter-
W1  W2
mined as ‘‘high’’ with respect to Table 3. Moreover, the flow diam-
W4 eter of GC10, GC20, GC25, GC30, GC40, GC50, GC75 and GC100
Saturated surface dry specific gravity ¼ ð4Þ
W4  W2 mixtures were higher in the ratio of 1.1, 2.8, 3.4, 3.9, 4.7, 5.0, 6.1
and 7.3% than that of GC0, respectively. In other words, the GCs
W4  W1
Water apsorption ð%Þ ¼ x100 ð5Þ incorporating LWFA achieved higher workability compared to the
W1
control mixture. It is known that aggregate properties and binder
components affect greatly the flowability of the GC. In addition,
W3  W1
Apparent porosity ð%Þ ¼ x100 ð6Þ using smooth surface of spherical aggregates in concrete/mortar
W3  W2
increases the flow diameter value by ball bearing effect of LWFA
where, W1: Drying weight of the sample (gr), W2: Weight of the [38,39]. Also, the test results were supported by EFNARC [40]
sample in water (gr), W3: Weight of the sample in the air (gr), reported that the workability of the mixture increased as the inter-
W4: Dry surface saturated weight of the sample (gr) and V: Volume nal friction decrease as a result of using spherical aggregate in the
of the mold (cm3). mixture.

2.3.4. Mechanical properties

According to ASTM C348 [35], prismatic samples in dimensions 3.2. SEM/EDX
of 40x40x160 mm3 were used to determine the flexural strength at
1st, 7th and 28th days. 3 prismatic samples were used for each test Fig. 5(a–f) represents EDX spectra the corresponding the SEM
age. The loading speed was 0.05 kN/s. At the end of flexural micrographs of GC0, GC25 and GC100 at 28 days. The EDX spec-
strength test, the broken six samples were used to determine the trums of GCs were obtained from the whole area of SEM images
compressive strength. Thus, the samples with two dimensions of presented in Fig. 5. The SEM/EDX analysis depicted the character-
40 mm were subjected to compressive strength test by the form istic morphologic changes of the GCs. It is known that the main
of a cube of 40 mm in three dimensions using a suitable testing compounds of the GC gel formation are Al and Si coming from FA
apparatus in the view of ASTM C349 [36]. The loading speed was with the alkali activation [41]. Similarly, as seen from Fig. 5(d,e,
applied as 2.4 kN/s during the compressive test. Ultrasonic pulse f), the GCs includes the Na2O, Al2O3, SiO2 as major components
velocity (UPV) test was carried out in accordance with ASTM with the trace of Fe2O3. EDX analysis results clearly showed the
C597-16 [37]. UPV test was applied to the 160 mm length of the variation of components that occurred in the GCs depending on
samples at 1st, 7th and 28th days. All mechanical properties were the increase in the LWFA ratio. According to the results, the Na2O
determined by calculating the average of the samples used. contents of GCs were almost equal to each other due to the fact
that Na2SiO3 and NaOH activators were used equally in all mix-
2.3.5. Water sorptivity tures. However, as the ratio of LWFA usage increased and quartz
In this study, water sorptivity coefficients of three samples of sand content reduced, the SiO2 contents of GCs decreased, while
50x50x50 mm for each mixture were determined on the 1st, 7th the Al2O3 and Fe2O3 contents increased. The reduction in SiO2 con-
and 28th days. Samples were kept in the oven for 24 h at tent was a result of the use of LWFA with a lower SiO2 content
100 ± 5 °C, only one side of the surface was covered with paraffin instead of quartz sand containing around 98% SiO2. Similarly, the
to ensure contact with water of sectional area. The water level was increment in Al2O3 and Fe2O3 contents for GC25 and GC 100 was
adjusted to be about 5 mm from the sample base. After the first a result of the use of LWFA incorporating Al2O3 and Fe2O3 instead
weights of the samples were weighed and placed in water, were of quartz sand without Al2O3 and Fe2O3. These findings indicated
removed from the water at the 1st, 4th, 9th, 16th, 25th, 36th, that the EDX data taken from the SEM image showed all the ele-
49th and 64th minutes, and their weight was measured after wip- ments in the area and the findings could not be attributed to a
ing into a towel. Water sorptivity coefficient values of the mixtures chemical reaction due to the quartz sand was not fineness enough
were determined by taking the average of the results obtained to react. Such that, according to apparent porosity, water absorp-
from 3 samples with the aid of following formula. tion, mechanical and water sorptivity test results, LWFA reacted
pffiffi chemically with alkali activators. Therefore, it can be said that
I ¼ S0 t ð7Þ
EDX data showed the elemental distribution in the whole area
0 1/2
where, S : sorptivity (mm/min ), I: cumulative infiltration (mm) and the higher Si content in GC0 can be explained with decreased
and t: time (min). quartz sand in other mixtures (GC25 and GC100).

