Cement and Concrete Research 95 (2017) 217–225

Contents lists available at ScienceDirect

Cement and Concrete Research

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

Understanding the interacted mechanism between carbonation and

chloride aerosol attack in ordinary Portland cement concrete
Jun Liu a, Qiwen Qiu b,⁎, Xiaochi Chen a, Feng Xing a, Ningxu Han a, Yijian He a, Yueshan Ma a
a
Guangdong Provincial Key Laboratory of Durability for Marine Civil Engineering, Shenzhen University, Shenzhen 518060, Guangdong, China
b
Department of Architecture and Civil Engineering, City University of Hong Kong, Hong Kong

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

Article history: An experimental study is carried out with the aim to understand the interacted mechanism between carbonation
Received 7 December 2016 and chloride aerosol attack in ordinary Portland cement (OPC) concrete. Effects of carbonation on the chloride
Received in revised form 17 February 2017 profile, the chloride binding capacity and the chloride diffusion coefficient are evaluated. Besides, effect of chlo-
Accepted 28 February 2017
ride aerosol attack on the carbonation rate is investigated. Concrete specimens with three water-to-cement ratios
Available online 10 March 2017
(0.38, 0.47 and 0.53) are fabricated in this work. Tested results demonstrate that carbonation remarkably affects
Keywords:
the chloride profile, reduces the chloride binding capacity, and also accelerates the rate of chloride ion diffusion of
Concrete concrete. Besides, the presence of chloride aerosol can lead to lower the carbonation depth and increase the pH
Carbonation value of carbonated concrete. Microscopic properties such as morphology, porosity, and pore size distribution for
Chloride aerosol attack the contaminated concretes are explored by scanning electron microscope and mercury intrusion porosimetry,
Water-to-cement ratio which provide strong evidence to these research findings.
Microstructure © 2017 Elsevier Ltd. All rights reserved.

1. Introduction investigate the effect of carbonation on chloride ingress for cement

Durability of reinforced concrete structures is of a great major con-
cern across the world especially in the coastal regions where structures
suffer from the steel rebar corrosion due to the ingress of carbon dioxide
and chloride ion. The penetration of carbon dioxide can initiate the car-
bonation reaction in concrete materials. Concrete carbonation is a neu-
tralization phenomenon which involves in the reaction between the
carbon dioxide and the hydrated cement compounds, resulting in a re-
duction of the pH value of concrete pore solution [1–3]. The alkalinity
reduction of concrete can lead to destroy the initially-formed passive
film around the steel reinforcement, causing the steel corrosion in the
presence of both moisture and oxygen [4–7]. In addition to carbon diox-
ide ingress, the steel corrosion can be occurred when a sufficient quan-
tity of chloride ions are accumulated on the steel rebar surface [8,9]. The
chloride-induced steel corrosion reduces the cross-sectional area of re-
inforcing bar, accelerates the crack formation in concrete, and even re-
sults in the spalling of concrete cover, which declines the residual
service life of concrete structures [10,11]. In some environmental re-
gions like atmospheric marine zone, both carbon dioxide and chloride
ion can be interacted and this coupled effect may lead to a faster mate-
rial degradation than if either one acts alone [12]. From the review of
previous research work, some studies have been conducted to
⁎ Corresponding author.
E-mail address: qwqiu3-c@my.cityu.edu.hk (Q. Qiu).
and concrete materials. For instance, Lee et al. [13] have reported that
a higher ratio of water-soluble (free) chloride to acid-soluble chloride
(total) content can be formed under the presence of carbonation.
Backus et al. [14] have indicated that carbonation causes the liberation
of bound chloride and a sudden increase of free chloride concentration
in concrete material. A recent study taken by Ye et al. [12] has also point-
ed out that carbonation can increase the chloride content at carbonation
fronts. Additionally, the rate of chloride diffusion in concrete material
can be significantly increased by carbonation, as demonstrated by
Ngala et al. [15], Chaoming et al. [16], and Ditao et al. [17] in their arti-
cles. On the other hand, it has been reported by Kuosa et al. [18] that
the presence of chloride ion lowers the carbonation depth. This research
finding is also supported by an experimental investigation concerning
the effect of chloride ion on carbonation of cement paste by using X-
ray computed tomography technique [19]. The researchers have found
that increasing the chloride ion in cement paste can refine the porous
structure and decrease the porosity. As a result, the resistance against
carbon dioxide penetration for cement paste is improved. Yoon et al.
[20] have stated that the risk of steel corrosion due to carbonation
(given that both the moisture and oxygen are present) is not as serious
as that caused by chloride attack, when concrete structures are subject-
ed to double environmental effects of carbonation and chloride ingress.
Besides the above experimental reports, the combined action of carbon-
ation and chloride ion penetration has been investigated by numerical
simulation work, which can be found in these literatures [21–24].

