Construction and Building Materials 166 (2018) 130–140

Contents lists available at ScienceDirect

Construction and Building Materials

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

Analysis of hydration and strength optimization of cement-fly

ash-limestone ternary blended concrete
Xiao-Yong Wang
Department of Architectural Engineering, Kangwon National University, Chuncheon-si 24341, South Korea

h i g h l i g h t s

Simulate the hydration of cement-fly ash-limestone ternary blends.

Consider interactions among cement hydration, fly ash reaction, and limestone reaction.

Evaluate the strength development of ternary blended concrete.

Find the optimum combinations of cement, fly ash, and limestone.

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

Article history: Limestone powder improves concrete early-age strength while fly ash improves concrete late-age
Received 25 September 2017 strength due to its pozzolanic reaction. The optimal mixture of cement, fly ash, and limestone is crucial
Received in revised form 6 January 2018 for material design of ternary blended concrete. This research presents a simulation program for evalu-
Accepted 9 January 2018
ating the hydration and strength optimization of ternary blended concrete. The simulation program
begins with a kinetic hydration model which simulates the hydration of cement-fly ash-limestone tern-
ary blends. The hydration model considers the mutual effects among reactions of cement, fly ash, and
Keywords:
limestone by means of the contents of calcium hydroxide and capillary water. The individual reaction
Fly ash
Limestone
degrees of components of ternary blends are calculated from the hydration model. Furthermore, the com-
Hydration pressive strength growth of hydrating ternary blended concrete is calculated by means of gel-space ratio
Model and Powers’ strength theory. Finally, based on parameter studies, the optimal combinations of cement, fly
Strength optimization ash, and limestone at different ages are determined. The proposed numerical procedure is valuable for
making composite cements as it pertains to compressive strength and environment regulations.
Ó 2018 Elsevier Ltd. All rights reserved.

1. Introduction
Fly ash and limestone are more and more utilized in producing
high performance concrete in the modern concrete industry. Fly
ash and limestone present different effects on strength develop-
ment of concrete. Fly ash can enhance the long-term age strength
of concrete due to a pozzolanic reaction while limestone can
enhance the young-age strength of concrete because limestone
can accelerate cement hydration. When limestone and fly ash are
used together, due to the synergy effect, the individual deficiencies
can be compensated [1].
Abundant experimental and theoretical studies have been per-
formed on cement-fly ash-limestone ternary blended concrete.
Weerdt et al. [2] found that the additional aluminum from the
fly ash pozzolanic reaction amplified the chemical reaction of lime-
stone powder. Scholer et al. [3] found that a limestone content of
2–5% leads to the stabilization of monocarbonate and ettringite
which can increase the volume of hydration products and increase
the strength. Bentz et al. [4] found that nano-limestone can accel-
erate the early age hydration of cement and reduce the setting time
of fly ash blended concrete. Thongsanitgarn et al. [5,6] found that
cement-fly ash-limestone ternary blended concrete presents
higher strength compared to cement-fly ash binary concrete. Lime-
stone with finer particle size is effective to speed up cement-fly ash
hydration. Celik et al. [7,8] found that concrete containing high vol-
ume fly ash and limestone has high workability, high late age
strength, high chloride resistance, and low global warming poten-
tial. Ghrici et al. [9] found that natural pozzolana did not change
the sulfate resistance of limestone cement, natural pozzolana
improved hydrochloric acid resistance, and limestone improved
sulfuric acid resistance.