5
Hatice Öznur Öz, Hasan Erhan Yücel, M. Günesß et al. Construction and Building Materials 272 (2021) 121963

Fig. 4. Flow diameters of GCs.

3.3. XRD for the increase in strength provided by LWFA. In addition,

although the strength characteristics of GC100 were lower than
XRD was used to investigate compositional variations (increase/ that of GC0, it can be considered as an interesting finding that N-
decrease) occurring in GC0, GC25 and GC100 samples when LWFA A-S-H contents of these samples were similar. However, the higher
was added to FA-based GCs [42]. The XRD patterns of GC0, GC25 apparent porosity value of GC100 than GC0 indicated that the high
and GC100 are shown in Fig. 6. The Albite (NaAlSi3O8-A), Nepheline porosity played a more effective role than gel strength during com-
(AlNaSiO4-N), Hematite (Fe2O3-H) peaks in addition to the crys- pression. Moreover, TGA/DTA reveals at the temperature range of
talline Quartz (SiO2-Q) and Mullite (Al2O3.2SiO2-M) for GCs are 400–800 °C, the phase formation was carbonates and sodium car-
seen in Fig. 6. The results have been showed the intensity peaks bonates [43,44]. Therefore, the carbonate and sodium carbonate
in the XRD spectrum of GC represented the presence of Q, M and contents of GCs were calculated from the mass losses drawn from
H in the FA [42]. In addition, the A and N peaks demonstrate that the TGA curves between approximately 550 °C (initial) and 575 °C
FA reacted with activators. According to Fig. 6, LWFA addition to (final) temperatures of the corresponding to peaks in the DTA
GC reduced the intensity of the Q peak. This statement supported curves. The mass losses between 550 and 575 °C of GC0, GC25
by the peaks at 21° 2h and 26.8° 2h where Q peak is seen intensely. and GC100 samples were 0.09%, 0.17% and 0.12%, respectively.
The obtained information confirmed obviously by the results of According to these results, it can be said that carbonate and sodium
SEM/EDX analysis. The intensity of the decreasing Q peak with carbonate contents of GCs were very close to each other. It is also
increasing LWFA can be explained by the reduced of quartz sand determined that the average total percentage of mass remaining
with high SiO2 content. Although the decrease in Q peak, the after 700 °C remained constant in all situations supported their
increase in strength can be attributed to the fact that FA-based thermal stability and that imported there was no other thermal
LWFA played a chemical role in the matrix like FA. decomposition formed upon heating.

3.4. TGA/DTA 3.5. FTIR

The TGA and DTA curves of GC0, GC25 and GC100 samples are FTIR spectra of GC0, GC25 and GC100 samples are presented in
represented in Fig. 7 and Fig. 8, respectively. These curves present a Fig. 10. The spectra in Fig. 10 exhibited major bands at approxi-
quite significant idea about the thermal stability of the GC matrix. mately 3380, 1456, 1059, 798, 779, 696 and 453 cm1 in GCs. There
Firstly, an endothermic peak around 50–150 °C with the sharp are two significant bands in GCs at 3380 cm1 and 1059 cm1. The
reduction in weight was determined which can be explained with H–O–H band is found at 1059 cm1 showed the existence of water
the evaporable (loss) of water content from the N-A-S-H binder gel molecules. The H–O–H band occurs after alkaline activation
for GCs [41]. Therefore, the N-A-S-H content in GCs could be because of the existence of water molecules bonded weakly
guessed from the mass losses calculated with the TGA curves [42,45]. The peaks at 3380 cm1 and 1059 cm1 represent to bend-
between 25 °C (initial) and 125 °C (final) temperatures of the cor- ing and stretching of the H-O–H band. Similar results have been
responding to peaks in the DTA curves by taking the theoretical stated by Hussin et al. [45]. Na2CO3 was determined to consist in
mass losses. Based on this information, the mass losses of GC0, control GC around at 1049 cm1 and 777 cm1 owing to carbona-
GC25 and GC100 samples in this range were determined as tion of unreacted Na2SiO3 and NaOH. N-A-S-H gel formation pro-
1.21%, 1.73% and 1.17% with the help of detailed drawings in vides to increment of 1049 cm1 peak intensity for GCs [42]. As
Fig. 7 and Fig. 8, respectively. These founds indicated that the N- seen in Fig. 10, this peak observed at 1059 cm1 in this study
A-S-H content of GC25 was higher than that of GC0 [41]. Also, this showed that N-A-S-H gel of GC25 was higher than that of GC0. It
case can be considered as evidence of the geopolymerization reac- also showed that N-A-S-H content of GC0 was slightly higher than
tion taking place between LWFA and alkali activators seen in Fig. 9. that of GC100. These findings confirmed the TGA/DTA results pre-
Although GC1000 s N-A-S-H content was lower than that of GC0, it sented in Fig. 7 and Fig. 8 and the reactions observed as the dark in
was fairly close to GC0. This significant increase observed in the Fig. 9. As a result, LWFA’s chemical activity increased N-A-S-H gel
binder gel for the GC25 sample can be thought as the main reason by reacting with alkali activators, thus, decreased the apparent
6
Hatice Öznur Öz, Hasan Erhan Yücel, M. Günesß et al. Construction and Building Materials 272 (2021) 121963