http://dx.doi.org/10.1016/j.cemconres.2017.02.032
0008-8846/© 2017 Elsevier Ltd. All rights reserved.
218 J. Liu et al. / Cement and Concrete Research 95 (2017) 217–225
Many studies on the subject of carbonation and chloride attack are
generally conducted by using carbon dioxide and chloride solution, con-
sidering the situation where the concrete structures are constructed
under seawater or in a splash zone. However, the relative humidity is
quite high in such marine regions, which does not allow the carbonation
to occur. Hence, the combined action of carbonation and chloride attack
is rarely existed in these coastal environments. Nevertheless, in the at-
mospheric marine zone, carbon dioxide and chloride aerosol do exist
concomitantly and the combined attack of carbonation and atmospheric
chloride on the offshore concrete structures can be occurred. In general,
chloride aerosol is generated along the seashore by breaking wave
movement and transported inland by wind [25]. The combination of
carbonation and chloride aerosol attack may cause a severe deteriora-
tion of concrete structures over their designed service life. However, lit-
tle research work has been done on concretes subjected to the
combined effect of carbonation and atmospheric chloride ingress. The
role of chloride aerosol in the presence of carbon dioxide and that of car-
bon dioxide in the presence of chloride aerosol on degradation of con-
crete material have not been well clarified. Understanding the coupled
physical-chemical interaction between carbonation and chloride aero-
sol attack in concrete is of great significance for prediction of service
life for reinforced concrete structures situated in atmospheric marine
environment.
The objective and originality of this research work is to investigate
the interacted mechanism between carbonation and chloride aerosol
attack in ordinary Portland cement concrete material. The ionic trans-
port characteristics of concrete under the combined action of carbon di-
oxide and chloride aerosol are explored by evaluating both the effect of
carbonation on the chloride profile, the chloride binding capacity and
the chloride diffusion coefficient, and the opposite, the effect of chloride
aerosol on the carbonation depth and the pH value. Microscopic proper-
ties of concrete under the coupled effect are also explored by using the
scanning electron microscope (SEM) and mercury intrusion
porosimetry (MIP) techniques.
2. Experimental work
2.1. Materials and specimen preparation
In this research work, type I ordinary Portland cement, made in
Shenzhen Haixing Onoda Cement Co. Ltd., was used as concrete compo-
nent. The chemical composition and the physical property of cement are
listed in Table 1. The contents of calcium silicate (C3S), dicalcium silicate
(C2S), tricalcium aluminate (C3A), and tetra-calcium aluminoferrite
(C4AF) in cement were 59.36%, 18.25%, 9.82%, and 10.18%, respectively.
Crush stone with the density of 2700 kg/m3 was used as coarse aggre-
gate, which was obtained from a quarry of AnTuo Mountain in
Shenzhen city, China. The maximum size and the minimum size for
the coarse aggregate were 20 mm and 5 mm, respectively. River sand
with the fineness modulus of 2.61 and the density of 2632 kg/m3 was
Table 1
Chemical composition and physical characteristics of ordinary Portland
cement.
Item Cement
Composition (mass % as oxide)
Calcium oxide (CaO) 64.67
Silica (SiO2) 18.59
Alumina (Al2O3) 4.62
Iron Oxide (Fe2O3) 4.17
Magnesium oxide (MgO) 2.35
Sulfur trioxide (SO3) 3.32
Potassium oxide (K2O) 0.92
Physical characteristics
Loss on ignition (LOI) 1.03
Specific surface area (m2/kg) 345
80 μm sieving fineness (%) 4.15
used as fine aggregate. Distilled water was used in the fabrication of
concrete mixtures. Three water-to-cement ratios of 0.38, 0.47 and 0.53
were considered. Table 2 shows the mix proportion of concrete
specimen.
All concrete mixtures were cast into 100 mm cube steel mold, and
then consolidated on a vibration table with the purpose to force out
the air bubbles. After demoulding, the mixtures were cured in a mois-
ture room at 20 °C and 95% relative humidity for 28 days. Prior to envi-
ronmental conditioning, only one surface of concrete was allowed to be
penetrated by carbon dioxide or chloride aerosol while the other sur-
faces were coated with epoxy resin. This was to realize one-dimensional
diffusion. The concrete specimens were separated into four groups
representing different exposure conditions, namely, group A (pure chlo-
ride aerosol attack), group B (combined carbonation and chloride aero-
sol attack), group C (pure carbonation) and group D (combined chloride
aerosol ingress and carbonation). More details of the experimental sce-
narios for these four groups of specimens are described in Table 3.
2.2. Accelerated carbonation test and measurement of pH value and car-
bonation depth
The carbonation test was conducted in the accelerated carbonation
chamber, following the Chinese National Standard GB/T 50082-2009
“Standard for test methods of long-term performance and durability
of ordinary concrete” [26]. The concrete samples were conditioned
with 20% CO2 concentration, 35 ± 2 °C temperature, and 70 ± 2% rela-
tive humidity. The tested period for carbonation was 28 days. In order to
evaluate the alkalinity variation of concrete under carbonation, the pH
value of carbonated concrete sample was measured. In this work, the
pH measurement was carried out within the 0–16 mm depth interval
of concrete. The method of pH measurement was based on mixing pow-
dered samples with distilled water as a solvent and measuring the pH of
the suspension, which was proposed and validated from a literature
[27]. The concrete specimen was powered by a grinding machine from
the surface to the interior, and the concrete powders were separately
collected with an interval of 1 mm for the first 10 mm away from ex-
posed surface and with an interval of 2 mm in the range from 10 mm
to 16 mm. Then, 1 g powder and 50 g distilled water were weighted,
and mixed together for 24 h prior to the pH value measurement. A de-
vice of pH meter was adopted to determine the pH value of carbonated
concrete layer at different depths. Additionally, a 1% phenolphthalein
solution was sprayed on the split cross section to determine the carbon-
