Case Studies in Construction Materials 17 (2022) e01583

Contents lists available at ScienceDirect

Case Studies in Construction Materials

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

Case study

Early hydration and compressive strength of steam cured

high-strength concrete based on simplex centroid design method
Panqi Wang, Xiaoguang Li *, Xiaoliang Huo, Saddam Ali, Jiazhen Sun, Jinyi Qin,
Kun Liang, Yunxiao Liu
School of Civil Engineering, Chang’an University, Xi’an 710061, China

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

Keywords: To obtain the precast concrete elements with high strength at an early stage, and strike a balance
High-strength concrete among mechanical properties, economic benefits and sustainable development. The heat of hy­
Simplex centroid design method dration, bound water content, calcium hydroxide content (CH), compressive strength and
Steam cured
porosity of high-strength concrete (HSC) with cement–silica fume–fly ash micro-bead binder at
Hydration product
Mechanical properties
85 ◦ C were investigated. The ternary binder formulations were the designed using simplex
centroid design method. The results show that the appropriate amount of silica fume and fly ash
micro-bead increases the total heat release and compressive strength of HSC, and reduce the CH
content and porosity. Silica fume incorporation advances the induction period end time, while fly
ash micro-bead increases the induction period end time. The pozzolanic reactivity of silica fume
was greater than that of fly ash micro-bead. The effect of silica fume on 28 days strength was
particularly obvious, and fly ash micro-bead have more significant improvement in early strength.
The porosity shows a strong correlation with the compressive strength. The findings of this study
are a worth reference for precast concrete manufacturers and can help to guide future studies.

1. Introduction

High-strength concrete (HSC) has gained unprecedented development in recent years with the increase of super high-rise buildings,
long-span bridges, offshore platforms and underground structures in China. HSC is a concrete with high compressive strength, strong
deformation resistance, low porosity and excellent durability [1,2,4]. The application of HSC in modern construction can not only
greatly reduce the amount of concrete and reinforcement, significantly reduce the dead weight of the structure, and obtain better
technical and economic effects [5,6].
The appropriate amount of mineral admixture has a positive impact on the hydration and microstructure of HSC under steam
curing [7–9]. In previous studies, the application of reactive pozzolanic materials (blast furnace slag, fly ash and silica fume) is of great
importance in the design of HSC. Compared with other pozzolanic materials, silica fume has high pozzolanic reactivity and large
specific surface area, which can effectively improve the mechanical properties of HSC [10,11]. However, silica fume has a negative
impact on the workability of HSC, and increases the tendency to form cracks at the early stage due to autogenous shrinkage [12,13]. In
order to improve the mechanical properties and fluidity of concrete at the same time, some researchers have used finer particles of fly

* Corresponding author.
E-mail addresses: 1025004825@qq.com (P. Wang), gxleee@chd.edu.cn (X. Li), 739076796@qq.com (X. Huo), 1048303193@qq.com (S. Ali),
2957406791@qq.com (J. Sun), jinyi.qin@chd.edu.cn (J. Qin), matkun@chd.edu.cn (K. Liang), 137685135@qq.com (Y. Liu).

https://doi.org/10.1016/j.cscm.2022.e01583
Received 15 August 2022; Received in revised form 4 October 2022; Accepted 16 October 2022
Available online 19 October 2022
2214-5095/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
P. Wang et al. Case Studies in Construction Materials 17 (2022) e01583

ash as supplementary cementitious material in concrete [14–17]. The average particle size of finer fly ash particles is about 3.4 µm
(much smaller than above 10 µm of ordinary fly ash), and finer fly ash particles has higher specific surface area and pozzolanic
reactivity than ordinary fly ash [18,19]. In addition, the morphology of finer fly ash is spherical, meaning that the fly ash can act as
“ball bearings”, thus increasing the workability of the concrete [20]. Compared with the concrete containing silica fume, the
autogenous shrinkage of the concrete containing ultrafine fly ash is significantly reduced [21].
In order to find an appropriate amount of mineral admixtures, the use of simplex centroid design method (also known as seven-
batch factorial design method) to design concrete mixtures has received more and more attention. For example, the relationships
between cement-fly ash-slag ternary cementitious composition and different properties were established based on the seven-batch
factorial design method [22]. Jiao et al. were used the simplex centroid design method to obtained the optimal content of cementi­
tious materials in concrete according to the relationship between workability, yield stress, plastic viscosity and the paste volume
fraction [23]. Chen et al. obtained the mix proportion, phase diagram components and mechanical properties of ultra-high perfor­
mance mortar (UHPM) based on simplex-centroid design method were investigated. The effects of material components, sand/binder
ratio and water/binder ratio on the strength and fluidity of UHPM were studied. The UHPM can be prepared based on simplex-centroid
experimental design method [24]. Shi et al. investigated the flowability, compressive strength, heat of hydration, porosity and calcium
hydroxide content of ultra-high-strength concrete (UHSC) with cement–silica fume–slag binder at 20 ◦ C. The results show that proper
silica fume content can improve the compressive strength of UHSC, reduce the porosity and calcium hydroxide content of UHSC, but
slag has a negative effect [25]. Therefore, the simplex-centroid design method can not only obtain the optimal composition of
cementitious materials, but also be used to explain the influence of different materials on concrete performance.
In this study, the simplex-centroid design method be used to design the cement-silica fume-fly ash micro-bead ternary binder. Then,
the relationship between the ternary binder and the hydration heat, hydration products, compressive strength and porosity of HSC was
obtained.

