Construction and Building Materials 332 (2022) 127273

Contents lists available at ScienceDirect

Construction and Building Materials

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

Mechanical property and microstructure development in alkali activated fly

ash slag blends due to efflorescence
Lakshmikanth Srinivasamurthy a, b, *, Venkata S. Chevali a, Zuhua Zhang c, d, e, *,
Márlon A. Longhi f, Thomas W. Loh g, Hao Wang a, *
a
Centre for Future Materials, University of Southern Queensland, Springfield Central, QLD 4300, Australia
b
Department of Civil Engineering, MS Ramaiah Institute of Technology, Bengaluru, Karnataka 560054, India
c
Key Laboratory for Green & Advanced Civil Engineering Materials and Application Technology of Hunan Province, College of Civil Engineering, Hunan University,
Changsha 410082, PR China
d
Key Laboratory of Building Safety and Energy Efficiency of Ministry of Education, Hunan University, Changsha 410082, China
e
International Science and Technology Innovation Centre for Green & Advanced Civil Engineering Materials of Hunan Province, Hunan University, Changsha 410082,
PR China
f
Building Innovation Research Unit, Universidade Federal do Rio Grande do Sul (NORIE/UFRGS), Av. Osvaldo Aranha, 99. Porto Alegre, Brazil
g
School of Engineering, RMIT University, GPO Box 2476, Melbourne, VIC 3001, Australia

A R T I C L E I N F O A B S T R A C T

Keywords: Efflorescence of alkali activated materials (AAMs) is caused by alkali leaching and precipitation of carbonated
Efflorescence salts, which occurs concurrently with leaching and natural carbonation. Efflorescence is specifically driven by
Fly Ash precursor and activator variability in AAMs vis-à-vis phase changes, microstructure, and mechanical properties
Alkali activated materials
are not well understood. To that end, this study analyses the effects of long-term (90 days) efflorescence on AAMs
Geopolymers
Slag
with eight varied calcium and activator contents and correlated with compressive and splitting tensile strengths.
MAS-NMR Microstructural features including N-A-S-H/C, N-A-S-H gel change are analysed using Fourier-transform infrared
spectroscopy (FTIR), magic-angle spinning nuclear magnetic resonance (MAS-NMR) and thermogravimetric
analysis (TGA). AAMs with 9 wt% Na2O and Ca/(Si + Al) ratio of 0.0 to 0.25 exhibit enhanced efflorescence and
dealumination of Al[IV] in N-A-S-H/C,N-A-S-H gels, higher mobility of Na+ ions, and natural carbonation. AAMs
with 5 wt% Na2O and Ca/(Si + Al) ratio of 0.0 to 0.25 exhibit lower efflorescence and higher stability, in
contrast. Under efflorescence, the binder with 5 wt% Na2O and Ca/(Si + Al) ratio of 0.25 exhibited deal­
umination and carbonation, and the binders with 5 wt% Na2O and Ca/(Si + Al) of 0.0 to 0.10 showed less
dealumination and higher concentration of Al[IV] in aluminosilicate gel, indicative of greater phase stability. This
study highlights the criticality of calcium and activator doses in controlling the stability of phase composition
and mechanical property, which is essential for the industrial application of AAMs.

1. Introduction consumption of ordinary Portland cement represents 36% of the CO2

emissions contributed by construction activities, which is ≈8% of all
Fly ash (FA), ground granulated blast-furnace slag (GGBFS), and global CO2 emissions [1-3]. Consequently, the cement industry has come
several other industrial byproducts [1] are increasingly used today for under scrutiny from environmental and sustainability perspectives,
developing novel eco-efficient binders such as calcium aluminate ce­ which is further intensified because of the considerable depletion of
ments, calcium sulfoaluminate cements, super-sulphated cements, and natural resources associated with the extraction of its raw materials and
alkali-activated binders today, especially in view of ordinary Portland energy-intensive production (≈10% of the total energy used) [4,5].
cement use increasing to over 4 gigatons per year [1]. This increased Overall, the construction industry is keen to address the energy

* Corresponding authors at: Centre for Future Materials, University of Southern Queensland, Springfield Central, QLD 4300, Australia (L. Srinivasamurthy). Key
Laboratory for Green & Advanced Civil Engineering Materials and Application Technology of Hunan Province, College of Civil Engineering, Hunan University,
Changsha 410082, PR China (Z. Zhang).
E-mail addresses: [email protected] (L. Srinivasamurthy), [email protected] (V.S. Chevali), [email protected] (Z. Zhang), Hao.Wang@
usq.edu.au (H. Wang).

https://doi.org/10.1016/j.conbuildmat.2022.127273
Received 6 October 2021; Received in revised form 28 February 2022; Accepted 21 March 2022
Available online 8 April 2022
0950-0618/© 2022 Elsevier Ltd. All rights reserved.
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

Fig. 1. A schematic of the experimental work conducted in this study.

consumption and greenhouse gas emissions towards attaining a net-zero ionic interaction between Al-O-Si and alkali ions, which weakens the Al-
carbon status by the year 2050, which necessitates the value chain in the O-Si bonds and the structural integrity of the aluminosilicate network
building and construction industry sector to pursue suitable sustainable [20] in turn. Also, the leached alkalis precipitated in the form of efflo­
alternatives to tackle these concerns. Among the feasible solutions, rescence or subflorescence salts, and the excessive precipitation of
alkali-activated materials (AAMs) are promising and commercially efflorescence salts can decrease the mechanical properties of AAMs. To
available. AAMs have carved a niche in building and construction ma­ this end, the partial inclusion of slag or calcium-rich precursors in FA
terials where traditional cements are predominant [6,7]. AAMs are based AAMs causes lower efflorescence potential [17]. Slag inclusion in
gaining substantial attention because of the possibility to utilize indus­ AAMs enhances mechanical properties and leads to the development of
trial by-products as precursor materials combined with 45–80% lower higher tortuosity in its porous network [21,22]. This increased tortu­
carbon emissions, enhanced mechanical strength, resistance to elevated osity impedes the movement of alkalis in AAMs and decreases the
temperatures and acids, and a potential to immobilize hazardous ma­ efflorescence potential.
terials, in comparison with ordinary Portland cements [8-10]. The high concentration of calcium in an aluminosilicate gel facili­
AAMs are synthesised by mixing solid aluminosilicate precursors tates the formation of a calcium aluminosilicate hydrate (C-A-S-H) gel
with an alkaline activator (e.g., NaOH, Na2SiO3, and KOH) to develop a [23]. The coexistence of N-A-S-H and C-A-S-H such as C-(N)-A-S-H/C,N-
binder paste. The hardened binder paste consists of a three-dimensional, A-S-H is often reported in hardened AAMs comprising of blended pre­
amorphous sodium-aluminosilicate-hydrate (N-A-S-H) gel structure cursors [24]. The formation of C,N-A-S-H gel in the hardened AAMs
interconnected with silica and alumina tetrahedral units resembling depends on the relative ratio of slag and FA, i.e., the ratio of Ca to Al. A
zeolitic structures [11], with alkali ions (Na+ or K+) neutralizing the high-Ca/low-Al combination in the precursors leads to the formation of
charge deficit developed due to the isomorphic substitution of Al3+ high-calcium C,N-A-S-H gel in the resultant AAMs [25,26]. Specifically,
instead of Si4+ in tetrahedral units of aluminosilicate framework [12]. a lower pH is realized because of the presence of calcium in slag, which
The resulting aluminosilicate framework creates a strong ionic interac­ favours the formation of C,N-A-S-H and hinders the evolution of N-A-S-H
tion by developing a Na-O-Al(Si) network, with a stoichiometric ratio of as the dominant gel. Conversely, a low-Ca/high-Al combination in a
Na/Al = 1.02 and Si/Al = 2.2 [13]. Also, because of the charge defi­ high-alkalinity environment favours the formation of C,N-A-S-H gel. A
ciency in the aluminosilicate network, the position of Na+ or K+ around higher pH caused by the activator in this case inhibits the dissolution of
Al-O-Si sites is more plausible compared with Si-O-Si sites. Furthermore, calcium from the precursor, and favours the formation of a low-calcium
the alkalis in the aluminosilicate network are in the state of Na(H2O)n+ C,N-A-S-H gel [27,28]. Though the inclusion of slag hinders alkali
in the pore solution [14,15] and weakly bound to the structural gel, movement in AAMs, the inclusion of a high slag fraction is cost-
resulting in unstable phases when the binder is exposed to efflorescence prohibitive, and is predisposed to carbonation. The challenge of how
and leaching. The precipitation of carbonate salts from the binder mixes the stability of N-A-S-H and C,N-A-S-H gels transforms under long-term
often leads to the formation of efflorescence in AAMs [16]. The complete efflorescence exposure has generated substantial interest in the research
understanding of efflorescence, which affects the durability of AAMs, is community, however, this topic still is relatively underexplored to date.
albeit ongoing. To that end, this work aims to understand the two aspects of efflo­
In partially-saturated AAM structures, efflorescence starts at the rescence in AAMs that have not met with a rigorous, systematic treat­
point of water contact assisted by capillary forces, followed by a) ment, i.e., i) effect of efflorescence on AAMs with different Ca content by
leaching of free alkalis to the outer surface under hydration, and b) varying slag and varying activator concentration, and ii) concomitant
reacting with atmospheric carbon dioxide, forming carbonate salts on phase evolution and stability when subjected to long term efflorescence.
the surface of the structures [16-18]. Also, atmospheric CO2 diffused A range of binder mixes were produced with variable resultant Ca and
into the binder mix reacts with water to form HCO3- , which further reacts pH levels, i.e., NaOH + Na2SiO3-activated FA and FA/slag-blended
with alkalis in the pore solution to develop salt crystals inside the porous AAMs with low Ca/(Si + Al) ratios (0 to 0.25) with two different
network causing a phenomenon termed subflorescence [16]. The Na2O content (5 wt% and 9 wt%), and were subjected to long term
excessive formation of salt crystals in the porous network is detrimental efflorescence up to 2160 h (90 days). Upon curing and efflorescence, the
to the tensile strength of AAMs [16]. Previous studies have reported an Al environment in aluminosilicate gel, Na stability, and mechanical
excessive amount of alkali leaching, i.e., up to 16% in FA- and 55% in properties including compressive and splitting tensile strengths were
MK-based AAMs [19]. The high leaching potential in AAMs affects the determined and analysed using standard methods. The phase changes

2
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

Table 1
Compositional details of the precursors used in this study.
Weight fraction (wt. %) LOI

