Construction and Building Materials 68 (2014) 650–658

Contents lists available at ScienceDirect

Construction and Building Materials

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

Electrochemical impedance behavior of mortar subjected to a sulfate

environment – A comparison with chloride exposure models
Mariana O.G.P. Bragança a,b, Kleber F. Portella b,⇑, Marcelle M. Bonato a, Cláudia E.B. Marino a
a
Programa Interdisciplinar de Pós Graduação em Engenharia e Ciência dos Materiais, Universidade Federal do Paraná, PO Box 19011, CEP: 81531-980, Curitiba, Paraná, Brazil
b
Departamento de Estruturas Civis, Institutos Lactec, PO Box 19067, CEP: 81531-980, Curitiba, Paraná, Brazil

h i g h l i g h t s

 Sulfate attack in mortar was studied by electrochemical impedance spectroscopy.

 Mechanisms that occur under the mortar pores during the aging were highlighted.
 Mortar resistance shifted and the system presented a capacitive behavior.
 Electrochemical impedance identified the sulfate attack by non-destructive method.
 Electrochemical results were comparable with physical chemical characterizations.

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

Article history: Samples of reinforced mortar specimens (Sps) were studied over 300 days of exposure in reference,
Received 24 April 2014 sulfate, and chloride environments. Electrochemical impedance spectroscopy (EIS) was used to investi-
Received in revised form 17 June 2014 gate external sulfate attack (ESA) in comparison with known chloride models. The study, based on the
Accepted 24 June 2014
mortar behavior and its measured electrical properties, highlighted the mechanisms that occur under
Available online 26 July 2014
the pores during the aging. The sulfate samples presented a capacitive behavior at lower frequencies,
and the mortar resistance shifted. It was confirmed that EIS is a qualified technique to identify the sulfate
Keywords:
attack in concrete by non-destructive methods.
Mortar
External sulfate attack
Ó 2014 Elsevier Ltd. All rights reserved.
Electrochemical impedance spectroscopy
Non-destructive diagnosis
Durability

1. Introduction thermodynamic conditions). This crystals growth promotes an

One of the primary aspects related to the durability of concrete
and mortar is the reaction of cement hydration products (alumina
and calcium hydroxide) with sulfate ions (SO24 ) derived from the
external environment, such as industrial wastes (water and air),
soil, and decomposing organic matter, among other sources. Exter-
nal sulfate attack (ESA) has been studied since the 1990s but there
are still difficulties in understanding the phenomenon. In general,
the mechanisms develop in a similar manner to reactions with chlo-
ride. The anions SO24 tend to penetrate into the concrete through
its surface, accumulating in the pores and near the reinforcement.
In the concrete or mortar, the sulfate reacts with the cementitious
compounds, promoting the formation and growth of ettringite,
gypsum, monosulfate, and thaumasite crystals (depending on the
⇑ Corresponding author. Tel.: +55 41 3361 6220; fax: +55 41 3361 6141.
E-mail address: kfportella@pq.cnpq.br (K.F. Portella).
increase in the pressure in the pores, thus leading to microcracking
of the matrix. In reinforced concrete, this distress also affect the
quality of the concrete/steel double layer, promoting the corrosion
of the reinforcement due to the bar de-passivation promoted by the
change in the environment alkalinity [1–5].
Diagnosis of the decrease in concrete durability due to ESA
involves the development of destructive tests, with the requirement
to extract specimens (Sps) for laboratory analysis using techniques
such as microscopy and X-ray diffraction. Moreover, specifically for
reinforced concrete under conditions exposed to environments con-
taining chloride ions, electrochemical methods and the electro-
chemical impedance spectroscopy (EIS) technique have been used
to analyze the corrosion potential and polarization resistance of
the metal bar. The broad use of these techniques is due to the pos-
sibility of obtaining electrical parameters that make it possible to
obtain, even in the field, the dielectric properties of the cementitious
material and its interfaces (cement paste/aggregate and cement

