Content uploaded by Warda Ashraf
Author content
All content in this area was uploaded by Warda Ashraf on Oct 12, 2017
Content may be subject to copyright.
Carbonation behavior of hydraulic and non-hydraulic
calcium silicates: potential of utilizing low-lime calcium
silicates in cement-based materials
Warda Ashraf
1,
* and Jan Olek
1
1
Lyles School of Civil Engineering, Purdue University, 550 Stadium Mall Drive, West Lafayette, IN 47907, USA
Received: 12 November 2015
Accepted: 17 March 2016
Published online:
1 April 2016
Springer Science+Business
Media New York 2016
ABSTRACT
This paper presents a study on the carbonation behaviors of hydraulic and non-
hydraulic calcium silicate phases, including tricalcium silicate (3CaOSiO
2
or
C
3
S), c-dicalcium silicate (c-2CaOSiO
2
or c-C
2
S), b-dicalcium silicate (b-
2CaOSiO
2
or b-C
2
S), rankinite (3CaO2SiO
2
or C
3
S
2
), and wollastonite
(CaOSiO
2
or CS). These calcium silicate phases were subjected to carbonation
reaction at different CO
2
concentration and temperatures. Thermogravimetric
analysis (TGA) tests were performed to quantify the amounts of carbonates
formed during the carbonation reactions, which were eventually used to mon-
itor the degree of reactions of the calcium silicate phases. Both hydraulic and
non-hydraulic calcium silicates demonstrated higher reaction rate in case of
carbonation reaction than that of the hydration reaction. Under specific car-
bonation scenario, non-hydraulic low-lime calcium silicates such as c-C
2
S, C
3
S
2
and CS were found to achieve a reaction rate close to that of C
3
S. Fourier
transformed infrared (FTIR) spectroscopy and scanning electron microscope
(SEM) tests were performed to characterize the carbonation reaction products of
the calcium silicate phases. The FTIR spectra during the early stage of carbon-
ation reaction showed formation of calcium silicate hydrate (C–S–H) from C
3
S,
c-C
2
S, b-C
2
S, and C
3
S
2
phases with a similar degree of polymerization as that of
the C–S–H that forms during the hydration reaction of C
3
S. However, upon
further exposure to CO
2
, these C–S–H phases decompose and eventually con-
verted to calcium-modified (Ca-modified) silica gel phase with higher degree of
silicate polymerization. Contradictorily, CS phase started forming Ca-modified
silica gel phase even at the early stage of carbonation reaction. This paper also
revealed the stoichiometry of the Ca-modified silica gel that formed during the
carbonation reaction of the calcium silicate phases using the SEM/energy dis-
persive spectroscopy (EDS) and TGA results.
Address correspondence to E-mail: washraf@purdue.edu
DOI 10.1007/s10853-016-9909-4
J Mater Sci (2016) 51:6173–6191
Introduction
Relatively high CO
2
footprint of ordinary Portland
cements (OPC) are caused by high production tem-
perature requirement (*1450 C) and the calcination
of limestone, both of which are dependent on the
required composition of cements. Conventional
cements contain nearly 80 % of calcium silicates,
namely, alite (C
3
S) and b-belite (b-C
2
S). C
3
S is well
known for its high hydraulic reactivity and hence,
considered as a desired component in conventional
cements for early strength gain properties. The
hydraulic reactivity of b-C
2
S is substantially slower
than that of C
3
S[1]. Among these calcium silicates,
C
3
S requires the highest amount of limestone and
temperature for its production. Thus, one of the most
viable options to reduce the CO
2
footprint of cement-
based materials is to decrease the amount of C
3
S and
increase the utilization of low-lime calcium silicates
such as C
2
S, C
3
S
2
, and CS. Utilization of these low-
lime calcium silicates will also reduce the energy
requirement because of their lower production tem-
perature [2–5]. However, lack of reactivity of low-
lime calcium silicate phases is the main concern that
needs to be addressed to ensure proper and opti-
mized usage of these phases in conventional cement-
based materials.
Several research initiatives have been undertaken
to obtain reactive forms of C
2
S by ‘remelting reaction’
of the a-C
2
S[6] or by burning the mixture of C–S–H
and portlandite [7]. Nonetheless, industrial applica-
tions of these methods are limited. High reactive
forms of C
2
S can also be stabilized by adding con-
trolled amounts of NaF [8], SO
3
, and B
2
O
3
[9–13].
While the industrial benefits of these later methods
are undeniable [14], the availability of raw materials
for the necessary impurities can be a concern. At this
point, development of C
2
S-rich cements as an alter-
native to OPC is an active field of research [15–20].
Contradictory to the vast research interests regarding
C
2
S, potential applications of CS and C
3
S
2
in cement-
based materials have not yet been thoroughly
explored. CS is commonly used in ceramics and
plastics as filler because of its good fluxing proper-
ties, less volatile component, and needle-like shapes
that give it a reinforcing capability. The application of
CS in cement-based materials so far was limited as
micro-fibers [21–25]. Accordingly, in order to explore
the potential utilization of the low-lime calcium sili-
cates (i.e., c-C
2
S, C
3
S
2
, and CS) as cementitious
materials, this research work has investigated the
reactivity of these phases in comparison to that of
hydraulic calcium silicates (i.e., C
3
S and b-C
2
S) that
are already being used in conventional cements.
A group of researchers [26–30] in 1970s revealed
that it is possible to activate the hydration reaction of
calcium silicates using CO
2
gas. Since then only a
limited progress has been made on this promising
means of developing a sustainable alternative to
conventional OPC system by carbonating low-lime
calcium silicates. The possibility of activating low-
lime calcium silicates has gained recent interest with
the development of carbonation-activated binders
[31–35]. The strength development of this type of
binder relies on the carbonation reaction of its
ingredients (i.e., calcium silicates). The carbonation
reaction of conventional cements and hydraulic cal-
cium silicates (i.e., C
3
S and b-C
2
S) can be considered
to be well explored [30,36–45]. A few researchers [46,
47] focused on the carbonation of CS, primarily, to
enhance the knowledge regarding the CO
2
seques-
tration capability of this mineral. In an attempt to
contribute to the development of the carbonation-
based alternative to traditional (i.e., hydration-based)
cementitious systems, this paper aims to compare the
reactivity and reaction products of the various pure
calcium silicate phases (i.e. C
3
S, b-C
2
S, c-C
2
S, C
3
S
2
,
and CS) while subjected to carbonation reaction.
Specific objectives of this research include: (i) com-
parison of the reactivity of hydraulic and non-hy-
draulic calcium silicates during carbonation
reactions, (ii) exploring the structure of the calcium-
silica gel (i.e., either C–S–H or Ca-modified silica gel)
phases that form during the carbonation reaction of
calcium silicates using FTIR, and (iii) determination
of the stoichiometric equations of the carbonation
reactions of calcium silicate phases.
Materials and methods
Sample preparation
Synthesis of calcium silicates
Several methods for synthesizing pure calcium sili-
cate phases can be found in the literature [3,41,42,
48,49], most of which involve sintering the stoichio-
metric mixture of lime and silica. In this study, the
synthetic calcium silicate phases were prepared by
6174 J Mater Sci (2016) 51:6173–6191
sintering the stoichiometric mixture of CaO and
amorphous (fumed) silica (SiO
2
). The C
3
S was pre-
pared by exposing the mixture of CaO and fumed
silica (at 3:1 molar ratio) to a temperature of 1510 C
for 4 h and then leaving the melt in the furnace until
it cooled down to a room temperature. The resulting
material was ground, sieved using # 200 (74 lm)
sieve, and refried (three times in total) to maximize
the chemical reaction of available lime and silica.
After each grinding, small amount of the powdered
sample was examined using X-ray diffraction (XRD)
method to check for the presence of any free lime.
