Thermochimica Acta 605 (2015) 37–42

Contents lists available at ScienceDirect

Thermochimica Acta
journal homepage: www.elsevier.com/locate/tca

Influence of fineness on hydration kinetics of supersulfated cement

Shuhua Liu a,b, * , Lu Wang a , Yuxin Gao c , Baoying Yu c , Wan Tang a
a
State Key Laboratory of Water Resources and Hydropower Engineering Science, Wuhan University, Wuhan 430072, China
b
Department of Civil, Environmental and Geomatic Engineering, University College London, Gower Street, London WC1E 6BT, UK
c
China Construction Ready Mixed Concrete Co., Ltd., Wuhan 430000, China

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

Article history: Hydration heat evolution rate and quantity of supersulfated cements (SSCs) with three different specific
Received 30 July 2014 surface areas (335 m2/kg, 445 m2/kg and 518 m2/kg) were measured at 25
C by isothermal calorimetry.
Received in revised form 8 November 2014 The hydration heat evolution quantity of SSCs improves with the increase of cement fineness, but is much
Accepted 17 February 2015
lower than that of ordinary Portland cement. The hydration of supersulfated cement is accelerated by the
Available online 19 February 2015
elevated fineness, and the hydration process of supersulfated cement after induction period was divided
into three periods, nucleation and crystal growth (NG), phase boundary reaction (I), and diffusion (D),
Keywords:
which are well simulated through Krstulovic–Dabic model. The connecting points of NG process and I
Supersulfated cement
Fineness
process, I process and D process both decrease with the increase of fineness which shows that the second
Hydration heat exothermic peak is in advance. I process also decreases with the increase of fineness, which indicates that
Kinetic parameters I process is accelerated by fineness. The hydration of supersulfated cement is sensitive to fineness,
especially for the specific surface area reaching up to 518 m2/kg.
ã 2015 Elsevier B.V. All rights reserved.

1. Introduction
Portland cement is the most popular and also one of the most
important cementing materials in the modern concrete [1], and
during the year of 2012, there are more than 2.2 billion tons cement
produced in China [2]. Because of the high hydration heat emission
of cement as well as the low thermal conductivity of concrete, the
hydration of cement always produces a large number of heat and
leads to a raise of internal temperature in massive concrete, which
may make concrete cracking during the cooling period [3]. In order
to reduce the hydration heat emission from the cementitious
materials, mineral admixtures are widely used in modern concrete
[3–6], which can reduce the hydration heat emission and the rate
of cementing materials [7,8].
So, in this paper, one kind of more environment-friendly
cement with three different specific surface areas will be studied.
Supersulfated cement (SSC) is a kind of cement with less clinker or
even without clinker, which contains almost 80% ground
granulated blast furnace slag, 15% gypsum as sulfate activator
and less than 5% clinker or lime as alkali activator [9–11]. SSC is
made directly by the mixtures of these three raw materials, often
added with 1% mass ratio of coagulant, and in ball mill to be ground
to suitable fineness. High content slag in SSC can make full use of
the potential activity of slag and also benefit for environment.
Studies show that granulated blast furnace slag is suitable for
producing SSC and have a reasonable strength development
process because of its high content of calcium and alumina (at least
14–15%) [11,12].
Some researchers have pointed out that the hydration heat of
SSC is much lower than that of ordinary Portland cement [9–12].
Grinding can affect the specific surface area, i.e. fineness, which
can influence the hydration process of the binders [13]. So not only
addition of mineral admixture can affect the hydration degree,
hydration heat evolution rate and total hydration heat of cement,
but also the fineness of raw materials has great influence on the
hydration properties. In order to understand the hydration process
of SSC with different finenesses, study of detailed heat evolution
behavior is needed. The hydration heat curves of SSCs with three
different finenesses were measured to analyze the interaction
among them on hydration heat emission and rate.
2. Experimental

2.1. Materials

* Corresponding author at: State Key Laboratory of Water Resources and
Hydropower Engineering Science, Wuhan University, Wuhan 430072, China.
Tel.: +86 27 68772233; fax: +86 27 68772310.
E-mail address: shliu@whu.edu.cn (S. Liu).
The SSCs are ground with three specific surface areas. SSC1,
SSC2 and SSC3 are short for the SSC with specific surface area of
335 m2/kg, 445 m2/kg and 518 m2/kg respectively. The chemical

http://dx.doi.org/10.1016/j.tca.2015.02.013
0040-6031/ ã 2015 Elsevier B.V. All rights reserved.
38 S. Liu et al. / Thermochimica Acta 605 (2015) 37–42
[(Fig._1)TD$IG]
Nomenclature