http://dx.doi.org/10.1016/j.conbuildmat.2014.06.040
0950-0618/Ó 2014 Elsevier Ltd. All rights reserved.
 M.O.G.P. Bragança et al. / Construction and Building Materials 68 (2014) 650–658
paste/reinforcement). In addition, it is possible to estimates of the
probability and rate of corrosion at specific points of the structure
[6–15].
EIS is a material characterization technique that involves the
analysis of the response of a specific material when applying an
alternating current (AC) signal to a sweep of frequencies. This
response is directly dependent on the applied frequency and makes
it possible to study several features of the electrode/electrolyte
interface, such as resistance (R), capacitance (C), and inductance
(I). These features, when analyzed mathematically, enable the anal-
ysis or estimation of the primary physical–chemical parameters of
the material, whose molecular structure and electrode charge allow
or limit the response to the electrical stimulus [16,17].
For reinforced concrete, the development of EIS measurements
is based on the principle that the material exhibits a supported-
type electrode with a high concentration of ions, which are not
adsorbed and do not react with the electrodes of the measurement
cell [17]. The analysis is normally based on the study of Bode dia-
grams (that relate the impedance modulus and the phase angle
with the frequency) and the complex plane (or Nyquist plots,
which relates the real and imaginary components of impedance)
for simulation with an equivalent electrical circuit. From these
spectra, the data for polarization resistance and solution resistance
(Rp and Rs), the capacitance of the electrical double layer (Cd) or the
constant phase element (CPE) and the diffusion or Warburg
coefficient (ZD) can be obtained. With these parameters, it is possi-
ble to determine a series of material properties related to corrosion
and durability [6,16].
The EIS technique for the analysis of concrete is typically used
for the prediction of the electrical properties of concrete during
the curing process or when exposed to chloride ions. Keddam
et al. [6] developed one of the first studies of the technique, direct-
ing these investigations toward the determination of polarization
time constants and the dielectric properties of the cement paste,
analyzed in the high-frequency region (103–108 Hz). This study
was developed with graphite electrodes to observe only the behav-
ior of the cement paste (avoiding the interfacial processes of the
paste/aggregate and paste/reinforcement double layers). The
results obtained make it possible to establish mathematical models
for analyzing the system and facilitate calculating the dielectric
constant for the cement paste when subjected to frequencies
higher than 108 Hz. Furthermore, it was possible to identify three
capacitive time constants for the wet cement paste when analyzed
in the region of intermediate frequencies (106–105 Hz), which cor-
respond to relaxation of the electrochemical processes that occur
in the paste at the interface with the electrode and are due to
interfacial effects—which, until then, was not understood by the
authors. These results demonstrate the possibility of using the
EIS technique to characterize cementitious materials.
Ford et al. [7] studied concrete and its simulated pore solution,
SPS, dividing the impedance spectra from the cement paste/
reinforcement system into distinct regions related to a specific
frequency range. EIS measurements were performed by applying
a signal with an excitation amplitude of 25 mV, in open-circuit
potential, to a frequency range of 20  106–2  10 3 Hz, thereby
developing control of the moisture of the Sp. For low frequencies
(from 10 3 Hz to 100 Hz), the mechanisms were independent of
the cement paste and corresponded to its interaction with the rein-
forcement, which was proven by the absence of the typical arcs
observed in the measurements with SPS. In this range of frequen-
cies, the high resistivity of the oxide film formed on the reinforce-
ment (1011–1012 X cm) was considered, with a thickness of
approximately 10 nm and a non-ideal capacitance (n less than
0.9) of approximately 0.5 mF cm 2. With the increase in chloride
concentration, a 3- to 4-fold decrease was observed in the diameter
of the arcs projected in the complex plane, which was attributed to
651
a possible dissolution of the passive oxide with active corrosion of
the metal. The region of frequencies between 10 and 100 Hz was
considered difficult to approximate due to the superposition of
processes and represented corrosion products such as iron oxides
deposited in the concrete pores or even the formation of a layer
of Ca(OH)2 at the metal/concrete interface. This curve tended to
be more defined with the aging of the reinforced concrete. The
region of frequencies higher than 100 Hz was attributed to the
resistance of the concrete, which increased with the curing time
and was approximately 103–106 X. Moreover, the behaviors of
the material regarding processes of diffusion and ion migration,
the sealing of pores, and the curing of the cement were also
characterized in this region.
The development of a methodology for calculating the corrosion
rate for reinforced concrete by EIS was performed by Ismail and
Ohtsu [9]. For this purpose, Sps were prepared with ordinary
Portland cement (OPC) and high-performance Portland cement
(HPC), which were subjected to different accelerated aging cycles
in a NaCl solution and dried in an oven. The influence of the corro-
sion rate measured by EIS for the type of cement and aging process
was proven by comparing the traditional Tafel and polarization
resistance methodologies, with a good correlation obtained for
the results from the different techniques.
In recent years, the studies developed have been based on the
determination of specific mathematical models for concrete and
mortar exposed to chloride ions, to interpret the degradation and
corrosion phenomena for both the concrete and the reinforcement
[8–15]. Specifically, Serdar et al. [13] studied the corrosion mech-
anisms of galvanized steel reinforcement in Portland cement mor-
tar exposed to chloride ions. The aging, which occurred over
24 months with immersion of the Sps in a 3.5% NaCl solution,
enabled monitoring of the evolution of the interfacial properties
of the cement matrix/reinforcement interface. For analysis of the
spectra, a methodology was proposed in which the system was
analyzed in terms of three polarization time constants: (1) the
high-frequency region (above 102 Hz), in which the impedance
modulus and the phase angle tended to zero relative to the fre-
quency representing the resistance properties of the mortar; (2)
the intermediate-frequency region (between 102 and 10 2 Hz), in
which the phase angle exhibited an increasing trend to values of
80° and reflected the capacitive properties of the bar-mortar
interface related to the formation of a passive film or the deposi-
tion of corrosion products; and (3) for the low-frequency region
(between 10 2 and 10 4 Hz), when the impedance modulus (rela-
tive to the frequency) was stabilized at a new level and the phase
angle was again close to zero, in which the processes of charge-
transfer resistance were identified. Regarding the results, two
mathematical models were proposed for analysis of the system,
with one for the formation of a passive film and the beginning of
pit formation (characterized by the increase in the interfacial resis-
tances) and the other for its propagation and the formation of cor-
rosion products (identified by the decrease in interfacial
resistances and consequent increase in capacitances). Both of the
models were similar and exhibited the characteristics of the sys-
tem in a separate manner, in terms of the properties of the mortar,
passive film, and metal. The results obtained with the electrochem-
ical technique and interpreted from the viewpoint of the
established models were also validated through the breaking of
the Sps and the surface analysis of the reinforcement.
The concrete exposure to sulfate solution was studied by
electrochemical characterization tests to determine the reinforce-
ment corrosion rate [18,19]. Cheng et al. [18] prepared samples
of reinforced concretes using two different types of cement
(Portland type I and V) and low carbon steel and exposed the spec-
imens to artificial seawater (Cl concentration around 20%) and
concentrated sulfate solution (SO24 concentration of 22%) during
652 M.O.G.P. Bragança et al. / Construction and Building Materials 68 (2014) 650–658
180 days. The deterioration of the immersed bar was measured by
open circuit potential (OCP) and EIS, which one was analyzed
through the Bode plot, considering three different regions of char-