E-mail address: wxbrave@kangwon.ac.kr

https://doi.org/10.1016/j.conbuildmat.2018.01.058
0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
 X.-Y. Wang / Construction and Building Materials 166 (2018) 130–140
Compared with abundant experimental studies of ternary
blended concrete [2–9], theoretical studies of ternary blended con-
crete are relatively limited. Gao et al. [10] put forward an analytical
modeling about the hydration of cement-slag-limestone blends.
The local water to cement ratio and water transport in the interfa-
cial transition zone was simulated. Maekawa et al. [11] proposed a
computational platform which can assess the time-dependent
mechanics and durability mechanics of concrete structures. The
couplings among hydration, mass transport, and damage evolution
were considered. However, Gao et al. [10] and Maekawa et al.’s
[11] studies do not consider the chemical reaction of limestone
in ternary blended concrete. Bentz [12] proposed a model which
simulated the dilution, nucleation, and chemical effects of lime-
stone additions in cement-limestone binary blends. Weerdt et al.
[13] made a thermodynamic modeling of hydration of cement-fly
ash-limestone ternary blends. The volumetric phase fractions of
the hydrating paste were calculated using Gibbs Energy Minimiza-
tion Software (GEMS) program. However, the thermodynamic
modeling in the Weerdt et al. study [13] mainly focuses on the
chemical aspects of ternary blended concrete. Regarding the opti-
mal combinations of cement, fly ash, and limestone in ternary
blended concrete, current models [10–13] still do not cover this
point.
To avoid the flaws of current studies, this research presents a
hydration based simulation program to assess the strength growth
and optimal combinations of ternary blended concrete. The
strength is evaluated considering the influences from reactions of
cement, fly ash, and limestone. The optimal combinations are
determined based on parameter studies of the hydration-
strength integrated model.
2. Simulation of the hydration of cement-fly ash-limestone
ternary blends
2.1. Simulation of the hydration of cement-fly ash binary blends
Our previous studies [14–16] presented a kinetic hydration
model for concrete containing fly ash. This kinetic hydration model
includes three sub-models, i.e. model for hydration of cement,
model for reaction of fly ash, and model for mutual effects between
reactions of cement and fly ash. The cement hydration model con-
siders the kinetic stages involved in the hydration of cement, such
as initial dormant stage, chemical-reaction related stage, and diffu-
sion related stage. The cement hydration model also considers the
water withdrawal on account of the lack of capillary water regard-
ing high strength concrete. The equation of cement hydration is
simplified and written as follows [14–16]:
da
¼ f ðkd ðTÞ; De ðTÞ; kr ðTÞ; r0 Þ  C wfree  ðSw =S0 Þ
dt
where ddta is rate of hydration, kd is rate of coefficient in initial dor-
mant stage, T is curing temperature, kr is rate of coefficient in reac-
tion related stage, De is rate of coefficient in diffusion related stage,
r 0 is unreacted cement particles radius, Sw means the effective con-
tact area between the surrounding capillary water and cement par-
ticles [14–16], S0 means the total area if cement hydration products
progress unconstrained, C wfree is capillary water content
 r
(C wfree ¼ W 0 0:4C 0a
where C 0 is cement content in concrete mix-
W0
ing proportions, W 0 is the content of water in the proportions of
concrete mix, r (r ¼ 2:6  4 W
C0
0
) is an empirical factor considering
the approachability of capillary water from outer hard shell to inner
anhydrous part of cement particles).
The rate coefficients kd , kr , and De can be determined based on
compound compositions of cement [14–16]. The effect of
131
temperature on rate of hydration of cement is recognized as fol-
lowing Arrhenius’s law [14–16]. For high strength concrete, degree
of hydration is reduced due to the withdrawal of capillary water.
The water withdrawal mechanism is considered through ðSw =S0 Þ
and C wfree in Eq. (1). ðSw =S0 Þ describes the decrease in contact area
between cement particle and ambient capillary water, and C wfree
describes the decrease in capillary water concentration.
Similarly with hydration of cement, fly ash reaction also con-
sists of three stages [14–16], initial dormant stage, chemical-
reaction related stage, and diffusion related stage. In addition, fly
ash reaction is dependent on calcium hydroxide content in
cement-fly ash binary blends. The equation of fly ash reaction is
simplified written as below [14–16]:
daFA mCH ðtÞ
¼ f ðkdFA ðTÞ; DeFA ðTÞ; krFA ðTÞ; r FA0 Þ  ð2Þ
dt P
where dadtFA is fly ash reaction rate, kdFA is rate of coefficient in initial
dormant stage of fly ash, DeFA is rate of coefficient in diffusion
related stage of fly ash, krFA is rate of coefficient in reaction related
stage of fly ash, r FA0 is unreacted fly ash particles radius, mCH ðtÞ is
the content of calcium hydroxide, P is fly ash content in proportions
of concrete mix.
The time-dependent cement or fly ash reaction degrees can be
calculated using the cement-fly ash binary hydration model. In
addition, the thermal properties, mechanical properties, and dura-
bility of fly ash blended concrete can be evaluated using reaction
degree of individual component of binders. The cement-fly ash bin-
ary hydration model is multiply validated using experimental data
for concrete with various proportions of mix and curing conditions.
However, because the cement-fly ash binary hydration model does
not consider the effect of limestone, binary hydration model can-
not analyze the hydration of cement-fly ash-limestone ternary
concrete.
2.2. Limestone powder reaction mode
Bentz [12] reported that the addition of limestone presents
dilution, nucleation, and chemical effects on cement hydration.
The dilution effect happens when limestone substitutes partial
cement, cement content decreases and water to cement ratio
increases correspondingly. The nucleation effect is the fact that
limestone may work as a nucleation site of hydrating
cement. Hydration of cement can accelerate due to the nucle-
ation effect. The chemical effect is the formation of monocar-
boaluminate due to the limestone reaction in preference to a
monosulfoaluminate.
In this research, the dilution effect of limestone powder can be
C0
considered by W term in C wfree of Eq. (1). Regarding the nucleation
0
ð1Þ effect, Maekawa et al. [11] and Wang [16] proposed that the nucle-
ation effect of limestone relates to the ratio of surface area of
