See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/316218691

The review of early hydration of cement-based

materials by electrical methods

Article in Construction and Building Materials · August 2017

DOI: 10.1016/j.conbuildmat.2017.04.073

CITATIONS READS

0 127

8 authors, including:

Shengwen Tang Xinhua Cai

Wuhan University Wuhan University
22 PUBLICATIONS 181 CITATIONS 18 PUBLICATIONS 89 CITATIONS

SEE PROFILE SEE PROFILE

Wei Zhou Hongyu Shao

Wuhan University The Hong Kong University of Science and Tec…
150 PUBLICATIONS 677 CITATIONS 17 PUBLICATIONS 145 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Micromechanics of granular materials and Simulation View project

Binary Mixture of Granular Materials View project

All content following this page was uploaded by Shengwen Tang on 24 April 2017.

The user has requested enhancement of the downloaded file.

 Construction and Building Materials 146 (2017) 15–29

Contents lists available at ScienceDirect

Construction and Building Materials

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

Review

The review of early hydration of cement-based materials by electrical

methods
S.W. Tang a,b, X.H. Cai a, Z. He a, W. Zhou a, H.Y. Shao b, Z.J. Li b, T. Wu c, E. Chen b,⇑
a
State Key Laboratory of Water Resources and Hydropower Engineering Science, Wuhan University, Wuhan, China
b
Department of Civil and Environment Engineering, The Hong Kong University of Science and Technology, Kowloon, Clear Water Bay, Hong Kong
c
School of Automation, China University of Geoscience, Wuhan, China

h i g h l i g h t s

The evaluation of early hydration by electrical methods is reviewed.

Merits and demerits of different electrical methods are commented.

Some hydration characterizations by electrical methods are presented.

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

Article history: This paper reviews the evaluation of hydration of cement-based materials by means of electrical methods
Received 5 September 2016 from the view point of direct and alternative current methods. The merits, demerits and application sce-
Received in revised form 30 March 2017 narios of electrical methods for hydration characterization are described. Influencing factors (pore solu-
Accepted 10 April 2017
tion conductivity, porosity, pore size distribution and pore connectivity) of electrical methods are briefly
commented. Besides, the predictions of setting time and shrinkage, and the determination of hydration
stages by electrical methods are presented. The combinations of some mainstream computer simulations
Keywords:
and electrical methods are reviewed.
Cement
Pore
Ó 2017 Elsevier Ltd. All rights reserved.
Hydration
Porosity
Shrinkage
Setting
Hydration stages
Electrical
Impedance
Computer simulation

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2. Experimental methods by electrical methods for hydration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.1. Direct current method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.2. Alternating current impedance spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.3. Non-contact electrical resistivity measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.4. Non-contact impedance measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.5. Four points Wenner probe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3. Influencing factors of electrical methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.1. Pore solution conductivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2. Porosity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.3. Pore size distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

⇑ Corresponding author.
E-mail address: victorymother@163.com (E. Chen).

http://dx.doi.org/10.1016/j.conbuildmat.2017.04.073
0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
16 S.W. Tang et al. / Construction and Building Materials 146 (2017) 15–29

3.4. Pore connectivity or tortuosity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

4. Hydration characterization by electrical methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.1. Prediction of setting time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.2. Prediction of shrinkage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.3. Determination of hydration stages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
5. Simulated methods for hydration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
6. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

1. Introduction tion, to examine the precision of non-contact electrical resistivity

measurement, Li and Wei employed standard KCl solutions with
Cement is the most used artificial material in the world, and different concentrations to replace ring-shaped cement-based
more than 2.5 tons of cement are consumed per person yearly materials. It was found that the maximal relative error of the mea-
[1,2]. Hydration of cement usually involves many complex reac- surement was around 0.1℅ [15]. Similar calibration was also
tions, consumes free water, and generates reaction products. Dur- implemented by Tang et al. to examine the precision of non-
ing this process one scale of porosity, namely, capillary porosity, is contact impedance measurement. The maximal relative error of
consumed and a finer scale of porosity, the so-called ‘‘gel pore” is this measurement was approximately 3℅ [5]. In addition, the reli-
produced. The continuously evolving network of pores governs ability of measured electrical data could be examined by Kramers-
the physical properties of cement-based materials during hydra- Kronig transformation [16].
tion, as well as the ultimate physical properties, including strength, In general, electrical methods can be classified into direct cur-
permeability, shrinkage, and creep. rent (DC) methods, alternating current impedance spectroscopy
A wide variety of methods have been utilized to determine the (ACIS) with two or four metallic electrodes, non-contact resistiv-
hydration in cement-based materials [3–6]. Electrical methods ity/impedance measurement and Wenner probe [17–19]. The
have distinct advantages over other microstructure-sensitive tech- working principles and evaluations of different electrical methods
niques, such as measurements of surface area using Brunauer- are demonstrated as follows:
Emmett-Teller analysis, mercury intrusion porosimetry, and elec-
tron microscopy, because cement paste is electrically conductive
2.1. Direct current method
by virtue of its interconnected pore network filled with water con-
taining mobile ions. It is not surprising that electrical properties,
Theoretically, the electrical response under the direct current
most notably conductance, have been employed for a long time
electrical field is able to reflect the ion transportation during the
to study the evolving hydration of neat pastes, pastes with various
hydration. However, very limited academic examples associated
admixtures, mortars and concretes. Electrical methods are nonin-
with the investigation of hydration characterization by direct cur-
vasive and nondestructive so that the features of microstructure
rent methods are found in the literatures, as direct current meth-
of cement-based materials can be captured in situ without
ods have distinct electrode polarization effects caused by contact
recourse to the water removal, usually by heating, which is known
issues although four-probe direct current methods are claimed to
to permanently alter paste microstructure. In addition, electrical
eliminate the effects [20,21].
methods may be performed rapidly and continuously during
Susanto et al. reported that the direct current brought about
hydration. In the past decades, plenty of works have focused on
additional microstructural changes in the bulk mortar matrix and
the resistivity/conductivity response of cement-based materials
thereby affected mechanical performance. The direct current flow
[7–11].
(at around the level of 10 mA/m2) was found to cause initial densi-
This work reviews some representative electrical methods
fication of the bulk matrix at early stage (until 14 days) by reducing
(direct current method, alternative current impedance spec-
porosity and critical pore size [22], whereas coarsening of the
troscopy, non-contact methods and Wenner probe) for hydration
material, calcium leaching, accelerated ion and water migration
study of cement-based materials. The influencing factors (pore
were observed at late hydration stage [22]. He et al. held the sim-
solution conductivity, porosity, pore size distribution, pore connec-
ilar view that the applied external direct current voltages could
tivity or tortuosity) of electrical methods are discussed emphati-
change the pH value of pore solution in cement-based materials
cally. Hydration features (setting time, shrinkage, determination
to some degree, and thus, had prompted the desorption of ion-
of hydration stages) characterized by electrical method are briefly
bonding [23].
discussed. Besides, some mainstream computer simulations based
on electrical methods are presented.
2.2. Alternating current impedance spectroscopy