2. Experiment

2.1. Materials

The test was conducted using ordinary P.O 52.5 portland cement (C) [26], silica fume (SF) and fly ash micro-bead (FAM), poly­
carboxylate superplasticizer and water.
Table 1 shows the chemical composition results of C, SF, and FAM determined by X-ray fluorescence (XRF).
River sand with fineness modulus of 2.81 was used as the fine aggregate in this experiment. The properties of sand, gravel 1 (5–20
mm) and gravel 2 (10–25 mm) were shown in Table 2.

2.2. Mixes proportions

The simplex-centroid design method used in this study was called seven-batch factorial design method [27]. The P-order regression
equation used in simplex-centroid design method was shown in Eq. (1) [28].
p
∑ ∑ ∑ ∑
Ey = βi xi + βij xi xj + βijk xi xj xk + βijk xi xj xk + β12⋯P x1 x2 ⋯xp (1)
i=1 i<j i<j<k i<j<k

Where Ey = responses, including any property characteristics of ternary binder, β12⋯P is the coefficient, x1 x2 ⋯xp is the mass pro­
portion.
The regression equation of the three component and three-order simplex-centroid design method used in this paper was shown in
Eq. 2. It can be seen that the number of test points for the design was 7. Through the axial point strengthening design, three groups were
added on the basis of tests to improve the reliability of the regression equation. The mixture design was not only limited by the
constraint conditions (Eq. 3), but also by upper and lower bound constraints [29,30]. According to a large number of previous basic
tests and studies, each component of the cementitious material should be controlled within a certain range, and the boundary con­
dition is the C 70~100 %, SF 0~30 %, FAM 0~30 % [31–34]. Therefore, a total of 10 groups of cement-silica fume-fly ash micro-bead
ternary binder system were designed. The design composition was shown in Table 3.
Ey = β1 x1 + β2 x2 + β3 x3 + β12 x1 x2 + β13 x1 x3 + β23 x2 x3 + β123 x1 x2 x3 (2)

Where xi is the mass proportion of C (i = 1), SF (i = 2) or FAM (i = 3).

Table 1
Chemical compositions of C, SF, FAM (wt%).
Composition CaO Fe2O3 SiO2 Al2O3 MgO K2 O SO3 TiO2

C 61.69 3.14 18.00 4.07 1.07 0.78 1.93 0.237

SF 0.942 3.589 93.94 0.841 1.37 1.65 0.282 0.015
FAM 6.983 6.19 57.93 16.6 1.28 2.73 0.369 0.631

2
P. Wang et al. Case Studies in Construction Materials 17 (2022) e01583

Table 2
The properties of sand, gravel 1 and gravel 2.
Mud content Apparent density Packing density

Sand 2.6 % 2573 kg/m3 1690 kg/m3

Gr 1 0.9 % 2710 kg/m3 1680 kg/m3
Gr 2 0.6 % 2720 kg/m3 1660 kg/m3

Table 3
Ternary binder compositions of C80, wt%.
Sample C SF FAM

C100 100 0 0
C70S30 70 30 0
C70F30 70 0 30
C85S15 85 15 0
C85F15 85 0 15
C70S15F15 70 15 15
C80S10F10 80 10 10
C90S5F5 90 5 5
C75S20F5 75 20 5
C75S5F20 75 5 20

{ }
xi ≥ 0 (i = 1, 2, ⋯p)
(3)
x1 + x2 + ⋯ + xp = 1

2.3. Methods

The HSC mix ratio: cementitious materials were 498 kg/m3, sand was 682 kg/m3, Gr 1 was 699 kg/m3, Gr 2 was 510 kg/m3, water
was 130 kg/m3. The amount of superplasticizer was added at 2.0% by mass of cementitious materials. The composition of cementitious
materials was shown in Table 3, and 10 groups of HSC were prepared. All the concrete specimens were sat in 20 ◦ C for 24 h after
casting, then put into steam curing room. The curing temperature was raised to 85 ◦ C with the rate of 20 ◦ C/h, and maintained at 85 ◦ C
for 8 h, then the curing temperature decreased to 20 ◦ C with the rate of 10 ◦ C/h. After steam curing, the samples were placed in an
environment with a temperature of 20 ± 2 ◦ C and a humidity of 100 % until testing. The 3，7 and 28 days compressive strength of
concrete were measured by Chinese Standards GB/T 50081–2019 [35]. The specimens were square (100 × 100 × 100 mm). The test
was carried out three times, and the average value of the test was taken as the compressive strength.
The hydration heat was determined using an isothermal calorimeter (TAM Air). Weigh the binder components accurately according
to the Table 3, and then put them into a high-speed mixer for quick mixing for 1 min. Water and superplasticizer were added to the
mixer and mixed at low speed for 3 min, and then quickly mix for 8 min. About 5.4 g paste samples was weighed and filled into a glass
bottle then placed into TAM Air immediately for measurement of heat of hydration. Finally, the computer was started to calculate the
heat flow at the same time. The test temperature of hydration heat was 85 ℃, and the test time was 48 h.
The crystalline component in HSC was characterized by X-ray diffraction (XRD, D/Max-2500, Rigaku, Japan). The specimens were

Fig. 1. Heat evolution of hydration of ternary binder: (a) Heat evolution rate and (b) Heat of hydration.

3
P. Wang et al. Case Studies in Construction Materials 17 (2022) e01583

scanned between 5◦ and 65◦ . The age of HSC in the test was 28 days.
The calcium hydroxide (CH) content and bound water (H) of HSC was measured by thermogravimetric analysis (TGA). First, the
cement stone with curing age of 3 days and 28 days were ground into powder, and then about 10 g of samples were put into TGA for
testing. The TGA was ranging from 25◦ to 1000◦ at 10 ◦ C/min, and the environment incoming gas was nitrogen. The thermogravi­
metric and differential scanning calorimetry curves were recorded.
The Mercury intrusion porosimetry (MIP) test used was Auto Pore IV from Micromeritics corporation. The age of HSC in the test was
28 days. The test specimens weighed about 1.5 g. The porosity of the specimen was calculated by the cumulative mercury intake.