cement particles to that of limestone powder. The nucleation effect
indicator of limestone powder can be written as follows [16]:
LS0  SLS
Lr ¼ ð3Þ
C 0  SC
where Lr denotes the indicator of the nucleation effect from lime-
stone addition, LS0 denotes the limestone mass in proportions of
concrete mix, SLS denotes the specific surface (Blaine) of limestone,
and SC denotes the specific surface (Blaine) of cement. Maekawa
et al. [11] reported that the nucleation effect of limestone is signif-
icant in the reaction related stage and the diffusion related stage. In
our former study [16], based on the experimental data of hydration
degree of cement in cement-limestone binary blends, Wang [16]
proposed that the nucleation effect of limestone powder can
described as follows:
132 X.-Y. Wang / Construction and Building Materials 166 (2018) 130–140
krLS ¼ kr ð1 þ 1:2Lr Þ
DeLS ¼ De ð1 þ 1:2Lr Þ
where krLS is the updated phase boundary reaction coefficient in
cement-limestone blends, 1.2 is enhance coefficients of kr [16],
DeLS is the updated diffusion coefficient in cement-limestone blends,
and 1:2 is enhance coefficients of De [16].
Ipavec et al. [17] measured the formation of carboaluminate
phases in limestone powder blended paste (the water to binder
ratio is 0.5, the limestone replacement ratio is 0.2, and curing tem-
perature is 20 °C). The contents of carboaluminate phases were
measured at 1, 3, 7, 15, 28, and 100 days. Monocarboaluminate is
the main hydration product of limestone at late ages. Based on
monocarboaluminate contents, we suggest that the limestone
reaction degree can be calculated as below:
aLS1 ¼ 0:0087 lnðtÞ  0:0265 ðt > 21Þ
where aLS1 is the reaction degree of limestone powder, and t is age
(hours). The evaluation results about reaction degree of limestone
are shown in Fig. 1. Fig. 1 shows that the reaction degree of
limestone is a logarithm function of time, which is similar to the
relation between hydration degree of cement and curing age
[14–16]. Fig. 1 also shows that the reaction of limestone starts after
about 21 h, not immediately after mixing. Bentz [12] proposed that
the limestone chemical reaction can be written as 3ðCaOÞ3 ðAl2 O3 Þ
CaSO4  12H2 O þ 2CaCO3 þ 18H2 O ! 2ðCaOÞ3 ðAl2 O3 Þ  CaCO3  11H2 O þ
ðCaOÞ3 ðAl2 O3 Þ  3CaSO4  32H2 O. Bentz’s [12] study shows that the
limestone reaction starts only when the initial calcium sulfate
is fully consumed and the formed ettringite phase begins to convert
to the AFm phase. Lothenbach [18] reported that after about 1 day,
the reaction of limestone starts. The starting time of limestone in
Lothenbach’s [18] study (1 day) is similar to our study (21 h). In
addition, by using Eq. (6), we find that at the age of 180 days, the
reaction degree of limestone is about 4.6%, which is similar to
the results of Bentz’s study [12] (Bentz [12] proposed that for
concrete with 20% limestone, at the age of 180 days, about 5%
limestone reacted). The reactivity of limestone is much lower
compared to cement or fly ash [14–16].
On the other hand, we should notice that Eq. (12) is only valid
for Ipavec et al.’s [17] study (water to binder ratio was 0.5, lime-
stone replacement ratio was 0.2, and curing temperature was
_20 °C). Limestone reaction is a complex process and relates to
abundant factors. The factors, such as limestone replacement
ratios, fly ash addition, limestone fineness, cement fineness, water
Fig. 1. Reaction of limestone.
ð4Þ to binder ratio, and curing temperature, will affect the limestone
reaction [19,20]. Considering these points, we propose a more
ð5Þ general equation for determining the reaction degree of limestone
as follows:
aLS ¼ aLS1  m1  m2  m3  m4  m5  m6 ð7Þ
where m1 considers the effect of limestone replacement ratios on
reaction degree of limestone, m2 considers the effect of limestone
fineness, m3 considers the effect of cement fineness, m4 considers
the effect of fly ash addition, m5 considers the effect of water to bin-
der ratio, and m6 considers the effect of curing temperature. For the
case of Ipavec et al.’s [17] study, all of these influencing factors
equal to 1.
2.2.1. Effect of limestone replacement ratios
Antoni et al. [20] reported that as limestone replacement ratio
ð6Þ increases, the reaction degree of limestone powder decreases.
Based on Antoni’s [20] results, we found that the reaction degree
of limestone is approximately an inverse proportional function of
limestone replacement ratio. Hence, we assume that m1 ¼ 0:2 LS0 .
C 0 þLS0
2.2.2. Fineness of limestone
Aqel and Panesar [19] reported that when average particle size
of limestone powder decreases, the reactivity of limestone
increases. Based on Aqel’s study [19] about the relation between
reaction degree and particle size of limestone, we assumed that
m2 ¼ 1:0131  0:0144  dLS where dLS is average diameter of
limestone.
2.2.3. Fineness of cement
Aqel and Panesar [19] found that when the Blaine surface of
cement increases, the average reaction degree of limestone also
increases. Based on Aqel’s [19] study, we assumed that that
m3 ¼ 0:55 SSC1C þ 0:45 where SC1 is the Blaine surface of cement used
in Ipavec et al.’s [17] study.
2.2.4. Fly ash additions
Weerdt et al. [2] reported that due to the high aluminum con-
tent in fly ash, the limestone reaction in cement-fly ash-
limestone ternary blends is enhanced. We proposed that the influ-
ence of fly ash on limestone addition can be described as
m4 ¼ 1 þ AlAlFAC aaCFA0P, where AlFA and AlC are aluminum contents in fly
ash and cement respectively. AlFA aFA P and AlC aC 0 are reacted alu-
minum contents from fly ash reaction and cement reaction
respectively.
2.2.5. Water to binder ratios
Similar with cement hydration, the reaction products of lime-
stone deposits in the pore space of concrete. We assumed that
when the water to binder ratio increases, the reaction degree of
limestone increase consequently. We proposed that m5 ¼ aa0:5
where a0:5 is reaction degree of cement for the case of water to bin-
der ratio 0.5.
2.2.6. Curing temperature
Curing temperature presents a two-sided effect on limestone
reactivity. First, as curing temperature increases, the solubility of
limestone decreases which will lower limestone reactivity. How-
ever, cement hydration is accelerated due to the curing tempera-
ture increasing. The increasing of cement reactivity may increase
limestone reactivity. Aqel and Panesar [19] found that compared
to other factors, such as moist curing duration, cement fineness,
and limestone content, the influence of curing temperature on
 X.-Y. Wang / Construction and Building Materials 166 (2018) 130–140 133