2. Experimental methods by electrical methods for hydration
Prior to electrical tests for hydration characterization, some cal-
ibrations are necessary to be performed in order to check the reli-
ability of measured reading [12], especially for cases in which
irregular stimulating mode of electrical fields is used [13]. The uti-
lization of standard solutions is known as the ideal calibration
choice. Soliman et al. utilized a standard NaCl solution with known
conductivity and measuring the electrical resistivity of the elec-
trode probe in alternating current impedance spectroscopy (ACIS)
tests to calculate the electrode constant [14]. Apart from NaCl solu-
It has been proved that electrical response of cement-based
materials is frequency-dependent [5]. Some literatures reported
that the electrical response behaved like resistor-capacitor circuits,
and there was a noticeable dispersion in impedance at different
frequencies of applied alternative current [24,25]. This dispersion
with frequencies was primarily due to the inertial motion or the
in-cage vibration of ions [26]. Kou et al. employed molecular
dynamics to study the electromanipulating transport phenomenon
for the net flux of water in the nanochannels and found that the ion
transportation in pore solution water was strongly affected by fre-
quencies of applied alternative current [27].
 S.W. Tang et al. / Construction and Building Materials 146 (2017) 15–29
Furthermore, alternating current (AC) method was capable of
eliminating the polarization encountered in the direct current
method to some degree because charge carriers moved back and
forth under AC condition [28–30], although it was reported that
alternative current methods would also refine the pore structure,
especially for small capillary and gel pores whose sizes were less
than 200 nm [23].
There were two types of frequently-used alternating current
techniques in the hydration investigation of cement-based materi-
als. One was ACIS with metallic electrodes; the other was non-
contact approaches based on the transformer principle.
With regard to ACIS, generally speaking, hydration evolution of
cement-based materials could not be monitored continuously per-
haps due to the intrinsic drawback of test instruments, and thus,
only limited discrete Nyquist plots were obtained during the
hydration process [31]. It was considered that the effective conduc-
tivity was represented by bulk resistance of the sample, which was
determined by the real coordinate of the meeting point of bulk and
electrode arcs in a Nyquist plot [32].
Hydration characterizations by ACIS were widely studied by
McCarter et al. [7–10,33], Gu et al. [34,35], Christensen et al.
[36–38], Novoa et al. [24,39–42]and Neithalath et al. [43–46]. ACIS
method was firstly used by McCarter et al. in 1990 [7,47], subse-
quently developed by Scuderi et al. in 1991 [48]. In their studies,
a Warburg impedance or two constant phase elements were
needed to simulate slow ionic diffusion effects at the electrode-
specimen interface at ultra-low frequencies [7]. The studies of
hydration characterization of cement-based materials by ACIS
methods were blooming later on [37,49].
Usually, equivalent circuit models are needed to interpret the
measured electrical response obtained from ACIS methods during
the hydration process. Due to the different assumptions adopted
in various equivalent circuit models and test conditions (such as
the electrode configuration, test-cell geometry, calibration proce-
dure, operating voltages and signal frequency), the results among
researchers cannot achieve consensus.
In conclusion, ACIS methods usually cover the wide frequency
range of interest, and thus, can extract valuable low and high fre-
quencies response of hydration if the contacting problems, espe-
cially the polarization effect and possible shrinkage or expansion
caused by hydration, can be solved by certain compensation and/
or elimination measures. Some measures has been adopted to
eliminate the negative effect caused by electrodes: wet sponges
[50,51], conductive gel [52], pieces of foam saturated with salt
solution [53] or lime saturated solution [54] was placed between
the specimen and metallic electrodes, and compressed with a
weight or suffered from a constant contact pressure to ensure
the intimate contact between specimen surface and electrodes
[55]. However, these wet substances or solutions also brought in
an additional impedance that played a significant role in the accu-
racy of experimental results [52]. For example, Bu and Weiss cali-
Fig. 1. Non-contact electrical resistivity measurement.
17
brated the resistance of specimens by subtracting the resistance of
sponges at two ends of measurement in ACIS tests [51].
Thanks to the comparably high sensitivity of metallic elec-
trodes, this method can measure discontinuously the electrical
response of cement-based materials at specified hydration age
with the aid of certain equivalent circuit models.
2.3. Non-contact electrical resistivity measurement
In order to eliminate the influence of electrodes, Li et al. in 2003
developed a non-contact electrical resistivity measurement
method to study the hydration mechanism of cement-based mate-
rials [56]. Fig. 1 is the picture of non-contact resistivity measure-
ment. The transformer principle was adopted in this
measurement. When an AC voltage from the generator was applied
to the primary coil of the transformer, a toroidal voltage would be
induced in the ring sample that was considered as the secondary
coil of the transformer. The toroidal current could be measured
by a leakage current meter and hence the resistance of the material
could be obtained based on Ohm’s law [57]. The resistivity was
then calculated from radii of mold, resistance and height of
samples.
The non-contact resistivity measurement has been used to
investigate the influences of chemical and mineral admixtures on
hydration process of cement-based materials [58,59]. These stud-
ies showed that non-contact resistivity measurement could give
a new insight to the evolution of hydration of cement-based
materials.
2.4. Non-contact impedance measurement
In order to overcome the intrinsic drawbacks of ACIS and non-
contact electrical resistivity measurement mentioned above, the
non-contact impedance measurement was developed recently, as
shown in Fig. 2 [1,60–63]. In this measurement, both modulus
and phase of impedance curves of cement-based materials were
recorded continuously. Therefore, experimental results were
slightly different from ones from ACIS and non-contact electrical
resistivity measurement.
Figs. 3–5 are modulus and phase of impedance development of
cement pastes with water to cement ratios 0.3, 0.4 and 0.5. The
data trends in Figs. 3–5 were verified by the proposed electro-
chemical theory. It was found that cement paste with the lower
water to cement ratio had larger impedance modulus and phase
lag. It was also revealed that the separation of curves at different
frequencies reflected the appearance of percolation of hydration
products [61]. This method was employed to study hydration pro-
cess of high-dosage fly ash blended cement pastes, for which gis-
mondine was particularly found in the zeolite formation stage[2].
Besides, the pore size distribution, permeability, chloride diffusiv-
ity and thermal conductivity could be predicted by non-contact
18 S.W. Tang et al. / Construction and Building Materials 146 (2017) 15–29
Fig. 2. Non-contact impedance measurement.
impedance measurement by combining with electrical chemical
theories [61–63].
Non-contact methods are able to monitor hydration evolution
of cement-based materials in-situ. Nevertheless, frequency ranges
of applied stimulus are restricted because the feasible frequency
range of magnetic materials and power amplifiers for transformer
in these methods are limited substantially, and thus, the
frequency-dependent properties of hydration have not been
revealed completely. Besides, these methods are only suitable for
real-time monitoring of hydration at the early age since electro-
magnetic conversion efficiency of transformer and current trans-
ducer in non-contact methods is very low at late hydration period.
2.5. Four points Wenner probe
Four points Wenner probe was a special electrical method to
measure the surface electrical resistivity of cement-based materi-
als [64,65], which was a commercially available equipment (see
Fig. 6). Placing the Wenner probe on one exposed surface of the
cement-based material, the probe produced a low frequency alter-
nating electrical current passing between the two outer electrodes
while the voltage drop between the two inner electrodes was mea-
sured [66].
Although this method is very easy to operate, this probe only
measures the electrical response of hardened cement-based mate-
rials, and is not able to monitor hydration behavior at early age.
Furthermore, experimental results from this method showed
remarkable variation due to heterogeneity of materials.
3. Influencing factors of electrical methods
The electrical response of cement-based materials can be
regarded as an indicator of the process of their hydration process
and micro-structure formation. A variety of factors play important
roles in the hydration characterization by electrical methods. Apart
from external environmental factors, such as sample geometry,
ambient temperature, sample conditioning, four intrinsic influenc-
ing factors of pore phase associated with hydration process are
pore solution conductivity, porosity, pore size distribution, pore
connectivity (or tortuosity) [67–69]. These factors will be dis-
cussed in the following:
3.1. Pore solution conductivity
When the electrical response of cement-based materials is dis-
cussed, pore solution conductivity should be considered firstly. The
pore solution is the primary conducting phase in cement-based
materials and usually contains Ca2+, Na+, K+, OH-, etc, and thus,
has a resistivity several orders of magnitude smaller than those
of solids and vapor phases [25].
An expression for electrical conductivity ðrc Þ for cement-based
materials was developed from the pore solution conductivity by
Weiss et al. [57]:
!
r0p 1
rc ¼ Sn ð1Þ
S F
where r0p was the conductivity of pore solution at saturation status,
F was the formation factor, S was the degree of saturation, n was fit-
ting parameter and called as the saturation coefficient. Apparently,
r0p was claimed to be attributable to the mixture design.
The second term on right-hand side of Eq. (1) accounts for the
total pore space, and the third term is related to the connectivity
of the fluid in the pore space. It seems that each term of Eq. (1)
is independent of the others, and all of the parameters are related
to well-defined properties of the matrix and the solution filling its
pores. It should be stated that the pore size distribution is not
taken into consideration in this simplified analysis.
It is difficult to extract the pore solution of cement-based mate-
rials, especially at the late age. Therefore, the determination of pore
solution conductivity is a challenge task. There are four typical
extraction methods for pore solution:
1) In the work by Ridha et al., pore solution was obtained
through an extraction of the cement sample at certain pres-
sure using Fann 175 ml HPHT API Filter Press. Pore solution
consisted of a filter paper connected to the orifice for fluid
expulsion. Once removed, it was collected directly into a
 S.W. Tang et al. / Construction and Building Materials 146 (2017) 15–29
Fig. 3. Modulus and phase of impedance of cement paste with water to cement ratio 0.3.
glass flask for the conductivity measurement (Core Test Sys-
temTM Auto Lab from New England Research) at specified
hydration periods [18].
2) In Xiao and Li tests, the pore solutions were extracted from
concrete specimens through a filter with a vacuum pump
at given hydration age and through squeezing by a compres-
sion machine at 24 and 48 h [70].The pore solution conduc-
tivity was determined by non-contact electrical resistivity
measurement.
3) Bu and Weiss utilized a high pressure apparatus to obtain
the pore solution from paste and mortar specimens. The
extracted pore solutions were stored in a small glass con-
tainer and capped immediately after the extraction to pre-
vent carbonation. Pore solution resistance was measured
using a Solartron 1260TM Impedance Gain-Phase analyzer
by a geometry factor [51].
4) Spragg et al. extracted the pore solution in the fresh state at
given age by a Millipore pressure filtering system, using
nitrogen gas at pressures up to 200 kPa. With respect to
19
hardened samples, extractions were conducted at ages of 1
d, 3 d, 5 d, and 7 d using a high pressure die at pressures
up to 380 MPa. The extracted solutions were then measured
for resistivity using a pore solution cell [25].
With regard to the evolution of pore solution conductivity, Har-
bec et al. studied pore solution in mortar incorporated with silica
fume and waste glass fume during the first 24 h after the initial
contact between the water and binder, as shown in Fig. 7[71].
The figure showed four distinct hydration phases. The first corre-
sponded to an increase of conductivity during the first 4–6 h. This
phase ended at the circles in Fig. 7. This phase was related to the
dissolution of the various species in the pore solution. The second
phase occurred in the circle labels, where conductivity values
kinked. This corresponded to the precipitation of the first crystals
of CH once the pore solution had reached an oversaturation level
in Ca2+ ions. This kink was associated with the end of the induction
period and the beginning of the mortar setting. The third phase
started past the circle labels and ended at the inflection point
20 S.W. Tang et al. / Construction and Building Materials 146 (2017) 15–29
Fig. 4. Modulus and phase of impedance of cement paste with water to cement ratio 0.4.
located in the square label. In this third phase, conductivity value
dropped rapidly and quasi-linearly. This phase corresponded to
the structuration of the hydrated cement paste in the mortar. At
the inflection point, conductivity value reached a minimum; the
first compressive-strength testing of the mortar could be per-
formed. In the fourth phase, a small conductivity value hump
was framed by the inserted square. This hump may relate to the
decomposition of AFt, the formation of AFm, and the release of
alkalis trapped in the C3A network.
The experimental results obtained from Harbec et al. exhibited
obvious decreasing tendency of conductivity vs. hydration time. In
contrast, Xiao and Li performed the similar pore solution conduc-
tivity test and illustrated that conductivity tended to be stable after
around 80 min [70]. More specifically, it could be seen in Fig. 8 that
the three curves had similar trends, exhibiting a sharp decrease in
resistivity at 40, 60 and 80 min after mixing for samples, followed
by a slight fluctuating trend during the rest time up to 48 h, corre-
sponding to the saturation of ion concentration in the pore
solution.
Experimental techniques have been developed to determine the
resistivity of the pore solution [72], but they are very time-
consuming to be performed at late ages due to the small amount
of pore fluid available. CEMHYD 3D, a numerical cement hydration
model that was developed at National Institute of Standards and
Technology (NIST), was used to simulate the reactions between
cement clinker minerals and water for predicting alkali and hydro-
xyl concentrations in cement pore water [72]. Subsequently, Sny-
der et al. developed an approach to determine the resistivity of
the pore solution using the concentration of different ionic species
[73]. This approach was programmed into a web application by
Bentz [74], which could be found at http://ciks.cbt.
nist.gov/poresolncalc.html. This website was especially powerful
in that it allowed the user to input some parameters, such as
masses of the water, cement and supplementary materials, the
chemical composition of those materials, and the probable degree
of hydration, so as to estimate the pore solution chemistry and
conductivity for different mixtures. Table 1 shows the estimated
pore conductivity simulated from NIST website when water to
cement ratio is fixed as 0.4, Na2O and K2O contents (mass%) in
cement are same as 0.1, hydrodynamic viscosity of pore solution
relative to water is assumed to be 1. The purpose of these selected
parameters is to obtain the suitable conductivities that lie in the
 S.W. Tang et al. / Construction and Building Materials 146 (2017) 15–29 21