3. Results and discussion

3.1. Hydration heat evolution

The heat evolution rate of ternary binder was shown in Fig. 1a. Since sulfate and aluminate dissolve rapidly in the binder within
minutes of mixing with water, the formation of AFt and syngenetic rocks should rapidly form a short and sharp exothermic peak.
However, it takes a while for the mixed slurry to be put into the isothermal calorimeter, so the initial reaction cannot be shown on the
Fig. 1a. The induction period, the acceleratory period, the decelerator period and the period of slow continued reaction can be clearly
seen in Fig. 1a.
According to Figs. 1 and 2, the peak heat flow of the binder of C85S15, C70S30 was appeared earlier than that of C100. Never­
theless, as long as FAM was added, the peak heat flow of the binder was later than that of C100. When the content of SF and FAM were
the same, the induction period of the slurry containing SF was shorter than that of the slurry containing FAM, and the former entered
the acceleration period earlier. The amount of SF and FAM increased from 0 % to 15 % and 30 %, and the end time of the induction
period for each decreased from 1.60 h to 1.02 h and 0.71 h, and from 1.60 h to 3.33 h and 2.30 h, respectively. The addition of SF
accelerates the hydration reaction of the binder, while the FAM slows down the hydration reaction [36–38], which is related to the
condition of the fly ash surface. A large amount of Ca2+ will be produced during cement hydration, but surface of fly ash particles will
absorb Ca2+. It causes the Ca2+ concentration in the solution to decrease, delays the nucleation of calcium hydroxide, and generates
unstable C-S-H gel with a low Ca/Si ratio. Finally, Ca2+ on the surface of fly ash particles will slowly release into the solution, forming a
stable C-S-H gel [39–42]. Thus, the ending time of the induction period is prolonged. On the contrary, silica fume particles can provide
nucleation sites for hydrates growth, and C-S-H hydrate products can precipitate on silica fume particles [43,44]. The hydration
exothermic peak of C90S5F5 was shifted to the right than that of C100, and the peak value increased. Compared with C90S5F5, the
exothermic hydration peak of C80S10F10 and C70S15F15 shifted to the left, and the peak value showed a decreasing trend. This shows
that 5 % of SF and FAM increase the hydration rate of silicate and aluminate in the binder, but the second and third peaks appear to be
delayed due to the extension of the induction period. When the amount of SF and FAM is 10 % or 15 %, the hydration rate of silicate
and aluminate will decrease, and the end of the induction period will be earlier.
Based on the induction period end time (t0) of the ten mixtures groups, the relationship between the t0 and C–SF–FAM binder can be
established by Eq. (4):
t0 = 1.64x1 + 1.32x2 + 18.55x3 − 3.67x1 x2 − 7.37x1 x3 − 15.96x2 x3 − 74.90x1 x2 x3 (4)
According to the hydration kinetic formula proposed by Knudson [45] and the hydration heat release curve of the C-SF-FAM
ternary system cement paste, the total hydration heat release at time t can be obtained by the following Eq. (5).

Fig. 2. The induction period end time t0 of ternary binder.

4
P. Wang et al. Case Studies in Construction Materials 17 (2022) e01583

（t − t0 ）
Qt = Qmax • (5)
（t − t0 ） + t50

Where Qt is the hydration heat release at time t; Qmax is the property when the time is infinitely long; t0 is the end time of induction
period; t50 is the time for 50% reaction degree.
Based on the Table 4, then the relationship between Qmax and the C–SF–FAM binder can be obtained as follows:
Qmax = 196x1 − 1927x2 − 1268x3 + 3026x1 x2 + 2155x1 x3 + 15467x2 x3 − 22764x1 x2 x3 (6)
It can be seen from Fig. 3 that with the increase of SF and FAM content, the Qmax of ternary binder composition increases signif­
icantly. When the amount of SF is less than 15 %, the Qmax of ternary binder composition increases, however when the amount of silica
fume exceeds 15 %, the Qmax decreases.

3.2. Thermal analysis

TGA and DSC of C-SF-FAM ternary binder were presented in the Figs. 4 and 5. The decomposition of the hydration product was
divided into three stages, corresponding to three endothermic peaks. The weight loss of samples at 25–200 ◦ C resulted from the
dehydration of C-S-H and AFt, the weight loss at 400–500 ◦ C was caused by the dehydroxylation of portlandite, and the weight loss at
550–870 ◦ C was the decomposition of CaCO3 [46–48]. The calculation method of the actual H content in the sample was shown in Eq.
(8). The decomposition reaction of CH at high temperature was shown in Eq. (7). The calculation method of the actual CH content in
the sample was shown in Eq. (9). The exact boundaries for the temperature intervals were read from the differential scanning calo­
rimetry curves (DSC) by using the tangent method. The calculation results of H and CH in C-SF-FAM ternary binder were listed in
Table 5.
Ca(OH)2 (74g/mol) → CaO + H2 O(18g/mol) (7)