List of acronyms and parameters identification

SSC Supersulfated cement
SSCs Supersulfated cements
NG Nucleation and crystal growth
I Phase boundary reaction
D Diffusion
SSC1 Supersulfated cement with specific surface area of
335 m2/kg
SSC2 Supersulfated cement with specific surface area of
445 m2/kg
SSC3 Supersulfated cement with specific surface area of
518 m2/kg
W/C Water to cement ratio
P The heat evolution quantity
a The hydration degree
Fig. 1. Hydration heat evolution rate curves of the three SSCs.
C—S—H Calcium silicate hydrate
CH Calcium hydroxide
K The reaction rate constant of hydration reaction As the hydration heat evolution rate curves of the three SSCs
n Reaction order shown in Fig. 1, in the first several minutes after mixing SSC
R The diameter of reacted particle powder with water, a sharp and transient exothermic peak occurs,
t0 The end time of the induction period which is formed by a quick formation of ettringite due to the rapid
dissolution of aluminates and sulfate from slag and gypsum, as
well as from the small amount of cement clinker. Then the heat
emission rate decreases rapidly, and the hydration heat increases
compositions of SSC are shown in Table 1. From Table 1, SSC very slowly. After that, the hydration goes to the induction period,
contains high content of SO3 up to 13.42% due to high content of and the hydration heat rate is closed to zero. As shown in Fig. 1,
gypsum. there is little difference of the hydration induction period between
SSC1 and SSC2, which lasts for about 24 h, ending at about 24.37 h
2.2. Test methods and 24.80 h (Fig. 3), which are much longer than that of ordinary
Portland cement. However, the induction period of SSC3 lasts for
The hydration heat evolution rate and total hydration heat about 18 h, and ending at about 18.74 h, which is ahead about 6 h
emission of SSC1, SSC2 and SSC3 were measured with an for SSC1 and SSC2. The induction period of hydration for SSC is
isothermal calorimeter (TAM Air) at 25
C within 96 h. TAM Air prolonged due to low clinker content in the cements. Since the
has eight parallel twin-chamber measuring channels maintained content of slag reaches up to 80% and it has high potential activity, a
at a constant temperature: one chamber containing the sample, large amount of Ca2+ produced during the hydration of cement
the other containing the reference. The temperature fluctuation is clinker and is absorbed on the surface of slag, which reduces Ca2+
less than 0.02
C, the precision can reach 20 mmW. The water to concentration in the solution, delays Ca(OH)2 (CH) nucleation and
cement ratio (W/C) was 0.4 and 10 g samples for each SSC were generates unstable C—S—H gel with low Ca/Si ratio (it will slowly
used. First take 4.00 g pure water into a bottle, and then pour change into a stable C—S—H gel). Thus the ending time of the
10.00 g sample powder into. At the same time, stirring quickly and induction period is prolonged [14].
putting it into the chamber to measure hydration heat. The Subsequently, the acceleration period comes with rapid
hydration heat evolution rate and total hydration heat of SSC can be hydration of C3S, fast formation of C—S—H gel and ettringite.
continuously monitored as a function of time.
[(Fig._2)TD$IG]
3. Results and discussions

3.1. Characteristics of hydration heat evolution

Fig. 1 shows the hydration heat emission rate curves of the three
SSCs and Fig. 2 shows the hydration heat emission quantity at the
constant temperature 25
C within 96 h. The acceleration period
and the end time of induction period of SSCs are shown in Fig. 3.
The rate of the second heat emission peak and the total heat
emission at different hydration ages determined from
heat evolution curves of SSCs are shown in Table 2.

Table 1
Chemical composition of SSC (mass fraction/%).