Fig. 5. Modulus and phase of impedance of cement paste with water to cement ratio 0.5.

although the determination of some simulated parameters is

uncertain.

3.2. Porosity

The electrical response of cement-based materials depends on

Fig. 6. Four points Wenner probe [66].
not only the conductivity of the pore solution but also the complex
pore structure, especially the porosity, as electrical resistance value
reflects the total free ion amount confined in pores [75]. An et al.
correlated the characteristic of bulk resistance ðRÞ and porosity
ð£Þ of cement pastes by ACIS methods [76]. The exponential rela-
tion between R and £ in Eq. (2) was fitted by a simple series-
parallel model and mercury intrusion porosimetry data. They sta-
ted that from this equation the amount of connected pores
increased remarkably when the porosity was larger than 24.6%.

R ¼ 7:55 þ 414:5eð0:246£Þ=0:084 ð2Þ

range presented in Figs. 7 and 8. From the comparison of conduc-
tivity values in Table 1, it is considered that the simulated results However, the validity of Eq. (2) needs to be further verified by a
are consistent with the ones obtained from work by Xiao and Li great deal of experiments.
22 S.W. Tang et al. / Construction and Building Materials 146 (2017) 15–29

Fig. 7. Electrical conductivity of pore solution as a function of hydration time [71].

2 3
£ð1  £Þ
b¼14 
r r
 5 ð4Þ
2£  effrp s ð3  £Þ

where reff was the effective conductivity of material saturated with

saline solution (solid phase and porous phase filled with elec-
trolyte), es was the dielectric constant of the solid phase, rp was
the conductivity of saline solution, rs was the conductivity of solid
phase, r0 ðxÞ was the real part of the complex conductivity with fre-
quency x, £ was the open porosity.
It was showed by them that the predicted porosity was sub-
stantially larger than those measured porosity obtained from
image and volumetric analysis methods [45]. Besides, with respect
to Eqs. (3) and (4), the adjustments of experimental data to these
equations were considerably sensitive to the selected frequency
range [78]. Sometimes, the upper and lower limits of the frequency
range remained unclear and may be determined by the intrinsic
performance of test instrument, which would cause wide variabil-
ity of porosity values.

Fig. 8. the electrical resistivity of pore solution in concrete with different fly ash 3.3. Pore size distribution
dosages at first 48 h [70].
Academic studies associated with the influence of pore size dis-
tribution on electrical response of cement-based materials are lim-
Neithalath et al. also established the relation of porosity and ited. In most of ACIS tests, Nyquist plot had an obvious depressed
real part of the complex conductivity according to the modified angle ðhÞ that was believed to be related to the pore size distribu-
Bergman equation and experimental fitting [32,45,77]: tion of cement-based materials [24,79]. The depression angle was
"   1b # expressed as the arc depression factor ðnÞ, as shown in Eq. (5):
reff  rs es x
r 0 ð xÞ ¼ r p þ ð3Þ
n ¼ 1  ð2h=pÞ ð5Þ
rp rp

Table 1
Estimated pore conductivity simulated from NIST website.