M50℃ − M550℃
H(%) = × 100 (8)
M550℃

CH TG × 74/18
CH(%) = × 100 (9)
M550℃

Where CHTG is % weight loss from water associated with calcium hydroxide, M550 ◦ C is the ignited weight at 550 ◦ C, M50 ◦ C is the
ignited weight at 50 ◦ C.
The first endothermic peak occurs near 100 ◦ C, which indicate that the dehydration reaction of C-S-H and AFt. The mass loss caused
by the dehydration reaction of AFt and C-S-H can be used to estimate the content of H. The H content of samples with ternary binder
were shown in Fig. 6. The H content of the samples containing SF and FAM was lower than that of C100. Moreover, the H content of the
binder pastes containing SF were higher than the binder pastes containing FAM at the same amount. When the SF content increased
from 0 % to 30 %, the 3 days H content decreased from 16.28 % to 14.64 %. When the FAM content increased from 0 % to 30 %, the 3
days H content decreased from 16.28% to 14.98%. When the curing age increases from 3 days to 28 days, the H content of most binder
were increased, which indicated the AFt and C-S-H produced by the hydration of binder were increased. Under the same amount of
cement, increasing the amount of FAM can increase the early hydration degree of ternary binder than SF. However, the opposite trend
was shown when the curing age reaches 28 days. It can be clearly seen that the H content formation rate of C100 was higher than that
of other ternary binder. The addition of SF and FAM reduces the amount of cement in the cementitious material. Although SF and FAM
will react with CH to form C-S-H, the amount of C-S-H produced by the same amount of cement hydration is more.
Based on the H content of the ternary binder, the relationship between the H content and C–SF–FAM binder can be described by
Eqs. (10–11):
H3d = 16.14x1 + 34.96x2 + 8.10x3 − 34.72x1 x2 + 6.86x1 x3 − 62.45x2 x3 + 87.03x1 x2 x3 (10)

Table 4
Values of Qmax , t0 , t50 with ternary binder.
No. t0 /h t50 /h Qmax / (J/g)

C100 1.60 0.6 188.68

C70S30 0.71 1.32 200.4
C70F30 5.30 0.88 210.08
C85S15 1.02 1.53 261.78
C85F15 3.33 0.36 244.50
C70S15F15 1.50 0.38 203.25
C80S10F10 1.74 0.38 223.7
C90S5F5 1.86 0.61 261.1
C75S20F5 1.04 1.38 203.3
C75S5F20 2.62 0.40 219.3

5
P. Wang et al. Case Studies in Construction Materials 17 (2022) e01583

Fig. 3. Total hydration heat of ternary binder.

Fig. 4. DSC curve of ternary binder at (a) 3 days and (b) 28 days.

Fig. 5. TGA curve of ternary binder at (a) 3 days and (b) 28 days.

6
P. Wang et al. Case Studies in Construction Materials 17 (2022) e01583

Table 5
The H and CH content of HSC.
% C100 C70S30 C70F30 C85S15 C85F15 C70S15F15 C80S10F10 C90S5F5 C75S20F5 C75S5F20

H3d 16.28 14.63 14.98 14.81 15.75 14.74 15.28 15.00 14.11 15.74
H28d 20.42 18.02 19.38 17.30 19.59 17.59 19.18 19.16 17.09 16.41
CH3d 13.99 2.24 5.38 5.64 7.63 4.60 6.14 8.13 3.55 5.94
CH28d 14.73 2.68 6.55 5.61 9.82 4.87 5.71 10.40 3.42 5.82

Fig. 6. Bound water content of ternary binder: (a) 3 days; (b) 28 days.

H28d = 20.6x1 + 69.2x2 + 29.9x3 − 81.7x1 x2 − 21.5x1 x3 − 479.6x2 x3 + 624.0x1 x2 x3 (11)

The second endothermic peak at 410 C was ascribed to the dehydroxylation of portlandite. The CH content of ternary binder were
◦

shown in Fig. 7. The CH content of ternary binder containing SF and FAM were lower than that of the C100 binder at 3 days and 28
days. The decrease in CH content was attributed to the conversion of CH into secondary C-S-H via pozzolanic reaction [49]. The lower
CH content means that more secondary C-S-H was formed in the ternary binder, which is undoubtedly beneficial to concrete strength.
The CH content of the binder depends on the proportion of cement content and the pozzolanic reaction of the auxiliary cementitious
material.
From the 3 days CH content results of the ternary binder, it can be seen that the amount of SF increases from 0 % to 30 %, and the
CH decreases from 13.99 % to 2.24 %, while the CH was decreased to 5.38% for FAM, indicating that the effect of SF on CH content was
more significant than FAM. It can be seen from Table 5, when the hydration age of HSC increases from 3 days to 28 days, the CH
content of almost formulations also increases, except C75S20F5 and C75S5F20. This indicates that CH consumed by the pozzolanic

Fig. 7. CH content of ternary binder: (a) 3 days; (b) 28 days.

7
P. Wang et al. Case Studies in Construction Materials 17 (2022) e01583

reaction of SF and FAM is less than that generated by cement hydration. It is worth noting that the CH content of C85S15, C80S10F10
and C75S20F5 decreased by 0.53 %, 7.00 % and 3.6 %, respectively, and the CH content of C80S10F10 decreased most obviously. By
comparison, when the content of SF was greater than 10 % and the content of FAM was less than 10 %, the CH content decreases with
the increase of age, and the pozzolanic reactivity will be further reflected. When the content of SF was 30 %, due to the adsorption of
free water in the binder by silica fume, the effective water-cement ratio decreases, and the initial content of CH in the binder was less.
However, the cement hydration becomes more and more sufficient with the increase of curing age, and the CH content increases
accordingly. Therefore, the consumption of CH through the pozzolanic reaction improves the properties of the transition zone of the
concrete interface, which is beneficial to improve the strength of the concrete [50,51].
Based on the CH content of the ternary binder, the relationship between the CH content Y and C–SF–FAM binder can be described
by Eqs. (12 and 13):
CH3d = 13.8x1 + 55.2x2 + 50.6x3 − 114.7x1 x2 − 92.2x1 x3 − 223.2x2 x3 + 370.9x1 x2 x3 (12)