Fig. 5. SEM images of (a) GC0, (b) GC25, (c) GC100 and EDX results of (d) GC0, (e) GC25, (f) GC100.

porosity for 25% replacement ratio of LWFA. However, with gravity of LWFA (1.76) was lower than the specific gravity of quartz
increased usage ratios of LWFA, the apparent porosity value of aggregate (2.65), the unit volume weight of the material decreased
GC gradually increased and reached the highest value for 100% as the rate of substitution of LWFA increased. Similarly, as can be
LWFA content. Therefore, according to FTIR analysis, it can be said seen from Fig. 12, the dry, apparent and saturated surface dry
that the strength of GC100, whose N-A-S-H content was very close specific gravity values of GCs gradually decreased as the ratio of
to GC0, was much lower than that of GC0 [42,45]. LWFA usage increased. According to the results of the tests, the
dry, apparent and saturated surface dry specific gravity values of
3.6. Physical properties GCs changed in the ranges of 1.87–1.99, 2.12–2.38 and 2.01–
2.15, respectively.
Fig. 11 and Fig. 12 show all the experimental results of unit vol- The water absorption and apparent porosity values of the GCs
ume weigth and dry, apparent and saturated surface dry specific are given in Figs. 13 and 14. The water absorption rates of GCs were
gravities at the age of 28 days. As can be seen from Fig. 11, as in the ranges of 7.5–11.2%, 5.6–9.4% and 4.4–8.6% at 1st, 7th and
the amount of LWFA in GCs increased, unit volume weights of 28th days, respectively. For the same test ages, the apparent poros-
the GCs decreased. For example, the highest unit weight ity values of GCs changed in the ranges of % 11.6–16.4%, 5.5–13.5%
(2105.5 kg/m3) was obtained from control GC, while GC100 had and 5.3–6.6%, respectively. As given in Figs. 13 and 14, the water
the lowest unit volume weight (2009.8 kg/m3). As it is known, absorption and apparent porosity values of GCs decreased gradu-
the specific weight of the materials used in the production of con- ally from GC0 to GC25, while these values increased gradually from
crete, mortar or geopolymer directly affects the unit volume GC25 to GC100. This can be attributed to the geopolymerization
weight of the produced material [46]. Therefore, since the specific reaction performed between LWFA and alkaline activators. It is
7
Hatice Öznur Öz, Hasan Erhan Yücel, M. Günesß et al. Construction and Building Materials 272 (2021) 121963

Fig. 6. XRD patterns of GC0, GC25 and GC100.

Fig. 7. TGA results of GC0, GC25 and GC100.

known that alkali-activated FA gains the strength from the poly- alumina from cold bonded LWFA and the alkaline activators. This
merization of aluminosilicate gels [5,47]. The visual of this reaction secondary reaction contributed to form more N-A-S-H products
between LWFA and alkali activators could be proven in Fig. 9. As leading to better polymerization. Due to porous structure of LWFA,
can be seen from Fig. 9, the regions with LWFA have become darker a geopolymerization reaction expected on the surrounding phase
than other regions due to the polymerization reaction. This darker of this aggregate causes fly ash to be a real binder. Similarly, Öz
region in GC25 can be explained by the increase in N-A-S-H et al. [48] reported that cold bonded artificial slag aggregates had
observed by TGA/DTA and FTIR analyzes. According to TGA/DTA the ability to react with Ca(OH)2 in self-compacting concrete due
analysis, the N-A-S-H content in GC0 and GC25 found as 1.21% to the dissolved powder materials of the aggregate surface. There-
and 1.73% from the mass losses curves in the range of 25–125 °C fore, it can be concluded that LWFA can be reacted with alkali acti-
temperatures due to a secondary reaction between LWFA and alka- vators due to its surface properties during the geopolymerization.
line activators [41]. This may be attributed to the repetitive poly- Thus, the presence of LWFAs in GCs increased N-A-S-H gels as a
merization reaction conducted between the dissolved silica and result of geopolymerization reaction and decreased the gap in
8
Hatice Öznur Öz, Hasan Erhan Yücel, M. Günesß et al. Construction and Building Materials 272 (2021) 121963