acterization (in terms of time constant): the concrete response, at
higher frequencies; the surface film (corrosion oxides) in interme-
diate frequency region; and at lower frequencies, the steel inter-
face. The results obtained to the sulfate exposure showed the
existence of two well-defined time constants to the surface film
and to the steel interface, which appears at frequencies of 0.1 Hz
and 0.1 mHz respectively. It seemed to suggest that the sulfate
exposure promoted the formation of a bigger layer of corrosion
products on the surface steel, which indicated that the reinforce-
ment deterioration in this environment was more severe than the
chloride one. Besides, the electrical resistances of the systems were
taken through the adjustment of equivalent circuits and it was ver-
ified an increase of the concrete resistance as the immersion time
increases, in both exposure solution. It was attributed to the chem-
ical property changes of the cement paste, which was not properly
discussed in the paper. Gerengi et al. [19] analyzed the corrosion
rate of three different traces of reinforced concrete exposed to sul-
phuric acid (to simulate the acid rain). In this research, diatomite
and zeolite-containing concrete were subjected to a solution of
0.5 M H2SO4 for 160 and every 15 days EIS measurements were per-
formed by applying a signal with an excitation amplitude of 10 mV,
in open-circuit potential, to a frequency range of 1  105–
1  10 2 Hz. The results, analyzed by the Nyquist plot, indicated a
decrease of the concrete and the charge transfer electric resistance
in function of the proposed aging, which was attributed to the rein-
forcement corrosion process. It was concluded that the acid rain is
one of the main sources of chemical degradation of concrete due the
corrosion destruction of the rebars; and that the porosity is the
main factor controlling the initiation of the corrosion in concrete.
The cement paste chemical modification due the sulfate attack
was neither explored in this paper.
Despite the wide use of EIS for analyzing the properties of the
cement paste and its modifications [6–8,15], as well as studying
the exposure of concrete and mortar to NaCl [9–14], little research
has been performed with regard to EIS application to contamina-
tion phenomena by sulfate ions [18,19]. Consequently, this study
aimed to investigate the electrochemical methods developed in
mortar exposed to SO2 (2%), based on the mathematical models
already established for chloride ions, to enable applying the meth-
odology for the non-destructive analysis of reinforced concrete.
2. Experimental methodology
2.1. Materials
The cement used in this study was CP II-Z 32 (composite Portland cement with
the addition of pozzolan) with crushed stone used as fine aggregates (particle size
of less than 4.8 mm). Both materials were characterized according to the Brazilian
technical standard [20,21]. Additionally, the chemical characterization was per-
formed using X-ray fluorescence (XRF). The test was performed with polycrystalline
samples in a PANalytical Axios Max spectrophotometer.
Regarding the fine aggregate, the characterization of the chemical phases pres-
ent was also performed by X-ray diffraction (XRD). For this purpose, pulverized
polycrystalline samples were used in a PANalytical EMPYREAN diffractometer using
Cu Ka radiation with a wavelength k of 1.54051 Å in a 2h scan mode between 0° and
70°, a 40 kV voltage, a 40 mA current, a 0.001 step, and an initial step velocity equal
to 0.002 p/s. The minerals were identified by comparison to the data bank of the
‘‘International Centre for Diffraction Data (ICDD)’’.
2.2. Dosage and exposure procedures
A mortar mixture was prepared from the characterized materials (cement and
aggregates), adopting the standard trace of 1:3.2:0.6 for cement: aggregate:
water/cement ratio on a mass basis, as established by the standard [22]. The mate-
rial was molded in prismatic molds containing carbon steel (CA-50) and graphite
electrodes for characterizing the properties based on electrochemical methods
(monitoring of the accelerated aging processes), in a similar manner to that devel-
oped by Kanning et al. [15]. Prior to molding, the steel electrodes were identified
and measured (mass and dimensions), and part of their surface area was protected
with a self-fusing anti-corrosion tape to facilitate studying the reaction mecha-
nisms that occur only at the interface of the mortar and reinforcement. The average
exposed area for each of the Sps was approximately 0.643 m2 and located 0.065 m
below the upper end and 0.025 above the lower end. Electrical contact was
maintained with a copper wire.
Once prepared, the mortars were air-cured for 24 h and then removed from the
molds. The curing of the Sps was performed in a humid chamber, according to the
standard [23] until the 28th day of age.
After curing, the Sps were divided into groups and subjected to different
accelerated aging processes, including the following:
 desiccator: a portion of the samples were kept as a reference sample in a
desiccator after wet curing for 28 days;
 sulfate chamber: exposure in a sulfate chamber under daily wetting (8 h at
40 °C in a humid atmosphere with a 2% SO2 concentration) and drying (16 h
at ambient conditions) cycles, according to a ‘‘Kesternich’’ procedure adapted
for cementitious materials from the Brazilian standard for metal plates [24];
 salt spray chamber: exposure in a chloride chamber under daily wetting (8 h at
40 °C in a humid atmosphere with a 5% NaCl concentration) and drying (16 h at
25 °C) cycles, according to the procedure adapted for cementitious materials
from the Brazilian standard for metallic plates [25].
After the exposure, some of the samples were broken and were characterized by
optical microscopy (OM) and scanning electron microscopy (SEM) with elemental
analysis by EDS using a Nikon SMZ800 (OM) and a TESCAN, model VEGA 3 (SEM/
EDS). The samples were prepared, prior to their characterization, with surface pol-
ishing (OM) and gold sputtering (SEM). Additionally, the samples were character-
ized with regard to surface and internal porosity by the Brunauer–Emmett–Teller
(BET) technique using the Quantachrome Autosorb Automated Gas Sorption equip-
ment with nitrogen gas as an adsorbate.
2.3. Electrochemical impedance spectroscopy (EIS)
Prior to and periodically during exposure, the reinforced mortar samples were
characterized by the EIS technique, according to the ASTM G106 standard [26].
The measurements were performed by applying an alternating signal with a
25 mV amplitude for a frequency interval from 106 Hz to 5  10 2 Hz in open-cir-
cuit potential, with 10 points obtained per decade. These analytical conditions were
established based on the literature [7,11,15] and previously confirmed in the
laboratory, obeying the criteria for linearity, causality, and stability, thus conferring
credibility and representativeness to the results obtained.
The measurements were performed in an electrochemical system from
Ecochemie, model Autolab – PGSTAT-100, interfaced with a microcomputer for
the acquisition and recording of data. The measurements were performed with
the wet Sps relative to a reference saturated calomel electrode.
The characterization of the samples was performed by interpreting the spectra
obtained from the Bode and Nyquist diagrams. The variation in the parameters of
complex components and the magnitude of the impedance were analyzed, as was
the phase angle for each of the frequency regions, relative to each type of cement
and aging interval, based on the literature [7,13,27]. The mathematical analysis of
the diagrams for adjusting the equivalent electrical circuit was performed with
the ZView software [13].
3. Results and discussion
3.1. Cement and aggregate characterization
The characterized materials exhibit physical–chemical proper-
ties in accordance with the Brazilian standard recommendations
and the manufacturer’s specifications [20,21]. The results from
the chemical characterization by XRF are presented in Table 1.
The aggregates were also characterized regarding their
predominant chemical phases. The results obtained indicate that
the material is composed of the following minerals: quartz (SiO2),
pyrite (FeS2), plagioclase ((Na,Ca)(Si,Al)4O8), and anatase (TiO2).
3.2. Impedance measurements of the reference mortar
The results obtained for the EIS of the reinforced mortar cured
in a humid chamber and kept in a desiccator after 28 days of age
are presented in the complex plane and Bode diagrams of Fig. 1.
The analysis of the diagrams for the high-frequency region
(above 102 Hz) reveals that the average resistance of the mortar
 M.O.G.P. Bragança et al. / Construction and Building Materials 68 (2014) 650–658 653