CH 28d = 15.0x1 + 67.0x2 + 10.0x3 − 133.3x1 x2 − 33.9x1 x3 − 417.8x2 x3 − 588.1x1 x2 x3 (13)

3.3. XRD analysis

The XRD patterns of the ternary binder were shown in Fig. 8. The major hydration products common to all samples were por­
tlandite, calcite, unreacted C2S and C3S. It is generally believed that delayed ettringite formation will occur in high temperature steam
curing [52,53]. However, the peak of the AFt phase was not observed in Fig. 8, which may be related to the type and content of the
incorporated mineral admixtures. Many studies have shown that pozzolanic materials can inhibit the generation of delayed ettringite
[54]. The pozzolanic materials make the concrete structure more compact and inhibit the water and ion transfer, which considerably
delays the onset of DEF expansion [55,56]. Compared with C100, the CH content of other binder was generally reduced, which was due
to the consumption of the pozzolanic reaction of SF and FAM. This result is consistent with the thermal analysis. In the ternary binder
with high SF content, the intensity of the silicate peaks was significantly reduced, suggesting that the pozzolanic reactivity of SF was
higher than that of FAM.

3.4. Compressive strength

The compressive strengths of HSC with ternary binder were shown in Fig. 9, and the compressive strength data of HSC were listed in
the Table 6.
It can be seen from Fig. 9a that when the SF content was less than 10 %, the addition of SF gradually increased the compressive
strength of HSC, while when it was greater than 10 %, the 3 days compressive strengths decreased with the increase of the SF content.
However, the trend of FAM was basically the same as that of SF. The composition of the ternary binder was C: SF: FAM = 0.77: 0.11:
0.12, and the maximum 3 days compressive strength of HSC was 101 MPa. With the addition of SF and FAM up to 15 %, the 7 days
compressive strength of HSC gradually increased, but the strength decreased when the addition exceeded 15 %. In particular, the
addition of FAM had a greater effect on HSC strength than SF. When the composition of the ternary binder was C: SF: FAM = 0.70: 0.15:
0.15, the maximum 7 days compressive strength of HSC was 104.4 MPa. As the SF content increased from 0 % to 30 %, the 28 days
compressive strength of HSC increased from 90.4 MPa to 102.3 MPa. However, excessive addition of FAM has a negative effect on the
28 days strength of HSC and this phenomenon shows a different regularity than the early strength of HSC. The composition of the
maximum 28 days compressive strength of HSC was C: SF: FAM = 0.70: 0.25: 0.05, and the corresponding compressive strength was
102.8 MPa. In summary, FAM can improve the early compressive strength of HSC more obviously than SF, and the improvement of 28
days compressive strength of SF was evident.
The synergistic effect of SF and FAM can improve the compressive strength of HSC, which is consistent with the views of many
scholars [57–60]. Firstly, SF and FAM can react with CH to form C-S-H, and secondly, the particle size of SF and FAM was smaller than
that of cement, playing a filling effect, thereby improving the compactness and mechanical properties of the structure [10,19].
Based on the compressive strength of the ternary binder, the relationship between the compressive strength (3 days, 7days, 28days)
Y and C–SF–FAM binder can be described by Eqs. (14–16):
Compressive Strength3d = 76.38x1 − 346x2 + 140x3 + 597x1 x2 − 25x1 x3 − 5906x2 x3 + 9192x1 x2 x3 (14)

Compressive Strength7d = 81.11x1 − 334x2 − 203x3 + 648x1 x2 + 460x1 x3 + 3838x2 x3 − 4737x1 x2 x3 (15)

Compressive Strength28d = 90.27x1 + 3x2 − 210x3 + 182x1 x2 + 431x1 x3 + 1473x2 x3 − 1835x1 x2 x3 (16)
In order to make HSC obtain higher compressive strength in both early and late stages, the 3 days, 7 days, and 28 days strength
contour plots of the ternary binder were superimposed, with the results shown in Fig. 10. The composition of the ternary binder should
meet the following requirements: the 3 days strength ≥ 98.5 MPa, the 7 days strength ≥ 100.5 MPa, and the 28 days strength
≥ 100.6 MPa. The area formed according to the above conditions is shown as the white area in Fig. 10. The composition of the best
ternary binder was: 72.5–84 % cement, 10–16.5 % silica fume and 6.5–14 % fly ash micro-bead. This result is obtained by analyzing the
compressive strength of HSC at different ages, and it is only a suggestion for manufacturers.

8
P. Wang et al. Case Studies in Construction Materials 17 (2022) e01583

Fig. 8. The XRD pattern of ternary binder at 28days.

Fig. 9. Compressive strengths of HSC with ternary binder: (a) 3 days; (b) 7 days; (c) 28 days.