Composition SiO2 Al2O3 Fe2O3 CaO MgO SO3 Na2O K2O P2 O 5

SSC 25.63 10.87 1.18 41.45 4.93 13.42 0.23 0.60 0.40
Fig. 2. Hydration heat quantity curves of the three SSCs.
 S. Liu et al. / Thermochimica Acta 605 (2015) 37–42
[(Fig._3)TD$IG]
Fig. 3. The acceleration period and end time of induction period of hydration
process.
The greater value of specific surface area, the higher peak value of
the second exothermic peak (Fig. 1). The early hydration of SSC is
accelerated with the increase of the fineness. The ending time of
induction period and the reaching time of the second peak are
shortened significantly. While the peak value increases greatly.
Fig. 3 shows that the ending time of induction period is almost the
same when the specific surface area increases from 335 m2/kg to
445 m2/kg; the time reaching the second exothermic peak of SSC is
shortened from 36.74 h to 34.43 h. The maximum value of the
second exothermic effect of SSC increases by 57.23% from 3.11 J/g h
to 4.89 J/g h (Table 2). When the specific surface area increases
from 445 m2/kg to 518 m2/kg, the ending time of induction period
is advanced from 24.80 h to 18.47 h; the time reaching the second
exothermic peak of SSC is shortened from 34.43 h to 25.75 h. The
maximum value of the second exothermic effect of SSC increases
by 51.00% from 4.89 J/g h to 7.38 J/g h. Quick consumption of
moisture in SSC composite binder pastes and large amount of
hydration products generated by the rapid hydration of binder
hinder the diffusion and migration of water and ions. As a result,
the hydration reaction rate decreases rapidly which forms the
second exothermic peak with a high but narrow shape especially
for SSC3. The reaction goes into the diffusion control phase after
about 50 h for SSC1 and SSC2, while for SSC3 it is about 40 h. The
reaction of slag is accelerated when the active phase of slag is
corroded by a strong alkaline solution because much CH is
generated with the increase of the fineness. The dilution effect of
slag at the early stage of hydration increases the effective water to
cement ratio of binder but shortens the duration of hydration
process of binder. Slag provides nucleation points for the hydration
products of the cement and reaction of slag increases the hydration
rate of binder. An obvious exothermic effect at the later period
appears on the hydration heat evolution curve of SSCs because of
the pozzolanic reaction of slag [15].
Table 2
Characteristic values of hydration heat evolution curves of SSCs at 25
C.
Sample Rate of the second heat emission peak qmax (J/g h)
SSC1 3.1075
SSC2 4.8859
SSC3 7.3777
39
Fig. 2 shows that the total hydration heat of SSC increases
disproportionally with the increase of the fineness. As slag is
gradually activated in the later period of hydration, the reaction of
slag is accelerated and it consumes CH generated by the cement
hydration, which in turns promotes SSC’s hydration. SSC1 has the
lowest hydration heat, and the hydration heat of SSC3 is
the highest.
The total hydration heat emission of SSC in the early period has
a substantial increase when the specific surface area rises from
335 m2/kg to 445 m2/kg, and to 518 m2/kg. It can be seen from
Table 2 that the 12 h heat emission increases by 3.13% from 1.60 J/g
to 1.65 J/g when the specific surface area rises from 335 m2/kg to
445 m2/kg, and by 41.82% from 1.65 J/g to 2.34 J/g when the specific
surface area rises from 445 m2/kg to 518 m2/kg. However, it is still
very low compared with that of ordinary Portland cement.
Fineness affects the early hydration of SSC. As shown in Table 2,
the heat emission increases by 23.84% from 60.66 J/g (SSC1) to
75.12 J/g (SSC2), by 26.12% from 75.12 J/g (SSC2) to 94.74 J/g (SSC3)
at 96 h. When hydration time from 12 h to 48 h, the total heat
emission increases by 2713.13% from 1.60 J/g to 45.01 J/g for SSC1,
by 3322.42% from 1.65 J/g to 56.47 J/g for SSC2, and by 3208.97%
from 2.34 J/g to 77.43 J/g for SSC3. The hydration of the SSC is
sustainably developed in the period from 12 h to 48 h. It leads to a
great increase of the total heat emission. There is little difference of
the hydration heat emission between SSC1 and SSC2 at the early
period. All the SSCs have much lower total heat emission due to the
small mass fraction of cement clinker, and much high content of
slag. And the hydration heat emission of the three SSCs does not
increase much at the later period. Fig. 1 shows that the reaction is
fast when the specific surface area reaches up to 518 m2/kg, the
induction period ends at about 18.47 h. And SSC3 reaches the
second exothermic peak at 25.75 h, which is much earlier than that
of SSC1 (36.74 h) and SSC2 (34.43 h). CH is generated quickly by the
hydration of cement and slag is activated significantly by the high
fineness.
3.2. The hydration process simulation of SSC
3.2.1. The exothermic process division of the cement-based materials
At present, the kinetic parameters of the hydration reaction of
the cementitious materials were achieved by testing the rate of