Table 1 properties of the double layer also tended toward stabilization

Chemical characterization of the materials (cement and aggregates) by XRF.
Materials CaO% SiO2%
CP II Z 32 54.8 21.53
Aggregates 9.1 47.1
Fe2O3% SO3%
CP II Z 32 2.94 2.66
Aggregates 18.3 0.3
tended to increase from the beginning of the process until 91 days
of age (approximately). This resistance R0, varied between 4  102
and 3  103 X cm, which represented an improvement in the prop-
erties of the cementitious matrix throughout the curing process.
This process occurs due to hydrate formation reactions, which tend
to fill the voids and pores, thus providing the isolation and conse-
quent protection of the reinforcement. According to Serdar et al.
[13], this process tends to occur during a period of up to 4 months
after molding of the Sp, even in accelerated aging conditions. After
91 days, the material properties remained constant. The results
obtained corroborate those shown in the literature [13].
For the intermediate-frequency region (102–10 2 Hz), in which
the phase angle tends to display an increasing curve with 1 slope,
the capacitive behavior of the bar/mortar double layer was
observed. Once again, curing of the matrix was evidenced until
91 days, with a small increase in the resistance R1 (from 2  102
to 7  102 X cm) and a decrease in the capacitance of the
associated double layer (from 500 to 50 lF cm 2). In a similar man-
ner to the analysis performed for the high-frequency region, the
after 91 days. This behavior reflects the formation of a passive film
Na2O% Al2O3% MgO% on the surface of the reinforcement, induced by the pH of the
0.19 4.73 5.43 cementitious matrix (approximately 12), which tends to protect
2.5 13.0 2.8 the material from corrosion [8,12]. With the advancement of stud-
TiO2% Others% Loss on Ignition% ies, the analysis of the phase angle curve relative to the frequency
for more advanced ages (greater than 200 days) revealed an
0.24 7.48 6.38
3.2 3.7 0.81
increasing trend toward capacitive behavior due to the progression
of this parameter from approximately 45° (at 205 days) to 70°
(beginning at 264 days), which again reflects the formation and
growth of a passive film of compact iron oxide [13,28].
Finally, for the low-frequency region (less than 10 2 Hz), in
which charge transfer processes are observed between the bar
and the matrix, an increase was again found for resistance R2, from
approximately 20  103 to 20  106 X cm, and a decrease was
found in the associated capacitance from approximately 5  10 3
to 20  10 5 F cm 2, primarily during the initial period, between
28 and 91 days of age (with maintenance of constant properties
until the end of the study). According to some authors [7,13,29],
the properties observed for this region may be related to the corro-
sive processes that tend to develop in the steel bar, with the
increase in R2 and the decrease in capacitance being associated
with an increase in corrosion resistance. Thus, the results found
for this region corroborate the remaining results (high- and inter-
mediate-frequency regions) because the obtainment of mortar
with a high electrical resistance and the formation of a compact
iron oxide film on the reinforcement tend to contribute to the pro-
tection of the metal, thus avoiding deterioration of the structure.

10
5
(a) (b)
R0 C1, n1 -70

R1 C2, n2
-60

4
R2 -50 28 days
10
93 days
log |Z| (Ohm)

-40 205 days

θ( )

264 days
o

-30 322 days

3
10 -20

-10

0
2
10
-3 -2 -1 0 1 2 3 4 5 6 -3 -2 -1 0 1 2 3 4 5 6
10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10
log f (Hz) log f (Hz)

1.8x10
4 (c) 2.0x10
3
(d)

4 28 days
1.5x10
93 days
3
1.5x10 205 days
4
1.2x10 264 days
322 days
-Z" (Ohm)
-Z" (Ohm)

3 3
9.0x10 1.0x10
0.01371 Hz

3
6.0x10
2
0.01371 Hz 5.0x10 2476.1 Hz
3
3.0x10
2476.1 Hz

0.0 0.0

3 3 3 4 4 4 2 2 3 3 3
0.0 3.0x10 6.0x10 9.0x10 1.2x10 1.5x10 1.8x10 -5.0x10 0.0 5.0x10 1.0x10 1.5x10 2.0x10
Z' (Ohm) Z' (Ohm)