The preparation process of c-C
2
S was the same as
that used to prepare C
3
S except in this case the molar
ratio of CaO and fumed silica was 2:1 and the sin-
tering temperature was 1400 C. b-C
2
S was obtained
by re-firing c-C
2
S up to 1000 C followed by a rapid
cooling process using de-ionized water (details of this
method can be found in [50]). The natural CS and
synthetic C
3
S
2
used in this study were supplied by
Solidia Technologies, NJ, USA.
The XRD patterns of the calcium silicate phases are
given in Fig. 1. These patterns match reasonably well
with the published XRD patterns for the C
3
S, b-C
2
S,
c-C
2
S, and CS [42,51,52]. The indexing of the XRD
patterns was performed using a simple search-match
software (Jade, MDI). The particle size distributions
(Fig. 2) of the calcium silicate phases were obtained
using Malvern Mastersizer 2000 particle size analyzer
for a refractive index of 1.63 of the powder samples.
Preparation of test specimens
Calcium silicate samples were subjected to four dif-
ferent exposure conditions (Table 1). In first two
scenarios, the objective was to compare the hydration
and carbonation behavior of the calcium silicate
phases. For these scenarios, the powdered calcium
C3S
β-C2S
γ-C2S
C3S2
CS
10 20 30 40 50 60 70 80
Two theta (degrees)
Intensity (arbitrary unit)
Figure 1 XRD patterns of the
calcium silicate phases.
J Mater Sci (2016) 51:6173–6191 6175
silicates were mixed with small amount of water
(water to solid ratio =0.40) and compacted into cir-
cular disks (0.75 inch in diameter and 0.25 inch in
height). For the hydration reaction, these disk sam-
ples were kept in an environment with 94 % RH and
at 23 C without the addition of any external CO
2
(however, it is difficult to completely eliminate the
presence of the atmospheric CO
2
). For carbonation
reaction according to scenario-2, disks samples were
placed in an enclosed carbonation chamber (Fig. 3)
which was kept on the laboratory bench and contin-
uously purged with 99.9 % pure CO
2
at 23 C.
Additional details of this carbonation setup can be
found in [53].
It is well known that the carbonation of cementi-
tious materials is a diffusion-based process, and the
extent of reaction can be substantially affected by
sample dimensions and level of compactions [54–56].
Hence, to avoid the effects of these parameters on the
reactivity and to obtain uniformly carbonated sam-
ples for characterization works, calcium silicate
phases were carbonated (scenarios-3 and 4, Table 1)
in paste form without any compaction. In these cases,
powder calcium silicates were first mixed with water
(water to solid ratio =0.40), and then these paste
samples were spread on plastic plates without any
compaction. These plates were then placed inside the
CO
2
chamber. Paste samples prepared in this proce-
dure had a maximum dimension of around 1–2 mm,
and hence, in this case, carbonation reaction was
expected to occur uniformly throughout the sample.
Exposure scenario-4 was designed to have faster
reaction rate (because of the higher CO
2
concentra-
tion and higher temperature) than that of the sce-
nario-3. For scenario-3, a commercial carbonation
chamber (VWR symphony CO
2
incubator) was used,
while scenario-4 had the same setup as shown in
Fig. 3.
Test methods
The TGA tests were performed using a commercially
available instrument (TA instruments, model: Q50).
The carbonated samples were first ground and then
sieved with # 200 sieve. These powdered samples
were then kept under isothermal conditions for
10 min followed by a raise in the temperature at the
rate of 10 C per minute up to 1000 C. Approxi-
mately 35–50 mg of powder was tested for each
specimen. Nitrogen (N
2
) was used as purge gas
during the TGA tests. TGA curves for one of the
Figure 2 Particle size distributions of calcium silicate phases.
Table 1 Details of the exposure scenarios
Exposure conditions Type of sample CO
2
concentration Temperature (C) Relative humidity
Scenario-1 Circular disk
(0.75 inch 90.25 inch)
None/atmospheric 23 94 %
Scenario-2 100 % 23
Scenario-3 Paste sample without any
compaction (1–2 mm thickness)
15 % 35
Scenario-4 100 % 55
Figure 3 Carbonation test setup used in exposure scenario-2 and
4.
6176 J Mater Sci (2016) 51:6173–6191
carbonated samples are given in Fig. 4as an example.
The mass losses in the range of 400–550 C and
550–900 C were used, respectively, to calculate the
mass fraction of portlandite (Ca(OH)
2
) and calcium
carbonate (CaCO
3
) present in the carbonated sam-
ples. The amount of chemically bound water in a
sample was calculated by subtracting the mass losses
due to the decomposition of Ca(OH)
2
, CaCO
3
, and
free water (mass loss below 100 C) from the total
mass loss up to 1000 C.
The FTIR spectra were collected for powder sam-
ples prepared following the same procedure as in the
TGA tests. The FTIR spectroscopy data were obtained
using the Thermo Nicolet Nexus 470 spectropho-
tometer which was fitted with an attenuated total
reflectance (ATR) accessory. The frequency range
was 800–4000 cm
-1
at a resolution of 4 cm
-1
. Each
spectrum presented in this paper is an average of 36
scans.
SEM images and the EDS information were
obtained using a FEI NOVA nanoSEM FESEM which
was operated in high vacuum mode. The instrument
was operated using accelerating voltage of 15 kV and
the working distance was 10 mm. For this setup, the
size of the microvolume affected by the electron beam
was determined using Monte-Carlo-based simulation
software Win X-ray [57] and found to be around
2lm. The SEM/EDS spectra were used to determine
the information related to the atomic fractions of
individual elements (i.e., C, Ca, Si, and O). Matrix
corrections were made using ZAF method. For SEM/
EDS data collection, first the carbonated samples
were impregnated with epoxy and then lapped/
polished to obtain a mirror like surface finish.
Results and discussions
Comparison of hydration and carbonation
behavior of calcium silicates
C
3
S, c-C
2
S, and CS samples were subjected to expo-
sure scenario-1 and scenario-2 to compare their
hydration and carbonation behavior. The relative
proportions of the reaction products formed during
these exposure conditions were determined using
TGA. The amount of chemically bound water can be
used to determine the amount of C–S–H for known
stoichiometric condition. However, the exact stoi-
chiometry of the C–S–H (or Ca-modified silica gel)
that forms during the carbonation reaction of calcium
silicates is yet to be confirmed, and hence, only the
relative amounts of the chemically bound water are
reported in this paper.
During the hydration reaction, calcium silicates
form C–S–H and Ca(OH)
2
as the reaction products.
On the other hand, during the carbonation reaction of
calcium silicates, the primary reaction products are
C–S–H and CaCO
3
[27]. Thus, to compare these two
different reaction systems, the total amounts of CaO
(address as CaO
total
for the remainder) present in
Ca(OH)
2
and CaCO
3
were calculated for scenario-1
and scenario-2 using TGA results.
Figure 5represents the relative amounts of CaO
total
and chemically bound water present in the hydrated
(scenario-1) and carbonated (scenario-2) calcium sili-
cate (C
3
S, c-C
2
S, and CS) samples. For all of the calcium
silicate phases, the amounts of CaO
total
(Fig. 5a, b)
present in the carbonated samples were higher than
that of the hydrated samples. Moreover, for c-C
2
S and
CS samples, the variations in the amounts of CaO
total
during the hydration reactions were negligible which
confirmed the non-hydraulic nature of these calcium
silicate phases. The relative amounts of chemically
bound water in carbonated c-C
2
S and CS samples were
also higher than that of the hydrated samples (Fig. 5c,
d). These findings suggest that higher fractions of C
3
S,
c-C
2
S, and CS phases have reacted during the car-
bonation reaction than that of the hydration reaction. It
is noteworthy that both the carbonated disks of C
3
S
and c-C
2
S were found to contain Ca(OH)
2
; this was
expected due to the non-uniformity of the carbonation
Figure 4 TGA curves of C
3
S paste exposed to scenario-2 for
48 h.