ation depth, in accordance with the testing standard [26]. The carbon-
ation depth was captured and recorded by a digital caliper on a split
face, perpendicularly from the carbonation front to the end edge.
2.3. Chloride aerosol conditioning and measurement of chloride content
To simulate the chloride aerosol exposure in marine atmospheric
environment, concretes were conditioned in an environmental cham-
ber (from Wuxi Sunan Experimental Equipment Limited Company)
which was designed to spray the chloride-contained aerosol from the
top down. Considering the deposition of salt fog in the environmental
chamber, the concrete specimens were positioned with a manner that
the top surface was upward and exposed to atmospheric chloride. A
Table 2
Mix proportions of OPC concrete.
Mixture ID Water-to-cement ratio Composition (kg/m3)
Cement Sand Crush stone Water
C38 0.38 454 729 1094 173
C47 0.47 409 720 1079 192
C53 0.53 396 718 1076 210
Note: The mass ratio of the fine aggregate to total aggregate (fine aggregate + coarse ag-
gregate) used in all concrete specimens was kept as 0.4.
 J. Liu et al. / Cement and Concrete Research 95 (2017) 217–225
Table 3
Experimental scenarios for concretes in group A, group B, group C and group D.
Group Specimens ID Tested period 1 Tested period 2
28 days 28 days
Group A A-C38, A-C47, A-C53 Control condition1 Chloride aerosol attack
Group B B-C38, B-C47, B-C53 Carbonation Chloride aerosol attack
Group C C-C38, C-C47, C-C53 Control condition2 Carbonation
Group D D-C38, D-C47, D-C53 Chloride aerosol attack Carbonation
Notes: Control condition 1 refers to 0% CO2 concentration, 35 ± 2 °C temperature, and
70 ± 2% relative humidity in the carbonation chamber (without CO2); Control condition
2
refers to 0% chloride aerosol concentration, 35 ± 2 °C temperature, and 70 ± 2% relative
humidity in the environmental chamber (without chloride aerosol).
solution of 50 g NaCl per liter of distilled water was supplied for gener-
ating the chloride aerosol environment. The NaCl solution was atomized
by using spray nozzles with pressurized air. In the environmental cham-
ber, the temperature and the relative humidity were controlled at 35 ±
2 °C and 70 ± 2%, respectively. The salt spraying deposition was con-
trolled as 1–2 mL/80 cm2·h. In accordance with the historical climate
data of the coastal zone of Shenzhen city, the ratio of rainy days to
sunny days is 2:3. In order to realistically simulate the exposure condi-
tion, the cyclic drying-wetting condition was set in a testing day (i.e.
10 h for chloride aerosol spraying and 14 h for drying condition).
After the chloride aerosol conditioning, the concrete samples were
powdered by a grinding machine for chloride ion content measurement.
The concrete powders were separately collected with an interval of
1 mm for the first 10 mm away from exposed surface then with an inter-
val of 2 mm in the range from 10 mm to 30 mm, as shown in Fig. 1. The
powdered concrete particles were filtered through a 0.63 mm sieve and
dried for 2 h. In this study, the free chloride content and the total chlo-
ride content were both measured, according to the AASHTO T260 stan-
dard [28]. The free chloride content was determined on the basis of
water-soluble extraction method, while the total chloride content was
measured by acid-soluble extraction in a nitric acid solution. After the
above chloride extraction, an automatic potentiometer titrator (version:
809 Titrando) was adopted to measure the chloride content. During po-
tentiometric titration, the silver nitrate (AgNO3) solution with a concen-
tration of 0.01 mol/L was used. The consumed volume of AgNO3 solution
Fig. 1. Concrete sampling for chloride content measurement: (a) only one surface of concrete was exposed to chloride aerosol, while the other surfaces were coated with epoxy resin.
30 mm thick concrete layer from exposed surface was cut and further powdered with grinding machine. The concrete powders were separately collected with an interval of 1 mm for
the first 10 mm away from exposed surface then with an interval of 2 mm in the range from 10 to 30 mm; (b) grinding machine used in the research; (c) the powdered concrete
particles were filtered through a 0.63 mm sieve and dried for 2 h.
219
was recorded and used to calculate the concentration (by mass of bind-
er) of chloride ion. By plotting the chloride ion content against the cor-
responding depth from where the test sample was collected, the
chloride profile of concrete was acquired.
2.4. Microscopy test
MIP has been commonly used to obtain the total porosity and pore
size distribution (PSD) of cement and concrete materials, which in-
volves in measuring the volume of mercury forced into a porous mate-
rial at given pressure levels [29–33]. In this research, an additional
investigation was taken by using a MIP machine (Micromeritics
AutoPore IV 9500) to further understand the effect of carbonation and
chloride aerosol attack on the microscopic properties of concrete. The
concrete sample with 10–15 mm diameter at the near surface region
(within 0–5 mm depth) was taken in the MIP measurement. A geomet-
rical model (known as Washburn equation) was applied for determin-
ing the pore diameter [34], as expressed in Eq. (1).
−4γ cosθ
D¼ ð1Þ
P
In Eq. (1), D represents the pore diameter (m), γ represents the sur-
face tension (N/m), θ represents the contact angle between mercury
and pore wall (°), and P represents the applied pressure (MPa). Based
on the general suggestions from the literatures [35,36], the γ was as-
sumed as 480 mN/m, and the θ was assumed as 130°.
SEM technique was used in this research to explore the morphology
of concretes subjected to the combined ingress of carbon dioxide and
chloride aerosol. Before SEM examination, the concrete specimen was
fractured and the small concrete fragment (with approximately 4 mm
thick × 7 mm long × 5 mm wide) was collected and placed in a vacuum
environment under 50 °C until constant weight. In order to obtain the
clear photo-micrograph, the top surface of the concrete fragment was
rendered conductive with a thin gold layer. By analyzing the SEM and
MIP results, a comprehensive understanding of microstructural proper-
ties of OPC concrete under the combined effect of carbonation and chlo-
ride aerosol attack can be obtained.
220 J. Liu et al. / Cement and Concrete Research 95 (2017) 217–225