Hydration degree K+ (mol/L) Na+ (mol/L) OH (mol/L) Estimated pore solution conductivity (S/m)
0 0.04 0.06 0.10 2.28
0.2 0.04 0.07 0.11 2.48
0.4 0.05 0.07 0.12 2.72
0.6 0.05 0.08 0.13 3.01
0.8 0.06 0.09 0.15 3.37
1 0.07 0.1 0.17 3.69
 S.W. Tang et al. / Construction and Building Materials 146 (2017) 15–29
where n was said to be between 0.7 and 0.9, depending upon the
type of cement, water to cement ratio, time and additives [80]. Nik-
lasson considered that n was related to the fractal dimension of the
pore structure [80]. Unfortunately, unified and explicit expressions
between depression angle and pore size distribution were absent.
Cabeza et al. utilized that ACIS electrical impedance measure-
ments with the wide frequency range (0.01–100 MHz) to derive
valuable information of pore network development as well as pore
occupancy by fitting the impedance data into the Debye model
[24], as illustrated by Eq. (6):
C2
CðxÞ ¼ C 1 þ a b
ð1 þ ðjxR2 C 2 Þ Þ
where CðxÞ was the complex capacitance, C 1 was defined by the
dielectric capacitance of the solids, C 2 was directly linked to the
double layer capacitance formed at the C-S-H gel layer limiting
occupied pores, a and b held for the so-called ‘‘symmetrical” disper-
sion of a fitting curve. However, how to correlate a and b to pore
size distribution remained a big challenge in this model.
Furthermore, based on non-contact impedance measurement,
Tang et al. proposed the pore fractal model to investigate the pore
size distribution of cement pastes with different water to cement
ratios [5]. The governing equation in the fractal analysis was illus-
trated by Eq. (7):
D Df 1
f ðdÞ ¼ Df  dmin
f
d
where f ðdÞ was the probability density function of pore size distri-
bution, d and dmin were the given and minimal pore diameters, Df
was fractal dimension for the pore space and derived from the com-
bination of fractal electrical and pore structure networks. dmin was
predicted as 6.2 nm by summing up the thickness of compact layer,
diffuse layer and bulk solution, based on a typical electrical double
layer model. It was well-documented that the pore size distribution
of cement-based materials may not reveal a single fractal behavior
[81,82], therefore, the pore fractal model should be further polished.
3.4. Pore connectivity or tortuosity
Pore connectivity or tortuosity is an intrinsic geometrical prop-
erty of pore structure, and remarkably affects the transport ability
of electrical pathways. With respect to Portland cement pastes,
before the first depercolation of capillary pores, ionic concentration
of pore solution significantly determined bulk paste electrical
resistivity. The reduction of pores connectivity affected the paste
resistivity much more than the decrease of capillary porosity until
fully depercolation time of capillary pores. After that, volume frac-
tion of C-S-H would mainly control the resistivity change of
cement paste [83]. Although a great deal of researchers has real-
ized the significance of tortuosity for hydration characterization
by electrical methods, the explicit equations of tortuosity are still
Table 2
Setting times obtained from ACIS tests and Vicat tests by Sanish et al. [32].
Mix Initial setting time
from Vicat test (min)
Plain cement pastes 210
Limestone blended cement pastes 195
Fly ash blended cement pastes 240
Silica fume blended cement pastes 180
Cement pastes with superplasticizer 0.2% 270
Cement pastes with superplasticizer 0.3% 360
Cement pastes with superplasticizer 0.4% 420
23
in a debate. The frequently-used equation in some works related
to electrical conductivity and pore tortuosity was [84–87]:
rc ¼ r0p  £  b ð8Þ
where b reflected the pore connectivity.
Nevertheless, the strict mathematical derivation of Eq. (8) is
missing. Besides, the solving of b is unknown all the same.
4. Hydration characterization by electrical methods
4.1. Prediction of setting time
ð6Þ
When the hydration procedure is discussed, the setting behav-
ior of cement-based materials cannot be bypassed [88,89]. Bentz
et al. considered that measurements of electrical resistance were
a promising potential for anticipating initial setting times in con-
crete mixtures [31].
Sanish et al. carried out ACIS tests and predicted initial and final
setting times of cement-based materials using electrical conductiv-
ity vs. time curve [32]. They claimed that the predicted setting time
were very close to the measured ones obtained from Vicat tests,
the maximal difference of setting time did not exceed 15 min, as
shown in Table 2.
Liao and Wei in 2014 investigated the setting process and evo-
lution of electrical resistivity of Portland cement pastes with con-
stant water to cement ratio 0.3 and with different dosages of
ð7Þ
naphthalene superplasticizer from 0 to 1.2% [90]. The setting time
and critical time determined by electrical resistivity method were
demonstrated in Table 3. The difference of th (occurring time of the
peak of the hyperbolic curve of electrical resistivity versus time)
and ti (initial setting time determined by Vicat needle tests) was
approximately 1 h, whereas the maximal difference of tr (occurring
time of the peak of rate of change of electrical resistivity versus
time) and tf (final setting time determined by Vicat needle tests)
was up to 3.33 h. The discrepancies from Vicat needle tests and
electrical methods were significantly large in Table 3, which was
completely different from those claimed by Sanish et al. [32]. The
discrepancies can be interpreted by the empirical operation of
Vicat needle test and different test conditions of these two
methods.
4.2. Prediction of shrinkage
Shrinkage of cement-based materials usually occurs at the early
age. It can be inferred that degree of hydration, chemical and auto-
genous shrinkage have a strong dependence on electrical resistiv-
ity. Autogenous shrinkage is defined as the macroscopic volume
contraction of cement-based materials at an early age (less than
one day after casting) occurring without moisture transfer from
materials to the surrounding environment. While chemical shrink-
age is taken as the reduction of the total absolute volume of the
cement system. Apparently, chemical shrinkage is equal to the
Initial setting time from Final setting time Final setting time from
conductivity method (min) from Vicat test (min) conductivity method (min)
200 330 345
200 315 310
230 390 380
175 300 310
285 420 410
350 510 510
425 540 535
24 S.W. Tang et al. / Construction and Building Materials 146 (2017) 15–29

Table 3 csðtÞ ¼ 0:0012qðtÞ þ 0:0068 ðw=c ¼ 0:3; R2 ¼ 0:9984Þ ð13Þ

The setting time and critical time for electrical resistivity method by Liao and Wei
[90].
csðtÞ ¼ 0:0023qðtÞ þ 0:0074 ðw=c ¼ 0:4; R2 ¼ 0:9984Þ ð14Þ
Notations Setting time (h) The occurring time of
peak (h)
ti tf th tr csðtÞ ¼ 0:0042qðtÞ þ 0:0125 ðw=c ¼ 0:5; R2 ¼ 0:9977Þ ð15Þ
SP-0 3.75 5.83 4.67 8.00 Few of research groups focus on the study of shrinkage and
SP-0.6% 6.00 7.92 7.17 11.25
electrical resistivity. The main reason may lie in the variation of
SP-0.8% 6.42 8.08 7.58 11.50
SP-1% 7.42 9.17 8.33 12.25 experimental results, which is proved by the comparison of Eqs.
SP-1.2% 7.92 9.75 8.83 12.75 (11) and (13) for the same water to cement ratio, and lack of com-
prehensive interpretation of correlative mechanism behind.
volumetric summation of autogenous shrinkage and empty cavi-
ties. As hydration proceeds, the decrease in the absolute volume 4.3. Determination of hydration stages
of cement paste and the increase in the volume of the solid phase
occur synchronously. The former will result in an increase in chem- The big achievement of electrical methods is to classify the
ical shrinkage, while the latter will lead to a decrease in the con- hydration stages of cement-based materials [12,88,93–96] as these
ductive channels, and thus, an increase in electrical resistivity. methods can track the structural changes that occur within the
Zuo and Wei studied the influence of water to binder ratio and hydrating cement-based materials from electrical response
fly ash content in cement pastes on electrical resistivity, chemical [97,98].
shrinkage and autogenous shrinkage [91]. They claimed that irre- In general, four or five hydration periods were identified by crit-
spective of water to binder ratio and fly ash content, the chemical ical points on the resistivity or conductivity curves [59,99]. W.J.
shrinkage ðcsðtÞÞ and autogenous shrinkage ðasðtÞÞ of fly ash McCarter, as a prestigious expert in characterizing the electro-
blended cement pastes had a linear relationship with electrical chemical features of hydration of cement-based materials, in
resistivity ðqðtÞÞ at early hydration age, as demonstrated by Eqs. 2013 utilized ACIS methods to monitor the hydration of near-
(9) and (10): surface concrete zone [12,13]. From electrical response (normal-
csðtÞ ¼ K cs qðtÞ þ a ð9Þ ized conductivity vs. time) four stages were clearly discernible
from their experimental results. More specifically, these stages
asðtÞ ¼ K as qðtÞ þ b ð10Þ comprised an initial stage (I) representing the dissolution stage;
a transition stage (II) representing the period of setting of the con-
where K cs , K as , a and b were fitting coefficients. crete and formation of a solid percolation within the binder; a
By fitting the experimental data, Eqs. (9) and (10) for the stage of rapid chemical activity (III) during which the binder devel-
cement paste with water to cement ratio 0.3 were rewritten as: oped rigidity; this was followed by a stage of reduced chemical
csðtÞ ¼ 1092qðtÞ þ 31173 ðR2 ¼ 0:9486Þ ð11Þ activity (IV) which was as a result of gradual infilling of the pore
network due to hydrate formation and pozzolanic reaction [12].
Meanwhile, Tang et al. utilized non-contact impedance measure-
asðtÞ ¼ 2:0992qðtÞ þ 299:76 ðR2 ¼ 0:9863Þ ð12Þ
ment to identify the hydration stages and phase transformation
Liao et al. also correlated the chemical shrinkage to electrical of calcium sulfoaluminate cement at early age [2]. Five hydration
resistivity via hydration degree [92]. It was found that the relation- stages for cement pastes with water to cement ratio 0.7 were
ship between chemical shrinkage ðcsðtÞÞ and electrical resistivity demonstrated in Fig. 9 and named as (I) dissolution stage
ðqðtÞÞ was a positive correlation, as demonstrated by Eqs. (13)–(15) (0–26 min); (II) transformation stage (26–180 min); (III) self-