J Mater Sci (2016) 51:6173–6191 6177
within the disk samples. But in the case of CS phase, no
Ca(OH)
2
was found to form either in hydration or
carbonation reaction.
The amounts of chemically bound water present in
the carbonated C
3
S samples were lower than that of
the hydrated C
3
S samples (Fig. 5c, d). This difference
in the amounts of chemically bound water can be
resulted from the different degree of polymerization
of C–S–H phases formed during hydration and car-
bonation reactions of C
3
S. The amounts of CaO
total
present in carbonated c-C
2
S samples were consis-
tently higher than that of the carbonated C
3
S samples
(Fig. 5b). This observation was unexpected consid-
ering the high reactivity of the C
3
S phase. However,
this might have resulted due to the different levels of
compaction and porosities of the samples resulting in
non-uniform carbonation reaction of the c-C
2
S and
C
3
S disks. Thus, for a better comparison of the reac-
tivity, next section (‘‘Comparison of the reactivity of
calcium silicates during carbonation reaction’’ sec-
tion) of this paper will present the carbonation
behavior of the calcium silicate paste samples pre-
pared without any compaction (i.e., scenario-3 and
scenario-4 of Table 1).
Comparison of the reactivity of calcium
silicates during carbonation reaction
From TGA results, Ca(OH)
2
was absent in all of the
samples carbonated according to the exposure sce-
nario-3 and 4. Hence, in these cases, CaCO
3
and cal-
cium-silica gel were the primary carbonation reaction
products of calcium silicate phases. The relative
amounts of CaCO
3
formed during the carbonation
reactions were used to calculate the normalized
degree of carbonation of the samples from the Eq. 1.
Normalized degree of carbonation DOCðÞat time t
¼CaCO3content mass;percentage
at time t
MaximumCaCO3content mass;percentage
100 %
ð1Þ
Figure 5 Relative mass fractions with exposure durations; a%
of CaO
total
after exposure to scenario-1, b% of CaO
total
after
exposure to scenario-2, camounts of chemically bound water
present in the samples after exposure to scenario-1, and damounts
of chemically bound water present in the samples after exposure to
scenario-2.
6178 J Mater Sci (2016) 51:6173–6191
Here, the maximum amounts of CaCO
3
that
formed during the carbonation reaction of calcium
silicates were determined experimentally. For this
purpose, the powder calcium silicate samples were
first mixed with water (w/c ratio =0.40) and car-
bonated at 100 % CO
2
(same as scenario-4) as long as
it took to achieve the plateau (i.e., increase in the
amount of CaCO
3
\2.00 %) of the percentage of
CaCO
3
formed. The maximum amounts of CaCO
3
formed from different calcium silicates are given in
Table 2. These maximum amounts of CaCO
3
are
expected to be dependent on both the particle size
distributions and reactivity of the calcium silicate
phases. D
50
(median particle size) and specific surface
area of the calcium silicate phases as obtained from
the laser particle size distribution are also given in
Table 2.
The DOC of pure calcium silicate phases during
the scenario-3 and scenario-4 are given in Fig. 6. Due
to the limited amount of available C
3
S
2
sample, it was
exposed to only carbonation scenario-3. From Fig. 6a,
it can be observed that at 15 % CO
2
concentration, a
reaction rate for b-C
2
S similar to that of C
3
S phase
was achievable. The reaction rate for c-C
2
S was
slower initially; however, after 82 h of reaction, it also
reached the 100 % DOC. But at this exposure condi-
tion, C
3
S
2
and CS did not reach the maximum pos-
sible degree of carbonation within the test duration
(82 h).
During the exposure scenario-4 (Fig. 6b), the car-
bonation reaction rates of all the calcium silicates
were increased. At this exposure condition, the
reactivity of c-C
2
S was similar to that of b-C
2
S and
C
3
S phases as the graphs for all of these phases are
overlapped (Fig. 6b). The reactivity of CS phase was
also found to be substantially improved as it
achieved 100 % DOC after 82 h of exposure duration.
Figure 6 Normalized degree of carbonations (DOC) of calcium
silicate phases with exposure durations; acarbonation according to
exposure scenario-3 and bcarbonation according to exposure
scenario-4.
Table 2 Maximum amounts of CaCO
3
formed and CO
2
stored during the carbonation reactions of calcium silicates
Calcium
silicate phase
Median particle
size, D
50
(lm)
Specific surface
area (m
2
/g)
Maximum amounts (weight
percentage) of CaCO
3
formed
Maximum amounts (weight
percentage) of CO
2
stored
C
3
S 35.311 0.37 44.41 19.54
b-C
2
S 9.676 1.14 66.07 29.07
c-C
2
S 15.264 1.01 46.07 20.27
C
3
S
2
10.113 1.44 45.93 20.21
CS 5.212 1.83 43.71 19.23
J Mater Sci (2016) 51:6173–6191 6179
Considering the variation of DOC with exposure
duration for scenario-3 and scenario-4, the carbona-
tion reactions of all calcium silicates can be consid-
ered as to be composed of two distinct stages. For all
of the calcium silicates, the first stage of carbonation
reaction was found to have a high reaction rate
(identified as (1) in Fig. 6a, b). The rapid carbonation
rate at this stage is attributed to the rapid dissolution
of calcium silicate phases. In the second stage (iden-
tified as (2) in Fig. 6a, b), the reaction rate became
slower than that observed during the first segment of
reaction. Considering the sudden slow rate, it is
suggested that the rate of carbonation reaction at this
segment is governed by the diffusion of ions through
the solid layers of reaction products formed during
the first stage of the reaction.
FITR spectroscopic observations
The FTIR spectra were utilized to investigate the
structure of the calcium-silica gel formed during the
carbonation reaction of the calcium silicate phases.
The FTIR spectra for silicate compounds usually
exhibit a large absorption between 800 and 1200 cm
-1
which correspond to the asymmetrical stretching
vibration (m
3
) of Si–O bond [31,49]. The absorption
bands of calcium silicates at 800 cm
-1
or below cor-
respond to the out-of-plane skeletal (m
4
) and in-plane
skeletal (m
2
)[58,59] vibrations of Si–O bond. In this
study, the FTIR spectra were collected in the range of
800–4000 cm
-1
, but the main focus was given on
800–1800 cm
-1
region to understand the structure of
the calcium-silica gel formed during the carbonation
reaction of calcium silicate phases. Thus, calcium-
silica gels were characterized based on the asym-
metrical stretching vibration (m
3
) of Si–O bond. The
exact and accurate structural analysis of calcium-sil-
ica gels based on the m
3
vibration of FTIR spectra can
be complicated as m
3
causes a broad absorption band
resulting from the overlapping of similar absorptions.
However, the m
3
band shifts to higher wave number
with increasing degree of polymerization of the sili-
cate compound due to the increase of the bond
strength of Si–O [58]. Thus, the exact positions of
these bands are dependent on the calcium to silica
(Ca/Si) atomic ratio of the calcium-silica gel phase as
this ratio also represent the degree of silicate poly-
merization of the C–S–H [60].