SiO2 Al2 O3 CaO MgO K2 O Na2 O Fe2 O3 P2 O5

FA 55.83 34.28 2.88 0.78 0.82 0.27 2.43 0.11 0.24

GGBFS 33.68 14.54 42.49 6.01 0.32 0.18 0.55 0.02 0.07

Table 2
Mix design details of AAMs.
Mix # FA (wt%) Slag (wt%) Na2O (wt%) Al/Si Ca/Si Na/Al Ca/(Si + Al) Curing time (days) (w/s) ratio

9A 100 0 9 0.57 0.0 0.37 0.0 28 0.35

9B 95 5 9 0.56 0.11 0.38 0.07
9C 90 10 9 0.55 0.16 0.39 0.10
9D 70 30 9 0.52 0.36 0.44 0.25

5A 100 0 5 0.61 0.0 0.20 0.0 28 0.35

5B 95 5 5 0.60 0.12 0.21 0.07
5C 90 10 5 0.59 0.17 0.22 0.10
5D 70 30 5 0.56 0.40 0.25 0.25

before and after efflorescence were assessed using FTIR, TGA, and 27Al efflorescence behaviour therefore, the ratios in the Table 2 show overall
and 23Na MAS-NMR spectroscopy. The detailed schematic of the current calcium, silica and aluminium content present in the binder mixes.
study is shown in Fig. 1.
2.3. Testing and characterization
2. Materials and methods
Efflorescence formation was assessed by visual observation of sam­
2.1. Materials ples after 28 days of curing according to references [16,17,29]. To attain
a partially saturated condition, the bottom portion of the specimens was
Class F fly ash or FA and a ground granulated blast furnace slag or immersed in water to a depth of 8 mm to 10 mm, and the top surface was
GGBFS with a basicity coefficient Kb of 1.005 and hydration modulus allowed to be exposed to air at standard laboratory conditions. The
(HM) of 1.9 were provided by Nielsen’s group, Queensland, Australia. X- water level was regularly replenished to match the initial level. Efflo­
ray fluorescence (XRF) was used to determine the chemical composition rescence tests were carried in a laboratory environment at 23 ± 2 ◦ C and
of precursors and LOI values were determined using TGA (Table 1). 55 ± 5% relative humidity.
GGBFS exhibited a higher fraction of CaO and a lower fraction of Al2O3. After 28 days of curing, the binder specimens were evaluated for
5 wt%, 10 wt%, and 30 wt% slag were mixed in a dry condition for 5 compressive strength, splitting tensile strength, and exposed to efflo­
min with FA to obtain a pre-determined range of compressive strength rescence conditions. After 28 days and 90 days of exposure to efflores­
after 28 days of curing (10 MPa – 50 MPa). Using the compositional cence, the mechanical properties were determined again. A universal
details obtained by the XRF data of the precursors, the weight fractions testing machine (MTS Insight 100) with a strain rate of 0.5 mm/min was
of SiO2, CaO, and Al2O3 were determined per unit weight of solid pre­ used to measure the mechanical properties of binder specimens,
cursors. From the obtained weight fraction of the SiO2, CaO, and Al2O3, following AS1012.10 and AS1012.9:2014 standards.
the weight of oxygen was excluded using molecular weight calculations The microstructure was analysed through chemical characterization
to obtain Ca/(Si + Al) ratios of 0.07, 0.10, and 0.25. The activating on a sample from the binder surface at depth of <2 mm. The collected
solutions were produced using laboratory-grade NaOH pellets with 99% samples were dried in an oven at 60 ◦ C to achieve an equilibrium mass.
purity obtained from Chemsupply, Australia. D-grade Na2SiO3 (Na2O = The dried samples were crushed and sieved through an ASTM E11, 200
14.7 wt%, SiO2 = 29.4 wt% and H2O = 52.7 wt%) with Ms = 2.0 (SiO2/ mesh or 75 µm sieve. The powdered samples were dried again in an oven
Na2O) was procured from PQ Australia. The obtained NaOH pellets were at 60 ◦ C to evaporate the residual water, and then stored in airtight
dissolved in deionised water and mixed with Na2SiO3 solution to obtain containers. Microstructural characterisation of the binder specimens
Ms = 1.4 with 5 wt% (5 wt.%/solid precursors) and 9 wt% Na2O (9 wt was conducted using X-ray computed tomography (CT) on a General
%/solid precursors) and cooled to room temperature. The activator so­ Electric Phoenix v|tome|xs instrument fitted with a micro-tube. Scans
lution was prepared 3 h before mixing with solid precursors. were performed at 140 kV and 50 µA with a voxel size of 17 µm. 2000
images of projections were recorded per specimen. The image pro­
2.2. AAM synthesis and efflorescence conditioning jections of the binder specimens were collected and merged into a three-
dimensional (3D) volumetric representation, which was then extracted
AAM specimens were prepared by mechanically mixing the pre­ and used to assess the porosity.
cursors with an alkaline activator (NaOH + Na2SiO3) for 5 min, using Shimadzu (IRAffinity-1S) ATR-FTIR was used for FTIR analyses at
additional water to achieve suitable workability and a final water/solid wavenumbers of 400 cm− 1 (25000 nm) to 4000 cm− 1 (2500 nm) before
(w/s) ratio of 0.35. The prepared binder mix was poured into cubical and after efflorescence. The spectral deconvolution was performed to
moulds (30 mm × 30 mm × 30 mm) and cylindrical moulds (Φ 25 mm × understand the changes in Si-O-Al/Si-O-Si stretching and Al-O-Si/Si-O-
25 mm height). The prepared AAMs specimens were wrapped with Si bending before and after efflorescence. The spectral deconvolution
plastic sheets, sealed in polycarbonate containers, and allowed to cure between 650 and 1300 cm− 1 and relative integral area under the
for 28 days under ambient conditions (~25 ◦ C). Table 2 shows the selected peaks was performed using Origin 2020 software. A vendor-
different mixes with their corresponding atomic ratios and Na2O con­ provided Gaussian multipeak fitting function was used to fit the
tent. The data presented in Tables 1 and 2 does not represent the reactive spectra and r2 = 0.98–0.99 was achieved throughout the deconvolution
content in the precursors. This work focused to understand the effect of analysis.
calcium content present in the slag on the alkali activation process and A TA Instruments SDT 650 thermogravimetric analyser was used to

3
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

Fig. 2. Cylindrical binder specimens with 28 days curing and showing efflorescence effects of exposure after 10 days (-10E), 60 days (-60E), and 90 days (-90E).

perform thermogravimetric analysis on binder samples using a heating surface of the binder specimens shifts the solid/liquid interface system
rate of 20 ◦ C/min and a temperature range between 30 ◦ C and 1000 ◦ C. to a more energetic solid/vapour interface system [34]. As a result of the
MAS-NMR analysis was conducted to understand the structural increase in the energy, the capillary forces increase the water absorption
properties and formation process in alkali activated materials. 27Al and and drive the flow of pore solution towards the exterior surface.
23
Na NMR were conducted on a JEOL (ECX 400) solid-state NMR (11.7 Concomitantly, water evaporates from the pore solution and increases
T) 400 MHz with a probe of 4 mm and a spinning speed of 10 kHz to the salt concentration at the surface of the binder specimens. Further­
study 27Al (transmitter frequency = 156 MHz, scans = 1024) and 23Na more, the diffusion of CO2 from the atmosphere reacts with salts in pore
(transmitter frequency = 132 MHz, scans = 1024) environments. A solution to precipitate as carbonate salts [35].
pulse width of 6 µs and a relaxation delay of 2 s was used for all samples. The optical characterization of efflorescence in binder specimens
The obtained MAS-NMR spectra of the 27Al environment were decon­ activated with 9 wt% and 5 wt% Na2O content with varying Ca/(Si + Al)
voluted using Origin 2020 software. The Gaussian multipeak fitting ratios are shown in Fig. 3. The binder mixes activated with 9 wt% Na2O
functions were used to fit the spectra and r2 = 0.98–0.99 was achieved content exhibited a substantial amount of salt deposition on the surface
throughout the deconvolution analysis. of the specimens, indicating a higher amount of free alkali leaching in
The leaching concentrations of Na+ ions from the bulk specimens binder mixes. After 10 days, slight efflorescence products on the surface
were analysed using a Shimadzu AA-7000 atomic absorption spec­ 9A, 9B, and 9C specimens with low Ca/(Si + Al) ratios were perceived.
trometer (AAS). The binder specimens (30 mm × 30 mm × 30 mm) after At the end of 60 days, significant efflorescence products on the surface of
28 days of curing were immersed in deionised water at the ratio of 1:50, the binder specimens were perceived. The binder mixes with Ca/(Si +
in a sealed plastic box. Throughout the leaching experiment, the boxes Al) = 0.25 showed no visible signs of efflorescence until 60 days of
were maintained at 23 ± 2 ◦ C with a relative humidity of ≥ 90% to avoid exposure. This trend of decreasing visible efflorescence is consistent
evaporation of water. A minimum of 10 ml of leachate was collected to with the literature, where an increase in slag content in binder mixes is
measure Na+, and Si4+ at 30, 60, and 90 days [6,31-33]. The calibration reported to decrease the efflorescence potential of AAMs [17]. This
curve was prepared by measuring varying concentrations of alkalis in behaviour is further attributed to the formation of C-A-S-H gel, which
the standard reagents in AAS before measuring the leached concentra­ has more bound water and a higher pore filling capacity compared with
tion alkalis from the binder specimens[32,33]. N-A-S-H gel [21]. Hence, a delayed efflorescence formation was
observed in binder mixes with higher Ca/(Si + Al) ratios. Nevertheless,
3. Results and discussion the presence of higher alkali content in 9 wt% Na2O mixes still facili­
tated the efflorescence at later stages. Due to the partial wetting-drying
3.1. Efflorescence formation efflorescence test condition (because the top surface is dry and the
bottom portion is wet), initiation of cracking was exhibited on binder
28-day partial saturation and ensuing efflorescence formation in specimens at 60 days of efflorescence, which transformed to a significant
binder mixes activated with 9 wt% and 5 wt% Na2O fractions are shown level of cracking at 90 days of efflorescence.
in Fig. 2. The binder mixes were exposed to partial saturation or efflo­ The substantial movement of free alkalis caused the development of
rescence conditions after 28 days of natural curing. Partial saturation salts in the pores. An increase in the salt deposition or subflorescence in
causes water to flow through the porous network of binder mixes and the pores would lead to increased crystallization pressure in the sur­
dilute the salt-rich solution. The evaporation of water at the exterior rounding regions, which in turn leads to crack formation, damage of

4
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

Fig. 3. FTIR spectra of Si-O-T (T = Al or Si) of AAMs with 9 wt% Na2O content and Ca/(Si + Al) ratios of a) 0.00, b) 0.07, c) 0.10, and d) 0.25.