Fig. 8. DTA results of GC0, GC25 and GC100.

strength values were obtained from GC25 mixture. Its compressive

strength was 38.8%, 43.4% and 39.6% higher than the control mor-
tar at 1st, 7th and 28th days, respectively. The compressive
strength of mixtures incorporating 30% and more LWFA were
lower than the GC25 mixture, and after this replacement ratio,
the compressive strength of GCs gradually decreased as the rate
of use of LWFA increased. Despite of decrement, the compressive
strength values of GC30 and GC40 mixtures were higher than that
of GC0 for all three test ages. Therefore, it can be said that the low
density of LWFAs was not enough to characterize the compressive
strength of GC [18,22].
Flexural strength test results of GCs suggested that flexural
strength of the GCs increased with increasing LWFA in the mix-
tures. As in the case of compression, the test results indicated that
the optimum replacement level being 25% for flexural strength.
However, using 30% and 40% replacement levels did not severely
affect the flexural strength, whereas there were 16, 23 and 28%
strength loss in 50, 75 and 100% replacement levels at 28 days,
respectively.
As seen Fig. 17, volume fraction of LWFA for UPV results had a
similar effect as compressive and flexural strengths. UPV values of
Fig. 9. Geopolymerization reactions thought to occurred between LWFA and alkali GCs changed in the range of 2590–3620 m/s. UPV value of GC0
activators. mixture was 3354 m/s while the effect of containing LWFA was
to increase that UPV value to as high as 3620 m/s. However, there
was a reduction in UPV value when LWFA was utilized in the
the matrix. In addition, the increasing percentage of water absorp- ranges of 30–100%.
tion and apparent porosity can be explained by the increasing The mechanical test results indicated that the compressive
amount of voids due to the LWFA in the GCs. strength has changed inversely proportional to apparent porosity.
It is known that, typically, higher apparent porosity leads to lower
3.7. Mechanical properties strength [49]. Therefore, the porosity of LWFA had a considerable
influence on the strength of GC [22]. In addition, according to the
The compressive strength, flexural strength and UPV values of results obtained from the tests, the increase in mechanical perfor-
GCs at 1st, 7th and 28th days are given in Figs. 15, 16 and 17, mance up to 25% LWFA content can be explained by the image
respectively. The compressive strength of the GCs changed in the given in Fig. 9. As can be seen from this picture, the red circles in
ranges of 25.1–50.6 MPa, 27.9–53.3 MPa and 35.0–60.8 MPa at the GC25 samples broken with flexural test had a darker appear-
1st, 7th and 28th days, respectively. The highest compressive ance than the other parts of the samples. This situation can be

9
Hatice Öznur Öz, Hasan Erhan Yücel, M. Günesß et al. Construction and Building Materials 272 (2021) 121963

Fig. 10. FTIR spectra of GC0, GC25 and GC100.

Fig. 11. Unit volume weights of GCs at 28th day.

attributed to the geopolymerization reaction between LWFA devel- concrete/mortar. The chemical process depends on pozzolanic
oped with FA and alkali activators [9,13,14] as proved by TGA/DTA characteristic of the aggregate; this process became influence only
and FTIR analyzes. at later age, beyond 28 days for concrete/mortar [48,50]. Therefore,
In the study investigated the reactions between sintered LWFAs depending on chemical reaction and physical interaction pho-
and the matrix by Wasserman and Bentur [50] stated that changes tographed in this study, it can be said that LWFA provided a better
in concrete/mortar strength could not always be explained by the interfacial transition zone with the activators compared to the
variations in the aggregate strength. These tendencies could be quartz aggregate [51]. LWFA exhibiting its chemical interaction
related to chemical and physical interfacial interactions which in concrete/mortar at later ages [16,50] has shown its chemical
had an effect on the overall strength beyond that of the aggregate reaction ability from the first day in GC. This situation can be
strength. The physical process defined was densification of the explained by the development of LWFA from FA, which earned
interfacial transition zone owing to absorption of the aggregates: strength to GC by reacting with alkalis. In mixtures designed
this process has significant effective already at early age for with higher LWFA content than 25%, the decrease in mechanical
10
Hatice Öznur Öz, Hasan Erhan Yücel, M. Günesß et al. Construction and Building Materials 272 (2021) 121963

Fig. 12. Dry, apparent and saturated surface dry specific gravity values of GCs at 28th day.