Fig. 9. Hydration stages of calcium sulfoaluminate cement paste with water to cement ratio 0.7 and relevant temperature development determined by non-contact
impedance measurement.
 S.W. Tang et al. / Construction and Building Materials 146 (2017) 15–29
desiccation stage (180–268 min); (IV) dynamic balance stage
(268–541 min); and (V) acceleration stage (541–1400 min). It
was reported that there may be a unique transformation between
AFt and AFm in three late hydration stages.
Besides, Tang et al. preliminarily analyzed the impedance
response of dissolution, competition and acceleration stages of
Portland cement pastes from the view point of physical-chemical
behaviors[16]:
1) In the dissolution stage, the current of the cement paste ðIÞ
was mainly associated with the applied voltage of electrical
signal ðEÞ and the parameter ðhÞ related to surface morphol-
ogy of cement particles. The complete differential form of
current could be written as:
dI ¼ ð@I=@EÞh dE þ ð@I=@hÞE dh ¼ dE=R þ Kdh ð16Þ
where R was the resistance of the solution, K was a coefficient
with the unit of ampere.
When some empirical data of fresh cement pastes were sub-
stituted into Eq. (16), the impedance response of the cement
paste ðZÞ in the dissolution stage was simplified as Eq. (17):
Z ¼ @E=@I ¼ R  KR=ðjxqÞ ð17Þ
where the value of R was between 250 and 300 O within the
dissolution stage, the term, KR=ðxqÞ, was equal to
2.7  107 X, x was angular frequency.
It could be seen that the estimated imaginary part (107 X
scale) in Eq. (17) was so small that it could be ignored, com-
pared with the real part of the impedance (250–300 O). It
would say that the impedance response of cement pastes
was approximately resistive in the dissolution stage, which
had a good agreement with data sets of Figs. 3–5 and 9,
viz., the flat plateau at the very beginning of hydration.
2) In the competition stage, the current of cement paste ðIÞ had
two governing components: ions dissolution current ðId Þ and
ions adsorption current ðIa Þ. The ions in the solution were
readily adsorbed by the formation of a thin layer of hydra-
tion products, which reduced ion concentration and mean-
while allowed additional ions from cement particles to be
dissolved into the solution. This competition of consuming
and releasing ions led to a dynamic balance of the impe-
dance response. The impedance response in this stage was
given by Eq. (18):
dI ¼ dId þ dIa ¼ dE=Rd þ K d dh þ dE=Ra þ K a dh ð18Þ
where dId and dIa were the change of current Id and Ia , Rd and
Ra were resistances corresponding to current Id and Ia , K d and
K a were coefficients related to the dissolution and adsorption
behavior, respectively.
Since ions dissolution and adsorption were opposite proce-
dures for surface of cement particles in the equilibrium sta-
tus with respect to the competition stage, the admittance
ðY c Þ of cement pastes could be derived as:
Y c ¼ dI=dE ¼ 1=Rd þ 1=Ra ð19Þ
It could be seen from Eq. (19) that the impedance response of
cement pastes in the competition stage was also
approximately resistive. Accordingly, a same flat plateau
was presented in imaginary part vs. time curves during
the competition stage, which accorded with the one in
Figs. 3–5 and 9.
25
3) With regard to the acceleration stage, the rapid reaction of
tri-calcium silicate and di-calcium silicate dominated and a
great amount of calcium silicate hydrates was formed sur-
rounding cement particles in this stage. The current of the
cement paste ðIÞ in the acceleration stage had two compo-
nents: ions transfer current ðIt Þ and chemical reaction cur-
rent ðIc Þ. The impedance response in this stage was given
by Eq. (20):
dI ¼ dIt þ dIc ¼ dE=Rt þ pdn ð20Þ
where Rt was the resistance corresponding to current It , p was
a coefficient related to the extent of chemical reaction ðnÞ in
the cement hydration system, which varied with hydration
time.
The admittance ðY a Þ in this stage was determined from the
further solving of Eq. (20) as:
Y a ¼ dI=dE ¼ 1=R þ 1=ðjxLÞ ð21Þ
where L was an inductive component.
It could be seen from Eq. (21) that the derived impedance
response of cement pastes in the acceleration stage corresponded
to the data trend in Figs. 3–5 and 9: imaginary parts increased as
frequency increased.
In a whole, although electrical methods have been used for
determination of hydration stages over several decades and some
literatures have described the physical and chemical behaviors in
each hydration stage [2,100], a recognized standard for hydration
stage determination is still absent until now. Usually, heat releas-
ing method, as a mature approach to identify the hydration stages
of cement-based materials, was used to compare results from elec-
trical methods. The hydration heat of cements was determined in
most European countries by using two standardized methods:
the solution method and semi-adiabatic calorimetry (also known
as the Langavant method) according to the EN196-8 [9] and
EN196-9 [10] standards, respectively. The heat of hydration was
obtained from the difference between the heat of solution of anhy-
drous cement and that of hydrated cement. The disadvantage of
this first method was that only small paste samples can be tested,
not mortar or concrete samples. The Langavant method consisted
of introducing into a semi-adiabatic calorimeter to measure the
temperature evolution of the fresh concrete just after mixing.
Before performing measurements of the hydration heat, the Langa-
vant calorimeter must be calibrated by electrical substitution, in
order to determine its coefficient of total heat loss and its heat
capacity [101].
Li et al. [59,99] and Zhang and Li [59,99] found that the resistiv-
ity development curves were similar to the curves of heat evolu-
tion of cement. In contrast, Bentz et al. claimed that results from
heat release and electrical resistance could not match each other
according to their findings [31]. The inconsistent conclusions
may be ascribed to different test conditions [93]. Wei and Xiao
derived logarithmic equations from non-contact electrical resistiv-
ity measurement and heat evolution at the same condition, respec-
tively [102]:
qðtÞ ¼ K m lnðDt Þ ð22Þ
Q ðtÞ ¼ K 0m lnðD0t Þ ð23Þ
where qðtÞ was the bulk resistivity of cement pastes, Q ðtÞ was inte-
gral heat, K m and K 0m were fitting parameters, Dt and D0t were
claimed to reflect the hydration rate of cement particles.
A positive linear relationship between K m and K 0m was found.
When the experimental data were substituted into Eqs. (22) and
(23), the correlation between electrical resistivity and integral heat
was expressed as:
26 S.W. Tang et al. / Construction and Building Materials 146 (2017) 15–29
Fig. 10. Simulated 2D and 3D images of capillary pores with w/c = 0.35 at different time (a) 0 h, (b) first de-percolated time (c) fully de-percolated time [103].
Q ðtÞ ¼ 69:499 ln qðtÞ þ 36:705 ðR2 ¼ 0:9648Þ
Apparently, Eq. (24) may not provide a general approach to cor-
relate Q ðtÞ to qðtÞ for all kinds of cement-based materials. Conse-
quently, the correlative mechanism between heat release and
electrical method is still unclear although some efforts, such as
the obtainment of regression relation, have been made.
5. Simulated methods for hydration
Although the non-contact resistivity measurement could char-
acterize the overall hydration process, it was overwhelmed to
investigate the microstructure morphology. For this case, com-
puter models, such as CEMHYD 3D and Hymostruc 3D, were excel-
lent complements [83].
The percolation of solid phases and depercolation of capillary
pores could be predicted by the comparison of results from CEM-
HYD 3D and non-contact electrical resistivity measurement
according to two characteristic points. Fig. 10 is the simulated 2D
and 3D images of capillary pores with w/c = 0.35 at different time.
Cement particles were surrounded by water-filled pore space, and
all interconnected pores in Fig. 10(a) mainly controlled transport
properties. As hydration proceeded, the pathway of the first perco-
lated solid phases was formed, as demonstrated in Fig. 10(b). While
Fig. 10(c) illustrated the fully depercolated of connected channels.