binder specimens, and decrease in tensile strength and hence, the load- aluminosilicate gel [38-42]. The peak or broad shoulder at wave­
bearing capacity of binder specimens under efflorescence [16,36]. numbers of between 1020 cm− 1 to 1150 cm− 1 is assigned to a Si-rich
The binder mixes activated with 5 wt% Na2O showed higher struc­ phase with low Al substitution [43-45], and this band overlaps with
tural stability compared with 9 wt% Na2O mixes. A mild deposition of the (>1150 cm− 1) band that is assigned to asymmetric stretching of Si-
the salt on the specimens was visualized with Ca/(Si + Al) = 0.0 at the O-T (T = Al or Si) in unreacted precursors [46]. Spectral deconvolution
end of 90 days of efflorescence, along with minimal cracking compared was applied between 600 cm− 1 to 1300 cm− 1, and the calculated rela­
with the 9 wt% Na2O mixes. The specimens activated with higher Ca/(Si tive integral areas of selected peaks are summarised in Fig. 6. The region
+ Al) ratios also assisted in the reduction of the salt formation on their between 972 – 999 cm− 1 and 1000 – 1010 cm− 1 were deconvoluted into
specimen surface. The use of a lower amount of alkali activator in these C,N-A-S-H, and N-A-S-H gel, respectively. The regions between 850 –
mixes along with the increase in Ca/(Si + Al) ratios facilitated the for­ 875 cm− 1 and 670 – 770 cm− 1 were deconvoluted as asymmetric
mation of a higher volume of C-A-S-H gels in these binder mixes because stretching of AlO-4 (Al-O-Si) groups and Al-O-Si/Si-O-Si stretching,
of a lower pH. The presence of higher fractions of Ca in these mixes respectively.
further hindered the movement of alkalis and caused a lower degree of Further analyses of FTIR spectra of binder mixes (9A-9D) with
efflorescence in the binder mixes. In addition to lowered efflorescence, varying Ca/(Si + Al) ratios and 9 wt% Na2O along with the deconvo­
the inclusion of slag also refines the pore structure and restricts the lution analysis are shown in Fig. 3(a-28d to d-28d), and the respective
movement of free alkalis. areas are compared in a stacked bar plot in Fig. 6. The band assigned to
Si-O-T stretching of N-A-S-H gel appeared around 1000 cm− 1 in 9A (Ca/
3.2. FTIR analysis (Si + Al) = 0.0). The increase in Ca/(Si + Al) ratios in 9B to 9D caused a
shift in the peak position of asymmetric stretching of Si-O-T bonds to­
The FTIR spectra of binder specimens before and after efflorescence wards a lower wavenumber (974 cm− 1) along with decreasing intensity
with varying Ca/(Si + Al) ratios and activated with 9 wt% Na2O and 5 and decreasing base width of the broad region between 600 and 1300
wt% Na2O are shown in Fig. 3, Fig. 4, and Fig. 5. The peaks centred at cm− 1, indicating the reduction in Al fraction in the binder mix, and the
1001 – 1010 cm− 1 are assigned to asymmetric stretching of Si-O-T (T = concomitant development of C,N-A-S-H phases [37,47]. The higher
Al or Si) in N-A-S-H gel, in binder mixes with Ca/(Si + Al) = 0. With the Na2O content in the binder mix suppresses the formation of the C-A-S-H
increase in the slag or Ca/(Si + Al) = 0.07, 0.10, and 0.25, the Si-O-T phase because of the higher pH in the mix, therefore, N-A-S-H gel is the
stretching vibration started shifting to lower wavenumbers (972 – dominant reaction product in these mixes [48]. Also, the inclusion of
999 cm− 1) indicating the formation of C,N-A-S-H phases in binder mixes slag increases the availability of calcium in the mix, and favours the
[37,38]. The shoulder between 851 and 874 cm− 1 (at 851, 858, 861, simultaneous formation of C-A-S-H and N-A-S-H gels. The faster disso­
863, and 874 cm− 1) was correlated to asymmetric stretching of AlO-4 (Al- lution of calcium in these mixes initially favours the formation of C-A-S-
O-Si) groups overlapping with a sharp peak at 874 cm− 1 that is attrib­ H, delaying the formation of N-A-S-H gel [48], causing the movement of
uted to the development of carbonates [37,38]. The peaks between 670 Si-O-T stretching to lower wavenumbers after 28 days of curing.
and 770 cm− 1 (specifically at 668, 715, and 740 cm− 1) are assigned to The increase in Ca/(Si + Al) ratios caused decreases in the intensities
symmetric stretching of Al-O-Si or Si-O-Si linkages, indicating the for­ of asymmetric stretching of AlO-4 groups (850–875 cm− 1) and Al-O-Si
mation of aluminosilicate gel, with Al[IV] as the main component in the bending (670 to 770 cm− 1) in the binder mix after 28 days of curing

5
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

Fig. 4. FTIR spectra of AAMs with 5 wt% Na2O content and varying Ca/(Si + Al) ratios of a) 0.00, b) 0.07, c) 0.10, and d) 0.25.

Fig. 5. FTIR spectra of Si-O-T (T = Al or Si) of AAMs with (a) 9 wt% Na2O content and Ca/(Si + Al) = 0.25, (b) 5 wt% Na2O content and Ca/(Si + Al) = 0.25.

[47,49,50]. The peaks attributed to asymmetric stretching of unreacted The FTIR spectra of the 9A-9D binder mix after efflorescence with 9
precursor in the binder mixes shifted to lower wavenumbers with the wt% Na2O content is shown in Fig. 3(a–d), and the spectral deconvo­
increase in the Ca/(Si + Al) ratios, indicating the presence of a higher lution is shown in Fig. 3(a-28d to d-90d) with associated data shown in
fraction of unreacted FA content in these mixes [51]. This shift is shown Fig. 6. The major changes after efflorescence occurred in the bands
by an increase in the area assigned to the asymmetric stretching of bonds attributed to asymmetric stretching of Si-O-T, stretching of AlO-4 groups,
assigned to unreacted precursor content in FA (1155 cm− 1) [31,51]. and bending of Al-O-Si groups. Fig. 3(a-90d to c-90d) shows the FTIR
Also, the area under the Al-O-Si bending peak (670–770 cm− 1) and an spectra of the binder mixes 9A, 9B and 9C with Ca/(Si + Al) ratios of 0.0,
asymmetric stretching peak of AlO4- (850–880 cm− 1) groups decreased 0.07, and 0.10 respectively. The wavenumbers assigned to the asym­
with increasing Ca/(Si + Al) ratio. metric stretching of Si-O-T shifted to lower wavenumbers, and peak

6
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

>1100 cm-1(Si-O-T) 975-1010 cm-1(Si-O-T, N-A-S-H/C-A-S-H)

945-955 cm-1 (Si-O-T, C-A-S-H) 840-880 cm-1 (Al-O-Si) 670-770 cm-1 (Al-O-Si/Si-O-Si)
100

1146

1134

1123

1130

1133

1123

1124
1124
1134

1058
1088

1115
1132

1024

1128

1105
80

60
1000
1002

1010

1001

1001
1003

1006

1006
1009

987

999

998

987
983

986

998
40

948

955
943

20
950

875

867

862

869

877

873
870

874
862

873
873

858

863
853

876

726

873

752

754
722
758

755

759

753

759
758

728
736
766
753

769

722
0
9B - Before
9B- After
9A - After

9C - Before
9C- After

9D - Before

5A - Before
5A- After

5B - Before
5B- After
9D- After

5C - Before
5C- After

5D - Before
5D- After
9A - Before

Fig. 6. Effect of efflorescence on the FTIR band intensities determined through deconvolution.

intensity decreased as well, indicating the degradation of geopolymer of AlO-4 groups after efflorescence in 9A, 9B, 9C, and 9D mixes were
gel and (1020 – 1150 cm− 1) in binder mixes [18,31,45]. With an in­ noted, which is representative of the dealumination of Al[IV] in C,N-A-S-
crease in the Ca/(Si + Al) ratio, the bands and relative integral areas H gel after efflorescence.
between 1020 and 1150 cm− 1 indicate the formation of a higher fraction The binder mixes (5A-5D) activated with 5 wt% Na2O content and
of Si-O-Si or silica-rich gel with low-Al substitution during efflorescence. varying Ca/(Si + Al) ratios before efflorescence are shown in Fig. 4(a–d)
This area overlaps with the integral areas assigned to the unreacted and their corresponding spectral deconvolution is shown in Fig. 4(a-28d
precursor content. Contrary to the Si-O-T decreasing peak shift to d-28d). The deconvolution analysis and integral areas are detailed in
measured for binder mixes 9A, 9B, and 9C, when the Ca/(Si + Al) ratio Fig. 6. The binder mixes showed wavenumber shifts similar to 9 wt%
was increased to 0.25 in binder mix 9D, a shift towards higher wave­ Na2O binder mixes after alkali activation, indicating asymmetric
number was realised for the Si-O-T asymmetric stretching peak. These stretching of Si-O-T in N-A-S-H gel. These bands moved towards lower
shifts are indicative of a) the formation of silica-rich gel with low-Al wavenumbers with an increase in Ca/(Si + Al) ratios, indicative of
substitution, b) release of aluminium under efflorescence condition, asymmetric stretching Si-O-T in C, N-A-S-H gel.
and c) decalcification of C-A-S-H in C,N-A-S-H environment [16,38]. The The deconvoluted FTIR spectra of binder specimens with 5 wt%
release of silica during efflorescence leads to the formation of more Si-O- Na2O content and varying Ca/(Si + Al) ratios after efflorescence are
Si bonds, a decrease in the level of Ca2+ ions, and the formation of silica- shown in Fig. 4. The deconvoluted peaks associated with asymmetric
rich, low-calcium C,N-A-S-H gel [37]. Furthermore, the peaks of silica- stretching of Si-O-T bonds of binder mixes with Ca/(Si + Al) ratio of 0.0,
rich gel (1020 – 1150 cm− 1) partly overlap with the unreacted precur­ 0.07, and 0.10 (Fig. 4(a–c) and Fig. 5(a-90d to c-90d)) shifted to higher
sor peaks, causing the integral areas of peaks beyond 1150 cm− 1 to wavenumbers with respect to the mixes before efflorescence, combined
decrease after efflorescence. with increased intensity. The relative integral areas of the peaks
During efflorescence, the development of carbonates because of CO2 attributed to C,N-A-S-H gel (960 – 980 cm− 1) moderately increased, and
diffusion concomitantly decalcifies the C-A-S-H gel and promotes the the integral area under unreacted precursor peak (>1050 cm− 1)
development of carbonate salts [16]. Therefore, the relative integral decreased after efflorescence. The increase in intensity and shift towards
areas after deconvolution indicate a) decrease in the area of Si-O-T higher wavenumbers were attributed to the formation of Si-O-Si bonds
asymmetric stretching peak of main aluminosilicate gel phase, and b) in the aluminosilicate gel. An increase in the integral areas associated
secondary Si-rich phases in the binder mix. A decrease in the band in­ with asymmetric stretching of AlO4- groups and Al-O-Si bending after
tensity and integral areas of Al-O-Si bending and asymmetric stretching efflorescence was also realized, indicating the possible inclusion of Al