Table 6
The compressive strength of HSC.
Compressive Strength/MPa C100 C70S30 C70F30 C85S15 C85F15 C70S15F15 C80S10F10 C90S5F5 C75S20F5 C75S5F20

3 days 76.7 75.6 88.9 90.0 81.8 93.8 99.6 89.5 92.5 100.4
7 days 81.7 93.0 90.8 102.4 96.0 102.9 97.7 94.1 101.1 106.1
28 days 89.9 104.5 89.0 102.2 98.1 101.2 100.8 99.1 95.7 103.5

3.5. Pore structure

Porosity and median pore diameter for ternary binder obtained from the MIP were provided in Fig. 11. With the increase of SF and
FAM adding amount, the porosity of ternary binder showed a decreasing trend, and the effect of SF was more obvious than that of FAM.
When the composition of the ternary binder was only cement, the porosity of HSC reaches the maximum value of 14.01 %. However,
C70S30 had the smallest porosity at 9.42 %. The pore structure of HSC containing SF and FAM was denser than that of C100, and
adding SF content can form a denser structure in HSC. Likewise, the median pore size of the ternary binder also showed roughly the
same trend as the porosity. The smallest median pore size of ternary adhesive composition was C70S30, which was 6.4 nm. Combined
with thermal analysis, it can be seen that SF has higher pozzolanic reactivity, it consumes more CH than SF, forming a large amount of
C-S-H gel. This phenomenon enables SF-containing HSC to have a denser structure.
Based on the porosity and median pore diameter of the ternary binder, the relationship between the Porosity or median pore
diameter Y and C–SF–FAM binder can be described by Eqs. (17 and 18):

9
P. Wang et al. Case Studies in Construction Materials 17 (2022) e01583

Fig. 10. Determination of ternary binder composition.

Fig. 11. Pore structure of ternary binder at 28 days: (a) Porosity; (b) Median pore diameter.

Porosity = 13.86x1 + 52.2x2 + 47.9x3 − 73.7x1 x2 − 64.1x1 x3 − 567x2 x3 + 775x1 x2 x3 (17)

Median pore diameter = 16.42x1 + 109.5x2 + 126.5x3 − 178.1x1 x2 − 193.2x1 x3 − 919x2 x3 + 1282x1 x2 x3 (18)
The contour plots of porosity and 28 days compressive strength were superimposed, and the results are shown in Fig. 12. The
porosity shows a strong correlation with the compressive strength, smaller porosity positively affects 28 days strength. When the 28
days strength is greater than 100.6 MPa, the probability of porosity less than 10.2 % is high.

4. Conclusions

This research studied the hydration, compressive strength and pore structure of steam cured HSC with C–SF–FAM binder. The
simplex centroid design method was used to characterize the interaction of the ternary binder. The relationships between the ternary
binder and the characteristics were expressed by contour lines. The following conclusions can be drawn:
SF and FAM altered the binder induction period, and the amount of SF was increased from 0 % to 30 %, while the end time of the
induction period was reduced from 1.60 h to 0.71 h. When the SF content increased from 0 % to 30 %, the end time of the induction
period increased from 1.60 h to 5.3 h. SF accelerates the hydration reaction of the binder, while FAM slow down the hydration re­
action. With the increase of SF and FAM content, the total hydration heat of the binder increases significantly. However, when the
content of the two exceeds 15 %, the total hydration heat begins to decrease, and with the increase of the content of silica fume and fly

10
P. Wang et al. Case Studies in Construction Materials 17 (2022) e01583

Fig. 12. Relationship between porosity and compressive strength of ternary binder at 28 days.

ash micro-bead, the CH content also decreased. The effect of SF on CH content is more significant than that of FAM, indicating that SF
has stronger pozzolanic reactivity than FAM. SF and FAM have a positive synergistic effect on the compressive strength of HSC. The
effect of SF on 28 days strength is particularly obvious, and the effect of FAM on the early strength is more significant. The porosity of
HSC decreased with the increase of SF and FAM content due to the filling and pozzolanic effects. The statistical analysis showed that
the compressive strength was positively correlated with the porosity.

Funding

This work was financially supported by the National Natural Science Foundation of China (No. 52272015), Shaanxi Provincial
Science and Technology Program Key Research and Development Program Project (No. 2020SF-360) and Xi’an Science and Tech­
nology Program Project (No. 2020KJRC0131 and No. SZJJ2019–11).

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

Data availability

Data will be made available on request.

References

[1] Naifah, M. Hasan, T.J.Jo.P.C.S. Saidi, The resistance of high strength concrete with diatomaceous earth as cement replacement to Nacl attack, Journal of Physics
Conference Series, 1933(1) (2021) 012104.
[2] M.A. Farooq, Y. Sato, K. Niitani, H. Yokota, Fatigue behavior of air-entrained high strength concrete with blast furnace slag fine aggregates under freezing and
thawing, Part of the Lecture Notes in Civil Engineering book series (LNCE,volume 101), EASEC16 101 (2021).
[4] H. Bayat, A. Ubysz, M. Maj, M. Chalecki, J. Wójt, A.J.M. Tamrazyan, Experimental research of ratio between residual and elastic strains eres/ee in high-strength
concrete beams subjected to bending, Materials 14 (20) (2021) 6007.
[5] I.D. Saikiran, C.S.J.I.C.S.M.S. Rao, Engineering, Performance of high strength concrete consisting of alccofine and metakaolin as a partial replacement of cement
and copper slag as fine aggregate, IOP Conference Series: Materials Science and Engineering 1185 (1) (2021), 012017-.
[6] I.H. Mohammed, A. Aamri, S. Javed, Y.J.M.S.F. Shamsi, A comparative investigation on normal and high strength concrete beams under torsion, Materials
Science Forum 1048 (2022).
[7] A. Amz, B. Mamj, C. Aa, D. Bat, The effect of steam curing regimes on the chloride resistance and pore size of high–strength green concrete, Construction and
Building Materials 280 (1) (2021).
[8] J. Zhang, T. Chen, X. Gao, Incorporation of self-ignited coal gangue in steam cured precast concrete, J. Clean. Prod. 292 (2021), 126004.
[9] M. Wang, Y. Xie, G. Long, C. Ma, X. Zeng, Microhardness characteristics of high-strength cement paste and interfacial transition zone at different curing regimes,
Constr. Build. Mater. 221 (OCT.10) (2019) 151–162.
[10] M. Mazloom, A.A. Ramezanianpour, J.J.J.C. Brooks, C. Composites, Effect of silica fume on mechanical properties of high-strength concrete, Cement &
Concrete Composites 26 (4) (2004) 347–357.
[11] A. Rhf, B. Afhs, B.J.Jo.B.E. Ad, Mechanical, fracture and durability properties of self-compacting high strength concrete containing recycled polypropylene
plastic particles, Journal of Building Engineering 25 (2019) 100808.
[12] C. Shi, Z. Wu, J. Xiao, D. Wang, Z. Huang, Z.J.C. Fang, B. Materials, A review on ultra high performance concrete: Part I. Raw materials and mixture design,
Construction and Building Materials, 101(DEC.30PT.1) (2015) 741-751.
[13] Y. Yang, R. Sato, K.J.C. Kawai, C. Research, Autogenous shrinkage of high-strength concrete containing silica fume under drying at early ages, Cement &
Concrete Research 35 (3) (2005) 449–456.