Pure calcium silicate phases
Figure 7shows the FTIR spectra of the pure calcium
silicate phases that were used in this study. C
3
S, b-C
2
S,
and c-C
2
S phases were found to contain major sharp
absorption band at around 846 cm
-1
. In addition to this
band, C
3
S phase contained a shoulder at around
980 cm
-1
.Inthecaseofb-C
2
Sandc-C
2
S phases, a sec-
ond sharp absorption band was located at around
925 cm
-1
. The shoulder at around 970 cm
-1
was pre-
sent in b-C
2
S but absent in c-C
2
S phase. The similarities
in the FTIR spectraof C
3
S, b-C
2
S, and c-C
2
Sphaseswas
expected as all of these calcium silicate phases were
primarily composed of isolated silicate tetrahedrons
(i.e., orthosilicate group). C
3
S
2
sample had four major
absorptions bands; locations of these bands were at
around 846, 906, 940, and 980 cm
-1
. The absorption
bands for pure CS sample were located at around 896,
963, 1010, and 1058 cm
-1
. Presence of major absorption
Figure 7 FTIR spectra of pure calcium silicate phases.
6180 J Mater Sci (2016) 51:6173–6191
bands at higher wavenumbers for both CS and C
3
S
2
samples are due to the fact that the polymerizations of
the silicate tetrahedrons in these phases are higher in
comparison to that of the C
3
SandC
2
S samples. The pure
C
3
S
2
sample is composed of dimer silicate tetrahedrons
(i.e., sorosilicates group); that is, one oxygen atom is
shared between two neighboring tetrahedra (Q1 species
in
29
Si NMR spectra [61]). Whereas, CS phase is com-
posed of bridging silicate tetrahedrons (i.e., single chain
inosilicate group); that is, one silica tetrahedron shares
two oxygen atoms with two other silicate tetrahedrons
(Q2 species in
29
Si NMR spectra [61]).
Hydrated calcium silicates
As observed from TGA results, the low-lime calcium
silicates (c-C
2
S and CS) did not exhibit any reactivity
to the pure hydration scenario (Fig. 5) within the
specified experimental duration (82 h). Thus, these
calcium silicate samples were not examined using
FTIR after the hydration reaction. However, C
3
S
phase is known for its hydraulic reactivity and this
was also evident from the changes in the FTIR spectra
(Fig. 8) of this phase collected after the hydration
reaction. Due to the hydration reaction of C
3
S sam-
ples, new bands were formed at around 3640 and
950 cm
-1
. The band at 3640 cm
-1
(not shown in fig-
ure) is caused by the stretching vibration of the
hydroxyl ion present in the Ca(OH)
2
[60,62–65]. The
band at around 950 cm
-1
is attributed to the m
3
vibration of the Si–O bond present in C–S–H. One
additional band can be observed in the FTIR spectra
of hydrated C
3
S samples (Fig. 8), in the range of
1300–1500 cm
-1
. This band indicates the presence of
CaCO
3
[60,65] in hydrated C
3
S samples which might
have formed due to the presence of atmospheric CO
2
.
Carbonated calcium silicates
The FTIR spectra of carbonated calcium silicate
samples are given in Figs. 9,10,11,12, and 13. Due to
the carbonation of calcium silicate samples, new
bands were observed to appear at around 875 cm
-1
,
1410 cm
-1
, and in the region of 940–1200 cm
-1
. The
band located at around 1410 cm
-1
is due to the
asymmetric stretching (m
3
) of C–O bond present in
CaCO
3
and the band located at around 875 cm
-1
corresponds to the out-of-plane bending vibration
(m
2
) of the same C–O bond [60,65].
For carbonated C
3
S samples, the positions of the m
3
vibration of Si–O bonds were observed to gradually
shift to the higher wavenumber with increasing degree
of carbonation (Fig. 9). In this case, after 0.5 h of car-
bonation, the Si–O bond appeared at around 954 cm
-1
which is same as for the hydrated C
3
Ssamples.After
this, the band height at 954 cm
-1
was observed to
decrease with increasing heights of bands in the range
of 1030–1180 cm
-1
. These successive increases in the
wavenumbers are due to the increasing polymerization
of the calcium-silica gel phase resulting from the car-
bonation-induced decalcification process. Thus, after
the initial carbonation reaction, C
3
S samples formed C–
Figure 8 FTIR spectra of hydrated C
3
S paste samples (exposure
scenario-1).
Figure 9 FTIR spectra of carbonated C
3
S samples with different
degree of carbonations.
J Mater Sci (2016) 51:6173–6191 6181
S–H phase with similar silicate polymerization as that of
the C–S–H formed during the hydration reaction of C
3
S.
With the additional carbonation, this C–S–H phase was
subjected to gradual decalcification processes which
lead to the formation of more polymerized Ca-modified
silica gel phase. Similar observations were made for
carbonated b-C
2
S, c-C
2
S, C
3
S
2
, and CS samples. For
these carbonated calcium silicate samples, the m
3
vibra-
tions of Si–O bonds present in calcium-silica gel phases
were identified at around 1070 and 1180 cm
-1
.
Polymerization of calcium-silica gel phase
The FTIR spectra of carbonated calcium silicate
samples were deconvoluted to obtain the accurate
wavenumber corresponding to the m
3
vibration of Si–
O bond. Deconvolution was performed using a sta-
tistical software (OriginPro
M
[66]). Then, the mean
wavenumber for the Si–O bond in calcium-silica gel
phase was determined using the Eq. 2for the band
range of 900–1200 cm
-1
.
Figure 10 FTIR spectra of carbonated b-C
2
S samples with
different degree of carbonations.
Figure 11 FTIR spectra of carbonated c-C
2
S samples with
different degree of carbonations.
Figure 12 FTIR spectra of carbonated C
3
S
2
samples with differ-
ent degree of carbonations.
Figure 13 FTIR spectra of carbonated CS samples with different
degree of carbonations.
6182 J Mater Sci (2016) 51:6173–6191
Im¼ðIiAiþINANÞ
ðAiþANÞð2Þ
Here, I
m
is the mean wavenumber of the asymmetric
stretching vibration (m
3
) of Si–O bond.
I
i
…I
N
are the wavenumber corresponding to the
bands i…N.
A
i
…A
N
represent the integrated area under the
bands at wavenumbers I
i
…I
N
.
The absorption bands below 900 cm
-1
wavenum-
ber were not considered in the calculation of Eq. 2to
avoid the effect of C-O bond which has an absorption
band at around 870 cm
-1
. Moreover, the m
3
absorp-
tion bands of Si–O bond in the range of 800–900 cm
-1
are primarily associated with the anhydrous part of
the calcium silicates phases. The anhydrous calcium
silicates can also have absorption bands over
900 cm
-1
wavenumber depending on their structure
(Fig. 7). However, with gradual progress of the
reaction, the influence of unreacted grains on the
mean wave number of the Si–O bond present in cal-
cium-silica gel phase will be lower. Aragonite,
vaterite, and amorphous forms of CaCO
3
also have
narrow absorption at around 1080 cm
-1
[67]. Con-
sidering the smaller amount of aragonite/vaterite
phases present in the carbonated calcium silicate
samples and the narrow band width, contribution of
this absorption band on the integrated area calcula-
tion (Eq. 2) should be negligible.
The deconvolution process was performed on all of
the FTIR spectra collected for the carbonated calcium
silicates samples. As examples, the deconvoluted FTIR
spectra of carbonated calcium silicate samples with
more than 80 % DOC are given in Fig. 14. For all of the
calcium silicate samples, the calcium-silica gel phases
formed after 80 % DOC contained three major m
3
absorption bands of Si–O bonds. For carbonated C
3
S
samples, the wavenumbers corresponding to these
bands were 960, 1030, and 1175 cm
-1
; where absor-
bance of the 1030 cm
-1
band was the highest (Fig. 14a).
For other carbonated calcium silicate (i.e., b-C
2
S, c-C
2
S,
C
3
S
2
, and CS) samples, these absorption bands were at
around 960, 1060, and 1180 cm
-1
; where absorbance of
the 1060 cm
-1
band showed the highest absorbance in
this case (Fig. 14b, c, d, e).