Fig. 1. Bode diagram (a/b) and Nyquist diagram (c/d) obtained through the EIS analysis of the reinforced mortar cured in a humid chamber until 28 days of age and kept in a
desiccator as a reference standard. Spectrum obtained for the frequency range from 1  106 Hz to 5  10 2 Hz with a 25 mV excitation signal in an open-circuit potential,
relative to a saturated calomel electrode.
654 M.O.G.P. Bragança et al. / Construction and Building Materials 68 (2014) 650–658
3.3. Impedance measurements of chloride-exposed mortar
The results obtained by EIS for the reinforced mortar cured in a
humid chamber and kept in a salt spray chamber (5% NaCl) after
28 days of age are presented in the Bode diagrams of Fig. 2.
The spectra obtained show a behavior for the curves that is
very similar to that of the reference standard for ages less than
91 days. The resistance of the mortar, analyzed in the high-fre-
quency region (above 102 Hz), tended to increase on average
from 5  102 to 5  103 X cm and is again related to the hydra-
tion processes of the components of the cementitious matrix,
which tend to decrease the porosity of the material for up to
4 months after molding, even in accelerated aging conditions
[8,13]. After 91 days, the resistance of the mortar continued
increasing in a more accentuated manner than for the reference
standard, with average values of approximately 106 X cm
obtained at 236 days. This behavior can be justified as a function
of the concentration of the solution on the interior of the pores
and the formation of Friedel’s salts, which tend to accumulate
in the voids, thus promoting a decrease in the preferential paths
for current passage [13,30–31]. At 264 days, the resistance of the
mortar showed an accentuated drop to values of approximately
2 X cm, which represented a characteristic profile of advanced
corrosive processes, with deterioration of the cementitious
matrix [7–13,29–31].
The bar/mortar double layer can be analyzed in the intermedi-
ate-frequency region (102–10 2 Hz) and tended to exhibit a resis-
tive behavior. An increase in resistance was observed, which was
attributed to the passive layer of iron oxide or to the corrosion
products formed (R1), on average from 100 X cm (at 28 days) to
108 X cm (264 days), and a decrease in the average capacitance
of the double layer (C2), from 3  10 8 F cm 2 (28 days) to
1.5  10 9 F cm 2 (264 days), was also observed. The increase in
the resistance and decrease in the capacitance were specifically
evident from the age of 205 days, when R1 increased from
approximately 1  103 X cm to 5  105 X cm and C1 decreased
from approximately 30  10 6 F cm 2 to 0.5  10 9 F cm 2. These
results may be related to the formation and deposition of oxides
and corrosion products on the bar/reinforcement interface, which
tend to promote the closing of pores and voids, thus leading to
an increase in the resistance and consequent decrease in
capacitance [13,29]. Furthermore, the beginning of the active cor-
rosion process, found from the age of 205 days using the corrosion
potential technique (decrease in the open-circuit potential [OCP]
vs. the saturated calomel electrode [SCE] from 298 mV to
473 mV), tended to increase the formation of deposited corrosion
9 R0
10
8 93 days
10
7
10
log |Z| (Ohm)
6
10
5 -40
10
4
10
3
10
2
10
-3 -2 -1 0 1 2 3 4 5
10 10 10 10 10 10 10 10 10
log f (Hz)
Fig. 2. Bode diagram, obtained for the EIS analysis of the reinforced mortar cured in a humid chamber until 28 days of age and then kept in a salt spray chamber (5% NaCl).
Spectrum obtained for the frequency range from 1  106 Hz to 5  10 2 Hz with a 25 mV excitation signal in an open-circuit potential, relative to a saturated calomel
electrode.
products, thereby modifying the parameters determined in the
intermediate-frequency region.
For evaluation of the low-frequency region (less than 10 2 Hz),
in which charge-transfer processes occur between the bar and
matrix and the corrosion resistance of the material is evaluated,
in an analogous manner to the reference standard, an increase in
the resistance R2 was observed from approximately 2  103 to
2  105 X cm, together with a decrease in the associated capaci-
tance from approximately 10 4 to 10 10 F cm 2, even after expo-
sure to 5% NaCl for 5500 h. Despite this finding, the analysis of
the phase angle showed a decreasing trend in the capacitive behav-
ior of the region, with variation of the parameter from approxi-
mately 40° to approximately 10°. Furthermore, the associated
capacitance showed a non-ideal behavior in this region, with vari-
ation of the parameter n of the CPE from 0.63 (28 days) to 0.19
(205 days). According to the literature, the increase in the resis-
tance R2 is normally related to the higher protection of the bar from
the corrosive processes. However, the deposition of Friedel’s salts in
the cementitious matrix and the high formation of corrosion prod-
ucts tend to increase the electrical resistance of the set, thereby hin-
dering the verification of the mechanisms under development by
analyzing only these parameters (R2 and C2). Furthermore, the
decrease in the parameter n, observed by the analysis of the phase
angle profile, was associated with the corrosion by chloride in the
studies by Feliu et al. [10]. These authors dosed concrete with the
addition of (3%) CaCl2 to induce internal contamination and found
values varying between 1.0 and 0.7. The predominance of the resis-
tive behavior, different from the reference standard, indicates the
occurrence of slow kinetic mechanisms, likely related to the attack
and dissolution of iron oxide from the passive layer.
Furthermore, regarding aging in the salt spray chamber, analyz-
ing the profiles in the entire spectrum revealed an alteration in the
phase angle curve beginning at 205 days and intensifying at
264 days. In these curves, a decrease from the level region of the
angle 0° was observed with a trend towards capacitive behavior, ini-
tially from the frequency of 104 Hz and changing to 103 Hz
(205 days) and 101 Hz (264 days). These changes in the profile of
the phase angle may be correlated to the steel corrosion process,
although R2 resistance values higher than 109 X cm have been
obtained [9,28].
3.4. Impedance measurements of sulfate-exposed mortar
The results obtained by EIS for the reinforced mortar cured in a
humid chamber and kept in a sulfate chamber (2% SO2) after
28 days of age are presented in the Bode diagram of Fig. 3.
C1, n1
-100
R1 C2, n2 28 days
R2
-80 205 days
R0 C1 , n1 C2 , n2 236 days
264 days
-60
R1 R2
θ( )
o
-20
0
6 -3 -2 -1 0 1 2 3 4 5 6
10 10 10 10 10 10 10 10 10 10 10
log f (Hz)
 M.O.G.P. Bragança et al. / Construction and Building Materials 68 (2014) 650–658
5
10
R0 C1, n1
R1 C2, n2
4 R2
10
log |Z| (Ohm)
3
10
2
10
-3 -2 -1 0 1 2 3
10 10 10 10 10 10 10 10
log f (Hz)
Fig. 3. Bode diagram, obtained for the EIS analysis of the reinforced mortar cured in a humid chamber until 28 days of age and kept in a sulfate chamber (2% SO2). Spectrum
obtained for the frequency range from 1  106 Hz to 5  10 2 Hz with a 25 mV excitation signal in an open-circuit potential, relative to a saturated calomel electrode.
Fig. 3 shows, for the high-frequency region (higher than 102 Hz)
in which it is possible to obtain mortar parameters, the increase of
the resistance property R0 of approximately 5  102 (28 days) to
1.5  103 X cm (91 days), and, after, the maintenance of R0 in
1.5  103 X cm until the end of the aging study. This value, which
is approximately half of that obtained for the reference standard at
ages greater than 91 days, represents a deficiency in the curing
process of the cementitious matrix, which tends to occur due to
the no stoichiometry of the reactions promoted by the presence
of sulfur compounds. These compounds favored the formation of
gypsite, ettringite, and monosulfate, with a predominance of the
cement hydrates, which led to decalcification of the material [3].
Despite these compounds tending to promote the filler effect, in
an analogous manner to the Friedel’s salts, they have a crystalline
(and conductive) and expansive characteristic, which tends to pro-
mote microcracks in the matrix [1–5], consequently hindering the
increase of the resistance of the mortar. Similar studies indicated
that the electrical resistance of mortar or concrete, after the cure
process ending and without any deterioration process occurrence,
can be higher than 3.0  103 X cm [13,32–34], the same result
obtained for the reference mortar. Although, Gerengi et al. [19]
did not discussed but proved that the exposure to sulphuric acid
reduced the concrete electrical resistance from 8.0  103 to less
than 2  103 X cm after 60 days of aging, which corroborate with
the results obtained in this research and are a consequence of
the sulfate attack occurrence.
For the intermediate-frequency region (102–10 2 Hz), in a sim-
ilar manner to the reference standard, the capacitive behavior of
the reinforcement-mortar double layer was observed, as shown
by the ascending curve with a 1 slope of the phase angle profile.
The passive film resistance (R1) exhibited constant values of
approximately 1  103 X cm; the associated capacitance (C1) also
had a constant value over the aging process, at approximately
4  10 5 F cm 2. This finding again revealed the formation of a
passive film on the metal resulting from the exposure of the metal
to the basic pH of the mortar. In this study, despite the exposure to
aggressive agents, up to 300 days of age, there were no signs of the
formation of corrosion products derived from the active deteriora-
tion process. This phenomenon can be explained by the size of the
sulfate ion (compared with chloride), which tends to difficult the
penetration through the material, leading to a slower occurrence
of corrosive processes [3]; a difference in the corrosion mecha-
nisms that tend to develop during exposure of the carbon steel
to SO24 (compared with Cl ), with the rupture of the passive film
by the decrease in the pH of the environment and the formation
of acid compounds, in a similar manner to the carbonation
655
-70
-60
28 days
-50 93 days
205 days
264 days
-40
θ( )
322 days
o
-30
-20
-10
0
4 5 6 -3 -2 -1 0 1 2 3 4 5 6
10 10 10 10 10 10 10 10 10 10 10 10
log f (Hz)
processes [3,30]; and the reactivity of the sulfate ion in a cementi-
tious environment, which leads to a greater formation of byprod-
ucts compared with the diffusion of the free form of the ion
(which tends to decelerate corrosion) [3].
In the low-frequency region (less than 10 2 Hz), a predomi-
nantly capacitive behavior was found for ages less than 91 days,
with an increase in the resistance R2 from approximately
20  106 X cm to 500  106 X cm, a decrease in the associated
capacitance from approximately 5  10 4 F cm 2 to 2  10 4 -
F cm 2, and maintenance of the phase angle at values greater than
60°. Analogous to the reference standard, these results represent
the protection of the reinforcement against corrosive processes
due to the formation of a passive oxide layer on the metal surface
as well as the curing process of the cementitious matrix (deficient
in a sulfate environment) [7,13,27]. After this age, however, the
resistance R2 decreased to values of 50  103 X cm, and the capac-
itive behavior was predominant, with a decrease in the phase angle
to approximately 45° (a value that may be related to diffusion
processes). The analysis of the behavior of these system parame-
ters throughout the aging period, together with the results from
OCP vs. SCE, maintained at values below 276 mV (region of prob-
ability of uncertain or low corrosion, in accordance with ASTM
C876 [36]) until 322 days of age, indicates the susceptibility of
the reinforcement, likely due to the microcracking of the matrix
in connection with the decrease in the pH due to exposure to sul-
fate ions [1–5]. The electrochemical results obtained, however, did
not enable confirming the active corrosion of the reinforcement
until the present.
3.5. Comparison between chloride and sulfate exposition
The average results obtained by simulation for the parameters
of the reinforced mortar exposed to the different environments
are presented in Table 2, namely, the reference, 5% NaCl, and 2%
SO2.
Comparison of the results obtained for the aging in salt spray
and sulfate showed a similar initial behavior for both environ-
ments, until 91 days, with an increase in the mortar resistance as
a function of the curing processes. This occurred in a more efficient
manner in the chloride environment because this salt tends to
accelerate curing processes [28,29]. After this age, there was a
marked increase in the solution resistance (R0), even in the cham-
ber with 5% NaCl, until 236 days, possibly due to the formation and
accumulation of Friedel’s salt on the interior of the pores. This phe-
nomenon was followed by a drop in this parameter as a function of
the deterioration of the matrix due to the initiation of the active
656 M.O.G.P. Bragança et al. / Construction and Building Materials 68 (2014) 650–658