7
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

Fig. 7. Differential thermograms of AAMs with (a) 9 wt% and (b) 5 wt% Na2O content.

tetrahedral units in the aluminosilicate gel. decalcification under efflorescence, which is because of the possible
The combination of lower Na2O fraction and low Ca fraction initially natural carbonation under efflorescence condition [50]. The reduction
inhibits the inclusion of Al during activation. However, during efflo­ in the integral areas assigned to AlO-4 groups and Al-O-Si bending
rescence, the rise in moisture content dissolves the unreacted alkalis indicated the dealumination in this binder mix. Furthermore, the 9D and
which causes an increase in the pH of the pore solution and further fa­ 5D is deconvoluted into five peaks as shown in Fig. 5(a and b). At 943
cilitates the dissolution and inclusion of unreacted Al and Si in alumi­ and 948 cm− 1, an additional peak representative of C-A-S-H gel in the
nosilicate gel. The binder mix with Ca/(Si + Al) ratio of 0.25 with 5 wt% binder mix presented in 9D and 5D, respectively. The peaks at 983 and
Na2O content (5D, Fig. 4(d-28d to d-90d)) behaved differently from 5A 988 cm− 1 are assigned to N-A-S-H gel in 9D and 5D binder mix,
to 5C binder mixes, where the intensity of bands assigned to Si-O-T respectively. The area under the C-A-S-H gel peak decreased after
stretching of C,N-A-S-H gel shifted to higher wavenumbers along with efflorescence, suggesting decalcification of the gel in 9D and 5D. How­
a corresponding reduction in the peak intensity. The integral area under ever, the areas assigned to N-A-S-H gel shifted to higher wavenumbers, i.
these peaks increased, indicating the formation of Si-O-Si bonds or a e., 986 and 998 cm− 1 in 9D and 5D overlapping with the region assigned
silica-rich gel. The 5D mix also showed a reduction in the integral area to silica-rich phase with low-Al content, indicating the formation of a
assigned to C,N-A-S-H gel after efflorescence, indicating the secondary geopolymeric phase under efflorescence. The areas assigned

8
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

to Al-O-Si/Si-O-Si bending at (at 720 – 770 cm− 1), asymmetric stretch­

ing of AlO4- (Al-O-Si) group (at 870 – 880 cm− 1) and unreacted precursor
content (at 1105 cm− 1) decreased significantly after efflorescence in
both 9D and 5D.

3.3. Thermogravimetric analysis

The differential thermograms of the binder mixes activated with 9 wt

% Na2O and 5 wt% Na2O before and after efflorescence are shown in
Fig. 7a and b, respectively. Predominantly, the mass loss shown is
associated with free water evaporation at T > 50 ◦ C in the binder mixes
[48]. The binder mixes with Ca/(Si + Al) ratio of 0.0 with 9 and 5 wt%
Na2O content, i.e., 9A (Fig. 7a) and 5A (Fig. 7b), exhibited minor mass
loss peaks centred at 582 ◦ C (9A), 716 ◦ C (9A) and 604 ◦ C (5A) after 28
days of curing, which are attributed to the presence of carbonate phases
in mixes developed during mixing and handling [48,52].
After efflorescence, 9A and 5A mixes showed a shift in free water loss
temperature from 96 ◦ C to 114 ◦ C in 9A and a shift from 124 ◦ C to 144 ◦ C
in 5A, respectively. This shift was accompanied by an increase in mass
loss peak intensity. The temperature shift indicates the possible forma­
27
tion of reaction products under efflorescence [48]. The secondary mass Fig. 8. Al MAS-NMR spectra of fly ash and slag.
loss peak between 500 ◦ C to 700 ◦ C for both the 9A and 5A mixes was
attributed to the decomposition of carbonates and the release of CO2 in Fig. 8. The resonances at δobs = 67.6 ppm in FA, and 66.3 ppm in the
[53]. A third peak, which occurred only in the 5A mix at 712 ◦ C was slag were attributed to the tetrahedral aluminium (Al[IV]) environment
attributed to the decomposition of sodium carbonates [53]. The evolu­ [57,58]. The resonance between 30 and 50 ppm was assigned to penta-
tion of carbonates was hence corroborated using TGA analysis. coordinated aluminium (Al[V]) in FA [59]. The resonances around 13.5
The specimens with a Ca/(Si + Al) ratio of 0.10, i.e., 9C and 5C show ppm in the slag, and 2.04 ppm in the FA represent an octahedral
the dehydration of gel through the evaporation of free water at 133 ◦ C in aluminium (Al[VI]) environment [57,58].
9C and 141 ◦ C in 5C, indicating the evaporation of strongly bonded The 27Al MAS-NMR spectra of binder mixes of the 9A-9D with
water from the gels and smaller pores [48]. The pristine 9C and 5C varying Ca/(Si + Al) ratios activated with 9 wt% Na2O content are
specimens, showed mass loss peaks at 609 ◦ C (9C), 558 ◦ C (5C), and shown in Fig. 9(a to d) with the deconvoluted NMR spectra (Fig. 9(a-28d
635 ◦ C (5C), attributed to the thermal decomposition of C,N-A-S-H and to d-28d)). The relative integral areas of deconvoluted peaks before and
calcite and sodium carbonate [53,54]. After efflorescence, the binder after efflorescence are shown in Table 3, later in this section.
mixes 9C and 5C showed a slight increase in the intensity of free water The resonances of Al[V] and Al[VI] environment of 9A (Fig. 9(a and a-
loss peaks. The 5C mix showed a major mass loss at 550 ◦ C, which is 28d)); Ca/(Si + Al) ratio of 0.0) binder significantly decreased after
attributed to the thermal decomposition of polymorphs of CaCO3 and activation. A narrow resonance at 60 ppm showed an increase of Al[IV]
the C,N-A-S-H gel [54,55]. The third mass loss peak centred at 724 ◦ C in indicating the presence of highly polymerised N-A-S-H gel in the Q4
5C and 726 ◦ C in 9C were attributed to the decomposition of CaCO3 and environment [53,60,61] and Al[IV] units present in the unreacted pre­
Na2CO3 [54-56]. cursors. The 9B, 9C and 9D binder mixes (Fig. 9(b–d) and Fig. 9(b-28d to
The binder mixed with a Ca/(Si + Al) ratio of 0.25 activated with 5 d-28d) with increased Ca/(Si + Al) ratios behaved similarly to the 9A
wt% Na2O (5D) showed a mass loss peak at 99 ◦ C correlated to the binder mix, exhibiting an Al[IV] resonance at higher chemical shifts, i.e.,
dehydration of water [53]. The 9D mixes showed three mass loss peaks 62 – 63 ppm, indicating the formation of the C,N-A-S-H gel after acti­
at 484 ◦ C, 597 ◦ C, and 705 ◦ C, which are assigned to the decomposition vation [53]. The deconvoluted NMR spectra of 9A-9D binder mixes
of C,N-A-S-H, traces of calcite, and sodium carbonates, respectively [53- yielded two distinct peaks after activation attributed to the Al[IV] envi­
55]. 5D showed a higher intensity compared with the pre-efflorescence ronment in N-A-S-H gel (9A, 60 ppm), C,N-A-S-H gel (9B-9D, 62–63
mass loss peak as well as a shift in the peak temperature from 524 ◦ C to ppm), and Al[VI] (-0.8 to 3.9 ppm) environment.
542 ◦ C, indicative of the decomposition of a C,N-A-S-H gel. After 90 days of efflorescence, the binder mix 9A (Fig. 9(a and a-
The mass loss peaks at 705 ◦ C in 9D and 713 ◦ C in 5D are assigned to 90d)) exhibited a shift in the Al[IV] resonances towards lower chemical
the decomposition of crystalline calcite and sodium carbonates phases shifts, i.e., 60 ppm from 61 ppm, indicating dealumination and degra­
[53,54]. After efflorescence, the broad differential thermograms detec­ dation of the N-A-S-H gel under efflorescence [62]. Also, broadening of
ted in the 5D mix exhibited free water mass loss peaks indicating the spectra was observed between 10 and 50 ppm in Al[V] after efflo­
dehydration of binder gel at 87 ◦ C and carbonate phases at 145 ◦ C [53]. rescence. The 9B, 9C, and 9D (Fig. 9(b- to d) and 9(b-90d to d-90d))
The intensity of mass loss peak at 728 ◦ C in the 9C and 5C specimens binder mixes showed similar behaviour where the Al[IV] resonances
increased after efflorescence, which indicates the formation of carbon­ shifted to lower values (59 – 60 ppm) and broadened in the Al[V] region.
ates under efflorescence. Overall, the differential thermograms corrob­ These spectral shifts suggest the degradation of the C,N-A-S-H under
orated the presence of carbonates after efflorescence. efflorescence. The deconvolution of the spectra (Table 3) also yielded in
three distinct peaks, i) attributed to Al[IV], ii) a broad section from 20 to
3.4. MAS-NMR analysis 50 ppm attributed to distorted Al-O-Si or Al[IV], and iii) Al[VI], respec­
tively. The relative integral areas attributed to Al[IV] region decreased,
23
Na and 27Al NMR were conducted to correlate the phases within and those between 20 and 50 ppm increased after efflorescence, indi­
the hardened AAM binders with Na sites in the crystalline, hydrated Na cating distinct changes in Al[IV] environment and degradation of N-A-S-
compounds, and changes in the oxygen coordination around the Al H/C,N-A-S-H gel [31], which are consistent with the trends detected in
atom, respectively. the FTIR analysis.
The broad section of the 27Al MAS-NMR spectra is an indicator of the
3.4.1
. 27Al NMR occurrence of pentahedral aluminium, distorted non-framework in
The 27Al MAS-NMR resonances of FA and slag precursors are shown Al[IV], and the presence of tetrahedral Al[IV] environment in

9
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

27
Fig. 9. Al MAS-NMR spectra of binder specimens with 9 wt% Na2O content and varying Ca/(Si + Al) ratios of a) 0.0, b) 0.07, c) 0.10, and d) 0.25.