3. Results and discussion

3.1. Chloride profiles of concretes against carbonation and chloride aerosol

ingress

The free chloride profiles of concretes against the single effect of

chloride aerosol ingress and the coupled effect of carbonation and chlo-
ride aerosol ingress are displayed in Fig. 2. By comparing the results of
tested samples between group A (exposed to pure chloride aerosol)
and group B (exposed to carbon dioxide and chloride aerosol), the effect
of carbonation on the behavior of chloride aerosol attack can be evaluat-
ed. Under the pure chloride aerosol penetration, the free chloride con-
tent shows a sharp increase in the near surface layer of concrete. The
chloride content increases up to its peak value at about 3 mm depth
from the exposed surface. In the interior region of concrete, the free
chloride content is shown to decrease. The presence of increasing
trend for chloride content in the first 3 mm depth is mainly due to the
capillary suction during the cyclic drying-wetting action. In addition to
the capillary suction, the wash-out effect on the surface and the evapo-
ration of water can result in a local maximal value of chloride concentra-
tion near the exposed surface. During the event of wetting condition,
the exposed surface region can have its chloride concentration reduced
due to the wash-out effect. In addition, during the drying process the
evaporation of water contributes to the accumulation of chloride ions
near the exposed surface. Beyond the convection zone, the effect of cap-
illary suction is less significant and thus the Fick's law of diffusion due to
a concentration gradient of chloride ion is followed [12]. It is worth not-
ing that water-to-cement ratio plays an important role in the concentra-
tion of free chloride ion, especially in deeper region of concrete. The
chloride ion content is decreased with lower water-to-cement ratio.
For concrete sample with the lowest water-to-cement ratio (i.e. 0.38),
the chloride ion content is reduced to nearly zero when the diffusion
depth reaches 18 mm. This can be explained by the fact that the
water-to-cement ratio influences the formation of pore structure and
subsequently affects the resistance of concrete against chloride
transport.
The presence of carbon dioxide has a remarkable impact on the
transport behavior of chloride aerosol in concrete. Fig. 2 also presents
the chloride profile of group B concrete samples subjected to both car-
bon dioxide and chloride aerosol. It is interesting to observe that the
free chloride content reduces at concrete surface layer (within 5 mm
from surface), instead of showing the same pattern as those of group
A specimens. Due to carbonation reaction, the capillary pores of con-
crete are partially filled and blocked by the deposition of calcium car-
bonate (CaCO3) [37–39]. As a result, the intensity of capillary suction
is decreased due to a reduction in surface capillary porosity and loss of
connectivity in the pore space [40–42]. Thus, the free chloride content
for carbonated concrete samples is found to be lower at the exposed
surface region. The microstructural modification is supported by SEM
micrographs as revealed in Fig. 3. Morphology of concrete specimen
A-C47 shows a more fractured structure and several crack spaces, as in-
dicated in Fig. 3(a). In contrast, a more compact structure for the speci-
men B-C47 is observed, as shown in Fig. 3(b). In addition to SEM
micrographs, the MIP tested results claim that the presence of carbon-
ation produces a displacement of pore size distribution curve towards
the smaller capillary pore zone, as shown in Fig. 4(a). Besides, Fig.
4(b) implies that concretes under combined ingress of carbon dioxide
and chloride aerosol generally have less porosity than the concretes ex-
posed to only chloride aerosol attack. Due to the change in physical pore
structure by the presence of carbonation, the penetration of chloride
aerosol in concrete surface layer is reduced.
Fig. 2. Chloride profile of OPC concretes under exposure to chloride aerosol with/without
Despite the lower free chloride content in the first 5 mm concrete carbonation. Under pure chloride aerosol attack, the chloride content increases up to its
layer, it is clearly found that the length of convection zone is increased peak value at about 3 mm depth from the exposed surface. The presence of carbonation
due to the presence of carbonation. In the interior region (5–10 mm can remarkably modify the chloride profile in concrete. The length of convection zone is
deep) the free chloride content of group B specimens appears to in- increased due to the presence of carbonation. Beyond 10 mm depth, the free chloride
content of group B concrete specimen is larger than that of group A specimen.
crease, and it is noticed that the free chloride content is larger than
 J. Liu et al. / Cement and Concrete Research 95 (2017) 217–225
Fig. 3. SEM micrographs of concrete (water-to-cement ratio = 0.47) against: (a) single
effect of chloride aerosol attack; (b) coupled effect of carbonation and chloride aerosol
attack. Specimen A-C47 shows a more fractured structure and several crack spaces,
while specimen B-C47 shows a more compact pore structure.
that of group A specimens beyond 10 mm depth. This phenomenon can
be related to the reduced chloride binding capacity in concrete which is
previously exposed to carbon dioxide. In the chloride-contaminated
concrete, the stability of Friedel's salt (3Cao∙ Al2O3·CaCl2 ∙10H2O) and
Kuzel's salt (3Cao∙ Al2O3 ∙ 0.5CaSO4 ∙ 0.5CaCl2 ∙ 10-11H2O) is affected by
concrete alkalinity, as confirmed by many studies [12,43,44]. Under
the carbonation reaction, the pH value in concrete pore solution is re-
duced which in turn increases the solubility of Friedel's salt and Kuzel's
salt. This can promote more quantity of free chloride ions inwards into
the specimen. From the engineering point-of-view, the combined effect
of carbonation and chloride aerosol poses a more critical risk of steel
corrosion in reinforced concrete structure because a higher free chloride
concentration can be formed in the interior region of concrete material.
3.2. Chloride diffusion coefficient of OPC concretes under carbonation and
chloride aerosol attack
The chloride diffusion coefficient is one of the most commonly used
parameters for characterizing the rate of chloride penetration into con-
crete [45]. In this work, the chloride diffusion coefficient is determined
by the following Eq. (2), the so-called “error function solution” for Fick's
second law of one-dimensional diffusion [46,47].
  
x ratio of concrete. For OPC concrete with low water-to-cement ratio,
C x ¼ C s 1−erf pffiffiffiffiffiffi ð2Þ
2 Dt
221
Fig. 4. MIP tested results for concrete subjected to pure chloride aerosol attack and coupled
effect of carbonation and chloride aerosol attack: (a) pore size distribution of concrete
(water-to-cement ratio = 0.47); (b) porosity of concretes in group A and group B. The
presence of carbonation initiates a movement of pore size distribution curve towards
the smaller capillary pore zone and a reduction in the proportion of large capillary
pores. In addition, concretes under combined ingress of carbon dioxide and chloride
aerosol generally have less porosity than the concretes exposed to individual chloride
aerosol attack.
In Eq. (2), D is the diffusion coefficient (mm2/s), Cx is the chloride
concentration at given depth x of concrete (%), Cs is the chloride concen-
tration at the boundary (%), t is the diffusion time (s), which is 28 days in
this study, erf() is the error function. Based on the measurement data of
chloride profile presented in Fig. 2, the chloride diffusion coefficient can
be calculated. Since the chloride ingress is very complicated at the con-
vection region of concrete owing to the capillary suction and the evap-
oration action, in this work, the chloride diffusion coefficient in the
interior region (in the 5–30 mm depth interval) is calculated. Chloride
ion diffusion coefficients for OPC concretes with different water-to-ce-
ment ratios under the chloride aerosol attack with/without exposure
to carbon dioxide are shown in Fig. 5. It is found that the presence of car-
bonation results in an increase of chloride ion diffusion coefficient in
concretes when compared to only chloride aerosol ingress. Due to the
decreased pH value in the pore solution after carbonation, the chloride
binding capacity is reduced which facilitates the diffusion of chloride
ion through the concrete material. It is also observed that the increasing
rate of chloride ion diffusion is depended upon the water-to-cement
the carbonation-induced increase of chloride ion diffusion coefficient
222 J. Liu et al. / Cement and Concrete Research 95 (2017) 217–225

