Construction and Building Materials 160 (2018) 117–126

Contents lists available at ScienceDirect

Construction and Building Materials

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

Effect of mineral admixtures on the structural build-up of cement paste

Qiang Yuan a,⇑, Dajun Zhou a, Baiyun Li a, Hai Huang a, Caijun Shi b
a
School of Civil Engineering, National Engineering Laboratory for High Speed Railway Construction, Central South University, Changsha 410075, China
b
College of Civil Engineering, Hunan University, Changsha 410082, China

h i g h l i g h t s

 Structural build-up of cement pastes was characterized by thixotropic area and growth of static yield stress.
 Structural build-up of cement pastes with various mineral admixtures was tested.
 Structural build-up of cement pastes was related with their calorimetric curve and zeta potential.

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

Article history: Structural build-up is of great importance in many concrete applications, such as formwork pressure,
Received 4 July 2017 multi-layers casting, slip form paving, stability, and 3D printing. In this paper, the growth of static yield
Received in revised form 6 November 2017 stress (ss) and thixotropic area were used to characterize the structural build-up of cement paste. The
Accepted 11 November 2017
effects of different mineral admixtures, i.e. silica fume (SF), ground slag (GS), fly ash (FA), attapulgite
(AG), nano calcium carbonate (NC), and nano silica (NS) on the structural build-up of cement pastes were
studied. Calorimetric curves, zeta potential and conductivity of the pastes were also measured. Results
Keywords:
showed that thixotropic behavior characterized by the growth of ss and thixotropic area gave different
Structural build-up
Thixotropy
ranking on cement pastes with various mineral admixtures. NS, AG, NC and SF increased the growth of
Mineral admixtures ss over 120 min, and NS was the most effective one. FA and GS decreased the growth of ss over 120
Cement min. Hydration of cement played a dominant role in the structural build-up of cement paste. However,
Static yield stress high hydration rate didn’t always lead to high structural build-up rate.
Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction 3D printing concrete [2–7]. However, high structural build-up rate

Structural build-up refers to the phenomenon that the
‘‘strength” of fresh paste/concrete increases with time due to
chemical reaction and/or physical actions. It is of great importance
to concrete technology [1]. Fresh concrete at rest tends to flow and
deform under the action of gravity, and it tends to segregate due to
the density difference of the constituents of concrete. High struc-
tural build-up rate benefits to the stability and shape-holding
capacity of concrete. When fresh concrete is at rest, the interac-
tions between particles develop because of colloidal flocculation
and cement hydration, and thus the structure of cement builds
up. As a consequence, cement paste can hold the aggregate against
gravity and prevent segregation from happening, and the shape-
holding capacity of concrete increases. Therefore, high structural
build-up rate is always desired in many applications, such as
reduction of formwork pressure, slip form paving, stability, and
⇑ Corresponding author.
E-mail address: 209176@csu.edu.cn (Q. Yuan).
may also be detrimental in some cases. For example, in the appli-
cation of multi-layers casting, high structural build-up rate reduces
the bonding strength between layers [8].
In concrete science, the terminology of thixotropy, instead of
structural build-up, is often used to describe the increase of the
‘‘strength” of fresh concrete. Thixotropy is generally defined as
the continuous decrease of viscosity with time when a shear rate
is applied to a sample that has been previously at rest, and the sub-
sequent recovery of viscosity in time when the shear stops [9,10]. It
has three characteristics: 1) the determination of thixotropy is
often based on the decrease of viscosity; 2) thixotropy is a time
dependent property; 3) thixotropy is a reversible process, and
related with the break-down and build-up of flocculates of suspen-
sion particles. It seems that the terminology of thixotropy is not
applicable to cement-based materials in most cases. On one hand,
evolution of static yield stress (ss), instead of viscosity, is more
often used in cement-based materials. On the other hand, once
cement is in contact with water, both reversible and irreversible
reactions happen simultaneously, and thixotropy is mainly dealing