10
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

Table 3
Pre- and Post-Efflorescence Chemical Shifts and Areas determined through MAS-NMR.

aluminosilicate structures [61,63]. The Al-O-Si sites in aluminosilicate indicating the occurrence of C,N-A-S-H gel [53].
gels exhibit a strong interaction with Na+ ions to balance the charge The 27Al MAS-NMR spectra of the binder mix with 5 wt% Na2O
deficiency of the Al ions [64]. The ingress and transport of moisture showed desirable characteristics compared with 9 wt% Na2O mixes,
during efflorescence can alter the mobility of the bound-Na+ ions in the because of moderate decrease in alkali content and the slag fraction in
aluminosilicate gel, which can, in turn, alter the interaction between both binder mixes are same. The addition of slag refines the overall
alkalis and oxygen atoms of Al[IV]. Leaching of Na+ ions during efflo­ porosity of the binder mixes and impedes the alkali movement. The 23Na
rescence can lead to weaker or cleaved Al-O-Si bonds, in addition to and 27Al MAS-NMR spectra of 5% Na2O binder mix with Ca/(Si + Al) =
causing distortion in these sites. Thus, constant leaching of Na+ ions 0.0, 0.07, and 0.10 (5A, 5B, and 5C) exhibited minor changes in the peak
during efflorescence can lead to the degradation of aluminosilicate gels. position and spectral shape, which indicated minor movement of Na+
This conjecture was supported by the 23Na NMR spectral analysis and ions and an unaffected Al-O-Si network, indicating stable C,N-A-S-H gel
indicated the changes in the Na-O distance. Since the Na+ ion (ionic under efflorescence. The 27Al MAS-NMR spectra of the 5D (Ca/(Si + Al)
radius = 0.97 Å) with a charge of +1 is leached because of the removal = 0.25) mix showed minor changes in the peak position and spectral
of alkalis from the aluminosilicate network, changes result in a) the Na- shape, which indicated the degradation of C,N-A-S-H gel (Fig. 3, 5D-
O distance, b) the Columbic interaction between Al-O-Si and Na ions, 90E). The 5D mix showed slight efflorescence formation (Fig. 3) and
23
and c) Al coordination geometry [63]. The broadening of MAS-NMR Na MAS-NMR spectra (Fig. 8b) showed a slight change in the peak
spectra in between δobs = 20 to 50 ppm region also indicates the position, which indicated slight Na+ ion movement under efflorescence.
distortion in Al-O-Si bonds in Al[IV] environment, which increases the Hence, natural carbonation of the 5D mix is the only possible cause for C,
electric field gradient at the Al nucleus and increases the quadrupolar N-A-S-H gel degradation due to prolonged exposure to natural carbon­
broadening [63]. Researchers have correlated the chemical shifts to Al- ation [53]. This conjecture is consistent with literature where the
O-Si bond angles of aluminosilicates, demonstrating that Al-O-Si angles effective movement of 27Al intensity to lower chemical shifts and
(δobs = 63 ppm has θ = 138◦ and δobs = 53 ppm has θ = 146◦ ) are higher removal of C-A-S-H gel component from 27Al environment was corrob­
for lower chemicals shifts [65]. Hence, the movement of the chemical orated as an effect of accelerated carbonation condition [53].
shift towards lower ppm values indicates the distortion of Al-O-Si bond After 90 days of efflorescence, the 5A-5C binder mixes (Fig. 10(a to c)
angles after efflorescence. The leaching of Na+ ions distort the Al-O-Si and Fig. 10(a-90d to c-90d)) showed a different trend compared with
network and slowly degrade the C,N-A-S-H gel, and concomitantly 9A–9D. The peak resonance of 5A shifted to higher values, i.e., from 59
reduce the mechanical performance of binder mixes. ppm to 60 ppm. Additionally, the narrowing of the spectra was wit­
The 27Al MAS-NMR spectra of the binder mixes (5A-5D) with 5% nessed, which indicated the ongoing reactivity and inclusion of Al[IV]
Na2O and varying Ca/(Si + Al) ratios after activation along with the units in the N-A-S-H environment. In contrast, the 5B and 5C binder
deconvoluted MAS-NMR spectra are shown in Fig. 10(a–d) and Fig. 10 mixes did not show any significant changes either in the Al[IV] peak
(a-28d to d-28d). The relative integral areas of the deconvoluted spectra positions or the 20 to 50 ppm broad region of distorted Al-O-Si linkage
are presented in Table 3. After activation, the 5A (Fig. 10(a and a-28d)); under efflorescence. The deconvolution of the MAS-NMR spectra and
Ca/(Si + Al) ratio of 0.0) showed an increased resonance intensity in the calculated relative integral areas displayed an increase in the Al[IV], and
Al[IV] environment, which indicates the existence of N-A-S-H gel. The 5A a decrease of the Al[V] and Al[VI] resulting from efflorescence. The
also exhibited broadening between δobs = 20 to 50 ppm, attributed to relative integral areas quantify the increases in the 20 to 50 ppm areas
unreacted Al[V] or distorted Al-O-Si linkage in Al[IV] environment, due to (attributed to Al[IV] or distorted Al-O-Si linkage) and decrease in the
the decrease in Na content in the binder mix [61,66]. The deconvolution Al[IV] areas after efflorescence. The removal of Na from the alumino­
of NMR spectra of 5A after activation showed three discrete peaks, i.e., i) silicate network is leading to a distorted Al-O-Si network. Also, in the
− 3.9 ppm (Al[VI]), ii) 48 ppm (Al[V]), and iii) 59 ppm (Al[IV]). Increasing binder mixes with slag, there is a slight movement of Al[IV] to lower
Ca/(Si + Al) or slag content in 5B, 5C, and 5D (Fig. 10(b to d) and Fig. 10 chemical shifts, indicating the release of Ca from the C,N-A-S-H gel, and
(b-28d to d-28d) shifted the Al[IV] to higher resonances to 60–62 ppm, an increase in the area distorted Al[IV] phases.

11
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

27
Fig. 10. Al MAS-NMR spectra of binder specimens with 5 wt% Na2O content and varying Ca/(Si + Al) ratios of a) 0.0, b) 0.07, c) 0.10, and d) 0.25.

12
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

23
Fig. 11. Na NMR spectra of (a) 9 wt% Na2O and (b) 5 wt% Na2O containing AAMs.

The presence of a minor quantity of slag content in the 5 wt% Na2O − 4.7 ppm (Fig. 11b). The increase in Ca/(Si + Al) ratios in 9A-9D also
binder mixes initially decelerates the formation of N-A-S-H gel at the caused a shift in the peak positions toward higher values (-3.2, − 3.2, and
initial stages and facilitates the formation of C-A-S-H gel, which can be − 3.1 ppm) for binder mixes 5B, 5C, and 5D, respectively, and exhibited
inferred to cause continuous activation at Ca/(Si + Al) ratios of 0.0, a trend similar to the 9% Na2O binder mixes after activation.
0.07, and 0.10. The increase in slag content increases the C-A-S-H gel After 90 days of efflorescence, 9A showed a large deviation in the
content and impedes Na+ ions movement. Thus, reducing the efflores­ chemical shifts towards 0, e.g., δobs = − 0.56 ppm (Fig. 11a). The 9A
cence formation in these mixes. As a result, there is a higher movement spectra exhibited an asymmetrical shape and extensive disorder of Na
for the 27Al environment of the binder specimens increasing with Ca/(Si sites in the binder mixes, indicating equivalent changes in the N-A-S-H
+ Al) ratios. Overall, the higher amount of alkali fraction in 9 wt% Na2O gel environment. The 9B, 9C, and 9D also exhibited noticeable deviation
binder mixes were responsible for more significant changes because of in chemical shifts towards positive values δobs = − 1.4, − 2.6, and − 1.6
efflorescence. ppm, respectively, indicating concomitant changes in their C,N-A-S-H
gel environment. However, the binder mixes with an increased Ca/(Si
3.4.2
. 23Na NMR + Al) showed a lower degree of variability in both resonance values and
Alkali ions in aluminosilicate gels neutralize the charges in two ways. shapes of their NMR spectra.
Firstly, the Na+ ions bound to aluminosilicate gel (-4 ppm) balance the The resonance of 5A after efflorescence shifted from − 4.7 ppm to
charge deficit developed by Al atoms [58]. Secondly, Na+ in pore so­ − 3.8 ppm (Fig. 11b), and the resonances of 5B, 5C, and 5D binder mixes
lution (0 ppm) balance the charge borne by the aqueous Al(OH)-4 groups shifted from − 3.2 ppm to − 2.9 ppm, − 3.2 ppm to − 3.7 ppm, and − 3.2
[58]. 23Na MAS-NMR analysis is an effective method in understanding ppm from − 3.1 ppm, respectively. These minor shifts indicated the in­
the effects of efflorescence on the N-A-S-H gel and C,N-A-S-H gel envi­ crease in Ca/(Si + Al) ratios for the 5 wt% Na2O binder mixes exhibited a
ronments of the AAM binder mixes as a function of Na2O content and weaker effect on chemical shift and the shape of the NMR spectra after
varying Ca/(Si + Al) ratios. efflorescence, driven by an increased slag fraction restricting the
9A (Ca/(Si + Al) = 0.0, 9 wt% Na2O) showed a peak resonance at movement of bound-Na+ ions in binder mixes, representative of a
− 4.5 ppm (Fig. 11a) that was associated with the charge-balancing Na+ possible refinement of porosity because of the development of C-A-S-H
ions surrounding Al-centred tetrahedra or Al-O-Si sites in the N-A-S-H gel in the mixes [21].
gel environment [67]. In 9B, 9C, and 9D, the increase in Ca/(Si + Al) The chemical shift in the 23Na environment occurs with hydration
ratios caused a shift in the peak positions towards − 2.8 ppm, − 4.3 ppm, and dehydration of aluminosilicates, where the dehydrated AAMs
and − 3.2 ppm, respectively. These shifts are attributed to the Na+ ions exhibit chemical shifts at higher negative values (-20 ppm), indicating
surrounding C,N-A-S-H gel after activation [67]. stronger interaction between oxygen atoms of Al tetrahedra and Na+
5A (Ca/(Si + Al) = 0.0, 5 wt% Na2O) showed a peak resonance at ions [64,68]. This movement of resonances towards positive values

13
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

3.0 28 days of curing 28 days of efflorescence

(b)
2.5

Split tensile strengthn (MPa)