Fig. 5. The chloride ion diffusion coefficients of concretes under chloride aerosol attack and
coupled effect of carbonation and chloride aerosol attack. It is found that the presence of
carbonation results in an increase of chloride ion diffusion coefficient in concretes when
compared to chloride aerosol ingress alone.

is little. However, the increase of the diffusion coefficient becomes

significant when the water-to-cement ratio is up to 0.47 and more.
The chloride ion diffusion coefficient for B\\C53 is almost 4 times
as large as that of A-C53. Larger water-to-cement ratio can result in
a less resistance to penetration of carbon dioxide, as demonstrated
by a study [1]. In this regard, reinforced concretes with higher ce-
ment-to-water ratio are more vulnerable to the combined attack of
carbonation and chloride aerosol, due to the greater chloride ion dif-
fusion coefficient.

3.3. The chloride binding capacity of OPC concrete under carbonation and
chloride aerosol attack

Despite that steel corrosion is directly resulted from free chloride ion Fig. 6. The relationship between the total chloride content and the free chloride content:
instead of bound chloride ion, the study of bound chloride ion is still es- (a) chloride aerosol attack without carbonation and (b) coupled effect of carbonation
sential because the originally bound chloride ion can be converted to and chloride aerosol attack. A linear trend between the total chloride content and the
the free chloride ion under certain conditions such as carbonation. In free chloride content can be clearly observed.

view of this issue, the binding capacity of chloride ion in concrete sub-
jected to carbonation and chloride aerosol attack is also studied in this an excellent fitting to the data points (with R2 value over the range
work. 0.9958 to 0.9987). By inserting Eq. (4) into Eq. (3), we have Eq. (5):
In concrete material, the total chloride content is regarded as the
summation of the free chloride content and the bound chloride content, C b ¼ ðα−1ÞC f þ β ð5Þ
as described in the following Eq. (3):

Ct ¼ C f þ Cb ð3Þ Based on the literature [54], the chloride binding capacity R in con-
crete is defined by the following expression:
where Ct represents the total concentration of chloride ion in con-
crete, Cf represents the quantity of free chloride ion, and Cb repre- ∂C b
sents the quantity of bound chloride ion. Up to present, some R¼ ð6Þ
∂C f
chloride binding isotherms (e.g. linear, Langmuir, Freundlich and
BET binding isotherm) have been adopted to characterize the rela-
tionship between the total chloride content and the free chloride
content [48–53]. Fig. 6 shows the relationship between the total
chloride content and the free chloride content for OPC concretes Table 4
The coefficient α and the corresponding binding capacity R for concretes under chloride
under the single chloride aerosol penetration and the coupled effect
aerosol attack with and without carbonation.
of carbonation and chloride aerosol penetration. From Fig. 6, we are
noticed that the total chloride content is linearly correlated with Water-to-cement ratio Chloride aerosol Chloride aerosol
attack without attack with
the amount of free chloride ion, as indicated by Eq. (4):
carbonation carbonation

C t ¼ αC f þ β ð4Þ α R α R

0.38 1.1139 0.1139 1.0195 0.0195

where the α and β are constants. By fitting the measurement data 0.47 1.1317 0.1317 1.0392 0.0392
0.53 1.1707 0.1707 1.0535 0.0535
through Eq. (4), it can be clearly found that the linear function performs
 J. Liu et al. / Cement and Concrete Research 95 (2017) 217–225 223

In view of Eq. (5), the binding capacity R can be given by

R ¼ α–1 ð7Þ

The coefficient α and the corresponding binding capacity R are ob-

tained and the results are listed in Table 4. From Table 4, we are noticed
that the binding capacity R increases with the increasing water-to-ce-
ment ratio. It is known that higher water-to-cement ratio can increase
the degree of cement hydration, resulting in more production of C-S-H
[55,56]. This can enhance the physical binding of chloride in OPC con-
crete [50,53,57]. The finding in our research is consistent with other
studies [48,58,59].
Results in Table 4 also indicate that chloride binding capacity under
combined effect of carbonation and chloride aerosol attack is signifi-
cantly lower than that under pure chloride aerosol attack. The binding
capacity R for concrete with water-to-cement ratio of 0.47 subjected
to chloride aerosol attack without carbonation is 0.1317, while the bind-
ing capacity becomes only 0.0392 under the combined action of carbon-
ation and chloride aerosol attack. This result demonstrates that the
chloride binding capacity R is lowered by carbonation reaction, resulting
in increasing the chloride ion concentration around the embedded steel
bar in concrete structures. In a concrete material, the hydration prod-
ucts can react with chloride ions with the production of Friedel's salt,
which is a beneficial aspect for retarding the chloride penetration. How-
ever, carbonation can consume the cement hydration products, which
weakens the chlorides binding. In addition, it has been mentioned that
carbonation is able to dissolve the Friedel's salts and release the chloride
ions. Due to these chemical mechanisms, the chloride binding capacity
is seriously reduced by the coupled effect of carbonation and chloride
aerosol attack.