Table 1 on experimental results of Weerdt et al. [13], we proposed that

Summary of influencing factors of limestone reaction. when 1 g of limestone reacts, 0.35 g of calcium hydroxide will be
Factor Equation Influencing trend consumed. Regarding cement-fly ash-limestone ternary blends,
Limestone m1 ¼ 0:2 As limestone replacement CH content can be determined as below:
LS0

ratio increases, aLS decreases

C 0 þLS0
replacement
ratios
Fineness of m2 ¼ 1:0131  0:0144 dLS As fineness of limestone
increases, aLS increases
limestone
Fineness of m3 ¼ 0:55 SSC1C þ 0:45 As fineness of cement
cement increases, aLS increases
Fly ash m4 ¼ 1 þ AlFA aFA P
AlC aC 0
As fly ash addition increases,
additions aLS increases
Water to binder m5 ¼ aa0:5 As water to binder ratio
ratios increases, aLS increases
Curing m6 ¼ 1 Curing temperature do not
temperature influence aLS
limestone reactivity is marginal. Based on Aqel and Panesar’s [19]
results, we approximately assumed the m6 ¼ 1:
The summaries of influencing factors of limestone reaction are
shown in Table 1. Our proposed model considers the dilution,
nucleation, and chemical effects of limestone. We consider dilution
C0
effect through capillary water concentration (W term in C wfree of
0
Eq. (1)). We consider nucleation effect through nucleation effect
indicator (Eq. (3)). We consider chemical effect using a logarithm
function and multiple modification factors (Eqs. (6) and (7)). The
modification factors reflect the general trends of reaction degree
of limestone powder. However, because the available experimental
results about reaction degree of limestone are very limited, the cal-
ibrations of modification factors still need further study. Other
influencing factors, such as the aluminum and gypsum content in
cement, also need more study. The heat liberation by limestone
reaction is not considered in the current study.
2.3. Interactions model among reactions of cement, fly ash, and
limestone
In this study, the contents of calcium hydroxide (CH) and capil-
lary water of hydrating ternary blends are adopted as fundamental
indicators to consider mutual interactions among reactions of
cement, fly ash, and limestone. Bentz [12] proposed that when 1
g limestone powder reacts, 1.62 g of water will be consumed.
The consumed water of limestone power is much higher compared
to cement or fly ash. This is because the reaction products of lime-
stone powder are monocarboaluminate and ettringite which con-
tain abundant water. For cement-fly ash-limestone ternary
blends, the content of capillary water W cap can be determined as
below:
W cap ¼ W 0  0:4  C 0  a  0:25  aFA  P  1:62  LS0  aLS
where 0:4  C 0  a, 0:25  aFA  P; and 1:62  LS0  aLS are the con-
sumed water from reactions of cement, fly ash, and limestone
respectively.
Cement hydration, fly ash reaction, and limestone reaction will
contribute to the formation of chemically bound water. The chem-
ically bound water W cbm can be determined as follows:
W cbm ¼ v  C 0  a þ 0:1  P  aFA þ 1:62  LS0  aLS
where v  C 0  a, 0:1  P  aFA ; and 1:62  LS0  aLS are the contents of
produced chemically combined water from reactions of cement, fly
ash, and limestone respectively.
Weerdt et al. [13] measured calcium hydroxide contents for
hydrating cement-limestone blends. They found that the calcium
hydroxide (CH) content in limestone blended concrete is lower
compared to control specimen. CH is consumed on account of
the production of hemicarbonate from limestone reaction. Based
CHðtÞ ¼ RCHCE  C 0  a  v FA  aFA  P  0:35  aLS  LS0
ð10Þ
where RCHCE means the produced CH as 1 unit mass of cement
hydrates, v FA means the consumed CH as 1 unit mass of fly ash
reacts. RCHCE  C 0  a is the produced CH from cement hydration.
v FA  aFA  P and 0:35  aLS  LS0 are the consumed CH from reactions
of fly ash and limestone respectively.
Summarily, from Sections 2.1–2.3, a simulation program is put
forward which simulates the hydration of cement-fly ash-
limestone blends. The interactions among reactions of cement, fly
ash, and limestone are taken into account by means of the contents
of calcium hydroxide and capillary water. This simulation program
considers the dilution, nucleation, and chemical effects from fly ash
and limestone additions. The input parameters of this numerical
procedure are concrete mixing proportions, physical and chemical
properties of binders, and curing conditions. Furthermore, by using
kinetic reaction equations, the time-dependent reaction degrees of
individual components can be determined. In addition, it should be
noticed that the reaction coefficients of proposed ternary hydration
model do not vary with proportions of concrete mix. As the combi-
nations of cement, fly ash, limestone, and water change from one
mix to another, the reaction coefficients of ternary hydration
model are constant.
2.4. Compressive strength development model
The products of the reaction of cement, fly ash, and limestone
powder will fill the pore space of concrete. Concrete compressive
strength develops with the progress of binders reaction. According
to Powers’ strength theory, concrete compressive strength can be
evaluated as below:
f c ðtÞ ¼ Axnc ð11Þ
where f c ðtÞ is concrete compressive strength, A is the intrinsic
strength of concrete, xc is gel-space ratio of concrete, and n is the
strength exponent. The gel-space ratio denotes the volumetric ratio
of binder hydration products to the sum of hydrated binders and
capillary pore. For cement-fly ash-limestone blends, 1 ml hydrated
cement occupies 2.06 ml of space [21–24], 1 ml reacted fly ash
occupies 2.52 ml space [21–24], and 1 ml reacted limestone occu-
pies 4.1 ml space [12]. Reacted products of 1 ml limestone can
occupy much higher space that that of cement (4.1 vs. 2.06). This
is because of the formation of ettringite and monocarboaluminate
from the limestone reaction. Considering the reactions of cement,
ð8Þ fly ash, and limestone, the gel-space ratio of cement-fly ash-
limestone ternary blended cement can be determined as follows:
2:06ð1=qC ÞaC 0 þ 2:52ð1=qFA ÞaFA P þ 4:1ð1=qLS ÞaLS LS0
xc ¼ ð12Þ
ð1=qC ÞaC 0 þ ð1=qFA ÞaFA P þ ð1=qLS ÞaLS LS0 þ W 0
where qFA and qLS are density of fly ash and limestone powder
respectively.
For ternary blends, individual binder component will affect the
ð9Þ intrinsic strength of concrete and strength exponent. We assume
that the intrinsic strength of concrete A and strength exponent n
can be written as functions of weight percentages of cement, fly
ash, and limestone in the proportions of mix as follows:
C0 P LS0
A ¼ a1 þ a2 þ a3 ð13Þ
C 0 þ P þ LS0 C 0 þ P þ LS0 C 0 þ P þ LS0
C0 P LS0
n ¼ b1 þ b2 þ b3 ð14Þ
C 0 þ P þ LS0 C 0 þ P þ LS0 C 0 þ P þ LS0
134 X.-Y. Wang / Construction and Building Materials 166 (2018) 130–140