hydration heat evolution with the isothermal calorimetry method
as the following equations,
1 1 t0:5
¼ þ (1)
P P1 P1 ðt  t0 Þ
PðtÞ
aðtÞ ¼ (2)
P1
In which P is the hydration heat evolution quantity; a is the
hydration degree; t0 is the end time of the induction period.
The curve of hydration degree with time can be plotted based
on the isothermal calorimetry testing data, which will be used to
set up the hydration kinetic model. After that, the constant n and
hydration rate constant K of the compound gel system will be
calculated according to the Krstulovic–Dabic model [16–18], and
Total heat emission Q (J/g)
12 h 48 h 60 h 72 h 96 h
1.60 45.01 53.89 58.09 60.66
1.65 56.47 65.90 71.34 75.12
2.34 77.43 86.11 91.14 94.74
 40 S. Liu et al. / Thermochimica Acta 605 (2015) 37–42
[(Fig._5)TD$IG]
the hydration mechanism of SSC at different hydration stages will
be obtained. The isothermal calorimetry method is very sensitive
for some low hydration degree of cement-based materials at early
hydration, especially for the acceleration period and reduction
period. Additionally, it is more convenient to analyze how different
factors affect the hydration rate before and during the acceleration
period of the cementitious material. In this paper, the hydration
heat evolution rate and quantity at different stages were tested
using this method.
Based on the characteristics of hydration heat, the hydration
process of cement-based materials generally can be divided into
five processes. Taking the SSC as an example is in this paper. As
shown in Fig. 4, it is divided into five processes [19,20]: rapid Fig. 5. Determination of maximum hydration emission heat Pmax from linear
response (0–A–B), which has just started contact cement with regression.
water, corresponds to the first exothermic peak of heat evolution
curve; induction period (B–C), which is relative dormant state, the the three processes to express the kinetic equation [16–18,23]
hydration rate is very low, and the hydration reaction is relatively between the hydration process and reaction time, that is,
inert; acceleration period (C–D), the hydration heat evolution rate Nucleation and crystal growth (NG):
of rapidly rising stage; deceleration phase (D–E), the hydration
slows down rapidly; recession (E–F), namely the end stage, the ½lnð1  aÞ1=n ¼ K 1 ðt  t0 Þ ¼ K 0 1 ðt  t0 Þ (3)
heat evolution rate is close to zero. Phase boundary reaction (I):
Due to the fact that the heat quantity released before the end of
the induction period (i.e., the area of the first exothermic peak in ½1  1  a1=3 1 ¼ K 2 r1 ðt  t0 Þ ¼ K 0 2 ðt  t0 Þ (4)
the heat evolution rate curve, 0–A–B) generally accounts for only Diffusion (D):
about 5% of the total hydration heat, it can be ignored relative to
the hydration process [21,22]. In practice, the mixing heat ½1  1  a1=3 2 ¼ K 3 r2 ðt  t0 Þ ¼ K 0 3 ðt  t0 Þ (5)
evolution tend to be included the initial casting temperature, In which a is the hydration degree, i.e., the hydration degree of
and SSC (especially in the absence of added coagulant) shows spherical particles of cement; K is the reaction rate constant of
relatively slow hydration process and very low hydration heat. hydration reaction; n is reaction order; R is the diameter of reacted
Therefore, the first exothermic peak is always ignored in the particle; t0 is the end time of the induction period.
hydration process; at the same time, due to the very low hydration At the same time, kinetic equation can be obtained from the
rate at induction period, the continuous heat quality of induction differential of above three types to express the hydration rate of the
period is also ignored. So the discussion of hydration heat starts three processes as following,
from the end of the induction period (the point C in Fig. 4), Nucleation and crystal growth (NG):
hydration heat evolution rate from the point C starts rising rapidly.
da
¼ F 1 ðaÞ ¼ K 01 nð1  aÞ½lnð1  aÞ1=n (6)
3.2.2. The introduction of Krstulovic–Dabic model dt
Hydration reaction of cement-based materials is divided into Phase boundary reaction (I):
three basic processes according to Krstulovic–Dabic model, namely
da
nucleation and crystal growth (NG), phase boundary reaction (I) ¼ F 2 ðaÞ ¼ 3K 02 ð1  aÞ2=3 (7)
dt
and diffusion (D). These three control processes may occur
simultaneously, and also can occur alone or in pairs, but the Diffusion (D):
hydration rate of the overall development process depends on the h i
da 3=2 K 03 ð1  aÞ2=3
one which reacts slowest, that is, the control process of the slowest ¼ F 3 ðaÞ ¼ (8)
reaction controls the reaction rate and mechanism of correspond- dt 1  ð1  aÞ1=3
ing stage. Krstulovic gave the integral and differential expression of
This model is a considerable representative hydration dynamic
[(Fig._4)TD$IG] model at the current academic, through which the hydration
process of cement-based materials can be controlled, and it will be
easier to understand the affecting factors at each stage. While there