https://doi.org/10.1016/j.conbuildmat.2017.11.050
0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
118 Q. Yuan et al. / Construction and Building Materials 160 (2018) 117–126
with a reversible process. Strictly speaking, thixotropy is not appli-
cable to cement-based materials. As stated above, structural build-
up at rest, which covers reversible and irreversible effects, is of
more interests in many applications. Thus, structural build-up at
rest is more applicable to cement-based materials than thixotropy.
However, reversible effects stemming from colloidal flocculation
may dominate for the first hour, and cement hydration, which is
irreversible effects, might be ignored [11,12]. In certain sense, the
terminology of thixotropy and structural build-up are often inter-
changeable in the early age of fresh cement-based materials. It
seems that most researchers preferred to use thixotropy, and a
few researchers used structural build-up. It is worth to note that
there are differences in the terms of thixotropy and structural
build-up. For the purpose of comparing with literature, these two
terms only refer to the process of structural build-up of cement
paste in this study, and thus they are interchangeable.
Numerous studies have been carried out on the thixotropy of
concrete. Generally, four methods have been used to measure the
thixotropy of materials, i.e. hysteresis, stepwise changes in shear
rate or shear stress, start-up or creep tests, and dynamic moduli.
Quanji et al. [13] used hysteresis area between up-curve and
down-curve to represent the thixotropy of cement-based materi-
als. Assaad [14] sheared concrete at different shear rate, and used
the area between the initial shear stresses and the equilibrium
stresses representing the thixotropy. Yuan et al. [15] used the stor-
age modulus obtained from small amplitude oscillatory test to
evaluate the structural build-up of cement paste. It was pointed
out that dynamic moduli method is the most suitable method to
follow the structural build-up [15]. However, it needs delicate
and expensive rheometer which is not available in most concrete
laboratories. Roussel et al. [1] proposed a thixotropic model for
concrete, and the increasing rate of shear stress at rest over time
was called thixotropic index (Athix). Roussel et al. [1] classified
self-compacting concrete (SCC) into three types based on Athix:
non-thixotropic SCC (Athix < 0.1 Pa/s) , thixotropic SCC (0.1 Pa/s 
Athix  0.5 Pa/s) and highly thixotropic SCC (Athix > 0.5 Pa/s). The
start-up test, also referred to as the static yield test or shear growth
test, is the most often used approach for cement-based materials.
Amziane et al. [16] developed a novel method to follow the struc-
tural build-up of cement-based materials, where a rough plate was
placed in a fresh paste and the weight of the plate was continu-
ously monitored. The weight of the plate decreased with time
due to the structural build-up of fresh paste. Omran et al. [17] used
portable vane method to evaluate the structural build-up at rest of
SCC. Khayat et al. [18] evaluated five field-oriented test methods
for the structural build-up at rest, including inclined plane (IP),
portable vane (PV), undisturbed slump spread (USS), cone penetra-
tion (CP), and K-slump test methods.
Assad et al. [2] studied the effect of viscosity enhancing admix-
ture (VEA) on the thixotropy of SCC, and found that the type, com-
bination, and addition level of VEA and high-range water-reducing
agent (HRWRA) had a significant effect on the thixotropy, and rel-
atively low addition level of VEA could result in high thixotropy.
Khayat et al. [3] reported the addition of HRWRA decreased the
structural build-up rate of SCC. Khayat et al. [19] used a propylene
carbonate-based thixotropy-enhancing admixture (TEA) to
increase the thixotropy of SCC. Assad et al. [4] reported that the
thixotropy of SCC with binary or ternary binder was much lower
than that of SCC without supplementary cementing materials,
and aggregate content decrease the structural build-up rate. Perrot
et al. [20] studied the structural build-up of rigid fiber reinforced
cement-based materials, and found that the structural build-up
rate was increased by the incorporation of fibers and aggregates.
Kawashima et al. [21] reported that purified attapulgite clays sig-
nificantly increased the structural build-up rate of pastes, espe-
cially at early ages. Quanji et al. [13] found that nano-sized
highly purified magnesium alumino silicate clay can significantly
increase the structural build-up rate of pastes. Ahari et al. [22]
examined the effect of supplementary cementitious materials
(SCM) on structural build-up of SCC, and found that except slag,
the use of fly ash, silica fume and metakaolin in SCC mixtures
increased thixotropy values in comparison with SCC without
SCM. Rahman et al. [23] found that the addition of silica fume
(2.5% replacement), limestone powder (15% replacement) and fly
ash (10% replacement) increased the thixotropic rate significantly.
Roussel et al. [11] ascribed high structural build-up to the CSH
bridging effect between cement particles. Thus, they believed that
the addition of particles known for their nucleation properties,
such as silica fume and limestone, will increase the nucleation rate
of CSH, and benefit structural build-up. Although there are a few
studies dealing with the factors affecting the structural build-up
of cement-based materials, the studies were based on different
testing methods, and the results may not be comparable. Thus, it
is necessary to comprehensively evaluate the effect of different
mineral admixtures on the structural build-up of cement paste
by using an appropriate method.
Although Jarny et al. [12] pointed out that the hydration of
cement in the first hour can be ignored by using MRI velocimetry,
it is well known that there is an initial heat liberation peak in the
first hour of cement hydration [24]. According to Taylor [24], the
initial heat liberation peak is ascribed to the exothermic wetting
and early-age hydration, and a gelatinous coating on cement parti-
cles and rods of AFt phase are formed. Thus, it would be very inter-
esting to see the relationship between hydration heat liberation
and structural build-up at very early age, which is seldom reported
in literature. In addition, a large amount of ions (Ca2+, K+, Na+, SO24 ,
OH ) are dissolved into the pore solution in the first hour. The elec-
trical double layer, which is very important for the interaction and
flocculation of particles, is instantaneously developed on the parti-
cle surfaces. Consequently, the zeta potential may have an influ-
ence on the structural build-up of cement paste. The use of
mineral admixtures makes the electrical double layer more com-
plicated. Thus, it is also interesting to investigate the evolution of
zeta potential of pastes containing various mineral admixtures
with time, and its relationship with structural build-up.
In this paper, the effects of different mineral admixtures (silica
fume, slag, fly ash, attapulgite, nano-calcium carbonate and nano-
silica) on the structural build-up rate of cement paste were studied
by static yield stress growth test and thixotropic area method. In
order to provide more insightful evidence of structural build-up,
calorimetric test was conducted to measure the hydration rate of
cement paste, and zeta potential and conductivity of cement paste
were also measured.
2. Materials and experimental programs
2.1. Materials and sample preparation
P II 52.5 portland cement which meets Chinese Standard
GB175-2007 and a polycarboxylate-based high-range water-
reducing agent (HRWRA) was used in this study. The mineral
admixtures, including silica fume (SF), ground slag (GS), fly ash
(FA), attapulgite (AG), nano calcium carbonate (NC) and nano silica
(NS), were used to partially replace cement. The chemical compo-
sitions and physical properties of the powders are given in Table 1.
The particle size distributions of cement, GS, FA and AG are deter-
mined by laser particle size distribution analyzer, as shown in
Fig. 1. It can be seen that GS, FA, and AG are finer than cement,
and cement is the coarsest powder. Since SF, NC, NS are very fine
powders, the particle sizes cannot be characterized by laser parti-
cle size distribution analyzer.
 Q. Yuan et al. / Construction and Building Materials 160 (2018) 117–126 119

Table 1
Chemical compositions and physical properties of the powders used in this study.

Cement (C) Silica fume (SF) Ground slag (GS) Fly ash (FA) Attapulgite (AG) Nano calcium carbonate (NC) Nano silica (NS)
CaO (%) 62.80 0.20 41.00 3.70 0.79 51.48 –
SiO2 (%) 20.60 93.00 33.10 52.7 58.50 – 99.8
Al2O3 (%) 4.13 0.41 15.40 25.8 9.92 – –
Fe2O3 (%) 2.99 0.80 0.50 9.70 6.23 0.05 –
SO3 (%) 2.56 0.39 1.95 0.20 – – –
MgO (%) 2.99 0.60 7.50 1.81 11.12 – –
Other minor oxides (%) 3.93 4.60 0.55 6.09 13.44 48.47 0.20
Loss on ignition (%) 0.31 2.01 0.40 1.76 10.61 0.50 5.01

C 2.2. Testing procedures

Cumulative percentage passing (%)

