Professional Documents
Culture Documents
http://dx.doi.org/10.1680/jmacr.15.00234
Volume 68 Issue 7
Paper 1500234
Effect of carbonation on release of Received 06/06/2015; revised 17/07/2015; accepted 28/07/2015
Published online ahead of print 23/09/2015
bound chlorides in chloride-contaminated
ICE Publishing: All rights reserved
concrete
Geng, Easterbrook, Liu and Li
In this paper an experimental study is presented on the interaction between concrete carbonation and chloride
attack. The experiments are carried out on chloride-contaminated cement pastes that have been exposed to a carbon
dioxide environment. Free and total chloride contents at different carbonation times are measured in the tested
specimens using traditional leaching and acid-soluble methods. Bound chloride contents are calculated from the
measured total and free chlorides. The experimental data show that the carbonation of cement paste results in a
release of bound chlorides, which is related to not only the decomposition of Friedel’s salt, but also the
decomposition of calcium-silicate-hydrate (C-S-H) gel. Based on the obtained experimental results, a numerical model
is also developed for simulating the carbonation process of the chloride-contaminated cement paste. The model can
be used to explain how the bound chlorides are released in a chloride-contaminated cement paste under the
influence of carbonation reactions.
353
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright © ICE Publishing, all rights reserved.
Magazine of Concrete Research Effect of carbonation on release of
Volume 68 Issue 7 bound chlorides in chloride-contaminated
concrete
Geng, Easterbrook, Liu and Li
of chloride ions in concrete as it can alter the pore volume and carbonation for almost the same period, showed that the solu-
pore structure and thus the transport properties of the con- bility of Friedel’s salt increases with the degree of concrete car-
crete. In certain cases the permeability of the carbonated con- bonation and the stability of both Friedel’s salt and ettringite
crete may increase, as in the case of concrete made with is pH dependent. Experimental study on the stability of
blended cements such as blast furnace slag and fly ash con- Friedel’s salt in calcium aluminate cement pastes containing
crete; in others, it may decrease as in ordinary Portland sodium chloride (NaCl) with respect to carbonation was con-
cement concrete (Dias, 2000; Saillio et al., 2014). In urban and ducted by Goñi and Guerrero (2003), in which the carbonation
industrial areas, where environmental pollution results in a sig- was carried out on a powdered sample in flowing 5% carbon
nificant concentration of carbon dioxide, carbonation-initiated dioxide gas at 65% relative humidity (RH) to accelerate the
reinforcement corrosion prevails (Yoon et al., 2007). However, process. It was found that the carbonation of Friedel’s salt
in highways and marine or coastal structures, chloride ions, did not produce a significant increase of soluble chloride,
originating from deicing salts or seawater, are the primary suggesting that the chlorides released by carbonation of
cause of reinforcing steel corrosion (Thomas and Jones, 1996). Friedel’s salt might be trapped by the solid. Similar exper-
The chlorides that are transported through the concrete pore iments in which the specimens were carbonated first and
network and micro-cracks depassivate the oxide film covering then followed by the chloride attack were conducted by
the reinforcing steel and accelerate the reaction of corrosion Hassan (2001) to investigate the chloride binding capacity of
(Ryou and Ann, 2008; Yu et al., 1993). The worst situation is carbonated concrete. In his experiments three different pre-
when a concrete structure is subjected to both carbonation and carbonated pastes were immersed in chloride solutions with
chloride attacks (Backus et al., 2013). Surveys on concrete different concentrations. He found that the pre-carbonated
structures have shown that the presence of moderate levels of pastes had almost no binding capacity.
chloride contamination (0·4–1·0% by mass of cement) can
have a significant influence on the rates of corrosion that may The above survey of the literature shows that, despite the con-
be induced in carbonated concrete (Aramaki et al., 2006). This siderable amount of work, both experimental and numerical,
illustrates the difficulty that arises when the presence of more that has been published on concrete carbonation and its influ-
than one corrosion-inducing agency contributes to the aggres- ence on chloride binding, there is not much work focusing on
siveness of the exposure conditions for a particular structure. the quantification of release of bound chlorides in chloride-
contaminated concrete caused by concrete carbonation. In this
Extensive research has been carried out over the last two paper an experimental study is presented on the release of
decades on the corrosion of reinforcing steel induced by con- bound chlorides in chloride-contaminated cement paste speci-
crete carbonation (Bertos et al., 2004; Papadakis et al., 1991; mens when they are exposed to a carbon dioxide environment.