Table 2
Parameters of the reinforced mortar exposed to the reference, 5% NaCl, and 2% SO2 environments, obtained through adjustment of the equivalent circuits in the EIS spectra.

EIS Parameters Exposition 28 days 91 days 205 days 236 days 264 days 322 days
3
R0 (10 X cm) Reference 0.43 2.9 3.1 – 2.9 2.4
5% NaCl 0.49 4.5 250 1200 0.002 –
2% SO2 0.46 1.6 1.5 – 1.5 1.6
R1 (103 X cm) Reference 0.17 0.67 1.2 – 1.1 0.93
5% NaCl 0.12 1.6 410 740 120,000 –
2% SO2 0.36 0.66 0.95 – 1.8 1.1
6 2
C1 (10 F cm ) Reference 440 44 32 – 29 31
5% NaCl 340 33 0.0005 0.0006 0.0014 –
2% SO2 320 47 52 – 40 27
nCPE1 Reference 0.5 0.7 0.5 – 0.75 0.71
5% NaCl 0.39 0.50 0.67 0.68 0.40 –
2% SO2 0.67 0.69 0.53 – 0.57 0.58
OCP (mV) Reference 580 140 320 – 270 210
5% NaCl 570 300 470 510 540 –
2% SO2 360 130 290 – 160 72
R2 (103 X cm) Reference 22 18,000 390 – 180,000 180,000
5% NaCl 1800 83 18,000 3800 160,000 –
2% SO2 18,000 500,000 64 – 18,000 180
6 2
C2 (10 F cm ) Reference 4300 170 150 – 120 120
5% NaCl 7400 200 0.69 0.099 0.00008 –
2% SO2 390 170 190 – 55 91
nCPE2 Reference 0.56 0.77 0.68 – 0.8 0.8
5% NaCl 0.64 0.79 0.19 0.39 0.75 –
2% SO2 0.77 0.78 0.73 – 0.67 0.69