100
80
60
40
20
0 from 0 to100 s 1 within 60 s, and then sheared the paste for 100
0.1
Fig. 1. Particle size distribution of powders.
Water and HRWRA (if any) were first added into the Hobart N50
mixer. This was followed by adding cement and other mineral
admixtures. The paste was mixed at low speed for 30 s, and then
the mixer was stopped for 60 s to manually homogenize the paste
sample. This was followed by mixing at high speed for 90 s and low
speed for another 30 s. The water-to-cement ratio (w/c) was 0.35
and 0.45, and HRWRA addition levels were 0, 0.1% and 0.2%. The
replacement levels (by mass) of the mineral admixtures were cho-
sen based on the most common addition levels of mineral admix-
tures used in literature [21,24,25], as follows: 5% (SF), 30% (GS),
20% (FA), 1% (AG), 2% (NC), 2% (NS), and 1% (NS). The sample was
named as follows: w/c– HRWRA addition lever-mineral admixture
type and replacement level. For example, sample 45-0-SF5 means
paste with w/c ratio of 0.45, HRWRA addition level of 0 and silica
fume replacement level of 5%.
FA
GS 2.2.1. Rheological tests
AG Anton paar Rheolab QC rheometer was used for rheological test.
The type of rotator was CC39 with a diameter of 4.000 cm, and the
inner diameter of the cylinder was 4.194 cm. In order to minimize
the slip, the surface of the container and rotator which are in con-
tact with cement paste were sandblasted. Dynamic shear test was
started at 7 min 30 s after the contact of cement and water, and the
test was conducted as follows: pre-sheared the paste at 100 s 1 60
s, and stopped for 5 s, followed by linearly increasing shear rate
s 1 for 15 s, followed by linearly decreasing shear rate from 100
1 10 100
to 0 s 1 within 60 s. Bingham model was used to fit the data
Particle diameter ( m) between the shear rate of 20 and 80 s 1, and dynamic shear stress
(sd), viscosity and thixotropic area can be obtained, as shown in
Fig. 2. According to [26], shear rate of 0.02 s 1 was chosen for test-
ing ss. At the time of 15 min, the paste was subjected to static shear
test which was conducted at a constant shear rate of 0.02 s 1 for
30 s. The static shear test was repeated every 15 min until 120
min. Since the capacity of shear force of the equipment was 450
Pa, some paste samples reached the limit earlier than 120 min,
and the test was stopped after reaching the limit. After 120 min,
the dynamic shear test was run again. During the entire test, the
temperature of paste was maintained at 25 °C by water bath.
2.2.2. Calorimetric test
TAM Air thermal activity micro-calorimeter was used to mea-
sure the hydration heat of cement pastes. The mass of sample
was calculated based on thermal capacity. The water of 7 g was
taken as a reference sample, and the specific heat capacity of pow-
der is approximately taken as 20% of water. Thus, 15.62 g sample of
cement paste with w/c of 0.45, and 17.18 g sample of cement paste

up curve
100 up curve 80 down curve
consistant curve Linear Fit of Shear stress
down curve
80
60
Pa
Shear rate (1/s)

Area is measure
60 thixotropy
40
Shear stress

40 .

20 .

0
20
yield
0
stress, 0
0
0 15 30 45 60 75 90 105 120 135 0 10 20 30 40 50 60 70 80 90 100
Time (s) Shear rate (1/s)
(a) shear history (b) curve fitting
Fig. 2. Dynamic shear test.
120 Q. Yuan et al. / Construction and Building Materials 160 (2018) 117–126

with w/c of 0.35 were used for calorimetric measurement. The
cups with paste were put into the chambers and the ambient tem-
perature around the samples was controlled to 25 °C. The data
were recorded every 30 s.
2.2.3. Zeta potential test
The zeta potential of suspension was measured by using the
zetaProbe equipment which is based on the electroacoustic tech-
nique. A diluted cement paste was made for zeta potential mea-
surement. The w/c of paste without HRWRA is 15, and the W/C
ratio of paste with 0.1% HRWRA is 12. The replacement levels of
mineral admixture were the same to those in rheological test.
Cement, mineral admixture and distilled water (if any HRWRA)
were mixed in a mixer for 2 min. About 300 ml of the paste was
poured in the sample cell of ZetaProbe. During the test, the sam-
ples were kept stirring at a speed of 350 rpm so that a uniform
cement suspension passes through the electrodes of the cell. The
zeta potential and the conductivity of the cement solution were
measured every minute. The zeta potential of each mineral admix-
ture was also measured.
3. Results and discussion
3.1. Dynamic shear testing results
The rheological parameters of cement pastes measured by
dynamic shear test are given in Table 2. It can be seen from Table 2
that the addition of very fine powders, such as SF, NC, NS signifi-
cantly increased the sd and viscosity of cement paste, and the addi-
tion of AG also increased the sd and viscosity of cement paste.
Although FA and GS were finer than cement, they slightly affected,
and may increase or decrease sd of cement paste. As observed in
[22,27,28], the effects of FA and GS on rheological properties
depend on their fineness of particles and chemical compositions.
After resting for 120 min, sd and viscosity of cement paste also
increased. The addition of NC, AG, NS, GS, and SF increased the
thixotropic area of cement paste. However, the addition of FA
decreased the thixotropic area of cement paste with low w/c of
0.35. The effect of mineral admixture on thixotropic area showed
different dependence on time. This might because some cement
pastes became too stiff after resting for 120 min, and severe slip-
page might happen during dynamic shearing of paste. Thus, due
to this measurement error, thixotropic area of cement paste at
120 min might decrease in comparison with initial cement paste.
3.2. Static yield stress testing results
Typical evolutions of shear time vs. shear stress are given in
Fig. 3. Since the measuring capacity of the rheometer was 450 Pa,
some pastes reached the limit earlier than 120 min, and the testing
number of ss was smaller than 8. Table 3 lists the number of effec-
tive testing. The smaller number of effective testing is, the faster
the ss develops. It can be seen from Fig. 3 that ss increased with
time. During the static yield stress test, the time to reach the peak
value also increased with time. This means that paste resting for
longer time can withstand larger deformation prior to the initia-
tion of the flow of paste.
The evolution of ss of cement pastes with different mineral
admixtures and w/c are given in Fig. 4. As can be seen, the mineral
admixtures affect the growth of ss differently. Growth of ss is due
to the structural build-up of paste at rest. Quick build-up of cement
paste leads to fast growth of ss. The growth rate of ss is given in
Table 3. A similar approach as literature [1] was used to calculate
the structural build-up rate of cement paste, which was repre-
sented by the growth rate of yield stress: (ss,t ss,15)/(t 15),
where ss,t is the last effective static yield stress at time t, ss,15 is
the static yield stress at 15 min. In this equation, the cement paste
was implicitly considered as a Bingham flow.
Take the pure cement paste with the same w/c ratio as reference
value, and the effect of mineral admixtures on the growth of ss can
be nominalized. When the nominal value of paste with mineral
admixture is greater (or smaller) than 1, the mineral admixture
has a positive (or negative) effect on the growth of ss. It can be seen
from Table 3 that FA and GS had negative effect on the growth of ss.
While SF, NC, NS, and AG had positive effect on the growth of ss, and
the positive effect was more obvious at low w/c ratio. NS was the
most effective mineral admixture to increase the growth rate of
ss, followed by AG, SF, and NC. Low w/c ratio led to more positive
effect of mineral admixture on the structural buildup. At the w/c
of 0.45, the structural buildup rate of cement with NS was 1.47
times as that of reference sample, while it reached 2.71 times at
the w/c ratio of 0.35. However, in comparison with the criteria pro-
posed by Roussel et al. [1] for SCC, the values of Athix obtained in this
study are very low. According to Billberg [29], the values of Athix
was 2.502, 2.99, 4.053, 5.131, 7.8 Pa/min for cement paste with
w/c ratio of 0.42, 0.4, 0.38, 0.36, and 0.34 respectively, and the val-
ues of Athix was 42–59 Pa/min. The structural build-up rates of
cement paste reached by Billberg [29] were comparable to the
results in this study. However, the structural build-up rate of con-
crete was much higher than that of cement paste. This may be