where coefficients a1 a2; and a3 in Eq. (13) denote the influences of

Se
cement, fly ash, and limestone towards the intrinsic strength of con-
crete, respectively, and the units of a1, a2, and a3 are MPa; the coef-
ficients b1, b2 and b3 in Eq. (14) denote the influences of cement, fly
ash, and limestone towards strength exponent, respectively. For
neat Portland cement concrete, the weight fractions of limestone
or fly ash are zero, the strength of concrete only relates to a1 and
b1. For fly ash blended binary concrete, the weight fraction of lime-
stone is zero, and the strength of concrete relates to coefficients a1,
a2, b1, and b2. For ternary blended concrete, concrete strength
depends to coefficients a1, a2, a3, b1, b2, and b3. These coefficients
a1, a2, a3, b1, b2, and b3 do not change for various concrete
mixtures.
The flowchart of the calculation is shown in Fig. 2. At every time
step, the reaction degrees of individual components are calculated
by using the ternary blended hydration model. The contents of CH,
chemically combined water, and capillary water are determined by
using reaction degrees of binders. Furthermore, the gel-space ratio
of hydrating concrete is calculated taking into account the contri-
butions from reactions of cement, fly ash, and limestone. By using
Powers’ strength theory, concrete compressive strength is
calculated.

3. Verifications of proposed model

Experimental results from references [2,13] about strength and

reaction degrees of binders in ternary blended concrete are used to
validate the proposed hydration-strength integrated model. Table 2
shows the mixtures of specimens. Table 3 shows the chemical
compositions and physical characteristics of cement, fly ash, and
limestone. Regarding the compressive strength, a total of 21 mix-
ing proportions are used including control specimen, limestone
blended binary mortar, fly ash blended binary mortar, and ternary
blended mortar. The replacement ratio for limestone or fly ash in
Fig. 2. Flowchart of simulation.
binary blended specimen ranged between 5% and 35%. The replace-
Table 2 ment ratio of mineral admixtures in ternary blended specimen was
Mixing proportions of mortar specimens. 35% (the fly ash or limestone content ranged between 5% and 30%,
and the sum of contents of fly ash and limestone was constant at
Mortar Portland cement (%) Fly ash Limestone Water to power ratio
(%) (%) 35%). The specimens for compressive strength measurement were
mortar specimen, and the specimens for degree of reaction were
1 100 0 0 0.5
2 95 0 5 0.5 paste specimen. For mortar specimen, the sand to power ratio
3 90 0 10 0.5 was 3:1. At the ages of 1, 28, 90, and 140 days, the compressive
4 85 0 15 0.5 strength was measured. For paste specimen (corresponding
5 80 0 20 0.5 to specimens No. 1, 2, 15, 16 in Table 2), the cement and fly ash
6 75 0 25 0.5
7 70 0 30 0.5
reaction degrees were measured at the age of 1, 7, 28, 90, and
8 65 0 35 0.5 180 days.
9 95 5 0 0.5
10 90 10 0 0.5
11 85 15 0 0.5 3.1. Hydration reaction of ternary blends
12 80 20 0 0.5
13 75 25 0 0.5 Based on the hydration model and specimen mixtures, the
14 70 30 0 0.5 development of reaction degree of cement, fly ash, and limestone
15 65 35 0 0.5
can be calculated. Fig. 3 shows hydration related properties of
16 65 30 5 0.5
17 65 25 10 0.5 cement-fly ash-limestone blends. As shown in Fig. 3a, compared
18 65 20 15 0.5 to the control specimen, reaction degree of cement in cement-
19 65 15 20 0.5 limestone binary blends (95% cement + 5% limestone) is slightly
20 65 10 25 0.5
higher. This is because of the dilution and nucleation effects from
21 65 5 30 0.5
limestone addition. In cement-fly ash binary blends (65% cement

Table 3
Properties of binders.

SiO2 Al2O3 Fe2O3 CaO MgO SO3 K2O Na2O Blaine surface
(%) (%) (%) (%) (%) (%) (%) (%) (m2/kg)
Cement 20.0 5.4 3.1 60.6 2.9 1.5 1.2 0.5 450
Fly ash 50.0 23.9 6.0 6.3 2.1 0.4 1.4 0.6 450
Limestone 12.9 2.7 2.0 42.3 1.8 – 0.6 0.5 810
 X.-Y. Wang / Construction and Building Materials 166 (2018) 130–140 135

100 50
control FA35%-cal
reaction degree of cement %

reaction degree of fly ash %

LS5% FA35%-expe
80 40
FA35% FA30%+5%LS-cal
FA30%+5%LS FA30%+5%LS-expe
60 exper. res 30

40 20

20 10

0 -2 0 2 4
0 0 2 4
10 10 10 10 10 10 10
curing age(hours) curing age(hours)

(a)Reaction degree of cement. (b) Reaction degree of fly ash.

limestone binary blended concrete
0.2

reaction degree of limestone powder

0.15 LS5%
LS15%
LS25%
LS35%
0.1

0.05

0
0 1000 2000 3000 4000 5000
curing age(hours)

(c)CH content in cement-limestone (d) Reaction degree of limestone in

binary blends. cement-limestone binary blends.
cement 65%+FA 20%+Limestone15%
1 unhydrous cement
unreacted fly ash
0.8 unreacted limestone
phase volume fractions

cement hydration products

fly ash reaction products
0.6 limestone reaction products
porosity

0.4

0.2

0 0 2
10 10
time(hours)

(e) Reaction degree of limestone in (f) Phase volume fractions of ternary

ternary blends
. blends
.
Fig. 3. Hydration of cement-fly ash-limestone ternary blended cement.