[(Fig._6)TD$IG]

Fig. 6. Determination of kinetic factors (n and K 01 ) of nucleation and crystal

Fig. 4. Hydration heat liberation rate and quantity of SSC. growth (NG) process from linear regression.
 S. Liu et al. / Thermochimica Acta 605 (2015) 37–42 41
[(Fig._7)TD$IG]
Table 3
Hydration kinetic parameters of SSC.

Sample Qmax n K 01 K 02 K 03 a1 a2 a2–a1

SSC1 75.76 2.1238 0.04549 0.0160 0.0047 0.235 0.376 0.141

SSC2 87.72 2.1980 0.05782 0.0215 0.0061 0.230 0.364 0.134
SSC3 108.70 2.2892 0.07160 0.0260 0.0061 0.187 0.306 0.119

(t  t0), so that it can obtain the kinetic parameters K 02 of phase

boundary reaction (I) parameter and K 03 of diffusion process (D)
parameter by linear fitting, as shown in Figs. 7 and 8.
Similarly, the hydration kinetic parameters of SSC2 and SSC3 are
Fig. 7. Determination of kinetic factors (n and K 02 ) of phase boundary reaction (I) obtained through this method, and all the parameters of the three
process from linear regression.
SSCs are shown in Table 3.
The kinetic parameters obtained above are substituted into the
are some problems about this model, such as the boundaries
Formulas (6)–(8), then it can obtain the kinetic curves of reaction
between the various processes is not clear at the station that the
rates: F1(a), F2(a) and F3(a) of NG process, I process and D process
model divides hydration reaction process into three basic
with the hydration degree a. Making the graph of F1(a), F2(a) and
processes, and the correlation between the experimental data
F3(a) with a. Such as the solid lines in Figs. 9–11, and the dotted
and models at the boundaries is poor. Even though, in this paper,
line indicates the degree of hydration with time of the calculated
Krstulovic–Dabic model is also used effectively to simulate the
hydration curve.
hydration process of SSC.
From the above figures, three processes after induction period
of SSC are NG process (before the hydration degree of a1), I process
3.2.3. Hydration process simulation of SSC
(between the hydration degree of a1 and a2) and D process (after
The hydration mechanism of SSC is different from that of
the hydration degree of a2). They are basically consistent with the
ordinary Portland cement. The hydration of ordinary Portland
SSC, and I process and D process fit a higher degree. Also
cement is based mainly on alkali-activated and the main hydration
the turning point (a1 and a2) at the intersection is also consistent
products are calcium hydroxide (CH) and calcium silicate hydrate
with the rate curves. Only the first process of NG is not that close
(C—S—H gel). But as for SSC, it takes gypsum as sulfate activator
with the rate curve, however, the underlying trend is consistent.
and clinker as alkaline activator, the hydration of slag is activated
Three curves of F1(a), F2(a) and F3(a) can basically segmented
by the excitation of these two substances, and the hydration
simulate the actual hydration rate curve da/dt that is obtained by
products of SSC includes C—S—H gel, ettringite and a small amount
heat-measurement experiments of SSC. The hydration reaction
of gypsum. Even so, the main hydration product of SSC is C—S—H
process of SSC is not a single, but also has complex process with
gel.
multiple reaction mechanisms. The control factors are various at
Taking Knudsen extrapolation equation (Formula (1)) to linear
different reaction stages. In the early hydration, the NG stage plays
fit the hydration heat evolution curve of SSC1 (the hydration heat
a leading role due to more adequate water and less hydration
curve in Fig. 1), as shown in Fig. 5, the maximum hydration
products; with prolonged hydration process, the hydration turns to
heat Pmax = 75.76 J/g can be calculated. Pmax is the heat released by
I stage and then turns to D stage because the quantity of the
some hydrated cementitious materials in the experiment process,
hydration products improve and ion mobility becomes difficult.
but not the theoretical heat evolution. The cementitious materials
The SSC’s clinker content is only 1–5%. At the early hydration,
cannot be fully hydrated in actual project, so it is reasonable to take
clinker is dissolved in water, and tricalcium aluminate first
part hydration here. The degree of hydration a(t) can be calculated
hydrates to generate hydrated calcium aluminates. While the
by Pmax according to Formula (2), putting a(t) into Formula (3) and
gypsum is dissolved in water to produce sulphate ion, and reacted
making the ln[ln(1  a)]/ln(t  t0) double logarithmic curve, then
with the hydrated calcium aluminates to generate ettringite. Then
through linear fitting to get kinetics parameters (n and K 01 ) of tricalcium silicate hydrates, indicating the end of the induction
nucleating and crystal growth process (Fig. 6). period, and produces more calcium hydroxide and C—S—H gel. At
And then, putting the hydration degree a(t) into Eqs. (4) and (5), the same time, active alumina in slag will react with CH and
to get the curves of [1(1  a)(1/3)]/(t  t0) and [1(1  a)(1/3)]2/ sulphate ion, and generate ettringite. This will promote the