Figure 15 shows the variation of the mean
wavenumbers corresponding to the m
3
vibration of Si–O
bond in calcium-silica gel with different DOC. From this
figure, it can be observed that the mean wavenumbers
are grouped within two different ranges along the
vertical axis. The first range of wavenumbers is
920–960 cm
-1
and this range can be attributed to the C–
S–H-like (in terms of silicate polymerization) phase. The
m
3
absorption band of the Si–O bond (i.e., 954 cm
-1
)inC–
S–H formed during the hydration reaction of C
3
Swas
also within this range. However, this range is lower than
the reported wavenumbers of m
3
vibration of C–S–H as in
[60]. The mean wavenumbers reported in [60]were
obtained for pure synthesized C–S–H samples. In com-
parison, carbonated samples in this study also contained
variable amounts (depending on the degree of reaction)
of unreacted calcium silicate grains which have lower
wavenumber than the C–S–H causing a decrease in the
mean wavenumber of Si–O bond. The second range of
the frequency observed in Fig. 15 is from 1030 to
1170 cm
-1
which is attributed to the Ca-modified silica
gel [68,69]phase.FromFig.15, the polymerization of
calcium-silica gel phase can also be correlated with the
DOC of the calcium silicate phases. It can be seen that CS
phase all through formed Ca-modified silica gel as the
carbonation reaction product irrespective of the amount
of calcium carbonate (i.e., DOC). Contradictorily, C
3
S, b-
C
2
S, c-C
2
S, and C
3
S
2
initially produced C–S–H during
the carbonation reaction with similar degree of poly-
merization compared to that of the C–S–H forms during
the hydration reaction of C
3
S. After a certain degree of
carbonation, this C–S–H decomposed to form Ca-modi-
fied silica gel. For these calcium silicate phases, the
decalcification of the C–S–H phase was found to occur at
around 50–60 % of DOC (around 20–30 % CaCO
3
for-
mation). The exact level of carbonation reaction at which
this decalcification of C–S–H occurs is expected to be
dependent on the initial particle size distribution (hence,
the specific surface area) of the starting calcium silicate
phases.
Microscopic observations
Morphological observations
Backscattered electron (BSE) and secondary electron
(SE) SEM images were collected from carbonated
calcium silicates samples (Figs. 16,17,18,19,20, and
21). During the carbonation reaction, C
3
S phase ini-
tially formed ‘foil’ like C–S–H which upon further
carbonation reaction changes to a ‘sheet’ or ‘laminar’-
like morphology (Fig. 16). Additionally, at low DOC,
the CaCO
3
crystals appeared to be intermingled with
the foil-like C–S–H phase (Fig. 17a). Needle-like C–S–
H phase was observed to form on the surface of c-C
2
S
J Mater Sci (2016) 51:6173–6191 6183
800 1000 1200 1400 1600 1800
-0.05
0.00
0.05
0.10
0.15
0.20
0.25
0.30
Absrobance (arbitrary unit)
800 1000 1200 1400 1600 1800
-0.05
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
Absorbance (arbitrary unit)
800 1000 1200 1400 1600 1800
0.00
0.04
0.08
0.12
0.16
0.20
0.24
0.28
0.32
Absorbance (arbitrary unit)
Wavenumber (cm-1) Wavenumber (cm-1)
Wavenumber (cm-1)
Wavenumber (cm-1) Wavenumber (cm-1)
800 1000 1200 1400 1600 1800
0.00
0.05
0.10
0.15
0.20
0.25
Absroption (arbitrary unit)
800 1000 1200 1400 1600 1800
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
Absorbance (arbitrary unit)
(a)
Carbonated C
3
S sample, DOC ≈ 100%.
(b)
Carbonatedβ-C
2
Ssample,DOC≈100%.
(c)
Carbonatedγ-C
2
Ssample,DOC≈90%.
(d)
CarbonatedC
3
S
2
sample, DOC ≈ 85%.
(e)
Carbonated CS sa mple, DOC ≈ 100 %.
Figure 14 Deconvolution of FTIR spectra of acarbonated C
3
S
samples after 100 % DOC, bcarbonated b-C
2
S samples after
100 % DOC, ccarbonated c-C
2
S samples after 90 % DOC,
dcarbonate C
3
S
2
samples after 85 % DOC, and eCS samples
after 100 % DOC. Here, line with symbol,solid line, and dashed
line represent experimental FTIR spectra, simulated spectra, and
deconvoluted absorbance bands, respectively.
6184 J Mater Sci (2016) 51:6173–6191
samples after around 40 % DOC (Fig. 17b). After
90 % DOC, for all types of carbonated calcium silicate
samples, the morphology of the Ca-modified silica
gel phases were ‘sheet’ or ‘laminar’ like. Presence of
both aragonite and calcite forms of CaCO
3
was
identified in carbonated C
3
S, b-C
2
S, and c-C
2
S sam-
ples. In contrast, based on the SE images, calcite
appeared to be the primary polymorph of the CaCO
3
for both carbonated C
3
S
2
and CS samples formed
during the carbonation reaction.
All of the carbonated calcium silicate systems
appeared to be very similar after 90 % DOC (BSE
images in Figs. 18b, 19b, 20b, 21b). In all of the BSE
images, four different regions can be identified based
on the gray levels; (i) pores (marked as ‘P’ on the
image) presented by the darkest area, (ii) Ca-modi-
fied silica gel phase (marked as ‘S’ on the image)
appeared as the second darkest regions, (iii) CaCO
3
(marked as ‘C’ on the image) represented by com-
paratively brighter region than that of the calcium-
silica gel, and (iv) the unreacted part of calcium sili-
cates grains (marked as ‘U’ on the image) appeared as
the brightest areas. Ca-modified silica gel phases
were observed to form as a rim around the unreacted
(or partially reacted) calcium silicate grains with a
characteristic dimension of around 1–3 lm. CaCO
3
appears to form in interparticle spaces without any
specific characteristic dimension. However, from SE
image observations, the size of the calcite crystals
appeared to be around a micron.
X-ray microanalysis
The purpose of the X-ray microanalysis was to
determine the Ca/Si atomic ratio of the Ca-modified
silica gel that formed during the carbonation reaction
of the pure calcium silicate phases. For all of the
samples, EDS were collected from at least 50 locations
on the Ca-modified silica gel phases which were then
used to calculate the average Ca/Si atomic ratio.
Samples only with more than 90 % DOC obtained
from the carbonation exposure scenario-4 (to ensure
Figure 15 Mean wavenumber of Si–O stretching band with
degree of carbonation for different calcium silicates phases.
Legends include the temperature and the CO
2
concentration used
for the carbonation.
Figure 16 SEM images of carbonated C
3
S samples showing the morphology of the calcium-silica gel; aafter 22 % of DOC and bafter
98 % of DOC. ‘S’ indicates Ca-modified silica gel, ‘U’ indicates unreacted/partially reacted C
3
S.
J Mater Sci (2016) 51:6173–6191 6185
Figure 17 Morphological observations; aintrusion of CaCO
3
crystals in C–S–H foils for carbonated C
3
S samples (around 40 % DOC),
and bformation of needle-like C–S–H from c-C
2
S surface after 24 h of carbonation (around 48 % DOC).
Figure 18 SEM images of carbonated b-C
2
S samples showing
the microscopic phases; aSE image and bBSE image (collected at
15 kV with 10 mm working distance). ‘S’ indicates Ca-modified
silica gel, ‘U’ indicates unreacted/partially reacted b-C
2
S, ‘C’
indicates CaCO
3
, and ‘P’ indicates pores.
Figure 19 SEM images of carbonated c-C
2
S samples showing the
microscopic phases; aSE image and bBSE image (collected at
15 kV with 10 mm working distance). ‘S’ indicates Ca-modified
silica gel, ‘U’ indicates unreacted/partially reacted c-C
2
S, ‘C’
indicates CaCO
3
, and ‘P’ indicates pores.