+ 35% fly ash), due to the dilution effect from fly ash addition, blends. The sequence of reaction degree of cement from high to
cement reaction degree is higher than control paste. The ternary low is ternary blends, fly ash binary blends, limestone binary
blends (65% cement + 30% fly ash + 5% limestone) has a higher blends, and control specimen. The proposed hydration model can
reaction degree of cement than the control specimen or binary reproduce this sequence of reaction degree of cement.
136 X.-Y. Wang / Construction and Building Materials 166 (2018) 130–140

Fig. 3b shows fly ash reaction degree in binary or ternary lower compared to binary mix (5% limestone + 95% cement). This is
blended concrete. When 5% limestone is added, the fly ash reaction because the cement to limestone ratio is lower in ternary mix
degrees in binary blends (65% cement + 35% fly ash) or ternary compared to binary mix (related to influencing factor m1 of
blends (65% cement + 30% fly ash + 5% limestone) are similar. First, limestone reaction).
the limestone addition can lower CH content which will decrease Based on reaction degree of binders and volumetric ratio
the rate of fly ash reaction. Second, when partial fly ash is replaced between reaction products and binders, the volumetric phase frac-
by limestone, the ratio between cement and fly ash increases tions of hydrating ternary paste can be calculated. Fig. 3f shows the
which will increase the rate of fly ash reaction. On account of the volumetric phase fractions of ternary paste (cement 65% + fly ash
simultaneous effects of increasing factor and decreasing factor, 20% + limestone 15%). Because cement hydrates much quicker
fly ash reaction degrees in binary blends or ternary blends are compared to fly ash, less cement is remained compared to fly
similar. ash. On account of the depositing of reaction products, capillary
Fig. 3c shows CH content in limestone binary blends (95% porosity decreases as hydration evolves.
cement + 5% limestone) and control concrete. On account of the
consumption of CH from limestone addition, limestone binary 3.2. Strength development and strength optimization of concrete
blends present lower CH contents compared to control
specimen. Fig. 3d shows that regarding cement-limestone binary Based on the reaction degree of binders, the gel-space ratio of
blends, the reaction degree of limestone decreases as limestone hardening concrete can be determined. Moreover, concrete com-
content increases. Fig. 3e shows that at late age, fly ash (5% pressive strength can be calculated using Powers’ strength theory.
limestone + 30% fly ash + 65% cement) can increase the reaction By using experimental data of concrete compressive strength with
degree of limestone compare to limestone-cement binary blends various mixing proportions at different ages, these coefficients a1,
(5% limestone + 95% cement). This is because of the high aluminum a2, a3, b1, b2, and b3 can be calibrated. The values of intrinsic
content in fly ash (related to influencing factor m4 of limestone strength of cement, fly ash, and limestone are 184.89 MPa,
reaction). While at early-age, limestone powder reaction degree 207.11 MPa, and 122.09 MPa respectively. The values of strength
in ternary mix (5% limestone + 30% fly ash + 65% cement) is slightly exponent of cement, fly ash, and limestone are 2.45, 3.04, and

limestone binary blended concrete limestone binary blended concrete

1 1.5
relative reaction degree of cement
reaction degree of cement

0.8
OPC
10%ls 1
0.6 20%ls
30%ls OPC
10%ls
0.4 20%ls
0.5 30%ls

0.2

0 0
0 50 100 150 200 0 50 100 150 200
curing age(days) curing age(days)

(a)Reaction degree of cement. (b) Relative reaction degree of cement.

limestone binary blended concrete limestone binary blended concrete

70 control 1.5
5%LS
relative compressive strength

60
compressive strength(MPa)

10%LS
15%LS
50
20%LS 1
25%LS
40
30%LS
35%LS
30
exper. res.
0.5 OPC
20
10%ls
20%ls
10
30%ls
0 -1 0 1 2
0 -1 0 1 2
10 10 10 10 10 10 10 10
curing age(days) curing age(days)

(c) Compressive strength. (d)Relative compressive strength.

Fig. 4. Strength evaluation of cement-limestone binary blends.
 X.-Y. Wang / Construction and Building Materials 166 (2018) 130–140 137

2.39 respectively. These coefficients do not change for various mix-
ing proportions of concrete (total 21 mixtures) and various curing
ages (from early age (1 day) to late ages (28, 90, and 140 days)). A
total of 84 experimental data about compressive strength (21 * 4 =
84) are used in this research.
The development of strength of cement-limestone binary
blends is shown in Fig. 4. Fig. 4a shows cement reaction degree
in cement-limestone binary blends. As limestone content
increases, due to dilution effect and nucleation effect, cement
hydration degree increases. In this study, the relative hydration
degree of cement means the ratio of hydration degree of cement
in binary blends to that in plain specimen, and the relative strength
means the ratio of compressive strength of binary blends to that of
plain specimen. As shown in Fig. 4b, as limestone content
increases, relative reaction degree of cement also increases.
Fig. 4c shows that the analysis results about compressive strength
generally conform to experimental data. Fig. 4d shows that at
early-age, limestone addition can increase concrete strength. This
is because of the acceleration of hydration of cement. While at late
age, due to dilution effect, limestone blended concrete presents
lower strength compared to control concrete. It should be noted
that the trend of relative degree of hydration (Fig. 4b) is different
from that of relative compressive strength (Fig. 4d). The compres-
sive strength relates to not only degree of hydration, but also the
contents of cement and limestone.
The analysis of strength development of cement-fly ash binary
blends is shown in Fig. 5. Fig. 5a shows fly ash reaction degree in
binary blends. As fly ash content increases, the alkali activation
effect becomes weaker, and fly ash reaction degree decreases.
Fig. 5b shows that the analysis results generally conform to exper-
imental data. Fig. 5c shows that at early-age, fly ash additions
lower concrete strength. This is because fly ash reaction rate is
much lower compared to cement (shown in Fig. 3b and a). While
at late age, due to the progress of the fly ash reaction, the strength
of concrete containing fly ash can surpass the strength of plain con-
crete. As fly ash content increases, fly ash reaction degree becomes
lower (shown in Fig. 5a), and age akin to exceed strength is going
to be delayed.
Fig. 6 presents the analyzed compressive strength of ternary
blends. At long term age, the specimens containing higher fly ash
content (Fig. 6a and b) have higher compressive strength than
those containing lower fly ash contents (Fig. 6e and f). The results
of analysis and tests are shown in Fig. 7. The relative coefficient
between analysis and experiment results is 0.99, and the root

fly ash binary blended concrete fly ash binary blended concrete
1 70 control
10%Fa
5%FA
20%Fa 60
compressive strength(MPa)

10%FA
reaction degree of fly ash

0.8 35%Fa
15%FA
50
20%FA
0.6 25%FA
40
30%FA
30 35%FA
0.4 exper. res.
20
0.2
10

0 -1 0 1 2
0 -1 0 1 2
10 10 10 10 10 10 10 10
curing age(days) curing age(days)

(a)Reaction degree of fly ash. (b)Compressive strength.

fly ash binary blended concrete

1.5
relative compressive strength

OPC
10%Fa
0.5 20%Fa
35%Fa

0 -1 0 1 2
10 10 10 10
curing age(days)

(c)Relative compressive strength.