2.0

1.5

1.0

0.5

0.0
9A 9B 9C 9D 5A 5B 5C 5D
Ca/(Si+Al) = 0 0.07 0.10 0.25 0 0.07 0.10 0.25

Fig. 12. A plot of (a) compressive strength and (b) splitting tensile strength of binder specimens as a function of curing and efflorescence.

suggests a) degradation of N-A-S-H/C,N-A-S-H gel, b) changes bound- On the other hand, the 5A-5D specimens showed increasing compressive
Na+ ions because of hydration, and c) changes in the Si/Al ratios of strength with increasing efflorescence exposure. After 90 days of efflo­
aluminosilicate gel. In general, Na+ ions show a strong inclination to the rescence, the mean compressive strengths for the 5A-5D specimens were
Al-O-Si sites rather than the Si-O-Si sites. The movement of Na+ ions in 12.81 MPa, 14.94 MPa, 17.99 MPa, and 26.25 MPa, respectively. The
the binder mixes increases the effective Na-O distance and coordination higher compressive strength after efflorescence was attributed to the
numbers, which causes a decrease in the ionic interaction between Na dissolution of silica and alumina under efflorescence [71], which is
and oxygen atoms of the Al tetrahedral units of N-A-S-H/C,N-A-S-H gels consistent with FTIR and MAS-NMR analysis conducted in this work.
[64,69]. The decrease in the electric field gradient at the Na+ nucleus The splitting tensile strength of the binder specimens after 28 days of
and decreased quadrupolar coupling exhibit a more isotropic Na+ curing is shown in Fig. 12(b). The binder mixes activated with 9 wt%
environment under hydration [70]. Hence, the bound Na+ ions move­ Na2O with varying Ca/(Si + Al) ratios showed increasing splitting ten­
ment due to hydration and increase in silica content in the N-A-S-H/C,N- sile strengths upon activation. After 28 days of efflorescence, a signifi­
A-S-H gel exhibits the major changes in the Na environment and in­ cant decrease in splitting tensile strength was observed in 9A-9D binder
dicates the depolymerisation of the aluminosilicate network. Binder mixes, attributed to the higher extent of efflorescence, which led to the
specimens 9A-9D showed a higher degree of chemical shifts, and the 5A- development of crystallization pressure in binder mixes [36,72] and a
5D binder mixes showed a lesser degree of variations in chemical shifts. concomitant decrease in tensile strength of binder mixes. Also, recent
The presence of a) low Na2O (5%), b) soluble silica, and c) increase in studies show a decreasing trend in tensile strength of binder mixes with
slag content in 5A-5D mixes restricted the movement of Na+ ions, and an increase in silica modulus [73]. The dissolution of silica during
thus exhibited greater stability under efflorescence, compared to the 9A- efflorescence is also a plausible reason for exhibiting a decrease in
9D mixes. splitting tensile strength.
The splitting tensile strength of binder mixes activated with 5% Na2O
and varying Ca/(Si + Al) showed an increasing trend after 28 days of
3.5. Mechanical properties
curing. After 28 days of efflorescence, the 5A-5C binder mixes showed
an increasing trend in splitting tensile strength, which is attributed to
The compressive strength of binder mixes before and after efflores­
the low efflorescence formation in these binder mixes. The FTIR and
cence is shown in Fig. 12(a). Overall, the compressive strength increased
MAS-NMR results also showed an ongoing alkali activation reaction and
with an increase in the Ca/(Si + Al) ratio. After 28 days of curing, the 9A
slight inclusion of Al and Si units under efflorescence.
– 9D specimens showed mean compressive strengths of 22.15 MPa,
27.23 MPa, 29.90 MPa, and 48.59 MPa, respectively. The 5A-5D binder
mixes in comparison showed lower mean compressive strengths values 3.6. Leaching and porosity analysis
after 28 days of curing, i.e., 9.30 MPa, 12.21 MPa, 14.74 MPa, and 24.0
MPa, respectively. These lower compressive strengths can be attributed The leaching concentrations of Na+ ions at end of 30, 60, and 90 days
to the lower activator content in 5 wt% Na2O binder mixes. The higher are shown for the 9% Na2O (Fig. 13a) and 5% Na2O (Fig. 13b) binder
alkali content in 9 wt% Na2O binder mixes caused a higher degree of mixes. The binder mixes with zero calcium or slag content showed a
activation, compared to the 5 wt% Na2O binder mixes. Also, an increase higher leaching potential compared with binder mixes with higher Ca/
in Ca/(Si + Al) ratios through the inclusion of slag led to the formation (Si + Al) ratios [29]. The increase in Ca/(Si + Al) significantly reduced
of C-A-S-H gel and facilitated the coexistence of N-A-S-H/C-A-S-H gels, the leaching of Na+ ions from binder mixes. Irrespective of the alkali
which brought about further improvements in the compressive strength content, the binder mixes with Ca/(Si + Al) = 0.25 showed the lowest
after activation. leaching potential of Na+ ions. The decrease in leaching potential is
After 28 days of efflorescence, the compressive strength of 9A – 9D attributed to the lower porosity values developed due to the slag
showed moderate changes (Fig. 12a). 9A and 9C showed an increase in addition.
compressive strengths, while 9B and 9D specimens showed a decreased The porosity values of selected binder mixes after activation are
compressive strength. Because efflorescence is a slow reaction, it is shown in Fig. 13(c). Irrespective of the alkali content, the binder mixes
difficult to observe any well-defined changes in compressive strengths. with a Ca/(Si + Al) ratio of 0.0 showed higher porosity values compared
After 90 days of efflorescence, it was not feasible to evaluate the with binder mixes with a Ca/(Si + Al) ratio of 0.25. Studies from other
compression strength of the 9A-9D specimens because of their extensive researchers have demonstrated that excess calcium occupies more pore
damage, and therefore they were assumed to have negligible strength. volume and refines porous network towards higher tortuosity in pore

14
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

Fig. 13. Concentration of sodium leached from AAM specimens at 30, 60, and 90 days (a) 9% Na2O specimens (9A-9D), (b) 5% Na2O (5A-5D) specimens and (c) the
porosity of binder specimens before and after efflorescence.

network [21] and inclusion of slag significantly decreases the pore this desirable range of properties and hence are suitable for future
diameter from 47.1 nm to 9.1 nm along with the reduction in pore commercial development.
volume from 33% to 8% [74].
The increase in slag content also hindered the Na+ leaching under CRediT authorship contribution statement
efflorescence exposure. However, the excessive alkali content in 9 wt%
Na2O binder mixes caused significant efflorescence formation and Na+ Lakshmikanth Srinivasamurthy: Conceptualization, Methodology,
movement after 60 days of exposure and severely reduced the dimen­ Investigation, Formal analysis, Writing – original draft, Visualization,
sional stability of the binder mixes. The presence of lower Na2O (5 wt%) Project administration. Venkata S. Chevali: Visualization, Writing –
and varying Ca/(Si + Al) ratios refined the pore structure, which review & editing, Supervision. Zuhua Zhang: Resources, Supervision,
reduced the leaching potential of alkalis and significantly reduced the Funding acquisition. Márlon A. Longhi: Methodology, Investigation.
efflorescence formation, and exhibited higher stability compared with 9 Thomas W. Loh: Methodology, Investigation, Formal analysis. Hao
wt% Na2O binder mixes. After efflorescence, there was an increase in Wang: Resources, Project administration, Supervision, Funding
porosity, which is attributed to the loss of alkalis from the binder mixes. acquisition.

4. Conclusions Declaration of Competing Interest

Efflorescence behaviour of FA/slag-based AAM binders was analysed The authors declare that they have no known competing financial
by varying Ca/(Si + Al) ratio and Na2O content. Overall, the binder interests or personal relationships that could have appeared to influence
mixes with 5 wt% Na2O with Ca/(Si + Al) ratios of 0.0, 0.07, and 0.10 the work reported in this paper.
showed lower efflorescence formation, enhanced stability, and
improved mechanical properties compared to the binder mixes with 9 wt Acknowledgments
% Na2O content.
The 9 wt% Na2O binder mixes activated with higher Ca/(Si + Al) The authors would like to acknowledge the financial support for this
showed significant efflorescence and considerable cracking. The 5 wt% project received through the ARC Discovery Project # DP160104149. LS
Na2O binder mixes showed lower efflorescence attributed to a lower would like to thank the Graduate Research School of the University of
activator content and a higher Ca/(Si + Al) ratio. Natural carbonation in Southern Queensland for the financial assistance received for con­
the higher Ca/(Si + Al) = 0.25 ratio led to C,N-A-S-H gel degradation. In ducting this work. Technical support and access to MAS-NMR at the
the 9% Na2O binder mixes, the splitting tensile strength was reduced to Indian Institute of Science, Bengaluru, India is appreciated. The partic­
25–50% of the pristine binders upon efflorescence and subflorescence in ipation of ZZ was partially supported by NSFC projects (51878263,
their pore networks. In the 5 wt% Na2O binder mixes, the splitting U2001225, 51638008).
tensile strength with Ca/(Si + Al) = 0.0, 0.07, and 0.10 increased
because of moderate efflorescence, and a slowly progressing alkali References
reaction.
Analysis of FTIR and MAS-NMR spectra showed major changes in the [1] G. Habert, S.A. Miller, V.M. John, J.L. Provis, A. Favier, A. Horvath, K.L. Scrivener,
Al[IV] environment of N-A-S-H/C,N-A-S-H gel, reinforcing the degrada­ Environmental impacts and decarbonization strategies in the cement and concrete
industries, Nat. Rev. Earth Environ. 1 (2020) 559–573, https://doi.org/10.1038/
tion in the Al[IV] environment of aluminosilicate gels. A high degree of s43017-020-0093-3.
mobility of Na+ ions and movement of the chemical shift towards lower [2] J.L. Provis, S.A. Bernal, Geopolymers and related alkali-activated materials, Annu.
ppm determined using 23Na and 27Al MAS-NMR supported the degra­ Rev. Mater. Res. 44 (2014) 299–327, https://doi.org/10.1146/annurev-matsci-
070813-113515.
dation of Al[IV] environment under efflorescence. FTIR and 27Al MAS- [3] Jannie S.J. van Deventer, John L. Provis. Alkali Activated Materials: State-of-the-
NMR findings supported the inclusion of Al[IV] units in the N-A-S-H/C, Art Report, RILEM TC 224-AAM, Springer, Dordrecht, the Netherlands, 2014.
N-A-S-H environment under efflorescence for 5% Na2O binder mixes [4] M.B. Ali, R. Saidur, M.S. Hossain, A review on emission analysis in cement
industries, Renew. Sustain. Energy Rev. 15 (2011) 2252–2261, https://doi.org/
with Ca/(Si + Al) = 0.0, 0.07, and 0.25. FTIR, TGA, and MAS-NMR
10.1016/j.rser.2011.02.014.
further revealed carbonation along with decalcification and deal­ [5] K.-H. Yang, J.-K. Song, K.-I. Song, Assessment of CO2 reduction of alkali-activated
umination of aluminosilicate gel in 5D binder. concrete, J. Cleaner. Prod. 9 (2013) 265–272.
Overall, a binder with lower activator content and high stability is [6] M.A. Longhi, Z. Zhang, E.D. Rodríguez, A.P. Kirchheim, H. Wang, Efflorescence of
alkali-activated cements (geopolymers) and the impacts on material structures: A
preferable for a lower CO2 footprint in production. In this study, the 5 wt critical analysis, Front. Mater. 6 (2019), https://doi.org/10.3389/
% Na2O binder mixes with Ca/(Si + Al) = 0.0, 0.07, and 0.10 showed fmats.2019.00089.