corrosion process. Regarding exposure to SO24 , the R0 parameter spray chamber), similar to what was reported in the literature for
tended toward stabilization after 91 days, at values even lower Friedel’s salts [31].
than the ones obtained for the reference standard, around For the intermediate-frequency region, the electrochemical
1.5  103 X cm, which represents the occurrence of sulfate attack study of aging in chloride showed the possible formation of oxide
with the formation of gypsum, ettringite and monosulfate crystal- corrosion products beginning at 205 days, evidenced by the signif-
line compounds and consequent microcrack of the matrix, compet- icant increase in the polarization resistance R1, decrease in the
ing with the curing of the cementitious material. This observation double-layer capacitance C1, and alteration of the phase angle pro-
could be proven based on the performance of complementary tests file. The increase in the polarization resistance and the decrease in
with the mortar, such as porosimetry as well as morphological the double-layer capacitance were also evidenced for the reference
analysis and elemental analysis by SEM/EDS. standard samples and those exposed to sulfate, although in a less
For the porosimetry test, according to the results shown in pronounced manner, possibly due to the formation of a passive
Table 3, a decrease in the surface area of the external region from oxide layer on the surface of the steel bar. The results shown in
13.9 m2 g 1 to 4.749 m2 g 1 was observed as a function of the Table 3 indicate that for all of the studies, there was an increase
aging applied (specifically for the sulfate chamber). Moreover, a in the surface area for the internal region of the Sp, compared to
reduction in pore size from 56.9 Å (reference standard) to the external region, related to the interface of the mortar with
43.43 Å (chloride) and 49.82 Å (sulfate), related to the accumula- the steel bar. The exception was the salt spray aging process, in
tion of byproducts in the pores, was also observed for both types which the pores tend to decrease in size as a function of the accu-
of aging [35,37]. The formed byproducts were confirmed by SEM/ mulation of iron oxides produced in the passivation phase of the
EDS, as presented in Figs. 4 and 5, in which microcracks and crys- steel. This characteristic was not observed for the chloride ions,
tals with a morphology and chemical composition similar to that of despite the increase in the polarization resistance R1, because the
gypsum and monosulfate (for aging in a SO2 chamber) can be visu- corrosion products tended to diffuse through the material, whereas
alized [35], as well as hexagonal-shaped plates and an elevated the oxide layer accumulated on the metal surface, thus protecting
concentration of the elements chlorine and aluminum (for the salt the reinforcement from the corrosive processes.
Finally, from the perspective of the reinforcement, the suscepti-
Table 3 bility to corrosion was evidenced electrochemically for the study in
Porosimetry results for the regions of mortar and oxide formation obtained using the a sulfate atmosphere, with the decrease in the charge transfer
BET technique for different types of accelerated aging. resistance R2 and the predominance of capacitive behavior
Parameter Exposition External region – Internal region – throughout the aging period. For the study with chloride, although
mortar oxide layer the variation of the parameters does not indicate the occurrence of
Surface area Reference 13.9 33.461 active corrosion, the alteration of the phase angle profile added to
(m2 g 1) 5% NaCl 22.88 38.885 the variation of the OCP vs. SCE, when evaluated in comparison
2% SO2 4.749 12.98 with similar studies [9,28], indicated deterioration of the reinforce-
Pore volume Reference 0.01977 0.04089 ment. These analyses, performed based on the EIS data, were pro-
(cm3 g 1) 5% NaCl 0.02484 0.04993 ven with the breaking of the Sps and the observation of the carbon
2% SO2 0.005915 0.01411
steel surface through OM. The images obtained in these tests are
Pore size (Å) Reference 56.9 48.889 shown in Fig. 6.
5% NaCl 43.43 51.403
According to the optical microscope images (Fig. 5), the study in
2% SO2 49.82 43.5
the chloride environment provided, after approximately 300 days
 M.O.G.P. Bragança et al. / Construction and Building Materials 68 (2014) 650–658 657

Fig. 4. SEM with 1000 magnification, obtained for the mortar samples after 300 days of aging with (a) 5% NaCl, external surface of the Sp; (b) 5% NaCl, internal region of the
Sp; (c) 2% SO2, external surface of the Sp; and (d) 2% SO2, internal region of the Sp.

0.5 0.9
(a) 0.8
(b)
0.4 0.7
Monosulfate
0.6 Ettringite
0.3 Gypsum
0.5
Cl / Ca

S / Ca

Friedel's Salt Surface Region

Surface region 0.4 Internal Region
0.2 Internal region
0.3
0.2
0.1
0.1

0.0 0.0
0.0 0.1 0.2 0.3 0.0 0.1 0.2 0.3 0.4 0.5
Al / Ca Al / Ca

Fig. 5. Elemental chemical analysis by EDS of the mortar samples after 300 days of aging with (a) 5% NaCl and (b) 2% SO2.