Fig. 5. Strength evaluation of cement-fly ash binary blends.
138 X.-Y. Wang / Construction and Building Materials 166 (2018) 130–140

mean square error (RMSE) between analysis and experiment is
1.48 MPa. The analysis results generally conform to the experi-
mental data. The proposed simulation program can evaluate the
strength of binary and ternary concrete at both early ages and late
ages.
Based on the hydration-strength integrated model, we perform
parameter studies about strength of ternary blended concrete with
various binder combinations at various ages. As shown in Fig. 8a
(0.3 day) and Fig. 8b (0.5 day), because limestone addition can
improve concrete early age strength, concrete with higher

LS 5%+ FA30% LS 10%+ FA 25%

70 70
analysis analysis
60 experiment 60 experiment
compressive strength(MPa)

compressive strength(MPa)
50 50

40 40

30 30

20 20

10 10

0 -1 0 1 2
0 -1 0 1 2
10 10 10 10 10 10 10 10
curing age(days) curing age(days)

(a) LS 5%+FA 30%. (b) LS 10%+FA 25%.

LS 15%+ FA 20% LS 20%+ FA15%
70 70
analysis analysis
60 experiment 60 experiment
compressive strength(MPa)

compressive strength(MPa)

50 50

40 40

30 30

20 20

10 10

0 -1 0 1 2
0 -1 0 1 2
10 10 10 10 10 10 10 10
curing age(days) curing age(days)

(c) LS 15%+FA 20%. (d) LS 20%+FA 15%.

LS 25%+ FA 10% LS 30%+ FA 5%
70 70
analysis analysis
60 experiment 60 experiment
compressive strength(MPa)

compressive strength(MPa)

50 50

40 40

30 30

20 20

10 10

0 -1 0 1 2
0 -1 0 1 2
10 10 10 10 10 10 10 10
curing age(days) curing age(days)

(e) LS 25%+FA 10%. (f) LS 30%+FA 5%.

Fig. 6. Strength evaluation of ternary blends.(a) LS 5%+FA 30%. (b) LS 10%+FA 25%. (c) LS 15%+FA 20%. (d) LS 20%+FA 15%.
 X.-Y. Wang / Construction and Building Materials 166 (2018) 130–140 139

70 12

optimum limestone replacement ratio %

LS binary blended concrete
60 FA binary blended concrete
10 0.3 day
ternary blended concrete
analysis results(MPa)

50
8
40
6
30 0.5 day
4 28 days 90 days
20

10 2

0 0
0 10 20 30 40 50 60 70 0 5 10 15 20 25
experimental results(MPa) optimum fly ash replacement ratio %

Fig. 7. Comparison between analysis and experimental results. Fig. 9. Optimum combinations of ternary blends.

limestone content and lower fly ash contents has the highest the highest strength of all mixes. As shown in Fig. 9, from early-
strength of all other mixes. As shown in Fig. 8d (90 days), because age to late age, the optimal combination for ternary blends shifts
fly ash can improve concrete late age strength, concrete with from high limestone-low fly ash zone to high fly ash-low limestone
higher fly ash contents and lower limestone contents presents zone.

11
25 25 15
fly ash replacement ratio %

fly ash replacement ratio %

10 14
20 20
13
9
15 15
12
8
10 10 11

7 10
5 5
9
6
0 0
0 2 4 6 8 10 12 0 2 4 6 8 10 12
limestone replacement ratio % limestone replacement ratio %
(a) 0.3 day. (b)0.5 day.

25 48 25 57
fly ash replacement ratio %

fly ash replacement ratio %

56
20 46 20
55

44 54
15 15
53
10 42 10 52
51
5 40 5
50

0 38 0 49
0 2 4 6 8 10 12 0 2 4 6 8 10 12
limestone replacement ratio % limestone replacement ratio %

(c) 28 days. (d) 90 days.