Table 2
Rheological parameters of cement pastes measured by dynamic shear test.

Mix Initial test Final test after resting for 120 min
s (Pa) g (Pa/s) Thixotropic area (Pa/s) s (Pa) g (Pa/s) Thixotropic area (Pa/s)
Measured value Nominal Measured Nominal Measured Nominal Measured Nominal Measured Nominal Measured Nominal
value value value value value value value value value value value
45-0-C 13.86 1.00 0.59 1.00 296.7 1.00 16.56 1.00 0.56 1.00 108.4 1.00
45-0-FA 15.80 1.14 0.61 1.03 349.6 1.18 18.93 1.14 0.57 1.02 187.2 1.73
45-0-GS30 12.39 0.89 0.84 1.42 559.5 1.89 14.53 0.88 0.86 1.54 169.1 1.56
45-0-SF5 25.85 1.87 0.94 1.59 558.6 1.88 30.89 1.87 1.25 2.23 1330.9 12.28
45-0-NC2 15.68 1.13 0.75 1.27 683.7 2.30 23.12 1.40 0.84 1.50 1429.1 13.18
45-0-AG1 14.77 1.07 0.71 1.20 519.8 0.44 20.92 1.26 0.72 1.29 466.94 4.30
45-0-NS2 36.58 2.64 3.72 6.31 2358.2 7.95 44.97 2.72 6.71 11.98 3074.7 28.36
35-0.1-C 0.78 1.00 0.73 1.00 457.1 1.00 21.22 1.00 0.96 1.00 1025.1 1.00
35-0.1-FA20 6.29 8.06 0.40 0.55 194.5 0.43 13.81 0.65 1.32 1.38 641.9 0.63
35-0.1-GS30 0.41 0.53 0.81 1.11 622.6 1.36 17.95 0.85 1.05 1.09 629.6 0.61
35-0.1-SF5 32.47 41.63 0.95 1.30 495.1 1.08 38.95 1.84 2.17 2.26 2216.6 2.16
35-0.1-NC2 5.20 6.67 0.85 1.16 717.6 1.57 31.17 1.47 1.23 1.28 1971.7 1.92
35-0.1-AG1 16.65 21.35 0.80 1.10 1234.0 2.70 26.79 1.26 1.21 1.26 587.9 0.57
35-0.2-NS1 22.31 28.60 0.84 1.15 728.9 1.59 54.41 2.56 1.35 1.41 1969.2 1.92
 Q. Yuan et al. / Construction and Building Materials 160 (2018) 117–126 121

200
400 180
160
120min

Shear stress (Pa)

140
Shear stress (Pa)

300 120min
120
100 105min
200 105min 80
90min 60 90min
100 75min 40 75min
60min 60min
45min 20 45min
30min 30min
0 15min 0 15min
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Time (s) Time (s)
(a) FA 20% w/c=0.45 (b) FA 20%+0.1%SP W/C=0.35
1
Fig. 3. Static yield stress test of cement pastes measured at 0.02 s every 15 mins.

Table 3 ate the hydration process of pastes. Two characteristic parameters

Number of effective testing of cement pastes. were used to evaluate the curve of heat liberation curve: (1) after
Mix Number of effective Growth rate of yield Nominal value the initial contact of cement and water, the time required to
testing stress (Pa/min) reach the first heat liberation peak, Tp; (2) the heat liberation rate
45-0-C 8 4.13 (Ref.) 1.00 at the peak, Hp. In order to evaluate the effect of mineral admix-
45-0-FA 8 3.53 0.85 tures on the hydration, pure cement pastes were taken as refer-
45-0-GS30 8 2.78 0.67 ence. If nominal Tp of paste with mineral admixture is greater
45-0-SF5 7 4.16 1.01
(or smaller) than 1, the mineral admixture decelerates (or accel-
45-0-NC2 7 4.23 1.02
45-0-AG1 8 4.60 1.11 erates) the hydration process. If nominal Hp of paste with mineral
45-0-NS2 5 6.09 1.47 admixture is greater (or smaller) than 1, the hydration intensity is
35-0.1-C 8 2.29 (Ref.) 1.00 strengthened (or weakened). Obviously, NS was the most effec-
35-0.1-FA20 8 1.58 0.69
tive mineral admixture, as shown in Table 4. It significantly accel-
35-0.1-GS30 8 1.87 0.82
35-0.1-SF5 7 4.23 1.85
erated the hydration process and intensified the hydration of
35-0.1-NC2 7 4.43 1.93 paste at the same time, and higher addition level led to more sig-
35-0.1-AG1 6 2.59 1.13 nificant effects. AG, NC and SF had a slight positive effect on the
35-0.2-NS1 5 6.21 2.71 hydration of cement. The use of GS postponed the occurrence of
the initial liberation peak, i.e. the hydration process of was decel-
erated, and the hydration intensity was also weakened. This is
explained by twofold reasons. First, differences in testing method
contrary to the results reached in [30], where the hydration of
for ss may be partially responsible for this. Secondly, aggregates
cement paste was improved by the addition of GS. Chemical com-
may strengthen the thixotropic behavior of concrete. Which factors
positions of GS vary much, and this may be responsible for the
play a dominant role needs further study.
difference. The results on FA are consistent with [31]. It was
found that nano-particles had positive effect on the hydration
3.3. Calorimetric testing results of cement, and this was consistent with [32]. Nucleation seeding
for CASAH provided by nano-particles may be responsible for
The curves of hydration heat liberation of pastes are given in this.
Fig. 5. The additions of mineral admixtures accelerate or deceler-