15
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

[7] J. Davidovits, Geopolymers – Inorganic polymeric new materials, J. Therm. Anal. [33] S.M. Rao, B.V. Venkatarama Reddy, G.C. Raju, S. Lakshmikanth, N.S. Ambika,
37 (1991) 1633–1656, https://doi.org/10.1007/BF01912193. Chemical stabilization of lead contaminated gypsum sludge, Int. J. Geotech. Eng. 3
[8] E. Adesanya, P. Perumal, T. Luukkonen, J. Yliniemi, K. Ohenoja, P. Kinnunen, (2009) 109–116, https://doi.org/10.3328/IJGE.2009.03.01.109-116.
M. Illikainen, Opportunities to improve sustainability of alkali-activated materials: [34] G.W. Scherer, Theory of Drying, Theory of Drying 73 (1) (1990) 3–14.
A review of side-stream based activators, J. Clean. Prod. 286 (2021) 125558. [35] M.A. Longhi, E.D. Rodríguez, B. Walkley, Z. Zhang, A.P. Kirchheim, Metakaolin-
[9] K.S. Getvoldsen, T.E. Butt, C. House, F. Ferreira, Sustainable Development and based geopolymers: Relation between formulation, physicochemical properties and
Climate Change, World Sustain. Ser. 38 (2018) 445–457, https://doi.org/10.1007/ efflorescence formation, Compos. Part B Eng. 182 (2020) 107671.
978-3-319-63007-6_27. [36] G.W. Scherer, Stress from crystallization of salt, Cem. Concr. Res. 34 (2004)
[10] B. Singh, G. Ishwarya, M. Gupta, S.K. Bhattacharyya, Geopolymer concrete: A 1613–1624, https://doi.org/10.1016/j.cemconres.2003.12.034.
review of some recent developments, Constr. Build. Mater. 85 (2015) 78–90, [37] B. Walkley, R. San Nicolas, M.A. Sani, G.J. Rees, J.V. Hanna, J.S.J. van Deventer, J.
https://doi.org/10.1016/j.conbuildmat.2015.03.036. L. Provis, Phase evolution of C-(N)-A-S-H/N-A-S-H gel blends investigated via
[11] C. Shi, A.F. Jiménez, A. Palomo, New cements for the 21st century: The pursuit of alkali-activation of synthetic calcium aluminosilicate precursors, Cem. Concr. Res.
an alternative to Portland cement, Cem. Concr. Res. 41 (2011) 750–763, https:// 89 (2016) 120–135, https://doi.org/10.1016/j.cemconres.2016.08.010.
doi.org/10.1016/j.cemconres.2011.03.016. [38] S.A. Bernal, J.L. Provis, V. Rose, R. Mejía De Gutierrez, Evolution of binder
[12] P. Duxson, J.L. Provis, G.C. Lukey, S.W. Mallicoat, W.M. Kriven, J.S.J. Van structure in sodium silicate-activated slag-metakaolin blends, Cem. Concr. Compos.
Deventer, Understanding the relationship between geopolymer composition, 33 (2011) 46–54, https://doi.org/10.1016/j.cemconcomp.2010.09.004.
microstructure and mechanical properties, Colloids Surfaces A Physicochem. Eng. [39] S. Kumar, R. Kumar, S.P. Mehrotra, Influence of granulated blast furnace slag on
Asp. 269 (2005) 47–58, https://doi.org/10.1016/j.colsurfa.2005.06.060. the reaction, structure and properties of fly ash based geopolymer, J. Mater. Sci. 45
[13] A. Fernandez-Jimenez, A. Palomo, I. Sobrados, J. Sanz, The role played by the (2010) 607–615, https://doi.org/10.1007/s10853-009-3934-5.
reactive alumina content in the alkaline activation of fly ashes, Microporous [40] I. García-Lodeiro, A. Fernández-Jiménez, M.T. Blanco, A. Palomo, FTIR study of the
Mesoporous Mater. 91 (2006) 111–119, https://doi.org/10.1016/j. sol-gel synthesis of cementitious gels: C-S-H and N-A-S-H, J. Sol-Gel Sci. Technol.
micromeso.2005.11.015. 45 (2008) 63–72, https://doi.org/10.1007/s10971-007-1643-6.
[14] M.R. Rowles, J.V. Hanna, K.J. Pike, M.E. Smith, B.H. O’Connor, 29Si, 27Al, 1H and [41] J.W. Phair, J.S.J. Van Deventer, J.D. Smith, Mechanism of polysialation in the
23Na MAS NMR study of the bonding character in aluminosilicate inorganic incorporation of zirconia into fly ash-based geopolymers, Ind. Eng. Chem. Res. 39
polymers, Appl. Magn. Reson. 32 (2007) 663–689, https://doi.org/10.1007/ (2000) 2925–2934, https://doi.org/10.1021/ie990929w.
s00723-007-0043-y. [42] A. Palomo, F. Glasser, Chemically-bonded cementitious materials based on
[15] M.R. Rowles, B.H. O’Connor, Chemical and structural microanalysis of metakaolin, Br. Ceram. Trans. J. 91 (1992) 107–112.
aluminosilicate geopolymers synthesized by sodium silicate activation of [43] A. Hajimohammadi, J.L. Provis, J.S.J. van Deventer, Time-resolved and spatially-
metakaolinite, J. Am. Ceram. Soc. 92 (2009) 2354–2361, https://doi.org/10.1111/ resolved infrared spectroscopic observation of seeded nucleation controlling
j.1551-2916.2009.03191.x. geopolymer gel formation, J. Colloid Interface Sci. 357 (2011) 384–392, https://
[16] Z. Zhang, J.L. Provis, X. Ma, A. Reid, H. Wang, Efflorescence and subflorescence doi.org/10.1016/j.jcis.2011.02.045.
induced microstructural and mechanical evolution in fly ash-based geopolymers, [44] I. García-Lodeiro, A. Fernández-Jiménez, A. Palomo, D.E. MacPhee, Effect of
Cem. Concr. Compos. 92 (2018) 165–177, https://doi.org/10.1016/j. calcium additions on N-A-S-H cementitious gels, J. Am. Ceram. Soc. 93 (2010)
cemconcomp.2018.06.010. 1934–1940, https://doi.org/10.1111/j.1551-2916.2010.03668.x.
[17] Z. Zhang, J.L. Provis, A. Reid, H. Wang, Fly ash-based geopolymers: The [45] A.A. Hoyos-Montilla, F. Puertas, J. Molina Mosquera, J.I. Tobón, Infrared spectra
relationship between composition, pore structure and efflorescence, Cem. Concr. experimental analyses on alkali-activated fly ash-based binders, Spectrochim. Acta
Res. 64 (2014) 30–41, https://doi.org/10.1016/j.cemconres.2014.06.004. – Part A Mol. Biomol. Spectrosc. 269 (2022) 120698. https://doi.org/10.1016/j.
[18] Z. Zhang, J.L. Provis, H. Wang, F. Bullen, A. Reid, Quantitative kinetic and saa.2021.120698.
structural analysis of geopolymers. Part 2. Thermodynamics of sodium silicate [46] C.A. Rees, J.L. Provis, G.C. Lukey, J.S.J. van Deventer, The mechanism of
activation of metakaolin, Thermochim. Acta. 565 (2013) 163–171, https://doi. geopolymer gel formation investigated through seeded nucleation, Colloids
org/10.1016/j.tca.2013.01.040. Surfaces A-Physicochemical Eng. Asp. 318 (2008) 97–105, https://doi.org/
[19] L.Y. Yang, Z.J. Jia, Y.M. Zhang, J.G. Dai, Effects of nano-TiO2 on strength, 10.1016/j.colsurfa.2007.12.019.
shrinkage and microstructure of alkali activated slag pastes, Cem. Concr. Compos. [47] X.C. Ping Yu, R. James Kirkpatrick, B. Poe, P.F. McMillan, Structure of Calcium
57 (2015) 1–7, https://doi.org/10.1016/j.cemconcomp.2014.11.009. Silicate Hydrate (C-S-H): Near-, Mid-, and Far-Infrared Spectroscopy, J. Am.
[20] H. Wan, L. Yuan, Y. Zhang, Insight Into the Leaching of Sodium Alumino-Silicate Ceram. Soc. 82 (1999) 742–748.
Hydrate (N-A-S-H) Gel: A Molecular Dynamics Study, Front. Mater. 7 (2020) 1–11, [48] I. Ismail, S.A. Bernal, J.L. Provis, R. San Nicolas, S. Hamdan, J.s.j., Van Deventer,
https://doi.org/10.3389/fmats.2020.00056. Modification of phase evolution in alkali-activated blast furnace slag by the
[21] J.L. Provis, R.J. Myers, C.E. White, V. Rose, J.S.J. Van Deventer, X-ray incorporation of fly ash, Cem. Concr. Compos. 45 (2014) 125–135, https://doi.
microtomography shows pore structure and tortuosity in alkali-activated binders, org/10.1016/j.cemconcomp.2013.09.006.
Cem. Concr. Res. 42 (2012) 855–864, https://doi.org/10.1016/j. [49] M. Criado, A. Fernández-Jiménez, A. Palomo, Alkali activation of fly ash, Part III:
cemconres.2012.03.004. Effect of curing conditions on reaction and its graphical description, Fuel. 89
[22] J. Xie, J. Wang, R. Rao, C. Wang, C. Fang, Effects of combined usage of GGBS and (2010) 3185–3192, https://doi.org/10.1016/j.fuel.2010.03.051.
fly ash on workability and mechanical properties of alkali activated geopolymer [50] S.A. Bernal, R.M. de Gutierrez, J.L. Provis, V. Rose, Effect of silicate modulus and
concrete with recycled aggregate, Compos. Part B Eng. 164 (2019) 179–190, metakaolin incorporation on the carbonation of alkali silicate-activated slags, Cem.
https://doi.org/10.1016/j.compositesb.2018.11.067. Concr. Res. 40 (2010) 898–907, https://doi.org/10.1016/j.
[23] I. García Lodeiro, A. Fernández-Jimenez, A. Palomo, D.E. Macphee, Effect on fresh cemconres.2010.02.003.
C-S-H gels of the simultaneous addition of alkali and aluminium, Cem. Concr. Res. [51] C.A. Rees, J.L. Provis, G.C. Lukey, J.S.J. Van Deventer, In situ ATR-FTIR study of
40 (2010) 27–32, https://doi.org/10.1016/j.cemconres.2009.08.004. the early stages of fly ash geopolymer gel formation, Langmuir. 23 (2007)
[24] C.K. Yip, G.C. Lukey, J.S.J. Van Deventer, The coexistence of geopolymeric gel and 9076–9082, https://doi.org/10.1021/la701185g.
calcium silicate hydrate at the early stage of alkaline activation, Cem. Concr. Res. [52] I. Majchrzak-Kucęba, W. Nowak, Thermal analysis of fly ash-based zeolites, Journal
35 (2005) 1688–1697, https://doi.org/10.1016/j.cemconres.2004.10.042. of Thermal Analysis and Calorimetry 77 (1) (2004) 125–131.
[25] M. Ben Haha, G. Le Saout, F. Winnefeld, B. Lothenbach, Influence of activator type [53] S.A. Bernal, J.L. Provis, B. Walkley, R. San Nicolas, J.D. Gehman, D.G. Brice, A.
on hydration kinetics, hydrate assemblage and microstructural development of R. Kilcullen, P. Duxson, J.s.j., Van Deventer, Gel nanostructure in alkali-activated
alkali activated blast-furnace slags, Cem. Concr. Res. 41 (2011) 301–310, https:// binders based on slag and fly ash, and effects of accelerated carbonation, Cem.
doi.org/10.1016/j.cemconres.2010.11.016. Concr. Res. 53 (2013) 127–144, https://doi.org/10.1016/j.
[26] S.A. Bernal, J.L. Provis, V. Rose, R.M. de Gutiérrez, J. Biernacki, High-resolution X- cemconres.2013.06.007.
ray diffraction and fluorescence microscopy characterization of alkali-activated [54] A.F. Abdalqader, F. Jin, A. Al-Tabbaa, Development of greener alkali-activated
slag-metakaolin binders, J. Am. Ceram. Soc. 96 (6) (2013) 1951–1957. cement: Utilisation of sodium carbonate for activating slag and fly ash mixtures,
[27] B. Qu, A. Martin, J.Y. Pastor, A. Palomo, A. Fernández-Jiménez, Characterisation of J. Clean. Prod. 113 (2016) 66–75, https://doi.org/10.1016/j.jclepro.2015.12.010.
pre-industrial hybrid cement and effect of pre-curing temperature, Cem. Concr. [55] C.L. Hwang, D.H. Vo, V.A. Tran, M.D. Yehualaw, Effect of high MgO content on the
Compos. 73 (2016) 281–288, https://doi.org/10.1016/j. performance of alkali-activated fine slag under water and air curing conditions,
cemconcomp.2016.07.019. Constr. Build. Mater. 186 (2018) 503–513, https://doi.org/10.1016/j.
[28] A. Roy, P.J. Schilling, H.C. Eaton, P.G. Malone, W.N. Brabston, L.D. Wakeley, conbuildmat.2018.07.129.
Activation of Ground Blast-Furnace Slag by alkali metal and alkaline earth [56] F. Jin, K. Gu, A. Al-Tabbaa, Strength and drying shrinkage of reactive MgO
hydroxides, J. Am. Ceram. Soc. 12 (1992) 3233–3240, https://doi.org/10.1111/ modified alkali-activated slag paste, Constr. Build. Mater. 51 (2014) 395–404,
j.1151-2916.1992.tb04416.x. https://doi.org/10.1016/j.conbuildmat.2013.10.081.
[29] L. Srinivasamurthy, V.S. Chevali, Z. Zhang, H. Wang, Phase changes under [57] S.A. Bernal, R. San Nicolas, J.L. Provis, R. Mejía de Gutiérrez, J.S.J. van Deventer,
efflorescence in alkali activated materials with mixed activators, Constr. Build. Mejía De Gutiérrez, J.S.J. Van Deventer, Natural carbonation of aged alkali-
Mater. 283 (2021) 122678. activated slag concretes, Mater. Struct. Constr. 47 (4) (2014) 693–707.
[31] M.A. Longhi, B. Walkley, E.D. Rodríguez, A.P. Kirchheim, Z. Zhang, H. Wang, New [58] P. Duxson, G.C. Lukey, F. Separovic, J.S.J. Van Deventer, Effect of alkali cations on
selective dissolution process to quantify reaction extent and product stability in aluminum incorporation in geopolymeric gels, Ind. Eng. Chem. Res. 44 (2005)
metakaolin-based geopolymers, Compos. Part B Eng. 176 (2019) 107172. 832–839, https://doi.org/10.1021/ie0494216.
[32] S.M. Rao, B.V.V. Reddy, S. Lakshmikanth, N.S. Ambika, Re-use of fluoride [59] X. Gao, Q.L. Yu, H.J.H. Brouwers, Apply 29Si, 27Al MAS NMR and selective
contaminated bone char sludge in concrete, J. Hazard. Mater. 166 (2009) 751–756, dissolution in identifying the reaction degree of alkali activated slag-fly ash
https://doi.org/10.1016/j.jhazmat.2008.11.115. composites, Ceram. Int. 43 (2017) 12408–12419, https://doi.org/10.1016/j.
ceramint.2017.06.108.