11
P. Wang et al. Case Studies in Construction Materials 17 (2022) e01583

[14] Y. Zhang, Y. Li, J. Zhou, J. Huang, Q.J.I.C.S.E. Wang, E. Science, Application of industrial solid waste in preparation of recycled concrete: a review, IOP
Conference Series: Earth and Environmental Science, 692(4) (2021) 042050 (5pp).
[15] K.V. Subramaniam, R. Gromotka, S.P. Shah, K. Obla, R.J.Jo.MiC.E. Hill, Influence of ultrafine fly ash on the early age response and the shrinkage cracking
potential of concrete, Journal of Materials in Civil Engineering 17 (1) (2005) 45–53.
[16] Aydin, S., Karatay, C., Baradan, B.J.P.T. -LAUSANNE-, The effect of grinding process on mechanical properties and alkali-silica reaction resistance of fly ash
incorporated cement mortars, Power Technology -Lausanne-, 197(1) (2010).
[17] A.N. Givi, S.A. Rashid, F. Aziz, M.J.C. Salleh, B. Materials, Assessment of the effects of rice husk ash particle size on strength, water permeability and workability
of binary blended concrete, Construction & Building Materials 24 (11) (2010) 2145–2150.
[18] F.U.A. Shaikh, S.W.M.J.C. Supit, B. Materials, Compressive strength and durability properties of high volume fly ash (HVFA) concretes containing ultrafine fly
ash (UFFA), Construction & Building Materials, 82(may 1) (2015) 192-205.
[19] L. Martin, F. Winnefeld, E. Tschopp, C. Müller, B.J.C. Lothenbach, C. Research, Influence of fly ash on the hydration of calcium sulfoaluminate cement, Cement
and Concrete Research (2017).
[20] Q. Wang, D. Wang, H.J.C. Chen, C. Composites, The role of fly ash microsphere in the microstructure and macroscopic properties of high-strength concrete,
Cement & Concrete Composites 83 (2017) 125–137.
[21] A.B. Hossain, S. Islam, K.D. Copeland, Influence of ultrafine fly ash on shrinkage and cracking tendency of concrete and the implications for bridge decks,
Transportation Research Board 86th, (2007).
[22] Shi, Caijun, Jiao, Dengwu, Zhang, Jian, Wang, Dehui, Yongjian, F.J. CONSTRUCTION, B. MATERIALS, Design of high performance concrete with multiple
performance requirements for #2 Dongting Lake Bridge, Construction & Building Materials, 165(Mar.20) (2018) 825-832.
[23] D. Jiao, C. Shi, Q. Yuan, X. An, Y.J.C. Liu, C. Composites, Mixture design of concrete using simplex centroid design method, Cement and Concrete Composites,
(2018) S0958946517306467.
[24] C. Ying, L. Peng, S.C. Fei, F. Sh, B. Gl, D. Zyb, E.J.Jo.M.R. Hc, Technology, Preparation and performance of the ultra-high performance mortar based on simplex-
centroid design method, Journal of Materials Research and Technology 15 (2021) 3060–3077.
[25] C. Shi, D. Wang, L. Wu, Z.J.C. Wu, C. Composites, The hydration and microstructure of ultra high-strength concrete with cement–silica fume–slag binder,
Cement & Concrete Composites 61 (2015) 44–52.
[26] C. Standard, Common Portland Cement (GB175-2020), China Environmental Science Press, 2020.
[27] R.J.T. Ziegel, Handbook of Statistical Methods for Engineers and Scientists, Technometrics 41 (1) (1999) 83.
[28] H. Scheffé, The simplex-centroid design for experiments with mixtures, J. R. Stat. Soc. 25 (2) (1963) 235–251.
[29] Dehuai, Wang, and, Zhaoyuan, C.J. Cement, C. Research, On predicting compressive strengths of mortars with ternary blends of cement, ggbfs and fly ash,
Cement and Concrete Research, 27(4) (1997) 487-493.
[30] C. Research, Cementitious properties of ladle slag fines under autoclave curing conditions, Cement and Concrete Research 33 (11) (2003) 1851–1856.
[31] V. Yogendran, B.W. Langan, M.N. Haque, M.A.J.A.M.J. Ward, Silica fume in high-strength concrete, Aci Materials Journal 84 (2) (1987) 124–129.
[32] C.H. Wu, Y.F. Lin, S.K. Lin, C.H.J.Se Huang, mechanics, Temperature development and cracking characteristics of high strength concrete slab at early age,
Structural engineering and mechanics 74 (6) (2020).
[33] A.R. Hariharan, A.S. Santhi, G.M.J.I.Jo.E.S. Ganesh, Technology, Study on Strength Development of High Strength Concrete Containing Fly ash and Silica fume,
International Journal of Engineering Science & Technology (2011).
[34] D. Wang, X. Zhou, Y. Meng, Z.J.C. Chen, B. Materials, Durability of concrete containing fly ash and silica fume against combined freezing-thawing and sulfate
attack, Construction and Building Materials, 147(aug.30) (2017) 398-406.
[35] C. Standard. Standard for test methods of concrete physical and mechanical properties（GBT 50081-2019）, China Environmental Science Press, 2019.