Fig. 11. Computer simulation proposed by Zuo et al. for the hydration evolution of cement paste with water to cement ratio 0.4 [107].
ð24Þ It was considered that the first percolation time of solid phases cor-
responded well to the first depercolation time of capillary pores
[103].
Nevertheless, the serious shortcoming of CEMHYD 3D model
stemmed from its inherent voxel resolution limit of 1 mm, which
came from the discrete approach which necessitated enormous
computation over a large set of volume elements of the grid. Con-
sequently, the model was not capable of representing small capil-
lary pores. Another drawback of this model was the highly
empirical dependencies on experimental data. Therefore, the study
for the combination of non-contact resistivity measurement and
CEMHYD 3D should be further developed.
Wei and Xiao made an attempt to utilize Hymostruc 3D to con-
struct the microstructure of concrete during hydration in the first
24 h to verify the validity of research finding from non-contact
resistivity measurement [104]. However, no exact and explicit
relations were obtained in this work. What’s more, Hymostruc
3D used the vector approach to generate resolution-free
microstructures. However, it was severely limited in so much as
it grew the microstructure as expanded spheres based on the orig-
inal cement particle distribution without taking into account inter-
actions with neighbors. In so doing, it was unable to fully capture
the microstructural complexity [105].
Besides, a computer simulation was conducted by Zuo et al.
based on electrical resistivity measurements to enable the visual-
 S.W. Tang et al. / Construction and Building Materials 146 (2017) 15–29
ization of the microstructural evolution of the hydration process,
as shown in Fig. 11 [106,107]. This simulation was capable of
studying the influence of water to cement ratios and citric acid
on the hydration process using the visualization and quantitative
results. The hydration degree determined using the computer sim-
ulation was in consistent with the results of the non-evaporable
water experiments, and the simulated porosity also helped to pre-
dict the trend of early strength development of cement pastes
[106]. Nevertheless, the considerations associated with features
of pore solution and pore structure morphology were scare in this
computer simulation.
6. Conclusions
In this work, research activities regarding to the evaluation of
hydration in cement-based materials based on electrical methods
are reviewed. Direct current methods have been rarely used for
hydration investigation due to distinct electrode polarization
effects, while alternative current methods (ACIS, non-contact
methods and Wenner probe) have widely used in hydration inves-
tigation in recent decades. The application scenarios of these alter-
native current methods are presented: ACIS tests are good methods
if certain compensation measures are adopted; non-contact meth-
ods are suitable for early-age hydration monitoring; Wenner probe
is preferred to hardened materials. Besides, influencing factors
(pore solution conductivity, porosity, pore size distribution and
pore connectivity) of electrical methods are briefly discussed. In
addition, the hydration characterization by electrical methods in
this work primarily focuses on the prediction of setting time and
shrinkage, and the determination of hydration stages. Some main-
stream computer simulations are promising complements for
hydration characterization by electrical methods.
Last but not the least, it is worth pointing out that the two
major disadvantages of electrical methods for the hydration char-
acterization are 1) the inability to differentiate coupling effects of
influencing factors responsible for conduction, and 2) absence of
adequate quantitative models that can relate the electrical
response to the hydration process.
Acknowledgement
The supports from Scientific Research Staring Foundation for
the introduction of talents under grant of 206-410100031, Inde-
pendent Scientific Research Project of Young Teachers under grant
of 2042016kf0065, the China Ministry of Science and Technology
under grant of 2015CB655101, National Natural Science Founda-
tion of China (NSFC) under grant of 51602229 and Hubei Province
Science Foundation for Youths under grant of 2016CFB241 are
gratefully acknowledged.
References
[1] S.W. Tang, X.H. Cai, W. Zhou, H.Y. Shao, Z. He, Z.J. Li, W.M. Ji, E. Chen, In-situ
and continuous monitoring of pore evolution of calcium sulfoaluminate
cement at early age by electrical impedance measurement, Constr. Build.
Mater. 117 (2016) 8–19.
[2] S.W. Tang, H.G. Zhu, Z.J. Li, E. Chen, H.Y. Shao, Hydration stage identification
and phase transformation of calcium sulfoaluminate cement at early age,
Constr. Build. Mater. 75 (2015) 11–18.
[3] S.W. Tang, X.H. Cai, Z. He, H.Y. Shao, Z.J. Li, E. Chen, Hydration process of fly
ash blended cement pastes by impedance measurement, Constr. Build. Mater.
113 (2016) 939–950.
[4] I. Odler, The BET-specific surface area of hydrated Portland cement and
related materials, Cem. Concr. Res. 33 (12) (2003) 2049–2056.
[5] S.W. Tang, Z.J. Li, E. Chen, H.Y. Shao, Impedance measurement to characterize
the pore structure in Portland cement paste, Constr. Build. Mater. 51 (2014)
106–112.
[6] R. Ylmén, U. Jäglid, B.-M. Steenari, I. Panas, Early hydration and setting of
Portland cement monitored by IR, SEM and Vicat techniques, Cem. Concr. Res.
39 (5) (2009) 433–439.
27
[7] W.J. Mccarter, R. Brousseau, The A.C. response of hardened cement paste,
Cem. Concr. Res. 20 (6) (1990) 891–900.
[8] W.J. Mccarter, T.M. Chrisp, G. Starrs, The early hydration of alkali-activated
slag: developments in monitoring techniques, Cem. Concr. Compos. 21 (4)
(1999) 277–283.
[9] W.J. Mccarter, M. Chrisp, Monitoring water and ionic penetration into cover-
zone concrete, ACI Mater. J. 97 (6) (2000) 668–674.
[10] W.J. Mccarter, G. Starrs, T.M. Chrisp, Electrical conductivity, diffusion, and
permeability of Portland cement-based mortars, Cem. Concr. Res. 30 (9)
(2000) 1395–1400.
[11] X. Wei, Z. Li, Study on hydration of Portland cement with fly ash using
electrical measurement, Mater. Struct. 38 (277) (2005) 411–417.
[12] W.J. Mccarter, T.M. Chrisp, G. Starrs, A. Adamson, P.A.M. Basheer, S.V.
Nanukuttan, S. Srinivasan, C. Green, Characterization of physio-chemical
processes and hydration kinetics in concretes containing supplementary
cementitious materials using electrical property measurements, Cem. Concr.
Res. 50 (2013) 26–33.
[13] W.J. Mccarter, G. Starrs, A. Adamson, T.M. Chrisp, P.A.M. Basheer, S.
Nanukuttan, S. Srinivasan, C. Green, Influence of different european
cements on the hydration of cover-zone concrete during the curing and
postcuring periods, J. Mater. Civil. Eng. 25 (9) (2013) 1335–1343.
[14] N.A. Soliman, K.H. Khayat, M. Karray, A.F. Omran, Piezoelectric ring actuator
technique to monitor early-age properties of cement-based materials, Cem.
Concr. Compos. 63 (2015) 84–95.
[15] X. Wei, Z. Li, Study on hydration of Portland cement with fly ash using
electrical measurement, Mater. Struct. 38 (3) (2005) 411–417.
[16] S.W. Tang, Z.J. Li, H.Y. Shao, E. Chen, Characterization of early-age hydration
process of cement pastes based on impedance measurement, Constr. Build.
Mater. 68 (2014) 491–500.
[17] W. Li, X. Zhu, J. Hong, W. She, P. Wang, W. Zuo, Effect of anionic emulsifier on
cement hydration and its interaction mechanism, Constr. Build. Mater. 93
(2015) 1003–1011.
[18] S. Ridha, S. Irawan, B. Ariwahjoedi, Prediction equation for permeability of
Class G oilwell cement under reservoir conditions, Arab. J. Sci. Eng. 39 (6)
(2014) 5219–5228.
[19] S. Ridha, S. Irawan, B. Ariwahjoedi, Strength prediction of Class G oilwell
cement during early ages by electrical conductivity, J. Petrol. Explor. Prod.
Technol. 3 (4) (2013) 1–9.
[20] S. Sharma, N.C. Kothiyal, Comparative effects of pristine and ball-milled
graphene oxide on physico-chemical characteristics of cement mortar
nanocomposites, Constr. Build. Mater. 115 (2016) 256–268.
[21] S. Sharma, N.C. Kothiyal, Synergistic effect of zero-dimensional spherical
carbon nanoparticles and one-dimensional carbon nanotubes on properties of