Fig. 13. Water absorption percentages of GCs.

properties can be considered to be due to the insufficiency of the between 20% and 40%. Similarly, in this study, the highest chemical
activators in the environment. Because of the higher water absorp- reaction of LWFA was observed on GC25. Moreover, Güneyisi et al.
tion and lower specific gravity of LWFA compared to quartz sand, reported that the highest pozzolanic interaction of LWFA was
the higher replacement of quartz sand by LWFA caused in a reduc- observed on self-compacting mortar containing 100% LWFA [16].
tion in compressive strength of GCs in comparison with GC25 [16]. Therefore, this development of GCs incorporating LWFAs can be
These findings indicated that strength of LWAs after 25% has been related to the formation of chemical bond (geopolimerization reac-
governed the compressive strength of GCs [22]. tion) between alkali activators and aggregate particle owing to the
rich alumina and silica content of the solid comprising the aggre-
3.8. Water sorptivity gate as seen in Fig. 18 [16,18,50,52]. In addition, Zhang and Gjorv
[52] reported that for aggregates without an outer layer or aggre-
The results of water sorptivity coefficients of GCs are shown in gates having a weaker and more porous outer layer, the interfacial
Fig. 18. According to the results given in Fig. 18, 1, 7 and 28-day transition zone is more homogenous and dense. The decreased
water sorptivity coefficients of GCs changed in the ranges of 0.00 water sorptivity coefficient may also be the result of this denser
65–0.0378 mm/min0.5, 0.006–0.0316 mm/min0.5 and 0.0037–0.02 interfacial transition zone. In addition, it is known that water sorp-
94 mm/min0.5, respectively. As can be seen from Fig. 18, the water tivity coefficient test depend on more on its porosity and pore tor-
sorptivity coefficients of GCs decreased gradually up to GC25 mix- tuosity. Therefore, the results have been changed in parallel with
ture. The interface transition zone enhanced as a result of this reac- the apparent porosity [53]. These findings indicated that the pres-
tion reduced the water permeability of GCs up to 25% LWFA. ence in the pore system of LWFA with chemical interaction have
According to Wasserman and Bentur [18,50], this improvement been affected the formality of the pore network tortuosity of GC.
due to pozzolanic activity of LWFA for concrete/mortar ranged Therefore, LWFA up to 25% strengthened the pore network
11
Hatice Öznur Öz, Hasan Erhan Yücel, M. Günesß et al. Construction and Building Materials 272 (2021) 121963

Fig. 14. Apparent porosity percentages of GCs.

Fig. 15. Compressive strengths of GCs.

Fig. 16. Flexural strengths of GCs.

12
Hatice Öznur Öz, Hasan Erhan Yücel, M. Günesß et al. Construction and Building Materials 272 (2021) 121963

Fig. 17. UPV values of GCs.

Fig. 18. Water sorptivity coefficients of GCs.

tortuosity against water permeability. Moreover, the increase in
Incorporating cold-bonded lightweight fine aggregate in
water sorptivity coefficients with higher LWFA can be attributed geopolymer composites increased the mechanical properties
to the increased average pore size and apparent porosity due to and also reduced permeability. According to these results,
the use of LWFA [54,55]. higher strength and lower permeability of geopolymer compos-
ites enhanced higher durability. The geopolymer composite
4. Conclusions with 25% cold-bonded lightweight fine aggregate content had
the highest physical, mechanic and durability performance. This
Depending on the findings of this research, the following con- improvement in strength and permeability characteristics with
clusions can be drawn: 25% cold-bonded lightweight fine aggregate was supported by
the increase in N-A-S-H observed in TGA/DTA and FTIR ana-

Geopolymer composites with cold-bonded lightweight fine lyzes. Thus, it was also proved that the dark regions observed
aggregate thanks to the spherical grain shape displayed higher after breaking from cold-bonded lightweight fine aggregate
workability in comparison to the control geopolymer composite containing samples were based on the geopolymerization reac-
and this increase in workability continued depending on the tions between cold-bonded lightweight fine aggregate and alka-
increase of cold-bonded lightweight fine aggregate. The oppo- line activators. In addition, using of cold-bonded lightweight
site way, unit volume weight, dry specific gravity, apparent fine aggregate decreased the SiO2 content with respect to
specific gravity and saturated surface dry specific gravity values SEM/EDX and XRD analysis. However, the reduction in SiO2 con-
of geopolymer composites decreased due to high porosity of tent performed with cold-bonded lightweight fine aggregate
cold-bonded lightweight fine aggregate. can be explained by the fact that cold-bonded lightweight fine