Fig. 6. Microscopic analysis (400 magnification) of the carbon steel bar after an exposure of approximately 300 days in the (a) reference standard, (b) 5% NaCl, and (c) 2% SO2
environments.
658 M.O.G.P. Bragança et al. / Construction and Building Materials 68 (2014) 650–658
of exposure, the formation of pits. Regarding sulfate, alteration of
the surface was observed, which is indicative of the susceptibility
of the reinforcement, without evidence of more advanced corrosive
processes.
4. Conclusions
The EIS results obtained for the reinforced mortar exposed for
approximately 300 days in a sulfate chamber indicated that this
technique can be used to monitor the decrease in the durability
of the system even in initial stages of deterioration, based on the
previously established models for the reference standard and the
samples exposed to chloride. From the analysis of the profiles
obtained and the adjustment of the equivalent circuits, it was iden-
tified the formation of gypsum and monosulfate byproducts in the
matrix, which competed with the formation of cement hydrates
during the curing process and yielded a material with a 50% less
electrical resistance (around 1.5  103 X cm) in comparison with
reference mortar. Moreover, it was verified the production of a
passive iron oxide layer on the surface of the steel throughout
the aging period, susceptibility of the reinforcement due to
microfissuring, and a decrease in the pH of the material. The elec-
trochemical results were proven by complementary assays (SEM/
EDS, BET, and OM), which evidenced the possibility of using the
EIS technique for the non-destructive analysis of cementitious
material susceptible to sulfate attack during the initial stages of
the corrosive process. Therefore, this technique can potentially be
used for monitoring corrosion in reinforced concrete.
Acknowledgments
The authors thank the PIPE-UFPR (Programa Interdisciplinar de
Pós-Graduação em Engenharia e Ciência dos Materiais da Univer-
sidade Federal do Paraná), COELBA (Companhia de Eletricidade
do Estado da Bahia), Institutos Lactec CNPq (Conselho Nacional
de Desenvolvimento Científico e Tecnológico), and ANEEL (Agência
Nacional de Energia Elétrica) for their financial support and infra-
structure. Finally, we thank the CNPq productivity fellowship from
Kleber F. Portella (grant number 305718/2011-8) and from Cláudia
Eliana Bruno Marino (grant number 301434/2013-1).
References
[1] Collepardi M. A state-of-art review in delayed ettringite attack on concrete.
Cem Concr Compos 2003;25:401–7.
[2] Zhang M, Chen J, Lv Y, Wang D, Ye J. Study on the expansion of concrete under
attack of sulfate and sulfate-chloride ions. Constr Build Mater 2013;39:26–32.
[3] Oliveira I, Cavalaro S, Aguado A. New kinetic model to quantify the internal
sulfate attack in concrete. Cem Concr Res 2013;43:95–104.
[4] Hime W, Mather B. ‘‘Sulfate attack’’, or is it? Cem Concr Res 1999;29:789–91.
[5] Famy C, Taylor H. Ettringite in hydration of Portland cement concrete and its
occurrence in mature concretes. ACI Mater J 2001;98(4):350–6.
[6] Keddam M, Takenouti H, Nóvoa XR, Andrade C, Alonso C. Impedance
measurement on cement paste. Cem Concr Res 1997;27(8):1191–201.
[7] Ford SJ, Shane JD, Mason TO. Assignment of features in impedance spectra of
the cement-paste/steel system. Cem Concr Res 1998;28(12):1737–51.
[8] Andrade C, Keddam M, Nóvoa XR, Pérez MC, Rangel CM, Takenouti H.
Electrochemical behavior of steel rebars in concrete: influence of
environmental factors and cement chemistry. Electrochim Acta 2001;46:
3905–12.
[9] Ismail M, Ohtsu M. Corrosion rate of ordinary and high-performance concrete
subjected to chloride attack by AC impedance spectroscopy. Constr Build
Mater 2006;20:458–69.
[10] Feliu S, González J, Miranda J, Feliu V. Possibilities and problems of in situ
techniques for measuring steel corrosion rates in large reinforced concrete
structures. Corros Sci 2005;47:217–38.
[11] Vedalaksmi R, Saraswathy V, Song H, Palaniswamy N. Determination of
diffusion coefficient of chloride in concrete using Warburg diffusion
coefficient. Corros Sci 2009;51:1299–307.
[12] Angst U, Elsener B, Larsen C, Vennesland O. Chloride induced reinforcement
corrosion: electrochemical monitoring if initiation stage and chloride
threshold values. Corros Sci 2011;53:1451–64.
[13] Serdar M, Zulj LV, Bjegovic D. Long-term corrosion behavior of stainless
reinforcing steel in mortar exposed to chloride environment. Corros Sci
2013;69:149–57.
[14] Cesen A, Kosec T, Legat A. Characterization of steel corrosion in mortar by
various electrochemical and physical techniques. Corros Sci 2013;75:47–57.
[15] Kanning RC, Portella KF, Bragança MOGP, Bonato MM, Santos JCM. Banana
leaves ashes as pozzolan for concrete and mortar of Portland cement. Constr
Build Mater 2014;54:460–5.
[16] Bondarenko A. Analysis of large experimental datasets in electrochemical
impedance spectroscopy. Anal Chim Acta 2012;743:41–50.
[17] Macdonald JR. Impedance spectroscopy. Ann Biomed Eng 1992;20:289–305.
[18] Cheng T-P, Lee J-T, Tsai W-T. Corrosion of reinforcements in artificial sea water
and concentrated sulfate solution. Cem Concr Res 1990;20:243–52.
[19] Gerengi H, Kocak Y, Jazdzewska A, Kurtay M, Durgun H. Electrochemical
investigations on the corrosion behavior of reinforcing steel in diatomite- and
zeolite-containing concrete exposed to sulphuric acid. Constr Build Mater
2013;49:471–7.
[20] NBR 11578. Cimento Portland composto – Especificação [Composite Portland
Cement– Specification], Brasil, 1991.
[21] NBR 7211. Agregados para Concreto – Especificação [Concrete Aggregates –
Specification], Brasil, 2009.
[22] NBR 13583. Cimento Portland: Determinação da Variação Dimensional de
Barras de Argamassa de Cimento Portland Expostas à Solução de Sulfato de
Sódio [Portland Cement: Determination of the Dimensional Variation of
Portland Cement Mortar Bars Exposed to a Sodium Sulfate Solution], Brasil,
1996.
[23] NBR 5738. Concreto: Procedimento para moldagem e cura de corpos de prova
[Concrete: Procedure for the molding and curing of specimens], Brasil, 2003.
[24] NBR 8094. Material metálico revestido e não revestido - Corrosão por
exposição à névoa salina – Método de ensaio [Coated and uncoated metallic
material – Corrosion by exposure to salt spray], Brasil, 1983.
[25] NBR 8096. Material Metálico Revestido e Não Revestido: Corrosão por
Exposição ao Dióxido de Enxofre – Método de Ensaio [Coated and Uncoated
Metallic Material: Corrosion through Exposure to Sulfur Dioxide – Test
Method], Brasil, 1983.
[26] ASTM G106. Standard Practice for Verification of Algorithm and Equipment for
Electrochemical Impedance Measurements, 2010.
[27] Christensen BJ. Impedance spectroscopy of hydrating cement-based materials:
measurement, interpretation and application. J Am Ceram Soc
1994;77:2789–804.
[28] Castela AS, Sena Da Fonseca B, Duarte RG, Neves R, Montemor MF. Influence of
unsupported concrete media in corrosion assessment for steel reinforcing
concrete by electrochemical impedance spectroscopy. Electrochim. Acta 2013.
http://dx.doi.org/10.1016/j.electacta.2013.11.157.
[29] Koleva DA, De Wit JHW, Van Breugel K, Veleva LP, Van Westing E, Copuroglu O,
et al. Correlation of microstructure, electrical properties and electrochemical
phenomena in reinforced mortar. Breakdown to multi-phase interface
structures. Part II: Pore network, electrical properties and electrochemical
response. Mater Charact 2008;59:801–15.
[30] Türkmen I, Gavgali M. Influence of mineral admixtures on the some properties
and corrosion of steel embedded in sodium sulfate solution of concrete. Mater
Lett 2003;57:3222–33.
[31] Talero R. Synergic effect of Friedel’s salt from pozzolan and from OPC co-
precipitating in a chloride solution. Constr Build Mater 2012;33:164–80.
[32] Duarte RG, Castela AS, Neves R, Freire L, Montemor MF. Corrosion behavior of
stainless steel rebars embedded in concrete: an electrochemical impedance
spectroscopy study. Electrochim. Acta 2013. http://dx.doi.org/10.1016/
j.electacta.2013.11.154.
[33] Choi Y-S, Kim J-G, Lee K-M. Corrosion behavior of steel bar embedded in fly ash
concrete. Corros Sci 2006;48:1733–45.
[34] Montemor MF, Simões AMP, Salta MM. Effect of fly ash on concrete
reinforcement corrosion studied by EIS. Cem Concr Compos 2000;22:175–85.
[35] Atahan HN, Dikme D. Use of mineral admixtures for enhanced resistance
against sulfate attack. Constr Build Mater 2011;25:3450–7.
[36] ASTM C876. Standard Test Method for Corrosion Potentials of Uncoated
Reinforcing Steel in Concrete. 2009.
[37] Dong B, Qui Q, Xiang J, Huang C, Xing F, Han N, et al. Electrochemical
impedance measurement and modeling analysis of the carbonation behavior
for cementitious materials. Constr Build Mater 2014;54:558–65.