500 35-0.1-SF 5
500 45-0-SF 5
45-0-FA 20 35-0.1-FA 20
Static yield stress (Pa)

45-0- GS 30 400 35-0.1- GS 30

Static yield stress (Pa)

400
45-0-NC 2 35-0.1-NC 2
45-0-AG 1 35-0.1-AG 1
300 45-0-NS 2 300 35-0.2-NS 1
45-0-C 35-0.1-C
200 200

100 100

0 0

0 15 30 45 60 75 90 105 120 0 15 30 45 60 75 90 105 120

Time (min) Time (min)
a w/c=0.45 (b)w/c=0.35

Fig. 4. Evolution of static yield stress of cement pates with time.

122 Q. Yuan et al. / Construction and Building Materials 160 (2018) 117–126

3.4. Zeta potential and conductivity testing results tures on the pH value of the diluted cement paste were little. These
results are reasonable. For most cement, the solution is saturated
The zeta potential of suspension with different mineral admix- with calcium hydroxide within 12 min, and the K+ and Na+ concen-
tures was measured, as shown in Fig. 6. The mineral admixtures trations increased rapidly during the first 12 min and remain rela-
were all negatively charged in aqueous solution, and the zeta tively constant afterwards [24]. The chemical reaction of mineral
potential remained constant with elapsed time, except AG. The admixture is negligible within 75 min. It can be seen that the con-
fluctuation of zeta potential of AG was probably due to the diffi- centrations of Ca2+, Na+, K+ remain relatively constant at the early
culty in dispersing AG powder in water. The evolution of zeta age, and they determine the concentration of OH . Thus, the pH
potential and electrical conductivity of diluted cement pastes are values of cements with various mineral admixtures at early age
given in Figs. 7 and 8. As can be seen, all the zeta potential of were similar. However, the zeta potential and conductivity
diluted cement pastes were positive. Cement paste is a compli- depends on the total ions in solution, which is significantly affected
cated suspension with particles differently charged. The initial zeta by the addition of mineral admixtures. Thus, the zeta potential and
potential of CSH phases is negative, but may become positive as conductivity of diluted paste were significantly affected by mineral
the Ca2+ concentration in the cement pore solution increases and admixtures. With the elevation of w/c, the conductivity of suspen-
Ca2+ adsorb onto the surface of CSH phases [33,34]. Ettringite sion increased and zeta potential decreased.
and monosulfate show positive zeta potentials, and syngenite,
portlandite and gypsum show zero or negative zeta potentials 3.5. Discussion
[35]. Thus, the measured zeta potential of cement suspension is
an average value of all the particles presented in the suspension. Thixotropy/structural build-up has been studied in numerous
Conductivity of cement paste depends on the ionic concentration literature. Thixotropic area and growth of yield stress are often
and ionic conductivity [35]. Under the effect of cement hydration, used to evaluate the thixotropy of cement paste. The results
conductivity and zeta potential of cement paste increased with obtained in this study are summarized in Table 6. Again, pastes
time. The average pH, conductivity and zeta potential of diluted without any mineral admixtures were taken as reference. It can
cement pastes with different mineral admixtures over 70 min are be seen that the growth of sd was two orders of magnitude lower
given in Table 5. It can be seen that the effects of mineral admix- than that of ss, and growth of sd gave different orders of thixotropic

0.007
0.006
Heat of release hydration rate (W/g)

Heat of release hydration rate (W/g)

35-0.1-FA 20
Heat of release hydration rate (W/g)

Heat of release hydration rate (W/g)

0.006
0.007 45-0-FA 20
0.005 35-0.1-GS 30 0.006 0.006
0.004 45-0-GS 30
0.005 35-0.1-SF 5 0.005
0.003 0.004 45-0-SF 5
0.002 35-0.1-NC2 0.005 0.003
45-0-NC 2
0.004
0.001 35-0.1-AG 1 0.002

0.000
35-0.2-NS 1
0.001 45-0-AG 1
0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.004 0.000
45-0-NS 1
Time (h) 35-0.1-NS 2 0.0 0.5 1.0 1.5 2.0 2.5 3.0
0.003 35-0.1-C
Time (h)
45-0-NS 2
0.003 45-0-C
0.002
0.002

0.001 0.001

0.000 0.000
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28
Time (h) Time (h)
a w/c=0.35 (b)w/c=0.45
Fig. 5. Hydration heat liberation curve of different cement pastes.

Table 4
Two characteristic parameters of hydration of different cement pastes.

Mix Time to reach the peak (Tp, h) Heat rate at the peak (Hp, J/h g)
Measured value Nominal value Measured value Nominal value
45-0-C 6.092 (Ref.) 1.00 0.00283 (Ref.) 1.00
45-0-FA20 6.567 1.08 0.00230 0.81
45-0-GS30 6.250 1.03 0.00294 1.04
45-0-SF5 5.902 0.97 0.00294 1.04
45-0-NC2 6.235 1.02 0.00282 1.00
45-0-AG1 6.414 1.05 0.00279 0.99
45-0-NS1 5.236 0.86 0.00323 1.14
45-0-NS2 4.422 0.73 0.00360 1.27
35-0.1-C100 6.747 (Ref.) 1.00 0.00299 (Ref.) 1.00
35-0.1-FA20 7.226 1.07 0.00249 0.83
35-0.1-GS30 6.745 1.00 0.00229 0.77
35-0.1-SF5 6.092 0.90 0.00311 1.04
35-0.1-NC2 6.589 0.98 0.00304 1.02
35-0.1-AG1 6.736 1.00 0.00305 1.02
35-0.2-NS1 5.911 0.88 0.00341 1.14
35-0.2-NS2 4.788 0.71 0.00376 1.26
 Q. Yuan et al. / Construction and Building Materials 160 (2018) 117–126 123

of hydration heat liberation and conductivity of paste, one will find

0
FA obviously decreased the hydration of cement and thus
-10 decreased the thixotropic property. Therefore, it can be concluded
that thixotropic area and growth of sd are inappropriate indicators
Zeta potential (mV)