cement-based ceramic matrix: microstructural perspectives and
crystallization investigations, Compos. Interface. 22 (9) (2015) 899–921.
[22] A. Susanto, D.A. Koleva, O. Copuroglu, K.V. Beek, K.V. Breugel, Mechanical,
electrical and microstructural properties of cement-based materials in
conditions of stray current flow, J. Adv. Concr. Technol. 11 (3) (2013) 119–134.
[23] Z. He, C. Shi, X. Hu, J. Zhang, Development on migration characteristic and
interactions of chloride ion in cement-based materials under applied
voltages, J. Chin. Ceram. Soc. 43 (8) (2015) 1111–1119.
[24] M. Cabeza, M. Keddam, X.R. Novoa, I. Sanchez, H. Takenouti, Impedance
spectroscopy to characterize the pore structure during the hardening process
of Portland cement paste, Electrochim. Acta 51 (8–9) (2006) 1831–1841.
[25] R. Spragg, C. Villani, K. Snyder, D. Bentz, J.W. Bullard, J. Weiss, Electrical
resistivity measurements in cementitious systems: observations of factors
that influence the measurements, J. Trans. Res. Board 2342 (2013) 90–98.
[26] T. Yamaguchi, T. Matsuoka, S. Koda, A theoretical study on the frequency-
dependent electric conductivity of electrolyte solutions. II. Effect of
hydrodynamic interaction, J. Chem. Phys. 130 (9) (2009) 094506.
[27] J. Kou, J. Yao, H. Lu, B. Zhang, A. Li, Z. Sun, J. Zhang, Y. Fang, F. Wu, J. Fan,
Electromanipulating water flow in nanochannels, Angew. Chem. Int. Ed. 54
(8) (2015) 2381–2385.
[28] L. Woo, S. Wansom, N. Ozyurt, B. Mu, S. Shah, T. Mason, Characterizing fiber
dispersion in cement composites using AC-Impedance Spectroscopy, Cem.
Concr. Compos. 27 (6) (2005) 627–636.
[29] F. Reza, J.A. Yamamuro, G.B. Batson, Electrical resistance change in compact
tension specimens of carbon fiber cement composites, Cem. Concr. Compos.
26 (7) (2004) 873–881.
[30] J. Xu, W. Yao, R. Wang, Nonlinear conduction in carbon fiber reinforced
cement mortar, Cem. Concr. Compos. 33 (3) (2011) 444–448.
[31] D.P. Bentz, K.A. Snyder, A. Ahmed, Anticipating the setting time of high-
volume fly ash concretes using electrical measurements: feasibility studies
using pastes, J. Mater. Civil. Eng. 27 (3) (2015) 04014129.
[32] 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.
[33] W.J. Mccarter, H.M. Taha, B. Suryanto, G. Starrs, Two-point concrete resistivity
measurements: interfacial phenomena at the electrode-concrete contact
zone, Meas. Sci. Technol. 26 (8) (2015).
[34] P. Gu, P. Xie, Y. Fu, J.J. Beaudoin, A.C impedance phenomena in hydrating
cement systems: Frequency dispersion angle and pore size distribution, Cem.
Concr. Res. 24 (1) (1994) 86–88.
[35] P. Xie, P. Gu, J.J. Beaudoin, Contact capacitance effect in measurement of a.c.
impedance spectra for hydrating cement systems, J. Mater. Sci. 31 (1) (1996)
144–149.
28 S.W. Tang et al. / Construction and Building Materials 146 (2017) 15–29
[36] R.T. Coverdale, B.J. Christensen, H.M. Jennings, T.O. Mason, D.P. Bentz, E.J.
Garboczi, Interpretation of impedance spectroscopy of cement paste via
computer modeling- Part I Bulk conductivity and offset resistance, J. Mater.
Sci. 30 (3) (1995) 712–719.
[37] B.J. Christensen, R.T. Coverdale, R.A. Olson, S.J. Ford, E.J. Garboczi, H.M.
Jennings, T.O. Mason, Impedance spectroscopy of hydrating cement-based
materials - Measurement, interpretation, and application, J. Am. Ceram. Soc.
77 (11) (1994) 2789–2804.
[38] B.J. Christensen, T.O. Mason, H.M. Jennings, Influence of silica fume on the
early hydration of Portland cements using impedance spectroscopy, J. Am.
Ceram. Soc. 75 (4) (1992) 939–945.
[39] I. Fayala, L. Dhouibi, X. Nóvoa, M.B. Ouezdou, Effect of inhibitors on the
corrosion of galvanized steel and on mortar properties, Cem. Concr. Compos.
35 (1) (2013) 181–189.
[40] B. Diaz, L. Freire, P. Merino, X.R. Novoa, M.C. Perez, Impedance spectroscopy
study of saturated mortar samples, Electrochim. Acta 53 (25) (2008) 7549–
7555.
[41] I. Sanchez, X.R. Novoa, G. De Vera, M.A. Climent, Microstructural
modifications in Portland cement concrete due to forced ionic migration
tests. Study by impedance spectroscopy, Cem. Concr. Res. 38 (7) (2008) 1015–
1025.
[42] A. Covelo, B. Diaz, L. Freire, X.R. Nóvoa, M.C. Pérez, Microstructural changes in
a cementitious membrane due to the application of a DC electric field, J.
Environ. Sci. Heal. A 43 (8) (2008) 985–993.
[43] N. Neithalath, J. Jain, Relating rapid chloride transport parameters of
concretes to microstructural features extracted from electrical impedance,
Cem. Concr. Res. 40 (7) (2010) 1041–1051.
[44] N. Neithalath, J. Persun, R.K. Manchiryal, Electrical conductivity based
microstructure and strength prediction of plain and modified concretes, Int.
J. Adv. Eng. Sci. Appl. Math. 2 (3) (2010) 83–94.
[45] N. Neithalath, Extracting the performance predictors of enhanced porosity
concretes from electrical conductivity spectra, Cem. Concr. Res. 37 (5) (2007)
796–804.
[46] N. Neithalath, J. Weiss, J. Olek, Characterizing enhanced porosity concrete
using electrical impedance to predict acoustic and hydraulic performance,
Cem. Concr. Res. 36 (11) (2006) 2074–2085.
[47] L. Xiao, Concrete Hydration Based on Electrical Resistivity Measurement,
VDM Verlag Dr. Müller, Saarbrucken, 2008.
[48] C. Scuderi, T. Mason, H. Jennings, Impedance spectra of hydrating cement
pastes, J. Mater. Sci. 26 (2) (1991) 349–353.
[49] P. Xie, P. Gu, Z. Xu, J. Beaudoin, A rationalized a.c. impedance model for
microstructural characterization of hydrating cement systems, Cem. Concr.
Res. 23 (2) (1993) 359–367.
[50] W. Mccarter, G. Starrs, T. Chrisp, P. Basheer, S. Nanukuttan, S. Srinivasan,
Conductivity/activation energy relationships for cement-based materials
undergoing cyclic thermal excursions, J. Mater. Sci. 50 (3) (2015) 1129–1140.
[51] Y. Bu, J. Weiss, The influence of alkali content on the electrical resistivity and
transport properties of cementitious materials, Cem. Concr. Compos. 51
(2014) 49–58.
[52] R. Spragg, Y. Bu, K. Snyder, D. Bentz, J. Weiss, Electrical Testing of Cement-
Based Materials: Role of Testing Techniques, Sample Conditioning, and
Accelerated Curing, Indiana Department of Transportation and Purdue
University, Indiana, 2013.
[53] K. Farzanian, K.P. Teixeira, I.P. Rocha, L.D.S. Carneiro, A. Ghahremaninezhad,
The mechanical strength, degree of hydration, and electrical resistivity of
cement pastes modified with superabsorbent polymers, Constr. Build. Mater.
109 (2016) 156–165.
[54] J.M. Cruz, I.C. Fita, L. Soriano, J. Paya, M.V. Borrachero, The use of electrical
impedance spectroscopy for monitoring the hydration products of Portland
cement mortars with high percentage of pozzolans, Cem. Concr. Res. 50
(2013) 51–61.
[55] K. Farzanian, K. Pimenta Teixeira, lL. Perdigão Rocha, L. De Sa Carneiro, A.
Ghahremaninezhad, The mechanical strength, degree of hydration, and
electrical resistivity of cement pastes modified with superabsorbent
polymers, Constr. Build. Mater. 109 (2016) 156–165.
[56] Z. Li, W. Li. Contactless, transformer-based measurement of the resistivity of
materials. US Patent 6,639,401; 2003.
[57] J. Weiss, K. Snyder, J.W. Bullard, D.P. Bentz, Using a saturation function to
interpret the electrical properties of partially saturated concrete, J. Mater.
Civil. Eng. 25 (8) (2013) 1097–1106.
[58] L.Z. Xiao, Z.J. Li, X.S. Wei, Selection of superplasticizer in concrete mix design
by measuring the early electrical resistivities of pastes, Cem. Concr. Compos.
29 (5) (2007) 350–356.
[59] J. Zhang, Z. Li. Hydration process of cements with superplasticizer monitored
by non-contact resistivity measurement. In: Proceedings of Advanced Testing
of Fresh Cementitious Materials. Stuttgart, Germany, 2006.
[60] Z. Li, S. Tang, Y. Lu. Pore structure analyzer based on non-contact impedance
measurement for cement-based materials. US Patent 20,120,158,333; 2012.
[61] S.W. Tang, Z.J. Li, H.G. Zhu, H.Y. Shao, E. Chen, Permeability interpretation for
young cement paste based on impedance measurement, Constr. Build. Mater.
59 (2014) 120–128.
[62] S.W. Tang, Z.J. Li, E. Chen, H.Y. Shao, Non-steady state migration of chloride
ions in cement pastes at early age, RSC Adv. 4 (89) (2014) 48582–48589.
[63] S.W. Tang, E. Chen, H.Y. Shao, Z.J. Li, A fractal approach to determine thermal