3.4. 3.4 effect of chloride aerosol attack on carbonation depth and pH value
of concrete

The carbonation depths of OPC concretes with and without exposure

to chloride aerosol are respectively listed in Table 5. It is found that a
lower water-to-cement ratio can result in lower carbonation depth of
concrete either exposed to pure carbonation or exposed to the com-
bined effect of chloride aerosol attack and carbonation. More interest-
ingly, it is observed that concretes under the coupled effect of chloride
aerosol attack and carbonation have lower carbonation depth compared
to concretes subjected to carbonation only. Fig. 7 shows the pH value of
concrete with different depths under the effect of chloride aerosol at-
tack and carbonation. The pH value for carbonated concrete without
chloride aerosol attack is significantly lower than that of concrete sub-
jected to the combined effect involving chloride aerosol attack and car-
bonation. These findings indicate that the carbonation process can be
delayed by the effect of chloride aerosol ingress. A reason for this phe-
nomenon is possibly related to higher moisture content and water
film thickness inside the pores due to chloride hygroscopy, which brings
about the decrease in carbonation rate [18]. Additionally, the large pores
of the carbonated concrete can be refined by the presence of chloride
aerosol ingress (as shown in Fig. 8), which contributes to retard the car-
bonation process. Fig. 9(a) illustrates the pore size distribution curve of
concrete specimen (water-to-cement ratio = 0.47) subjected to pure
Fig. 7. The pH profiles of concretes subjected to two environmental conditions: (a) single
effect of carbonation and (b) coupled effect of chloride aerosol attack and carbonation. An
increasing water-to-cement ratio can result in lower pH value of concrete. Besides, the
Table 5
presence of chloride aerosol attack can lead to higher pH value for carbonated concrete.
Carbonation depth (unit: mm) of OPC concretes with and without chloride aerosol ingress.