-20
-30
-40
-50
-60
-70
-80
0 2
Total elasped time (min)
Fig. 6. Zeta potential of suspensions with different mineral admixtures.
behavior for different pastes. This is consistent with the results by
Billberg [29]. Thus, the growth of sd is inappropriate to character-
ize the thixotropy of pastes. When comparing the nominal value of
growth of ss with that of thixotropic area, these two methods gave
different ranking on thixotropy of cement pastes. For instance,
both FA and GS increased the thixotropic area of cement pastes,
while growth of ss gave the opposite results. By analyzing the data
of structural build-up behavior of cement paste.
Hydration is the most important factor affecting the thixotropy
of cement paste. Hydration heat and conductivity indirectly reflect
the hydration process of cement paste. In order to evaluate the
relationship between these parameters and thixotropy, the
increase rate of ss, conductivity and the cumulated heat liberation
are compared in Fig. 9.
It can be seen from Fig. 9 that increase rates of conductivity and
heat liberation are good indicators of the structural build-up, just
4 6 8 10 12 with few exceptions. Heat liberation at early age mainly depends
on the formation of CSH which plays a significant role in the struc-
tural build-up of cement. However, the reaction of C3A also con-
tributes to part of the heat liberation, which seems have little
effect on structural build-up of cement. What’s more, the same
amount of CSH may have different effects on the structural
build-up for different binders. Conductivity depends on the ionic
types and concentrations in the suspension. It is generally believed
that faster hydration might lead to higher conductivity. However,
some reactions may decrease conductivity. For example, the reac-
tion of silica fume with calcium hydroxide decreases the conduc-
tivity of pore solution [36].

4.6 35-0.1-FA 20 35-0.1-AG 1

7.0
45-0-FA 20 45-0-AG 1 35-0.1-GS 30 35-0.1-NS 1
6.8 4.4
45-0-GS 30 45-0-NS 2 35-0.1-SF 5 35-0.1-NS 2
6.6
4.2 35-0.1-NC 2 35-0.1-C
6.4 45-0-SF 5 45-0-C
Zeta potential (mV)
Zeta potential (mV)

6.2 45-0-NC 2 4.0

6.0 3.8
5.8 3.6
5.6
5.4 3.4
5.2 3.2
5.0 3.0
4.8
2.8
4.6
4.4 2.6
0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70
Total elasped time (min) Total elasped time (min)
(a) w/c=15 (b) w/c=12+HRWRA 0.1%
Fig. 7. The evolution of zeta potential with time of cement pastes with different mineral admixtures.

35-0.1-FA 20 35-0.1-AG 1
11.5 11.0 35-0.1-NS 2
45-0-FA 20 35-0.1-GS 30
45-0-GS 30 35-0.1-SF 5 35-0.2-NS 1
11.0 10.5
45-0-SF 5 35-0.1-NC2 35-0.1-C
Conductivity (mS/cm)

Conductivity (mS/cm)

10.5 45-0-NC2
10.0
10.0 45-0-AG 1
45-0-NS 2
9.5 9.5
45-0-C
9.0 9.0
8.5
8.5
8.0
7.5 8.0
7.0
7.5
0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70
Total elasped time (min) Total elasped time (min)
(a) w/c=15 (b) w/c=12+HRWRA 0.1%
Fig. 8. The evolution of conductivity with time of cement pastes with different mineral admixtures.
124 Q. Yuan et al. / Construction and Building Materials 160 (2018) 117–126

Table 5
Averaged pH, conductivity and zeta potential of different diluted pastes.

w/c Mix pH Zeta Potential (mV) Conductivity (mS/cm)

15 45-0-C 12.38 5.4 9.506
15 45-0-FA20 12.38 5.3 7.891
15 45-0-GS30 12.38 5.1 7.166
15 45-0-SF5 12.46 5.8 8.823
15 45-0-NC2 12.38 5.3 9.418
15 45-0-AG1 12.32 5.4 9.656
15 45-0-NS2 12.30 5.8 9.619
12 35-0.1-C 12.39 3.5 9.956
12 35-0.1-FA20 12.36 3.5 8.229
12 35-0.1-GS30 12.37 3.0 8.399
12 35-0.1-SF5 12.40 3.9 10.078
12 35-0.1-NC2 12.43 4.0 10.193
12 35-0.1-AG1 12.42 3.8 10.001
12 35-0.1-NS1 12.43 3.5 9.749
12 35-0.1-NS2 12.40 3.8 9.780

Table 6
Thixotropic area and growth of dynamic yield stress and static yield stress.

Mix Dynamic yield stress slope (pa/min) Static yield stress slope over 120 min (pa/min) Thixotropic area (pa/s)
Measured value Nominal value Measured value Nominal value Measured value Nominal value
45-0-C 0.02 (Ref.) 1.00 4.13 (Ref.) 1.00 296.7 (Ref.) 1.00
45-0-FA20 0.06 3.00 3.53 0.85 349.6 1.18
45-0-GS30 0.02 1.00 2.78 0.67 559.5 1.89
45-0-SF5 0.06 3.00 4.16 1.01 558.6 1.88
45-0-NC2 0.06 3.00 4.23 1.02 683.7 2.30
45-0-AG1 0.05 2.50 4.60 1.11 519.8 1.75
45-0-NS1 0.27 13.50 6.09 1.47 2358.2 7.95
35-0.1-C 0.17 (Ref.) 1.00 2.29 (Ref.) 1.00 457.1 (Ref.) 1.00
35-0.1-FA20 0.03 0.18 1.58 0.69 194.5 0.43
35-0.1-GS30 0.15 0.88 1.87 0.82 622.6 1.36
35-0.1-SF5 0.04 0.24 4.23 1.85 495.1 1.08
35-0.1-NC2 0.22 1.29 4.43 1.93 717.6 1.57
35-0.1-AG1 0.03 0.18 2.59 1.13 1234.0 2.70
35-0.1-NS2 0.07 0.41 6.21 2.71 728.9 1.59