conductivity in cement pastes, Constr. Build. Mater. 74 (2015) 73–82.
[64] E. Ghiasvand, A.A. Ramezanianpour, A.M. Ramezanianpour, Influence of
grinding method and particle size distribution on the properties of Portland-
limestone cements, Mater. Struct. 48 (5) (2015) 1273–1283.
[65] A.M. Ramezanianpour, K. Esmaeili, S.A. Ghahari, A.A. Ramezanianpour,
Influence of initial steam curing and different types of mineral additives on
mechanical and durability properties of self-compacting concrete, Constr.
Build. Mater. 73 (2014) 187–194.
[66] S. Teng, T.Y.D. Lim, D.B. Sabet, Durability and mechanical properties of high
strength concrete incorporating ultra fine Ground Granulated Blast-furnace
Slag, Constr. Build. Mater. 40 (2013) 875–881.
[67] S. Tang, Y. Yao, C. Andrade, Z. Li, Recent durability studies on concrete
structure, Cem. Concr. Res. 78 (2015) 143–154.
[68] S.W. Tang, X.H. Cai, Z. He, W. Zhou, H.Y. Shao, Z.J. Li, T. Wu, E. Chen, The
review of pore structure evaluation in cementitious materials by electrical
methods, Constr. Build. Mater. 117 (2016) 273–284.
[69] S.W. Tang, E. Chen, Z.J. Li, H.Y. Shao, Assessment of steady state diffusion of
volatile organic compounds in unsaturated building materials based on
fractal diffusion model, Build. Environ. 84 (2015) 221–227.
[70] L.Z. Xiao, Z.J. Li, Early-age hydration of fresh concrete monitored by non-
contact electrical resistivity measurement, Cem. Concr. Res. 38 (3) (2008)
312–319.
[71] D. Harbec, A. Tagnit-Hamou, F. Gitzhofer, Waste-glass fume synthesized
using plasma spheroidization technology: reactivity in cement pastes and
mortars, Constr. Build. Mater. 107 (2016) 272–286.
[72] R.S. Barneyback, S. Diamond, Expression and analysis of pore fluids from
hardened cement pastes and mortars, Cem. Concr. Res. 11 (2) (1981) 279–
285.
[73] K.A. Snyder, X. Feng, B.D. Keen, T.O. Mason, Estimating the electrical
conductivity of cement paste pore solutions from OH, K+ and Na+
concentrations, Cem. Concr. Res. 33 (6) (2003) 793–798.
[74] D.P. Bentz, A virtual rapid chloride permeability test, Cem. Concr. Compos. 29
(10) (2007) 723–731.
[75] M. Davraz, The effect of boron compound to cement hydration and
controllability of this effect, Acta Phys. Pol., A 128 (2B) (2015) 26–33.
[76] X. An, C. Shi, Z. Cao, D. Wang, F. He, Z. Hu, Relation between bulk resistance
and pore structure of hardened cement pastes, J. Chin. Ceram. Soc. 42 (8)
(2014) 1011–1017.
[77] N. Schwarz, M. Dubois, N. Neithalath, Electrical conductivity based
characterization of plain and coarse glass powder modified cement pastes,
Cem. Concr. Compos. 29 (9) (2007) 656–666.
[78] V. Sánchez-Fajardo, M. Torres, A. Moreno, Study of the pore structure of the
lightweight concrete block with lapilli as an aggregate to predict the liquid
permeability by dielectric spectroscopy, Constr. Build. Mater. 53 (2014) 225–
234.
[79] P. Xie, P. Gu, Y. Fu, J.J. Beaudoin, A.C. impedance phenomena in hydrating
cement systems: Detectability of the high frequency arc, Cem. Concr. Res. 24
(1) (1994) 92–94.
[80] G.A. Niklasson, A. Berg, K. Brantervik, B. Hedberg, L.O. Nilsson, Dielectric
properties of porous cement mortar: fractal surface effects, Solid State
Commun. 79 (1) (1991) 93–96.
[81] Q. Li, S. Xu, Q. Zeng, The effect of water saturation degree on the electrical
properties of cement-based porous material, Cem. Concr. Compos. 70 (2016)
35–47.
[82] S.W. Tang, Z. He, X.H. Cai, R.J. Cai, W. Zhou, Z.J. Li, H.Y. Shao, T. Wu, E. Chen,
Volume and surface fractal dimensions of pore structure by NAD and LT-DSC
in calcium sulfoaluminate cement pastes, Constr. Build. Mater. 143 (2017)
395–418.
[83] Z. Liu, Y. Zhang, Q. Jiang, Continuous tracking of the relationship between
resistivity and pore structure of cement pastes, Constr. Build. Mater. 53
(2014) 26–31.
[84] F. Rajabipour, J. Weiss, Electrical conductivity of drying cement paste, Mater.
Struct. 40 (10) (2007) 1143–1160.
[85] G. Sant, F. Rajabipour, P. Fishman, P. Lura, J. Weiss. Electrical conductivity
measurements in cement paste at early ages: a discussion of the contribution
of pore solution conductivity, volume, and connectivity to the overall
electrical response. Rilem Workshop on Advanced Testing of Fresh
Cementitious Materials. 2006.
[86] B.J. Christensen, T. Coverdale, R.A. Olson, S.J. Ford, E.J. Garboczi, H.M. Jennings,
T.O. Mason, Impedance spectroscopy of hydrating cement-based materials:
measurement, interpretation, and application, J. Am. Ceram. Soc. 77 (11)
(1994) 2789–2804.
[87] E.J. Garboczi, Permeability, diffusivity, and microstructural parameters: A
critical review, Cem. Concr. Res. 20 (4) (1990) 591–601.
[88] R. Rajamma, L. Senff, M.J. Ribeiro, J.A. Labrincha, R.J. Ball, G.C. Allen, V.M.
Ferreira, Biomass fly ash effect on fresh and hardened state properties of
cement based materials, Compos. Part B-Eng. 77 (2015) 1–9.
[89] A. Ghirian, M. Fall, Coupled behavior of cemented paste backfill at early ages,
Geotech. Geol. Eng. 33 (5) (2015) 1141–1166.
[90] Y. Liao, X. Wei, Penetration resistance and electrical resistivity of cement
paste with superplasticizer, Mater. Struct. 47 (4) (2014) 563–570.
[91] Y. Zuo, X. Wei, Relations among electrical resistivity, chemical and
autogenous shrinkage of cement pastes, Adv. Cem. Res. 27 (3) (2015) 175–
183.
[92] Y. Liao, X. Wei, Relationship between chemical shrinkage and electrical
resistivity for cement pastes at early age, J. Mater. Civil. Eng. 26 (2) (2014)
384–387.
 S.W. Tang et al. / Construction and Building Materials 146 (2017) 15–29
[93] J.-B. Champenois, C. Cau Dit Coumes, A. Poulesquen, P. LeBescop, D. Damidot,
Beneficial use of a cell coupling rheometry, conductimetry, and calorimetry to
investigate the early age hydration of calcium sulfoaluminate cement, Rheol.
Acta 52 (2) (2013) 177–187.
[94] A.Z. Morshed, A. Azar, Y. Shao, Characterising cement carbonation curing
using non-contact electrical resistivity measurement, Adv. Cem. Res. 27 (4)
(2015) 214–224.
[95] A. Coulon, D. Laurencin, A. Grandjean, C. Cau Dit Coumes, S. Rossignol, L.
Campayo, Immobilization of iodine into a hydroxyapatite structure prepared
by cementation, J. Mater. Chem. A2 (48) (2014) 20923–20932.
[96] Y. Deng, C. Zhang, X. Wei, Influence of lithium sulfate addition on the
properties of Portland cement paste, Constr. Build. Mater. 50 (2014) 457–462.
[97] Y. Wang, M. Fall, A. Wu, Initial temperature-dependence of strength
development and self-desiccation in cemented paste backfill that contains
sodium silicate, Cem. Concr. Compos. 67 (2016) 101–110.
[98] W. Li, M. Fall, Sulphate effect on the early age strength and self-desiccation of
cemented paste backfill, Constr. Build. Mater. 106 (2016) 296–304.
[99] Z.J. Li, X.S. Wei, The electrical resistivity of cement paste incorporated with
retarder, J. Wuhan. Univ. Technol. 18 (3) (2003) 76–78.
[100] J. Zhang, L. Qin, Z. Li, Hydration monitoring of cement-based materials with
resistivity and ultrasonic methods, Mater. Struct. 42 (1) (2009) 15–24.
View publication stats
29
[101] B. Hay, J. Hameury, G. Davee, M. Grelard, Assessment of uncertainties in
calibration of langavant calorimeters, Int. J. Therrmophys. 35 (9) (2014)
1757–1769.
[102] X. Wei, L. Xiao, Kinetics parameters of cement hydration by electrical
resistivity measurement and calorimetry, Adv. Cem. Res. 26 (4) (2014) 187–
193.
[103] Z. Liu, Y. Zhang, Q. Jiang, W. Zhang, J. Wu, Solid phases percolation and
capillary pores depercolation in hydrating cement pastes, J. Mater. Civil. Eng.
26 (12) (2014) 04014090.
[104] X. Wei, L. Xiao, Electrical resistivity monitoring and characterisation of early
age concrete, Mag. Concrete Res. 65 (10) (2013) 600–607.
[105] M. Zalzale, P.J. Mcdonald, Lattice Boltzmann simulations of the permeability
and capillary adsorption of cement model microstructures, Cem. Concr. Res.
42 (12) (2012) 1601–1610.
[106] Y. Zuo, J. Zi, X. Wei, Hydration of cement with retarder characterized via
electrical resistivity measurements and computer simulation, Constr. Build.
Mater. 53 (2014) 411–418.
[107] Y. Zuo, X. Wei, Portland cement hydration by electrical resistivity method and
computer simulation, J. Test. Eval. 41 (3) (2013) 451–457.