Water-to-cement ratio

Tested group 0.38 0.47 0.53

Group C (without chloride aerosol 0.4 1.1 4.2 carbonation and coupled effect of chloride aerosol ingress and carbon-
ingress) (σ = 0.05) (σ = 0.09) (σ = 0.12) ation. It is observed that the combined effect decreases the proportion
Group D (with chloride aerosol 0.3 1.0 3.8 of medium capillary pore (diameters between 10 nm to 50 nm) as
ingress) (σ = 0.08) (σ = 0.09) (σ = 0.17) well as the large capillary pores (diameters between 50 nm to
Note: σ refers to standard deviation of measured carbonation depth. 1000 nm), compared with the individual action of carbonation. Porosity
224 J. Liu et al. / Cement and Concrete Research 95 (2017) 217–225
Fig. 8. SEM micrographs of concrete (water-to-cement ratio = 0.47) against: (a) single
effect of carbonation; (b) coupled effect of chloride aerosol attack and carbonation. The
SEM image for pure carbonation case shows some obvious large pores while the
micrograph for the coupled effect reveals more uniform and denser pore structure.
of concrete under the coupled effect is also shown to be reduced, as il-
lustrated in Fig. 9(b). The densification of pore structure caused by the
additional ingress of chloride aerosol results in an enhanced resistance
of concrete against penetration of carbon dioxide, which leads to a
higher pH value and less carbonation depth in concrete. Nevertheless,
it is reminded that this combined effect should not be regarded as a ben-
efit to steel-reinforced concrete structures owing to the ingress of chlo-
ride ion. The net effect of coupled chloride aerosol ingress plus
carbonation is still highly detrimental when it refers to steel reinforce-
ment corrosion.
4. 4. Conclusion
In this research, the interaction between carbonation and chloride
aerosol attack on OPC concrete is experimentally investigated. Based
on the findings in this work, the following conclusions can be drawn.
The presence of carbonation has a profound influence on the penetra-
tion of chloride aerosol through concrete. A build-up of chloride concen-
tration can be initiated by the pure chloride aerosol ingress and the
coupled effect of carbonation and chloride aerosol ingress. But the
build-up of chloride content in the latter is occurred in a more interior
region of concrete. Besides, the carbonation can reduce the chloride
binding capacity and also accelerate the rate of chloride ion diffusion
in concrete. On the other hand, the chloride aerosol penetration can re-
sult in lower carbonation rate when compared to the case without chlo-
ride aerosol exposure. Lower carbonation depth and higher pH value are
Fig. 9. MIP tested results for concretes subjected to single effect of carbonation and
coupled effect of chloride aerosol attack and carbonation: (a) pore size distribution of
concrete (water-to-cement ratio = 0.47); (b) porosity of concretes in group C and
group D. The presence of chloride aerosol decreases the proportion of medium capillary
pore as well as the large capillary pores of carbonated concrete. Porosity of concrete
under the coupled effect is also shown to be reduced, compared to the case of only
carbonation.
observed under the interacted effect of chloride aerosol exposure and
carbonation. An increase of water-to-cement ratio can lead to the in-
crease in free chloride content and chloride ion diffusion coefficient.
Coupled effect of carbonation and chloride aerosol attack can densify
the microstructure of OPC concrete when compared to the single effect
of carbonation or chloride aerosol attack.
Acknowledgments
The financial supports from Natural Science Foundation of China
(No. 51578340) and Shenzhen City Science and Technology Project
(No. JCYJ20140418182819141, JCYJ20160422102136542) are greatly
acknowledged.
References
[1] B. Dong, Q. Qiu, J. Xiang, C. Huang, F. Xing, N. Han, Y. Lu, Electrochemical impedance
measurement and modeling analysis of the carbonation behavior for cementititous
materials, Constr. Build. Mater. 54 (2014) 558–565.
[2] P.F. Marques, C. Chastre, Â. Nunes, Carbonation service life modelling of RC struc-
tures for concrete with Portland and blended cements, Cem. Concr. Compos. 37
(2013) 171–184.
[3] I. Yoon, O. Çopuroğlu, K. Park, Effect of global climatic change on carbonation prog-
ress of concrete, Atmos. Environ. 41 (2007) 7274–7285.
 J. Liu et al. / Cement and Concrete Research 95 (2017) 217–225
[4] B. Dong, Q. Qiu, J. Xiang, C. Huang, H. Sun, F. Xing, W. Liu, Electrochemical imped-
ance interpretation of the carbonation behavior for fly ash–slag–cement materials,
Constr. Build. Mater. 93 (2015) 933–942.
[5] B. Dong, Q. Qiu, Z. Gu, J. Xiang, C. Huang, Y. Fang, F. Xing, W. Liu, Characterization of
carbonation behavior of fly ash blended cement materials by the electrochemical
impedance spectroscopy method, Cem. Concr. Compos. 65 (2016) 118–127.
[6] A. Nasser, A. Clément, S. Laurens, A. Castel, Influence of steel–concrete interface con-
dition on galvanic corrosion currents in carbonated concrete, Corros. Sci. 52 (2010)
2878–2890.
[7] B. Dong, Q. Qiu, J. Xiang, C. Huang, F. Xing, N. Han, Study on the carbonation behavior
of cement mortar by electrochemical impedance spectroscopy, Materials 7 (2014)
218–231.
[8] S.R. Tripathi, H. Ogura, H. Kawagoe, H. Inoue, T. Hasegawa, K. Takeya, K. Kawase,
Measurement of chloride ion concentration in concrete structures using terahertz
time domain spectroscopy (THz-TDS), Corros. Sci. 62 (2012) 5–10.
[9] M.F. Montemor, A.M.P. Simões, M.G.S. Ferreira, Chloride-induced corrosion on rein-
forcing steel: from the fundamentals to the monitoring techniques, Cem. Concr.
Compos. 25 (2003) 491–502.
[10] S. Han, Influence of diffusion coefficient on chloride ion penetration of concrete
structure, Constr. Build. Mater. 21 (2007) 370–378.
[11] C.L. Page, K.W.J. Treadaway, Aspects of the steel electrochemistry of steel in con-
crete, Nature 297 (1982) 109–115.
[12] H. Ye, X. Jin, C. Fu, N. Jin, Y. Xu, T. Huang, Chloride penetration in concrete exposed to
cyclic drying-wetting and carbonation, Constr. Build. Mater. 112 (2016) 457–463.
[13] M.K. Lee, S.H. Jung, B.H. Oh, Effects of carbonation on chloride penetration in con-
crete, ACI Mater. J. 110 (2013) 559–566.
[14] J. Backus, D. McPolin, M. Basheer, A. Long, N. Holmes, Exposure of mortars to cyclic
chloride ingress and carbonation, Adv. Cem. Res. 25 (2013) 3–11.
[15] V.T. Ngala, C.L. Page, Effects of carbonation on pore structure and diffusional proper-
ties of hydrated cement pastes, Cem. Concr. Res. 27 (1997) 995–1007.
[16] P. Chaoming, G. Meirong, X. Jian, W. Lun, L. Guanguo, Q. Honggen, Influence of test
methods and carbonation on chloride ion diffusion coefficient of concrete, J. South-
east Univ. (Nat. Sci. Ed.) 41 (2011) 1313–1318.
[17] N. Ditao, S. Congtao, Study on interaction of concrete carbonation and chloride cor-
rosion, J. Chin. Ceram. Soc. 41 (2013) 1094–1099.
[18] H. Kuosa, R.M. Ferreira, E. Holt, M. Leivo, E. Vesikari, Effect of coupled deterioration