13
Hatice Öznur Öz, Hasan Erhan Yücel, M. Günesß et al. Construction and Building Materials 272 (2021) 121963

aggregate contained lower SiO2 than that of quartz aggregate. [15] E. Güneyisi, M. Gesoglu, O.A. Azez, H.Ö. Öz, Physico-mechanical properties of
self-compacting concrete containing treated cold-bonded fly ash lightweight
Therefore, chemically active cold-bonded lightweight fine
aggregates and SiO2 nano-particles, Constr. Build. Mater. 101 (2015) 1142–
aggregate increased the performance properties of GCs despite 1153.
its porous structure. [16] E. Güneyisi, M. Gesoğlu, I._ Altan, H.Ö. Öz, Utilization of cold bonded fly ash

Fly ash-based geopolymer design and obtaining quietly success- lightweight fine aggregates as a partial substitution of natural fine aggregate
in self-compacting mortars, Constr. Build. Mater. 74 (2015) 9–16.
ful results with usage of artificial cold-bonded lightweight fine [17] M. Gesoglu, E. Güneyisi, T. Ozturan, H.O. Oz, D.S. Asaad, Shear thickening
aggregate instead of quartz aggregate is very important for a intensity of self-compacting concretes containing rounded lightweight
sustainable environment management. In addition, the usage aggregates, Constr. Build. Mater. 79 (2015) 40–47.
[18] R. Wasserman, A. Bentur, Effect of lightweight fly ash aggregate
of fly ash as aggregate can be considered as a new finding for microstructure on the strength of concretes, Cem. Concr. Res. 27 (4) (1997)
the production of new environmentally friendly materials in 525–537.
terms of workability, physical, mechanical and durability [19] V. Corinaldesi, G. Moriconi, Behaviour of cementitious mortars containing
different kinds of recycled aggregate, Constr. Build. Mater. 23 (1) (2009) 289–
performance. 294.
[20] M. Gesoğlu, E. Güneyisi, T. Özturan, H.Ö. Öz, D.S. Asaad, Self-consolidating
characteristics of concrete composites including rounded fine and coarse fly
CRediT authorship contribution statement ash lightweight aggregates, Compos. B Eng. 60 (2014) 757–763.
[21] O. Kayali, Fly ash lightweight aggregates in high performance concrete, Constr.
Hatice Öznur Öz: Methodology, Validation, Investigation, Build. Mater. 22 (12) (2008) 2393–2399.
[22] M. Gesoğlu, E. Güneyisi, H.Ö. Öz, Properties of lightweight aggregates
Resources, Writing - original draft, Writing - review & editing, produced with cold-bonding pelletization of fly ash and ground granulated
Supervision. Hasan Erhan Yücel: Methodology, Validation, Investi- blast furnace slag, Mater. Struct. 45 (10) (2012) 1535–1546.
gation, Resources, Writing - original draft, Writing - review & edit- [23] A. Wongsa, V. Sata, P. Nuaklong, P. Chindaprasirt, Use of crushed clay brick and
pumice aggregates in lightweight geopolymer concrete, Constr. Build. Mater.
ing. Muhammet Günesß: Methodology, Validation, Investigation, 188 (2018) 1025–1034.
Resources, Writing - original draft, Writing - review & editing, [24] F. Colangelo, G. Roviello, L. Ricciotti, V. Ferrándiz-Mas, F. Messina, C. Ferone,
ß evki Köker: Methodology, Validation, Inves-
Visualization. Turan S O. Tarallo, R. Cioffi, C.R. Cheeseman, Mechanical and thermal properties of
lightweight geopolymer composites, Cem. Concr. Compos. 86 (2018) 266–
tigation, Resources. 272.
[25] M. Priyanka, M. Karthikeyan, M.S.R. Chand, Development of mix proportions of
geopolymer lightweight aggregate concrete with LECA, Mater. T.: Proceed. 27
Declaration of Competing Interest (2020) 958–962.
[26] M.U. Rehman, K. Rashid, I. Zafar, F.K. Alqahtani, M.I. Khan, Formulation and
The authors declare that they have no known competing finan- characterization of geopolymer and conventional lightweight green concrete
by incorporating synthetic lightweight aggregate, J. Build. Eng. 31 (2020)
cial interests or personal relationships that could have appeared 101363.
to influence the work reported in this paper. [27] J. Shi, B. Liu, Y. Liu, E. Wang, Z. He, H. Xu, X. Ren, Preparation and
characterization of lightweight aggregate foamed geopolymer concretes
aerated using hydrogen peroxide, Constr. Build. Mater. 256 (2020) 119442.
References [28] ASTM C127, Standard test method for specific gravity and absorption of coarse
aggregate, Annual Book of ASTM Standards, 2007.
[29] P. Zhang, Y. Zheng, K. Wang, J. Zhang, A review on properties of fresh and
[1] Y.C. Choi, B. Park, Effects of high-temperature exposure on fractal dimension of
hardened geopolymer mortar, Compos. B Eng. 152 (2018) 79–95.
fly-ash-based geopolymer composites, J. Mater. Res. Technol. 9 (4) (2020)
[30] J.G.S. Van Jaarsveld, J.S.J. Van Deventer, The potential use of geopolymeric
7655–7668.
materials to immobilise toxic metals: Part II. Material and leaching
[2] M.M. Al-mashhadani, O. Canpolat, Y. Aygörmez, M. Uysal, S. Erdem,
characteristics, Miner. Eng. 1 (1999) 75.
Mechanical and microstructural characterization of fiber reinforced fly ash