Gauffinet-Garrault [37] examined the rheological properties

during the setting process of cement paste, and argued that CSH 4.0
gel is responsible for the structural build-up of cement paste. Thus,
3.5
they proposed mechanical efficiency of CSH to address the contri-
bution of CSH gel to the structural build-up. The mechanical effi- 3.0
Nominal value of structural buildup rate
ciency of CSH was defined as the increase in elastic modulus,
Nominal value

2.5 Nominal value of heat liberation increase rate

determined by small amplitude oscillatory shear test, vs. the for- Nominal value of conductivity increase rate
mation of a certain amount of CSH, estimated by the hydration 2.0
heat liberation. It is worth to mention that the contribution of 1.5
CSH to structural build-up is a physical effect which results from
the preferential precipitation of CSH gel at the contact points 1.0
between cement particles. With the increase of the amount of 0.5
CSH gel, a percolating network will be formed in the mixture and
35 .1-G 20
-0 FA C
45 0-N 5

35 -0.1 S30
-0 G1

35 .1- C2

S1
45 0-F -C
45 GS 0
45 -0-S 30

45 -A 2

S2

35 0.1 SF5
- 0 --

.1 1

the ‘‘strength” of the network develops progressively with the

-0 AG
- F
-0 A2

-0 C

35 0.1- .1-

-N
- -0

-N

-0 -N
- -
45 45

increase in contact points. Faster cement hydrates, and more CSH

-
-

35 35

are generated, and faster cement builds up. However, the contribu-
tion of physical interaction between particles due to Van der Waal
force or electrostatic force may also contribute to the structural
build-up of cement.
In this study, the growth of ss over the cumulated heat libera-
tion was used to evaluate the contribution of cement hydration
to structural build-up, as shown in Fig. 10. Since ss was measured
every 15 min and some pastes reached the capacity limit of the
rheometer within 75 min, the data between 30 and 75 min were
used. Mechanical efficiency of CSH of cement pastes with different
mineral admixtures based on ss and hydration liberation over 30–
75 min are given in Table 7. It can be seen that the effect of mineral
admixtures on mechanical efficiency of cement pastes depends on
w/c ratio and the addition of HRWRA. SF and NC significantly
Fig. 9. Nominal increase rate of static yield stress, heat flow and conductivity over
30–75 min.
increased the mechanical efficiency of paste without HRWRA,
and little effects were observed in the case of paste with HRWRA.
FA generally decreased the mechanical efficiency, and NS and AG
always increased the mechanical efficiency.
Low mechanical efficiency of paste means that unit hydration
heat results in low growth of yield stress. According to Gauffinet-
Garrault [37], CSH is the main hydration products at early age.
Thus, it can be roughly said that low mechanical efficiency means
that unit CSH results in low growth of yield stress. The factors
 Q. Yuan et al. / Construction and Building Materials 160 (2018) 117–126 125

0.45-0-NS2 of ss. However, high mechanical efficiency of cement paste

400 0.35-0.2-NS1 is not always lead to high growth rate of ss, because both
regression line physical and chemical effects contribute to the growth of
ss, while only chemical effect contributes to the hydration
Static yield stress (Pa)

regression line
300
heat.