6186 J Mater Sci (2016) 51:6173–6191
uniform carbonation throughout the samples) were
subjected to this X-ray microanalysis investigation.
Figure 22 presents the average Ca/Si atomic ratio
of the Ca-modified silica gel that formed during the
carbonation reaction of the pure calcium silicate
phases. For all of the calcium silicate phases after
90 % DOC, the average Ca/Si atomic ratios of the Ca-
modified silica gel phases were in the range of
0.40–0.70. Formation of this phase during the car-
bonation of OPC concrete and C
3
S pastes has also
been reported in some other researches [40,70]. The
relatively higher Ca/Si atomic ratio of the Ca-modi-
fied silica gel (Fig. 22) present in carbonated C
3
S
samples in comparison to that of other calcium sili-
cate phases is also consistent with the observation
made from FTIR spectra (the highest absorbance for
the Ca-modified silica gel present in carbonated C
3
S
Figure 20 SEM images of carbonated C
3
S
2
samples showing the microscopic phases; aSE image and bBSE image. ‘S’ indicates Ca-
modified silica gel, ‘U’ indicates unreacted/partially reacted C
3
S
2
grains, ‘C’ indicates CaCO
3
, and ‘P’ indicates pores.
Figure 21 SEM images of carbonated CS samples showing the microscopic phases; aSE image and bBSE image. ‘S’ indicates Ca-
modified silica gel, ‘U’ indicates unreacted/partially reacted CS grains ‘C’ indicates CaCO
3
, and ‘P’ indicates pores.
Figure 22 Average Ca/Si atomic ratios of Ca-modified silica gel
phase formed during the carbonation reaction of the calcium
silicate samples.
J Mater Sci (2016) 51:6173–6191 6187
sample was at around 1030 cm
-1
, whereas for other
calcium silicates, it was around at 1070 cm
-1
).
Stoichiometry of the Ca-modified silica gel
The stoichiometries of the Ca-modified silica gel
formed during the carbonation reactions of hydraulic
and non-hydraulic calcium silicates were determined
using the Ca/Si atomic ratio of this phase and TGA
results. For each carbonated calcium silicate samples,
the average ratio of CaCO
3
and chemically bound
water was determined from around 10 to 15 samples
using the TGA results. These values were then used
in the generalized form of equation (Eq. 3) to deter-
mine the stoichiometry of the Ca-modified silica gel
phase. To ensure the uniformity of carbonation
within the samples, only the samples from scenario-3
and 4 were used in this calculation. Finally, the bal-
anced carbonation equations for C
3
S, b-C
2
S, c-C
2
S,
C
3
S
2
, and CS phases were determined as shown in
Eqs. 4–7.
xCaO SiO2þxyðÞCO2þnH2O
!xyðÞCaCO3þyCaO SiO2nH2O:ð3Þ
Tricalcium silicate (3CaOSiO
2
):
10 3CaO SiO2
ðÞþ23CO2þ6H2O
!23CaCO3þ7CaO 10SiO26H2O:ð4Þ
Dicalcium silicate (b-2CaOSiO
2
and c-2CaOSiO
2
):
10 2CaO SiO2
ðÞþ16CO2þ3:5H2O
!16CaCO3þ4CaO 10SiO23:5H2O:ð5Þ
Rankinite (3CaO. 2SiO
2
):
10 3CaO 2SiO2
ðÞþ11CO2þ2:5H2O
!11CaCO3þ4CaO 10SiO22:5H2O:ð6Þ
Wollastonite (CaOSiO
2
):
10 CaO SiO2
ðÞþ6CO2þ1:3H2O
!6CaCO3þ4CaO 10SiO21:3H2O:ð7Þ
It is important to note that these equations were
formulated considering only the initial and final
products of carbonation reaction. As observed from
FTIR study, there were intermediate stages involved in
carbonation reactions of C
3
S, b-C
2
S, c-C
2
S, and C
3
S
2
where C–S–H were formed. These intermediate stages
were not included in Eqs. 4–7because of the unsta-
ble stoichiometry of the C–S–H phase. The amounts of
chemically bound water in the Ca-modified silica gel
were found to vary slightly depending on the starting
calcium silicate phases (Eqs. 4–7). Additionally, based
on the balanced carbonation reactions, after same
molar reaction of the calcium silicate phases, C
3
S pro-
duces the highest amount of CaCO
3
(also stores the
highest amount of CO
2
) and CS produces the lowest
amount of CaCO
3
. However, for similar particle size
distribution, carbonation reaction of both C
3
S and CS
samples were observed to cease after nearly same
amount of CaCO
3
formation (Table 2). Thus, for
specific particle size distribution and after the same
extent of carbonation reaction (in terms of CaCO
3
content), a higher proportion of unreacted grains is
likely to exist in case of carbonated C
3
S samples than
that of the CS samples. This can be a noteworthy issue
for optimizing the reaction process of the carbonation-
based cementitious binders.
Conclusions
This paper presented a study on the reactivity of pure
calcium silicate phases and their reaction products
during the carbonation reaction. In summary, fol-
lowing observations can be made from this study:
1. Based on the TGA results, the amounts of CaO
total
(as in Ca(OH)
2
and CaCO
3
) formed during the
carbonation reactions of C
3
S, c-C
2
S, and CS
phases were found to be higher than that gener-
ated during the hydration reaction. This finding
suggests that the carbonation reaction for the
tested exposure scenario (i.e., 82 h in 99.9 % pure
CO
2
purging environment) was more effective
than the hydration reaction, even for nominally
highly hydraulic material (i.e., C
3
S).
2. C
3
S, c-C
2
S, and b-C
2
S samples were almost
equally reactive to carbonation at both, 35 and
55 C temperatures. In comparison, the reactivity
of C
3
S
2
and CS samples were lower than that of
the C
3
S and C
2
S samples.
3. The maximum amounts of CaCO
3
formed during
the carbonation reaction of calcium silicates were
in the range of 43–66 % (given in Table 2). These
values did not increase upon further exposure to
CO
2
containing environment. It indicates that
further carbonation reaction was not possible (or
largely hindered) due to the formation of CaCO
3
and Ca-modified silica gel phases. Thus, for
similar particle size distribution, the maximum
amounts of CO
2
that can be stored in these
calcium silicate phases are nearly the same,
6188 J Mater Sci (2016) 51:6173–6191
though they have a different percentage of CaO
content. Nonetheless, it is expected that these
values of maximum amounts of CO
2
storage will
change depending on the particle size distribu-
tions of the calcium silicate phases.
4. Based on the reaction rate, the carbonation
reactions of calcium silicate phases were
observed to occur in two different stages. In the
first stage, the carbonation reaction was rapid and
nominally controlled by the dissolution rate of
calcium silicate phases. In the second stage, the
reaction rate was slower and expected to be
dependent on the diffusion of ions through the
already formed reaction products.
5. FTIR spectra of the carbonated samples revealed
that after initial stage of carbonation reaction,
C
3
S, b-C
2
S, c-C
2
S, and C
3
S
2
samples produced C–
S–H with a similar degree of silicate polymeriza-
tion as that of the C–S–H formed during the
hydration reaction of C
3
S. Upon further carbon-
ation reaction, this C–S–H phase was decom-
posed and Ca-modified silica gel was formed
with a higher degree of silicate polymerization.
6. In the case of the carbonation reaction of CS, Ca-
modified silica gel was formed even at the early
stage of carbonation; no significant variation in
the polymerization of this phase due to the
increased DOC could be observed.
7. From the microstructure of carbonated C
3
S sam-
ples, it was observed that the C–S–H at low
degree of carbonation had either foil or needle-
like morphology which gradually changed to
laminar morphology with increasing DOC. The
Ca-modified silica gel had laminar-like structure
irrespective of the types of starting calcium
silicate phases.