[(Fig._8)TD$IG] [(Fig._9)TD$IG]

Fig. 8. Determination of kinetic factors (n and K 03 ) of diffusion (D) process from

linear regression. Fig. 9. Hydration rate curves for SSC1.
 42 S. Liu et al. / Thermochimica Acta 605 (2015) 37–42
[(Fig._10)TD$IG]
Fig. 10. Hydration rate curves for SSC2.
[(Fig._1)TD$IG]
Fig. 11. Hydration rate curves for SSC3.
formation of hydroxyl radicals and then more C—S—H gel and
ettringite will be generated. Also the ettringite can enhance early
strength of SSC.
So for SSC hydration, the Krstulovic–Dabic model can simulate
its hydration at a certain extent. It is able to act the trend of
hydration and control the process of various stages. The total heat
emission of cementitious materials improves with the increase of
fineness, just as from SSC1 to SSC3. In this paper, measured
maximum hydration heat is Pmax = 75.76 J/g for SSC1, 87.72 J/g for
SSC2, 108.70 J/g for SSC3, and all these values indeed are lower than
that of ordinary Portland cement. SSC is low heat cement and
conducive to mass concrete. With the increase of fineness, the NG
process and I process are shortened gradually, and the D process is
prolonged gradually.
4. Conclusions
(1) The hydration heat evolution rate of SSC increases gradually
with the increase of fineness, and forms a high but very narrow
second exothermic peak as it reaches up to 518 m2/kg. The
hydration heat emission of SSC1 and SSC2 has little difference
at the early period, while SSC3 has much higher hydration heat
emission than that of SSC1 and SSC2.
(2) The hydration of SSC can be accelerated by grinding. With the
increase of fineness, the end time of induction period and the
reaching time of the second peak are shortened significantly,
and the peak value is increased gradually. Even so, the highest
total heat emission is only 108.70 J/g which is still much lower
than that of ordinary Portland cement.
(3) The simulation curves with the Krstulovic–Dabic model can
simulate the SSC hydration in a certain extent and it is able to
reflect the trend of hydration and control the process of various
stages. The NG process and I process are shortened gradually
and the D process is prolonged gradually with the increase of
SSC fineness.
Acknowledgement
This Project is funded by the National Natural Science
Foundation of China(51208391).
References
[1] J. Stark, Recent advances in the field of cement hydration and micro structure
analysis, Cem. Concr. Res. 41 (7) (2011) 666–678.
[2] China Cement Association, China’s cement industry review in 2012 and
outlook in 2013, Chin. Cem. 3 (2013) 26–29.
[3] B. Lothenbach, K. Scrivener, R.D. Hooton, Supplementary cementitious
materials, Cem. Concr. Res. 41 (12) (2011) 1244–1256.
[4] S.H. Liu, Y.N. Kong, L. Wang, A comparison of hydration properties of cement–
low quality fly ash binder and cement–limestone powder binder, J. Therm.
Anal. Calorim. 116 (2) (2014) 937–943.
[5] B.W. Langan, K. Weng, M.A. Ward, Effect of silica fume and fly ash on heat of