[31] A. Fernández-Jiménez, A. Palomo, Composition and microstructure of alkali
based geopolymer composites, Constr. Build. Mater. 167 (2018) 505–513.
activated fly ash binder: Effect of the activator, Cem. Concr. Res. 35 (10) (2005)
[3] Z. Yahya, M.M.A.B. Abdullah, K. Hussin, K.N. Ismail, R.A. Razak, A.V. Sandu,
1984–1992.
Effect of solids-to-liquids, Na2SiO3-to-NaOH andcuring temperature on the
[32] ASTM C1437, Standard Test Method for Flow of Hydraulic Cement Mortar,
palm oil boiler ash (Si plus Ca) geopolymerisation system, Mater. 8 (2015)
Annual Book of ASTM Standards, 2017.
2227–2242.
_ [33] K. Mermerdasß, S. Manguri, D.E. Nassani, S.M. Oleiwi, Effect of aggregate
[4] C.D. Atisß, E.B. Görür, O. Karahan, C. Bilim, S. Ilkentapar, E. Luga, Very high
properties on the mechanical and absorption characteristics of geopolymer
strength (120MPa) class F fly ash geopolymer mortar activated at different
mortar, Eng. Sci. Technol. 20 (6) (2017) 1642–1652.
NaOH amount, heat curing temperature and heat curing duration, Constr.
[34] K. Ghosh, P. Ghosh, Effect of Na2O/Al2O3, SiO2/Al2O3 and w/b ratio on setting
Build. Mater. 96 (2015) 673–678.
time and workability of fly ash based geopolymer, Int. J. Eng. Res. App. 2 (2012)
[5] S. Chuah, W.H. Duan, Z. Pan, E. Hunter, A.H. Korayem, X.L. Zhao, F. Collins, J.G.
2142–2147.
Sanjayan, The properties of fly ash based geopolymer mortars made with dune
[35] ASTM C348-14, Standard Test Method for Flexural Strength of Hydraulic-
sand, Mater. Des. 92 (2016) 571–578.
Cement Mortars, Annual Book of ASTM Standards, 2017.
[6] A. Bagheri, A. Nazari, Compressive strength of high strength class C fly ash-
[36] ASTM C349-14, Standard test method for compressive strength of hydraulic-
based geopolymers with reactive granulated blast furnace slag aggregates
cement mortars (using portions of prisms broken in flexure), Annual Book of
designed by Taguchi method, Mater. Des. (1980-2015) 54 (2014) 483–490.
ASTM Standards, 2017.
[7] Z. LI, S. LI, Carbonation resistance of fly ash and blast furnace slag based
[37] ASTM C597-16, Standard Test Method for Pulse Velocity Through Concrete,
geopolymer concrete, Constr. Build. Mater. 163 (2018) 668–680.
Annual Book of ASTM Standards, 2016.
[8] C. Carreño-Gallardo, A. Tejeda-Ochoa, O.I. Perez-Ordonez, J.E. Ledezma-Sillas,
[38] G.H. Tattersall, Workability and Quality Control of Concrete, CRC Press, 1991.
D. Lardizabal-Gutierrez, C. Prieto-Gomez, J.A. Valenzuela-Grado, F.C. Robles
[39] P.N. Quiroga, The Effect of Aggregate Characteristics on the Performance of
Hernandez, J.M. Herrera-Ramirez, In the CO2 emission remediation by means
Portland Cement Concrete, PhD Dissertation University of Texas at Austin,
of alternative geopolymers as substitutes for cements, J. Environ. Chem. Eng. 6
2003.
(4) (2018) 4878–4884.
[40] EFNARC, European Federation of National Associations Representing for
[9] O.F. Nnaemeka, N.B. Singh, Durability properties of geopolymer concrete made
Concrete, Specification and Guidelines for Self-Compacting Concrete, 2005.
from fly ash in presence of Kaolin, Mater. T.: Proceed. https://doi.org/10.1016/
[41] M. Sivasakthi, R. Jeyalakshmi, N.P. Rajamane, R. Jose, Thermal and structural
j.matpr.2020. 04.696.
micro analysis of micro silica blended fly ash based geopolymer composites, J.
[10] J.S.J. Van Jaarsveld, J.G.S. Vandeventer, G.C. Lukey, A comparative study of
Non-Cryst. Solids 499 (2018) 117–130.
kaolinite Versus Metakaolinite in FA based geopolymers containing
[42] S. Luhar, S. Chaudhary, I. Luhar, Thermal resistance of fly ash based rubberized
immobilized metals, Chem. Eng. Comm. 191 (2004) 531–549.
geopolymer concrete, J. Build. Eng. 19 (2018) 420–428.
[11] T. Bakharev, Resistance of geopolymer materials to acid attack, Cem. Concr.
[43] V.K.J. Bohra, R. Nerella, S.R.C. Madduru, P. Rohith, Microstructural
Res. 35 (4) (2005) 658–670.
characterization of fly ash based geopolymer, Mater. Trans.: Proceed.
[12] T. Bakharev, Durability of geopolymer materials in sodium and magnesium
https://doi.org/10.1016/j.matpr. 2020.03.338.
sulfate solutions, Cem. Concr. Res. 35 (6) (2005) 1233–1246.
[44] G.M. Kim et al., Heavy metal leaching, CO2 uptake & mechanical characteristics
[13] S. Mani, B. Pradhan, Investigation on effect of fly ash content on strength and
of carbonated porous concrete with alkali-activated slag & bottom ash, Int. J.
microstructure of geopolymer concrete in chloride-rich environment, Mater.
Concr. Struct. Mater. 9 (3) (2015) 283–294.
Today:. Proc. 32 (2020) 865–870.
[45] M.W. Hussin, M.A.R. Bhutta, M. Azreen, P.J. Ramadhansyah, J. Mirza,
[14] P. Nath, P.K. Sarker, Use of OPC to improve setting and early strength
Performance of blended ash geopolymer concrete at elevated temperatures,
properties of low calcium fly ash geopolymer concrete cured at room
Mater. Struct. 48 (3) (2015) 709–720.
temperature, Cem. Concr. Compos. 55 (2015) 205–214.