by freeze–thaw, carbonation and chlorides on concrete service life, Cem. Concr.
Compos. 47 (2014) 32–40.
[19] J. Liu, M. Ba, Y. Du, Z. He, J. Chen, Effects of chloride ions on carbonation rate of hard-
ened cement paste by X-ray CT techniques, Constr. Build. Mater. 122 (2016)
619–627.
[20] I.S. Yoon, Deterioration of concrete due to combined reaction of carbonation and
chloride penetration: experimental study, Key Eng. Mater. 348-349 (2007)
729–732.
[21] D. Conciatori, F. Laferrière, E. Brühwiler, Comprehensive modeling of chloride ion
and water ingress into concrete considering thermal and carbonation state for real
climate, Cem. Concr. Res. 40 (2010) 109–118.
[22] I.-S. Yoon, Simple approach to calculate chloride diffusivity of concrete considering
carbonation, Comput. Concr. 6 (2009) 1–18.
[23] A. Delnavaz, A.A. Ramezanianpour, The assessment of carbonation effect on chloride
diffusion in concrete based on artificial neural network model, Mag. Concr. Res. 64
(2012) 877–884.
[24] X. Zhu, G. Zi, W. Lee, S. Kim, J. Kong, Probabilistic analysis of reinforcement corrosion
due to the combined action of carbonation and chloride ingress in concrete, Constr.
Build. Mater. 124 (2016) 667–680.
[25] M.E.R. Gustafsson, L.G. Franzén, Dry deposition and concentration of marine aerosols
in a coastal area, Atmos. Environ. 30 (1996) 977–989.
[26] GB/T 50082-2009, Standard for Test Methods of Long-Term Performance and Dura-
bility of Ordinary Concrete, 2009.
[27] V. Räsänen, V. Penttala, The pH measurement of concrete and smoothing mortar
using a concrete powder suspension, Cem. Concr. Res. 34 (2004) 813–820.
[28] AASHTO, T260, Standard Method of Test for Sampling and Testing for Chloride Ion in
Concrete and Concrete Raw Materials, American Association of State Highway and
Transportation Officials, Washington, United state, 2009.
[29] S. Diamond, A critical comparison of mercury porosimetry and capillary condensa-
tion pore size distributions of Portland cement pastes, Cem. Concr. Res. 1 (1971)
531–545.
[30] M. Wyrzykowski, R. Kiesewetter, J. Kaufmann, R. Baumann, P. Lura, Pore structure of
mortars with cellulose ether additions – mercury intrusion porosimetry study, Cem.
Concr. Compos. 53 (2014) 25–34.
225
[31] J. Liu, F. Xing, B. Dong, H. Ma, D. Pan, Study on surface permeability of concrete
under immersion, Materials 7 (2014) 876–886.
[32] J. Liu, F. Xing, B. Dong, H. Ma, D. Pan, Study on water sorptivity of the surface layer of
concrete, Mater. Struct. 47 (2013) 1941–1951.
[33] J. Liu, K. Tang, Q. Qiu, D. Pan, Z. Lei, F. Xing, Experimental investigation on pore struc-
ture characterization of concrete exposed to water and chlorides, Materials 7 (2014)
6646–6659.
[34] K.K. Aligizaki, Pore Structure of Cement-Based Materials: Testing, Interpretation and
Requirements, CRC Pres, Boca Raton, United States, 2005.
[35] D.K. Panesar, J. Francis, Influence of limestone and slag on the pore structure of ce-
ment paste based on mercury intrusion porosimetry and water vapour sorption
measurements, Constr. Build. Mater. 52 (2014) 52–58.
[36] H. Ma, Mercury intrusion porosimetry in concrete technology: tips in measurement,
pore structure parameter acquisition and application, J. Porous. Mater. 21 (2013)
207–215.
[37] H.-W. Song, S.-J. Kwon, Permeability characteristics of carbonated concrete consid-
ering capillary pore structure, Cem. Concr. Res. 37 (2007) 909–915.
[38] B. Johannesson, P. Utgenannt, Microstructural changes caused by carbonation of ce-
ment mortar, Cem. Concr. Res. 31 (2001) 925–931.
[39] C.A. García-González, A. Hidalgo, C. Andrade, M.C. Alonso, J. Fraile, A.M. López-
Periago, C. Domingo, Modification of composition and microstructure of Portland
cement pastes as a result of natural and supercritical carbonation procedures, Ind.
Eng. Chem. Res. 45 (2006) 4985–4992.
[40] W.P.S. Dias, Reduction of concrete sorptivity with age through carbonation, Cem.
Concr. Res. 30 (2000) 1255–1261.
[41] V. Rostami, Y. Shao, A.J. Boyd, Durability of concrete pipes subjected to combined
steam and carbonation curing, Constr. Build. Mater. 25 (2011) 3345–3355.
[42] J.J. Chang, W. Yeih, R. Huang, C.T. Chen, Suitability of several current used concrete
durability indices on evaluating the corrosion hazard for carbonated concrete,
Mater. Chem. Phys. 84 (2004) 71–78.
[43] J.G. Jang, H.J. Kim, H.K. Kim, H.K. Lee, Resistance of coal bottom ash mortar against
the coupled deterioration of carbonation and chloride penetration, Mater. Des. 93
(2016) 160–167.
[44] A.K. Suryavanshi, R.N. Swamy, Stability of Friedel's salt in carbonated concrete, Cem.
Concr. Res. 26 (1996) 729–741.
[45] R. van Noort, M. Hunger, P. Spiesz, Long-term chloride migration coefficient in slag
cement-based concrete and resistivity as an alternative test method, Constr. Build.
Mater. 115 (2016) 746–759.
[46] M. Funahashi, Predicting corrosion free service life of a concrete structure in a chlo-
ride, ACI Mater. J. 87 (1998) 581–587.
[47] J. Liu, K. Tang, D. Pan, Z. Lei, W. Wang, F. Xing, Surface chloride concentration of con-
crete under shallow immersion conditions, Materials 7 (2014) 6620–6631.
[48] C. Arya, N.R. Buenfeld, J.B. Newman, Factors influencing chloride-binding in con-
crete, Cem. Concr. Res. 20 (1990) 291–300.
[49] B. Martı́n-Pérez, H. Zibara, R.D. Hooton, M.D.A. Thomas, A study of the effect of chlo-
ride binding on service life predictions, Cem. Concr. Res. 30 (2000) 1215–1223.
[50] Q. Yuan, C. Shi, G.D. Schutter, K. Audenaert, D. Deng, Chloride binding of cement-
based materials subjected to external chloride environment-a review, Constr.
Build. Mater. 23 (2009) 1–13.
[51] L. Tang, Chloride transport in concrete-measurement and prediction(PhD thesis)
Department of Building Materials, Chalmers University of Technology, Goteborg,
Sweden, 1996.
[52] H. Zibara, Binding of External Chlorides by Cement Pastes(PhD-Thesis) University of
Toronto, Canada, 2001.
[53] L. Tang, L.-O. Nilsson, Chloride binding capacity and binding isotherms of OPC pastes
and mortars, Cem. Concr. Res. 23 (1993) 247–253.
[54] L. Tang, L. Nilsson, P. Basheer, Resistance of Concrete to Chloride Ingress: Testing and
Modelling, Spon Press, Abingdon, United Kingdom, 2012.
[55] R.A. Cook, K.C. Hover, Mercury porosimetry of hardened cement pastes, Cem. Concr.
Res. 29 (1999) 933–943.
[56] D.P. Bentz, Three-dimensional computer simulation of Portland cement hydration
and microstructure development, J. Am. Ceram. Soc. 80 (1997) 3–21.
[57] H. Hirao, K. Yamada, H. Takahashi, H. Zibara, Chloride binding of cement estimated
by binding isotherms of hydrates, J. Adv. Concr. Technol. 3 (2005) 77–84.
[58] P.S. Mangat, B.T. Molloy, Chloride binding in concrete containing PFA, GBS or silica
fume under sea-water exposure, Mag. Concr. Res. 47 (1995) 129–141.
[59] T. Cheewaket, C. Jaturapitakkul, W. Chalee, Long term performance of chloride bind-
ing capacity in fly ash concrete, Constr. Build. Mater. 24 (2010) 1352–1357.