8. Carbonated C
3
S, b-C
2
S, and c-C
2
S samples con-
tained both aragonite and calcite forms of CaCO
3
,
whereas, for both carbonated C
3
S
2
and CS samples,
the primary morphology of the CaCO
3
phase was
calcite.
9. The stoichiometries of the Ca-modified silica gel
phases were determined from the SEM/EDS and
TGA results. Although the Ca/Si atomic ratios of
these phases were similar for all starting materi-
als, the amounts of chemically bond water varied
slightly. Thus, the exact arrangements of the
silicate tetrahedrons present in the Ca-modified
silica gel phases may not be the same, even
though the mean wavenumber of Si–O bond (m
3
vibration) for these phases were in the same
range (1030–1070 cm
-1
) for all of the calcium
silicate samples.
The observations made from this study open up
the opportunity to utilize the low-lime non-hydraulic
calcium silicates as a partial replacement of conven-
tional cement or as a sole binder. Nonetheless, as a
developing technique, several aspects of this ‘car-
bonation activated binder system’ are yet to be
studied, which include the kinetics of the carbonation
reaction of calcium silicate phases, mechanical
properties, and long-term durability performances.
Acknowledgements
Solidia Technologies Inc., NJ, USA is gratefully
acknowledged for providing the financial support
required for this study.
Compliance with ethical standards
Conflict of Interest The authors declare that they
have no conflict of interest.
References
[1] Fukuda K, Ito S (1999) Improvement in reactivity and
grindability of belite-rich cement by remelting reaction.
J Am Ceram Soc 82:2177–2180
[2] Engel’sht VS, Muratalieva VZ (2013) Thermal interaction
between limestone and silica. High Temp 51:769–775
[3] Kaifi FMZ, Aurangzeb M, Ahmed B, Khan M (2004)
Manufacture of synthetic wollastonite. J Chem Soc 26:1–4
[4] Kotsis L, Balogh A (1989) Synthesis of wollastonite. Ceram
Int 15:79–85
[5] Treiman AH, Essene EJ (1983) Phase equilibria in the Sys-
tem CaO-SiO2-CO2. Am J Sci 283:97–120
[6] Fukuda K, Wakamatsu N, Ito S (2001) Improvement in
hydration reactivity of alpha-phase belite by remelting
reaction. J Am Ceram Soc 84:639–641
[7] Kurdowski W, Duszak S, Trybalska B (1997) Belite pro-
duced by means of low-temperature synthesis. Cem Concr
Res 27:51–62
[8] Kacimi L, Simon-Masseron A, Salem S, Ghomari A, Der-
riche Z (2009) Synthesis of belite cement clinker of high
hydraulic reactivity. Cem Concr Res 39:559–565
[9] Stane
ˇk T, Sulovsky
´P (2015) Active low-energy belite
cement. Cem Concr Res 68:203–210
J Mater Sci (2016) 51:6173–6191 6189
[10] Ziemer B, Altrichter B, Jesenak V (1984) Effect of SO
3
on
formation and hydraulic reactivity of belite. Cem Concr Res
14:686–692
[11] Ghosh SN, Rao PB, Paul AK, Raina K (1979) The chemistry
of dicalcium silicate mineral. J Mater Sci 14:1554–1566
[12] Cuesta A, Losilla ER, Aranda MAG, Sanz J, DeLaTorre AG
(2012) Reactive belite stabilization mechanisms by boron-
bearing dopants. Cem Concr Res 42:598–606
[13] Jelenic I, Bezjak A, Bujan M (1978) Hydration of B
2
O
3
stabilized a– and b– modifications of dicalcium silicate.
Cem Concr Compos 8:173–180
[14] Popescu CD, Muntean M, Sharp JH (2003) Industrial trial
production of low energy belite cement. Cem Concr Compos
25:689–693
[15] Song F, Yu Z, Yang F, Lu Y, Liu Y (2015) Microstructure of
amorphous aluminum hydroxide in belite-calcium sulfoalu-
minate cement. Cem Concr Res 71:1–6
[16] Quillin K (2001) Performance of belite-sulfoaluminate
cements. Cem Concr Res 31:1341–1349
[17] Martin-Sedeno MC, Cuberos AJM, DeLaTorre AG, A
´lvarez-
Pinazo G et al (2010) Aluminum-rich belite sulfoaluminate
cements: clinkering and early age hydration. Cem Concr Res
44:6855–6862
[18] Chen IA, Juenger MCG (2011) Synthesis and hydration of
calcium sulfoaluminate-belite cements with varied phase
compositions. J Mater Sci 46:2568–2577. doi:10.1007/
s10853-010-5109-9
[19] Trauchessec R, Mechling JM, Lecomte A, Roux A, Rolland BL
(2015) Hydration of ordinary Portland cement and calcium
sulfoaluminate cement blends. Cem Concr Com pos56: 106–114
[20] Chatterjee AK (1996) High belite cements - present status
and future technological options: Part I. Cem Concr Res
26:1213–1225
[21] Soliman AM, Nehdi ML (2014) Effects of shrinkage
reducing admixture and wollastonite microfiber on early-age
behavior of ultra-high performance concrete. Cem Concr
Compos 46:81–89
[22] Low NMP, Beaudoin JJ (1994) Stability of Portland cement-
based binders reinforced with natural wollastonite micro-fi-
bres intrgduetion. Cem Concr Res 24:874–884
[23] Low NMP, Beaudoin JJ (1993) Flexural strength and
microstructure of cement binders reinforced with wollas-
tonite micro-fibres. Cem Concr Res 23:905–916
[24] Low NMP, Beaudoin JJ (1994) The flexural toughness and
ductility of portland cement-based binders reinforced with
wollastonite micro-fibres. Cem Concr Res 24:250–258
[25] Low NMP, Beaudoin JJ (1994) Mechanical properties and
microstructure of high alumina cement-based binders rein-
forced with natural wollastonite micro-fibres. Cem Concr
Res 24:650–660
[26] Berger RL, Young JF, Leung K (1972) Acceleration of
hydration of calcium silicates by carbon dioxide treatment.
Nat Phys Sci 240:16–18
[27] Goodbrake CJ, Young JF, Berger RL (1979) Reaction of
hydraulic calcium silicates with carbon dioxide and water.
J Am Ceram Soc 62:488–491
[28] Bukowski JM, Berger RL (1979) Reactivity and strength
development of CO
2
activated non-hydraulic calcium sili-
cates. Cem Concr Res 9:57–68
[29] Berger RL, Klemm WA (1972) Accelerated curing of
cementitious systems by carbon dioxide part-II: hydraulic
calcium silicates and aluminates. Cem Concr Res 2:647–652
[30] Young JF, Berger RL, Breese J (1974) Accelerated curing of
compacted calcium silicate mortars on exposure to CO
2
.
J Am Ceram Soc 57:394–397
[31] Riman RE, Gupta S, Atakan V, Li Q (2013)
US20130122267 A1: Bonding element, bonding matrix and
composite material having the bonding element, and method
of manufacturing thereof. US Patent
[32] Sahu S, DeCristofaro N (2013) Solidia Cement TM. Solidia
Technol, White Paper
[33] Ashraf W, Olek J, Tian N (2015) Nanomechanical charac-
terization of the carbonated wollastonite system. In: 5th
Nanotechnology in construction (NICOM5), pp 71–77
[34] Riman RE, Nye TE, Atakan V, Vakifahmetoglu C, Li Q,
Tang L (2015) US 9216926 B2: Synthetic formulations and
methods of manufacturing and using thereof. US patent.