Yuan et al., 2009). The work includes the experimental investi- The free and total chloride contents at various different carbo-
gation and numerical modelling of the carbonation process in nation times were measured in the tested specimens using
reinforced concrete structures (Chang and Chen, 2006; Isgor leaching and acid-soluble methods. The bound chloride con-
and Razaqpur, 2004; Saetta and Vitaliani, 2004, 2005; Saetta tents were calculated from the measured total and free chlor-
et al., 1993), experimental investigation of the effect of con- ides. Based on the obtained experimental results, a numerical
crete carbonation on pore structure and diffusion properties of model was also developed for simulating the carbonation
hydrated cement pastes (Anstice et al., 2005; Ngala and Page, process of the chloride-contaminated cement paste.
1997) and the experimental and material analyses of the effect
of carbonation on chloride binding capacity in cementitious
materials (Arya et al., 1990; Chávez-Ulloa et al., 2013; Experimental programme
Delagrave et al., 1997; Fabbri et al., 2009; Glass and Buenfeld, The cement used in this experiment was Type 42·5 ordinary
2000; Goñi and Guerrero, 2003; Hassan, 2001; Ioannou et al., Portland cement. In order to reduce the experimental running
2015; Jirickova and Cerny, 2006; Mesbah et al., 2012; Reddy time but still be able to achieve good and representative results,
et al., 2002; Suryavanshi and Swamy, 1996; Suryavanshi et al., chloride binding was achieved by using 0·5 mol/l sodium chlor-
1996). Note that in the service life prediction of a concrete ide solution as the mixing water for the casting of the cement
structure it is necessary to distinguish the free and bound paste specimens. The water to cement ratio used in the mix
chlorides due to the retardation effect of bound chlorides was 0·5.
(Mangat and Molloy, 1995). If the release of bound chlorides
due to concrete carbonation were ignored in the prediction In total, ten cement paste specimens of dimensions 40 mm
model, the service life predicted would be inaccurate and 40 mm 160 mm were cast and then cured at a standard
unsafe. The X-ray diffraction (XRD) and differential thermal curing condition (20 ± 2°C and 95% RH) for a period of 56 d.
analysis (DTA) results reported by Suryavanshi and Swamy After this, all specimens were dried at room temperature
(1996), obtained from the experiments of four different types (20 ± 2°C and 60% RH) for 1 d. After the standard curing and
of concrete slabs subjected to chloride penetration over a total drying processes, two specimens were analysed and the remain-
period of about 3 years and then exposed to atmospheric ing eight specimens were used for further accelerated
354
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright © ICE Publishing, all rights reserved.
Magazine of Concrete Research Effect of carbonation on release of
Volume 68 Issue 7 bound chlorides in chloride-contaminated
concrete
Geng, Easterbrook, Liu and Li
carbonation tests. In order to investigate the effect of carbona- cement paste. However, owing to the transport of free chloride
tion depth on the release of bound chlorides, each specimen ions taking place in the specimen during the carbonation
was sealed on five of its six surfaces by paraffin wax and only period, the total chloride content measured at each time in each
one 40 mm 40 mm surface was left untouched. All eight layer may not be equal to this initial total chloride content.
specimens were placed into an accelerated carbonation
chamber, which was set to about 20% carbon dioxide gas with The XRD analysis was carried out using a D8 Advance instru-
a temperature of 20–25°C and a RH of 65–75%. During the ment (Bruker AXS) with a Cu Kα radiation generated with
accelerated carbonation test, two specimens were removed for 40 kV and 30 mA.
analysis after the carbonation times of 14, 28, 56 and 90 d.