hydration of Portland cement, Cem. Concr. Res. 32 (7) (2002) 1045–1051.
[6] Q. Wang, M. Miao, P.Y. Yan, The influence of high-temperature curing on the
hydration characteristics of a cement-GGBS binder, Adv. Cem. Res. 24 (1)
(2012) 33–40.
[7] Q. Wang, P.Y. Yan, Hydration properties of basic oxygen furnace steel slag,
Constr. Build. Mater. 24 (7) (2010) 1134–1140.
[8] P.Y. Yan, G.D. Mi, Q. Wang, A comparison of early hydration properties of
cement–steel slag binder and cement–limestone powder binder, J. Therm.
Anal. Calorim. 115 (1) (2014) 193–200.
[9] M.C.G. Juenger, F. Winnefeld, J.L. Provis, Advances in alternative cementitious
binders, Cem. Concr. Res. 41 (12) (2011) 1232–1243.
[10] L. Li, Q.L. Zhao, T. He, Study on performance of supersulphated cement
masonry mortar, Proceedings of the third nationwide academic symposium on
commercial mortar, Wuhan, China, 2009, pp. 51–54.
[11] L. Li, Research of Enhance Mechanism and Application on Supersulphated
Cement Using Metallurgical Slag, Wuhan University of Technology, 2010.
[12] A. Gruskovnjak, Hydration mechanisms of sulphated slag cement, Cem. Concr.
Res. 38 (7) (2008) 983–992.
[13] S.H. Liu, L.H. Li, Influence of fineness on the cementitious properties of steel
slag, J. Therm. Anal. Calorim. 117 (2) (2014) 629–634.
[14] M. Narmluk, T. Nawa, Effect of fly ash on the kinetics of Portland cement
hydration at different curing temperature, Cem. Concr. Res. 41 (6) (2011)
579–589.
[15] N.Y. Mostafa, P.W. Brown, Heat of hydration of high reactive pozzolans in
blended cements: isothermal conduction calorimetry, Thermochim. Acta 435
(2) (2005) 162–167.
[16] J.J. Thomas, A new approach to modeling the nucleation and growth kinetics of
tricalcium silicate hydration, J. Am. Ceram. Soc. 90 (10) (2007) 3282–3288.
[17] J.J. Thomas, J.J. Biernacki, J.W. Bullard, et al., Modeling and simulation of
cement hydration kinetics and microstructure development, Cem. Concr. Res.
41 (12) (2011) 1257–1278.
[18] J. Cheung, A. Jeknavorian, L. Roberts, et al., Impact of admixtures on the
hydration kinetics of Portland cement, Cem. Concr. Res. 41 (12) (2011)
1289–1309.
[19] X.Y. Wang, H.S. Lee, K.B. Park, et al., A multi-phase kinetic model to simulate
hydration of slag?cement blends, Cem. Concr. Compos. 32 (6) (2010) 468–477.
[20] F. Lin, C. Meyer, Hydration kinetics modeling of Portland cement considering
the effects of curing temperature and applied pressure, Cem. Concr. Res. 39 (4)
(2009) 255–265.
[21] P.Y. Yan, F. Zheng, Kinetics model for the hydration mechanism of cemetitious
materials, J. Chin. Ceram. Soc. 34 (5) (2006) 555–559.
[22] S. Kourounis, S. Tsivilis, P.E. Tsakiridis, et al., Properties and hydration of
blended cements with steelmaking slag, Cem. Concr. Res. 37 (6) (2007)
815–822.
[23] S. Swaddiwudhipong, D. Chen, M.H. Zhang, Simulation of the exothermic
hydration process of Portland cement, Adv. Cem. Res. 15 (4) (2003) 61–69.