[36] F. Wei, M.W. Grutzeck, D.M. Roy, The retarding effects of fly ash upon the hydration of cement pastes: the first 24 hours, Cem. Concr. Res. 15 (1) (1985)
174–184.
[37] F. Han, R. Liu, D. Wang, P. Yan, Characteristics of the hydration heat evolution of composite binder at different hydrating temperature, Thermochim. Acta 586
(2014) 52–57.
[38] W. Nocuń-Wczelik, Heat evolution in hydrated cementitious systems admixtured with fly ash, J. Therm. Anal. Calorim. (2001).
[39] M. Narmluk, T.J.C. Nawa, C. Research, Effect of fly ash on the kinetics of Portland cement hydration at different curing temperatures, Cement and Concrete
Research 41 (6) (2011) 579–589.
[40] G. Baert, S. Hoste, G.D. Schutter, N.D.J.Jo.T.A. Belie, Calorimetry, Reactivity of fly ash in cement paste studied by means of thermogravimetry and isothermal
calorimetry, Journal of Thermal Analysis & Calorimetry 94 (2) (2008) 485–492.
[41] Analysis, Calorimetry, Heat evolution in hydrated cementitious systems admixtured with fly ash, Journal of Thermal Analysis&Calorimetry (2001).
[42] M. Kumar, S.K. Singh, N.P.J.T.A. Singh, Heat evolution during the hydration of Portland cement in the presence of fly ash, calcium hydroxide and super
plasticizer, Thermochimica Acta, 548(none) (2012).
[43] El, Hadj, Kadri, Roger, D.J.A.M. Journal, Effect of ultrafine particles on heat of hydration of cement mortars, ACI Materials Journal, (2002).
[44] E.H. Kadri, R.J.C. Duval, B. Materials, Hydration heat kinetics of concrete with silica fume, Construction & Building Materials 23 (11) (2009) 3388–3392.
[45] T. Knudsen, The dispersion model for hydration of portland cement I. General concepts, Cem. Concr. Res. 14 (5) (1984) 622–630.
[46] B. Ma, T. Zhang, H. Tan, X. Liu, J. Mei, H. Qi, W. Jiang, F. Zou, Effect of triisopropanolamine on compressive strength and hydration of cement-fly ash paste,
Constr. Build. Mater. 179 (DEC.20) (2018) 89–99.
[47] S. Adu-Amankwah, M. Zajac, C. Stabler, B. Lothenbach, L. Black, Influence of limestone on the hydration of ternary slag cement, Cem. Concr. Res. 100 (2017)
96–109.
[48] S. El-Gamal, S.A. Abo-El-Enein, E.D. Ahmed, A.A. Abdelnaby, Effect of novel binaphthol admixture on the hydration characteristics and mechanical properties of
OPC pastes, Adv. Cem. Res. (2018).
[49] R.J. Olek, Investigation into the synergistic effects in ternary cementitious systems containing portland cement, fly ash and silica fume, Cem. Concr. Compos.
(2012).
[50] M. Nili, A.J.M. Ehsani, Design, Investigating the effect of the cement paste and transition zone on strength development of concrete containing nanosilica and
silica fume -, ScienceDirect, Materials & Design 75 (2015) 174–183.
[51] K.Y. Liaoa, P.K. Changb, Y.N. Penga, C.C.J.C. Yangc, C. Research, A study on characteristics of interfacial transition zone in concrete, Cement & Concrete
Research 34 (6) (2004) 977–989.
[52] X. Brunetaud, R. Linder, L. Divet, D. Duragrin, D. Damidot, Effect of curing conditions and concrete mix design on the expansion generated by delayed ettringite
formation, Mater. Struct. 40 (6) (2007) 567–578.
[53] Batic Oscar, Milanesi Carlos, Pedro, Secondary ettringite formation in concrete subjected to different curing conditions, Cem. Concr. Res. (2000).
[54] Y. Fu, Delayed ettringite formation in Portland cement products.1996.
[55] Y. Kawabata, H. Takahashi, S.J.C. Watanabe, B. Materials, The long-term suppression effects of fly ash and slag on the expansion of heat-cured mortar due to
delayed ettringite formation, Construction and Building Materials 310 (2021) 125235.
[56] C. Famy, C.J.I.C.L. Famy. Expansion of heat-cured mortars, Imperial College, London, 1999.
[57] Q. Wang, D. Wang, H. Chen, The role of fly ash microsphere in the microstructure and macroscopic properties of high-strength concrete, Cem. Concr. Compos.
83 (2017) 125–137.
[58] J.D. Brito, Properties of fibre-reinforced high-strength concrete with nano-silica and silica fume, Appl. Sci. 11 (2021).
[59] D. Pedro, J. De Brito, L. Evangelista, Evaluation of high-performance concrete with recycled aggregates: Use of densified silica fume as cement replacement,
Constr. Build. Mater. 147 (aug.30) (2017) 803–814.
[60] Silva Digo, Oliveira de, Raiz Victor, Carrazedo Ricardo, Experimental study on normal-strength, high-strength and ultrahigh-strength concrete confined by
carbon and glass FRP laminates, J. Compos. Constr. 23 (1) (2019).

12