200

Acknowledgements
100
Financial supports by National Key R&D Program of China (con-
tract No. 2017YFB0310100) and National Natural Science Founda-
0
tion of China (contract No. 51778629) are greatly appreciated.
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
Heat flow (J/g)
References
Fig. 10. Cumulated heat flow vs. static yield stress.
[1] N. Roussel, A thixotropy model for fresh fluid concretes: theory, validation and
applications, Cem. Concr. Res. 36 (2006) 1797–1806.
[2] J. Assaad, K.H. Khayat, Effect of viscosity-enhancing admixtures on formwork
Table 7 pressure and thixotropy of self-consolidating concrete, ACI Mater. J. 103 (2006)
Mechanical efficiency of CSH of pastes with various mineral admixtures based on 280–287.
static yield stress and hydration liberation over 30–75 min. [3] K.H. Khayat, J. Assaad, Effect of w/cm and high-range water-reducing
admixture on formwork pressure and thixotropy of self-consolidating
Mix Mechanical efficiency Nominal value of concrete, ACI Mater. J. 103 (2006) 186–193.
of CSH (Pa.g/J) mechanical efficiency [4] J. Assaad, K.H. Khayat, Formwork pressure of self-consolidating concrete made
45–0-C 63.8 (Ref.) 1.00 with various binder types and contents, ACI Mater. J. 102 (2005) 215–223.
45–0-FA20 36.6 0.57 [5] K.H. Khayat, M. Saric-Coric, F. Liotta, Influence of thixotropy on stability
characteristics of cement grout and concrete, ACI Mater. J. 99 (2002) 234–241.
45–0-GS30 65.8 1.03
[6] K. Wang, S.P. Shah, G. Lomboy, N. Taylor, P. Wiegand, J. Grove, Self-
45–0-SF5 90.1 1.41
Consolidating Concrete—Applications for Slip-Form Paving: Phase II, Iowa
45–0-NC2 92.2 1.45
State University, 2011 136.
45–0-AG1 66.1 1.04 [7] Z. Malaeb, H. Hachen, A. Tourbah, T. Maalouf, N.E. Zarwi, F. Hamzeh, 3D
45–0-NS2 132.4 2.08 concrete printing: machine and mix design, Int. J. Civil. Eng. 6 (2015) 14–22.
35–0.1-C 106.7 (Ref.) 1.00 [8] N. Roussel, F. Cussigh, Distinct-layer casting of SCC: the mechanical
35–0.1-FA20 96.5 0.90 consequences of thixotropy, Cem. Concr. Res. 38 (2008) 624–632.
35–0.1-GS30 96.6 0.91 [9] J. Mewis, N.J. Wagner, Thixotropy, Adv. Colloid Interface Sci. 147–148 (2009)
35–0.1-SF5 106.0 0.99 214–227.
35–0.1-NC2 105.5 0.99 [10] H.A. Barnes, Thixotropy — a review, Non-Newtonian Fluid Mech. 70 (1997) 1–
35–0.1-AG1 158.7 1.49 33.
35–0.1-NS1 145.5 1.36 [11] N. Roussel, G. Ovarlez, S. Garrault, C. Brumaud, The origins of thixotropy of
fresh cement pastes, Cem. Concr. Res. 42 (2012) 148–157.
[12] S. Jarny, N. Roussel, S. Rodts, R. Le Roy, P. Coussot, Rheological behavior of
cement pastes from MRI velocimetry, Cem. Concr. Res. 35 (2005) 1873–1881.
[13] Z. Quanji, G.R. Lomboy, K.J. Wang, Influence of nano-sized highly purified
magnesium alumino silicate clay on thixotropic behavior of fresh cement
pastes, Constr. Build. Mater. 69 (2014) 295–300.
affecting the mechanical efficiency of CSH are still an open ques- [14] J. Assaad, Formwork pressure of self-consolidating concrete influence of
tion. It might be associated with the physical properties of particle, thixotropy. PhD Thesis, University of Sherbrooke, Canada, 2004.
and needs further study. [15] Q. Yuan, X. Lu, K.H. Khayat, et al., Small amplitude oscillatory shear technique
to evaluate structural build-up of cement paste, Mater. Struct. 50 (2) (2017) 1–
12.
4. Conclusions [16] S. Amziane, A. Perrot, T. Lecompte, A novel settling and structural build-up
measurement method, Meas. Sci. Technol. 19 (10) (2008) 105702.
[17] A.F. Omran, Portable vane test to assess structural build-up at rest of self-
Based on studies carried out above, the following conclusions consolidating concrete, ACI Mater. J. 108 (2011) 628–637.
can be drawn: [18] K.H. Khayat, A.F. Omran, S. Naji, P. Billberg, A. Yahia, Field-oriented test
methods to evaluate structural build-up at rest of flowable mortar and
concrete, Mater. Struct. 45 (2012) 1547–1564.
(1) Among NS, NC, FA, AG, SF, and GS, NS is the most effective [19] K.H. Khayat, J.J. Assaad, Use of thixotropy-enhancing agent to reduce
mineral admixture to improve the structural build-up rate. formwork pressure exerted by self-consolidating concrete, ACI Mater. J. 105
Low w/c ratio leads to more positive effect of mineral admix- (2008) 88–96.
[20] A. Perrot, T. Lecompte, P. Estellé, S. Amziane, Structural build-up of rigid fiber
ture on the structural build-up. At the w/c of 0.45, the struc- reinforced cement-based materials, Mater. Struct. 46 (2013) 1561–1568.
tural build-up rate of cement with NS is 1.47 times as that of [21] S. Kawashima, M. Chaouche, D.J. Corr, S.P. Shah, Rate of thixotropic rebuilding
reference sample, while it reaches 2.71 times at the w/c ratio of cement pastes modified with highly purified attapulgite clays, Cem. Concr.
Res. 53 (2013) 112–118.
of 0.35. [22] R.S. Ahari, T.K. Erdem, K. Ramyar, Thixotropy and structural breakdown
(2) Thixotropic behavior characterized by the growth of yield properties of self consolidating concrete containing various supplementary
stress and thixotropic area give different ranking on cement cementitious materials, Cem. Concr. Compos. 59 (2015) 26–37.
[23] M.K. Rahman, M.H. Baluch, M.A. Malik, Thixotropic behavior of self-
pastes with various mineral admixtures. The growth of sd is compacting concrete with different mineral admixtures, Constr. Build.
two orders magnitude lower than that of ss. Thixotropic area Mater. 50 (2014) 710–717.
is also not suitable to characterize the structural build-up of [24] P.-C. Aïtcin, High-performance concrete, E & FN SPON, 2004.
[25] S. Kawashima, P.K. Hou, D.J. Corr, S.P. Shah, Modification of cement-based
cement paste at rest.
materials with nanoparticles, Cem. Concr. Res. 36 (2013) 8–15.
(3) The liberation rate of hydration heat and increase rate of [26] J.J. Assaad, J. Harb, Y. Maalouf, Effect of vane configuration on yield stress
conductivity are good indicators of the growth rate of ss. measurements of cement pastes, J. Non-Newtonian Fluid Mech. 230 (2016)
(4) Mechanical efficiency of CSH based on the growth of yield 31–42.
[27] A.I. Laskar, S. Talukdar, Rheological behavior of high performance concrete
stress over cumulated hydration heat can be used to charac- with mineral admixtures and their blending, Constr. Build. Mater. 22 (2008)
terize the contribution efficiency of CSH to the growth rate 2345–2354.
126 Q. Yuan et al. / Construction and Building Materials 160 (2018) 117–126
[28] C.K. Park, M.H. Noh, T.H. Park, Rheological properties of cementitious materials
containing mineral admixtures, Cem. Concr. Res. 35 (2005) 842–849.
[29] P. Billberg, Form Pressure Generated by Self-Compacting Concrete – Influence
of Thixotropy and Structural Behaviour at Rest, School of Architecture and the
Built Environment Division of Concrete Structures, Royal Institute of
Technology, Stockholm, Sweden, 2006.
[30] E. Gruyaert, N. Robeyst, N.D. Belie, Study of the hydration of Portland cement
blended with blast-furnace slag by calorimetry and thermogravimetry, J.
Therm. Anal. Calorim. 102 (2010) 941–951.
[31] N.B. Singh, M. Kalra, M. Kumar, S. Rai, Hydration of ternary cementitious
system: Portland cement, fly ash and silica fume, J. Therm. Anal. Calorim. 119
(2015) 381–389.
[32] L. Raki, J. Beaudoin, R. Alizadeh, J. Makar, T. Sato, Cement and concrete
nanoscience and nanotechnology, Materials 3 (2010) 918–942.
[33] H. Viallis-Terrisse, A. Nonat, J.C. Petit, Zeta-potential study of calcium silicate
hydrates interacting with alkaline cations, J. Colloid Interface Sci. 244 (2011)
58–65.
[34] J. Plank, C. Hirsch, Impact of zeta potential of early cement hydration phases on
superplasticizer adsorption, Cem. Concr. Res. 37 (2007) 537–542.
[35] K.B. Sanish, N. Neithalath, M. Santhanam, Monitoring the evolution of material
structure in cement pastes and concretes using electrical property
measurements, Constr. Build. Mater. 49 (2013) 288–297.
[36] C. Shi, Effect of mixing proportions of concrete on its electrical conductivity
and the rapid chloride permeability test (ASTM C1202 or ASSHTO T277)
results, Cem. Concr. Res. 34 (2004) 537–545.
[37] S. Gauffinet Garrault, The Rheology of Cement During Setting, in
Understanding the Rheology of Concrete, Woodhead Publishing, 2012.