Fig. 8. Parameter studies of strength of ternary blends.
140 X.-Y. Wang / Construction and Building Materials 166 (2018) 130–140
4. Conclusions
This research presents a simulation program for analyzing the
hydration and strength optimization of ternary blended concrete.
First, a kinetic hydration model is put forward for simulating
the hydration reaction of cement-fly ash-limestone ternary blends.
The interactions among reactions of cement, fly ash, and limestone
are taken into account by means of the contents of calcium hydrox-
ide and capillary water. This simulation program considers the
dilution, nucleation, and chemical effects from limestone addi-
tions. The input parameters of this numerical procedure are con-
crete mixing proportions, physical and chemical properties of
binders, and curing conditions.
Second, by using kinetic hydration reaction equations, the time-
dependent cement, fly ash, and limestone reaction degrees are
determined. The compressive strength development of hardening
ternary blended concrete is evaluated by means of gel-space ratio
and Powers’ strength theory.
Third, through parameter studies, the optimal combinations of
cement, fly ash, and limestone at different ages are determined.
From early-age to late age, the optimal combination for ternary
blends shifts from high limestone-low fly ash zone to high fly
ash-low limestone zone. The proposed hydration-strength inte-
grated model is valuable for making composite cements regarding
compressive strength.
Acknowledgements
This research was supported by the Basic Science Research Pro-
gram through the National Research Foundation of Korea (NRF)
funded by the Ministry of Science, ICT & Future Planning (No.
2015R1A5A1037548), and a NRF grant (NRF-2017R1C1B1010076).
References
[1] M.F. Carrasco, G. Menéndez, V. Bonavetti, E.F. Irassar, Strength optimization of
tailor-made cement” with limestone filler and blast furnace slag, Cem. Concr.
Res. 35 (2005) 1324–1331.
[2] K. De Weerdt, K.O. Kjellsen, E. Sellevold, H. Justnes, Synergy between fly ash
and limestone powder in ternary cements, Cem. Concr. Compos. 33 (2011) 30–
38.
[3] Axel Schöler, Barbara Lothenbach, Frank Winnefeld, Maciej Zajac, Hydration of
quaternary Portland cement blends containing blast-furnace slag, siliceous fly
ash and limestone powder, Cem. Concr. Compos. 55 (2015) 374–382.
[4] Dale P. Bentz, Taijiro Sato, Igor de la Varga, W. Jason Weiss, Fine limestone
additions to regulate setting in high volume fly ash mixtures, Cem. Concr.
Compos. 34 (2012) 11–17.
[5] Pailyn Thongsanitgarn, Watcharapong Wongkeo, Arnon Chaipanich, Hydration
and compressive strength of blended cement containing fly ash and limestone
as cement replacement, J. Mater. Civ. Eng. 26 (2014) 1–5.
[6] Pailyn Thongsanitgarn, Watcharapong Wongkeo, Arnon Chaipanich, Chi Sun
Poon, Heat of hydration of Portland high-calcium fly ash cement incorporating
limestone powder: effect of limestone particle size, Constr. Build. Mater. 66
(2014) 410–417.
[7] Kemal Celik, Cagla Meral, A. Petek Gursel, P. Kumar Mehta, Arpad Horvath,
Paulo J.M. Monteiro, Mechanical properties, durability, and life-cycle
assessment of self-consolidating concrete mixtures made with blended
Portland cements containing fly ash and limestone powder, Cem. Concr.
Compos. 56 (2015) 59–72.
[8] K. Celik, M.D. Jackson, M. Mancio, C. Meral, A.-H. Emwas, P.K. Mehta, P.J.M.
Monteiro, High-volume natural volcanic pozzolan and limestone powder as
partial replacements for portland cement in self-compacting and sustainable
concrete, Cem. Concr. Compos. 45 (2014) 136–147.
[9] M. Ghrici, S. Kenai, M. Said-Mansour, Mechanical properties and durability of
mortar and concrete containing natural pozzolana and limestone blended
cements, Cem. Concr. Compos. 29 (2007) 542–549.
[10] Yun Gao, Geert De Schutter, Guang Ye, Haoliang Huang, Zhijun Tan, Wu. Kai,
Characterization of ITZ in ternary blended cementitious composites:
experiment and simulation, Constr. Build. Mater. 41 (2013) 742–750.
[11] K. Maekawa, T. Ishida, T. Kishi, Multi-scale modeling of structural concrete, J.
Adv. Concr. Technol. 1 (2003) 91–126.
[12] Dale P. Bentz, CEMHYD3D: A Three-Dimensional Cement Hydration and
Microstructure Development Modeling Package.Version 3.0. NISTIR 7232,
National Institute of Standards and Technology, Gaithersburg, MD, 2005.
[13] K. De Weerdt, M. Ben Haha, G. Le Saout, K.O. Kjellsen, H. Justnes, B.
Lothenbach, Hydration mechanisms of ternary Portland cements containing
limestone powder and fly ash, Cem. Concr. Res. 41 (2011) 279–291.
[14] Xiao-Yong Wang, Effect of fly ash on properties evolution of cement based
materials, Constr. Build. Mater. 69 (2014) 32–40.
[15] Xiao-Yong Wang, Simulation of temperature rises in hardening Portland
cement concrete and fly ash blended concrete, Mag. Concr. Res. 65 (2013) 930–
941.
[16] Xiao-Yong Wang, Modeling of hydration, compressive strength, and
carbonation of portland-limestone cement (PLC) concrete, Materials 10
(2017) 115–131.
[17] A. Ipavec, R. Gabrovsek, T. Vuk, V. Kaucic, J. Macek, A. Meden, Carboaluminate
phases formation during the hydration of calcite-containing portland cement,
J. Am. Ceram. Soc. 94 (2011) 1238–1242.
[18] B. Lothenbach, G.L. Saout, E. Gallucci, K. Scrivener, Influence of limestone on
the hydration of Portland cements, Cem. Concr. Res. 38 (2008) 848–860.
[19] M. Aqel, D.K. Panesar, Hydration kinetic and compressive strength of steam-
cured cement pastes and mortars containing limestone filler, Constr. Build.
Mater. 113 (2016) 359–368.
[20] M. Antoni, J. Rossen, F. Martirena, K. Scrivener, Cement substitution by a
combination of metakaolin and limestone, Cem. Concr. Res. 42 (2012) 1579–
1589.
[21] L. Lam, Y.L. Wong, C.S. Poon, Degree of hydration and gel/space ratio of high-
volume fly ash/cement systems, Cem. Concr. Res. 30 (2000) 747–756.
[22] K. van Breugel, Numerical simulation of hydration and microstructural
development in hardening cement-based materials (I) Theory, Cem. Concr.
Res. 25 (1995) 319–331.
[23] K. van Breugel, Numerical simulation of hydration and microstructural
development in hardening cement-based materials (II) Applications, Cem.
Concr. Res. 25 (1995) 522–530.
[24] D.P. Bentz, V. Waller, F. de Larrard, Prediction of adiabatic temperature rise in
conventional and high-performance concretes using a 3-D microstructural
model, Cem. Concr. Res. 28 (1998) 285–297.