16
L. Srinivasamurthy et al. Construction and Building Materials 332 (2022) 127273

[60] F. Škvára, L. Kopecký, L. Myšková, V.Í.T. Šmilauer, L. Alberovská, L. Vinšová, [70] M. Criado, A. Fernández-Jiménez, A. Palomo, Alkali activation of fly ash: Effect of
Aluminosilicate polymers – Influence of elevated temperatures, efflorescence, the SiO2/Na2O ratio, Part I: FTIR study, Microporous Mesoporous Mater. 106
Ceram. – Silikaty. 53 (2009) 276–282. (2007) 180–191, https://doi.org/10.1016/j.micromeso.2007.02.055.
[61] B. Walkley, J.L. Provis, Solid-state nuclear magnetic resonance spectroscopy of [71] P. Duxson, J.S.J. Van Deventer, 17 - Commercialization of geopolymers for
cements, Mater. Today Adv. 1 (2019) 100007. construction – opportunities and obstacles, in: J.L. Provis, J.S.J.B.T.-G. van
[62] J. Pires, A. Carvalho, M. Pinto, J. Rocha, Characterization of Y zeolites Deventer (Eds.), Woodhead Publ. Ser. Civ. Struct. Eng., Woodhead Publishing,
dealuminated by solid-state reaction with ammonium hexafluorosilicate, J Porous 2009: pp. 379–400. https://doi.org/https://doi.org/10.1533/
Mater. 13 (2) (2006) 107–114. 9781845696382.3.379.
[63] J.A. Van Bokhoven, A.L. Roest, D.C. Koningsberger, J.T. Miller, G.H. Nachtegaal, A. [72] G. Huang, Y. Ji, L. Zhang, J. Li, Z. Hou, The influence of curing methods on the
P.M. Kentgens, Changes in structural and electronic properties of the zeolite strength of MSWI bottom ash-based alkali-activated mortars: The role of leaching
framework induced by extraframework Al and la in H-USY and La(x)NaY: A 29 Si of OH− and free alkali, Constr. Build. Mater. 186 (2018) 978–985, https://doi.org/
and 27 AI MAS NMR and 27 A1 MQ MAS NMR study, J. Phys. Chem. B. 104 (2000) 10.1016/J.CONBUILDMAT.2018.07.224.
6743–6754, https://doi.org/10.1021/jp000147c. [73] J. Chang, W. Li, D. Wang, Y. Zhang, Effect of silicate modulus on tensile properties
[64] A. Fernández-Jiménez, A. Palomo, Nanostructure/microstructure of fly ash and microstructure of waterproof coating based on polymer and sodium silicate-
geopolymers, Geopolymers. (2009) 89–117, https://doi.org/10.1533/ activated GGBS, Constr. Build. Mater. 252 (2020), 119056, https://doi.org/
9781845696382.1.89. 10.1016/j.conbuildmat.2020.119056.
[65] E. Lippmaa, A. Samoson, M. Mägi, High-Resolution 27Al NMR of Aluminosilicates, [74] T. Yang, X. Yao, Z. Zhang, Quantification of chloride diffusion in fly ash-slag-based
J. Am. Chem. Soc. 108 (1986) 1730–1735, https://doi.org/10.1021/ja00268a002. geopolymers by X-ray fluorescence (XRF), Constr. Build. Mater. 69 (2014)
[66] C. Ruiz-Santaquiteria, A. Fernandez-Jimenez, J. Skibsted, A. Palomo, Clay 109–115, https://doi.org/10.1016/j.conbuildmat.2014.07.031.
reactivity: Production of alkali activated cements, Appl. Clay Sci. 73 (2013) 11–16,
https://doi.org/10.1016/j.clay.2012.10.012.
[67] B. Walkley, A. Kashani, M.A. Sani, T.D. Ngo, P. Mendis, Examination of alkali-
Further reading
activated material nanostructure during thermal treatment, J. Mater. Sci. 53
(2018) 9486–9503, https://doi.org/10.1007/s10853-018-2270-z. [75] AS 1012.10, Methods of testing concrete Determination of indirect tensile strength
[68] V.F.F. Barbosa, K.J.D. MacKenzie, C. Thaumaturgo, Synthesis and characterisation of concrete cylinders (’Brazil’ or splitting test), Standards Australia (2000).
of materials based on inorganic polymers of alumina and silica: sodium polysialate [76] AS 1012.9, Methods of testing concrete Compressive strength tests - Concrete,
polymers, International Journal of Inorganic Materials 2 (4) (2000) 309–317. mortar and grout specimens, Standards Australia (2014).
[69] S.K. Lee, J.F. Stebbins, The distribution of sodium ions in aluminosilicate glasses: a
high-field Na-23 MAS and 3Q MAS NMR study, Geochim. Cosmochim. Acta. 67
(2003) 1699–1709, https://doi.org/10.1016/S0016-7037(03)00026-7.

17