[35] Farnam Y, Villani C, Washington T, Spence M, Jain J, Weiss
WJ (2016) Performance of carbonated calcium silicate based
cement pastes and mortars exposed to NaCl and MgCl
2
deicing salt. Cons Build Mater 111:63–71
[36] Morandeau A, Thie´ry M, Dangla P (2015) Impact of
accelerated carbonation on OPC cement paste blended with
fly ash. Cem Concr Res 67:226–236
[37] Shao Y, Rostami V, He Z, Boyd A (2013) Accelerated carbon-
ation of Portland limestone cement. J Mater Civ Eng 26:117–124
[38] Klemm WA, Berger RL (1972) Accelerated curing of
cementitious systems by carbon dioxide part I. Portland
cement. Cem Concr Res 2:567–576
[39] Fang Y, Chang J (2015) Microstructure changes of waste
hydrated cement paste induced by accelerated carbonation.
Constr Build Mater 76:360–365
[40] Groves GW, Brough A, Richardson IG, Dobson CM (1991)
Progressive changes in the structure of hardened C
3
S cement
pastes due to carbonation. J Am Ceram Soc 74:2891–2896
[41] Shtepenko O, Hills C, Brough A, Thomas M (2006) The
effect of carbon dioxide on b-dicalcium silicate and Portland
cement. Chem Eng J 118:107–118
[42] Goto S, Suenaga K, Kado T (1995) Calcium silicate car-
bonation products. J Am Ceram Soc 78:2867–2872
6190 J Mater Sci (2016) 51:6173–6191
[43] Hidalgo A, Domingo C, Garcia C, Petit S, Andrade C, Alonso
C (2008) Microstructural changes induced in Portland cement-
based materials due to natural and supercritical carbonation.
J Mater Sci 43:3101–3111. doi:10.1007/s10853-008-2521-5
[44] Short NR, Purnell P, Page CL (2001) Preliminary investi-
gations into the supercritical carbonation of cement pastes.
J Mater Sci 36:35–41. doi:10.1023/A:1004870204162
[45] Ashraf W, Olek J, Atakan V (2015) Chemo-mechanical
comparison of the carbonation and hydration reaction
products of synthetic tricalcium silicate (C
3
S). In: 11th brittle
matrix composites (BMC), Warsaw, Poland
[46] Daval D, Martinez I, Guigner JM, Hellmann R, Corvisier J,
Findling N, Dominici C, Goffe B, Guyot F (2009) Mechanism
of wollastonite carbonation deduced from micro- to nanometer
length scale observations. Am Miner 94:1707–1726
[47] Daval D, Martinez I, Corvisier J, Findling N, Goffe´ B, Guyot
F (2009) Carbonation of Ca-bearing silicates, the case of
wollastonite: experimental investigations and kinetic mod-
eling. Chem Geol 265:63–78
[48] Berliner R, Ball C, West P (1997) Neutron powder diffrac-
tion investigation of model cement compounds. Cem Concr
Res 27:551–575
[49] Sa´ezDelBosque IF, Martı´nez-Ramı´rez S, Blanco-Varela MT
(2014) FTIR study of the effect of temperature and nanosilica
on the nano structure of C-S-H gel formed by hydrating tri-
calcium silicate. Constr Build Mater 52:314–323
[50] Shibata S, Kishi K, Asaga K, Daimon M (1984) Preparation
and hydration of b-C
2
S without stabilizer. Cem Concr Res
14:1689–1699
[51] National Research Council (1972) Guide to Compounds of
Interest in Cement and Concrete Research. Highway
Research Board, Division of Engineering
[52] Puertas F, Trivin˜o F (1985) Examinations by infra-red
spectroscopy for the polymorphs of dicalcium silicate. Cem
Concr Res 15:127–133
[53] Ashraf W, Olek J, Atakan V (2015) A comparative study of
the reactivity of calcium silicates during hydration and car-
bonation reactions. In: 14th International congress on cement
chemistry, Beijing, China
[54] Bertos MF, Simons SJR, Hills CD, Carey PJ (2004) A
review of accelerated carbonation technology in the treat-
ment of cement-based materials and sequestration of CO
2
.
J Hazard Mater 112:193–205
[55] Silva PD, Bucea L, Moorehead DR, Sirivivatnanon V (2006)
Carbonate binders: reaction kinetics, strength and
microstructure. Cem Concr Compos 28:613–620
[56] Cui H, Tang W, Liu W, Dong Z, Xing F (2015) Experimental
study on effects of CO
2
concentrations on concrete carbon-
ation and diffusion mechanisms. Constr Build Mater
93:522–527
[57] Gauvin R, Lifshin E, Demers H, Horny P, Campbell H
(2006) Win X-ray: a new Monte Carlo program that com-
putes X-ray spectra obtained with a scanning electron
microscope. Microsc Microanal 12:49–64
[58] Okada Y, Masuda T, Takada M, Xu L, Mitsuda T (1998)
Relationship between NMR
29
Si chemical shifts and FT-IR
wave numbers in calcium silicates. In: Colombet P, Grimmer
AR, Zanni H, Sozzani P (eds) Nuclear magnetic resonance of
spectroscopy of cement-based materials. Springer, pp 69–78
[59] Yousuf M, Mollah A, Pargat JR, Cocke DL (1992) An
infrared spectroscopic examination of cement-based solidi-
fication/stabilization systems - Portland types V and IP with
zinc. J Environ Sci Health Part A 27:1503–1519
[60] Yu P, Kirkpatrick RJ, Poe B, McMillan PF, Cong X (1999)
Structure of calcium silicate hydrate (C-S-H): near-, mid-, and
far-infrared spectroscopy. J Am Ceram Soc 82:742–748
[61] Hansen MR, Jakobsen HJ, Skibsted J (2003)
29
Si chemical
shift anisotropies in calcium silicates from high-field
29
Si
MAS NMR spectroscopy. Inorg Chem 42:2368–2377
[62] Hughes TL, Methven CM, Jones TGJ, Pelham SE, Fletcher
P, Hall C (1995) Determining cement composition by
Fourier transform infrared spectroscopy. Adv Cem Based
Mater 2:91–104
[63] Arnold T, Rozario-Ranasinghe M, Youtcheff J (2006)
Determination of lime in hot-mix asphalt. Transp Res Rec
1962:113–120
[64] Chollet M, Horgnies M (2011) Analyses of the surfaces of
concrete by Raman and FT-IR spectroscopies: comparative
study of hardened samples after demoulding and after
organic post-treatment. Surf Interface Anal 43:714–725
[65] Ferna´ ndez-Carrasco L, Torrens-Martı´n D, Morales LM,
Martı´nez-Ramı´rez S (2012) Infrared spectroscopy in the
analysis of building and construction materials. In: Infrared
spectroscopy: materials science, engineering and technology,
pp 369–382
[66] Origin Lab (2015) OriginPro. Northampton, MA. http://
www.originlab.com. Accessed 10 Oct 2015
[67] Andersen FA, Brec
ˇevic
´L (1991) Infrared spectra of amor-
phous and crystalline calcium carbonate. Acta Chem Scand
45:1018–1024
[68] Ibanez J, Artus L, Cusco R, Lopez A, Menendez E, Andrade
MC (2007) Hydration and carbonation of monoclinic C
2
S
and C
3
S studied by Raman spectroscopy. J Raman Spectrosc
38:61–67
[69] Sevelsted TF, Skibsted J (2015) Carbonation of C–S–H and
C–A–S–H samples studied by 13C, 27Al and 29Si MAS
NMR spectroscopy. Cem Concr Res 71:56–65
[70] Castellote M, Fernandez L, Andrade C, Alonso C (2009)
Chemical changes and phase analysis of OPC pastes carbon-
ated at different CO
2
concentrations. Mater Struct 42:515–525
J Mater Sci (2016) 51:6173–6191 6191
A preview of this full-text is provided by Springer Nature.
Content available from Journal of Materials Science
This content is subject to copyright. Terms and conditions apply.