14
Hatice Öznur Öz, Hasan Erhan Yücel, M. Günesß et al. Construction and Building Materials 272 (2021) 121963

[46] D. Hardjito, S.E. Wallah, D.M.J. Sumajouw, B.V. Rangan, On the development of [51] A. Gümüsß, Effect of Thermal Curıng Process on Geopolymer Concrete
fly ash-based geopolymer concrete, ACI Mater. J. 6 (2004) 467–472. Properties Master Thesis, Afyon Kocatepe University, 2016.
[47] N. Jambunathan, J.G. Sanjayan, Z. Pan, G. Li, Y. Liu, A.H. Korayem, W.H. Duan, F. [52] M.-H. Zhang, O.E. Gjørv, Microstructure of the interfacial zone between
Collins, The role of alumina on performance of alkali-activated slag paste lightweight aggregate and cement paste, Cem. Concr. Res. 20 (4) (1990) 610–
exposed to 50°C, Cem. Concr. Res. 54 (2013) 143–150. 618.
[48] H.Ö. Öz, M. Gesoğlu, E. Güneyisi, S.F. Mahmood, Combined use of natural and [53] H.Ö. Öz, M. Günesß, The effects of synthetic wollastonite developed with
artificial slag aggregates in producing self-consolidating concrete, A. Mater. J. calcite and quartz on high performance mortars, Struct. Concr. 21 (2) (2020)
113 (5) (2016) 599–608. 1–16.
[49] L.-P. Qian, Y.-S. Wang, Y. Alrefaei, J.-G. Dai, Experimental study on full-volume [54] X. Liu, K.S. Chia, M.-H. Zhang, Water absorption, permeability, and resistance
fly ash geopolymer mortars: Sintered fly ash versus sand as fine aggregates, J. to chloride-ion penetration of lightweight aggregate concrete, Constr. Build.
Clean. Product. 263 (2020) 121445. Mater. 25 (1) (2011) 335–343.
[50] R. Wasserman, A. Bentur, Interfacial interactions in lightweight aggregate [55] E. Güneyisi, M. Gesoğlu, Ö. Pürsünlü, K. Mermerdasß, Durability aspect of
concretes and their influence on the concrete strength, Cem. Concr. Compos. concretes composed of cold bonded and sintered fly ash lightweight
18 (1) (1996) 67–76. aggregates, Compos. Part B.: Eng. 53 (2013) 258–266.

15