Four slices, each 1 cm thick, were cut from the carbonated
specimens, starting from the exposed surface (see Figure 1). Experimental results
Each slice was broken into small blocks, which were then Chloride binding in cementitious materials is dominated
immersed in anhydrous ethanol for 7 d to terminate hydration. by the content of tricalcium aluminate (C3A) and tetra-
These small blocks were ground into fine powder (passing calcium aluminoferrite (C4AF) to form Friedel’s salt
through a 0·15 mm mesh sieve). The powders were then stored (3CaO•Al2O3•CaCl2•10H2O) (Florea and Brouwers, 2012). In
in a desiccator with silica gel and soda lime at 11% RH to this study, sodium solution was used as the mixing water, so
minimise any further carbonation before they were used for the Friedel’s salt is formed from the chemical reactions not only
determination of total and free chloride contents and other between sodium chloride and tricalcium aluminate but also
material characterisation analyses. between sodium chloride and mono-sulfoaluminate, which is
the hydration product of tricalcium aluminate. The forming
The free chloride content of the fine powder was determined by process of Friedel’s salt can be interpreted by following
using the leaching method. The total chloride content of the reactions
fine powder was determined using the acid-soluble method. The
bound chloride content was calculated based on the measured CaðOHÞ2 þ 2NaCl ! CaCl2 þ 2Naþ þ 2OH
total and free chloride contents. Note that the initial total chlor-
ide content in each specimen cured after the standard curing 3CaO Al2 O3 þ CaCl2 þ 10H2 O
condition can be calculated based on the mixing water of ! 3CaO Al2 O3 CaCl2 10H2 O
0·5 mol/l sodium chloride solution, which is 8·863 mg/g of
1 2 3 4
Exposed surface for
carbonation experiment
160 mm
4
3
2
1
(b)
355
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright © ICE Publishing, all rights reserved.
Magazine of Concrete Research Effect of carbonation on release of
Volume 68 Issue 7 bound chlorides in chloride-contaminated
concrete
Geng, Easterbrook, Liu and Li
3CaO Al2 O3 CaCl2 10H2 O Figure 2. The mass concentration of free, bound and total
! 4Ca2þ þ 2Cl þ 4OH þ 2AlðOHÞ3 chlorides in surface layer
12
AlðOHÞ3 þOH ! AlðOHÞ
4
Total chlorides
Mass concentration of chlorides: mg/g
Free chlorides
10
Bound chlorides
Carbonation changes the nature of the hydration products and
thus affects the chloride binding ability of cement pastes. 8
Figure 2 shows the variation of the measured bound, free and
total chlorides with the carbonation time, obtained from the 6
surface layer of the tested specimens. It can be seen from the
figure that there is almost no change in bound chlorides
4
during the first 14 d of carbonation. This is mainly because the
thickness of layers chosen in the tests is relatively large (1 cm),
while the penetration of carbon dioxide in cement paste is very 2
slow. However, with the further increase of carbonation time,
there is a clear decrease of bound chlorides. The longer the 0
carbonation time is, the more the bound chlorides reduce. This 1 2 3 4
356
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright © ICE Publishing, all rights reserved.
Magazine of Concrete Research Effect of carbonation on release of
Volume 68 Issue 7 bound chlorides in chloride-contaminated
concrete
Geng, Easterbrook, Liu and Li
12
Total chlorides C
Free chlorides
Mass concentration of chlorides: mg/g
10 Bound chlorides
8
V
V 90 d
6 V C C
V C
4 56 d
F CH
2 28 d
E C-S-H 0d
0
1 2 3 4
Layer 5 10 15 20 25 30 35 40 45 50 55 60
2θ
Figure 4. The mass concentration of free, bound and total
chlorides at 90 d Figure 5. XRD patterns of surface layer at 0, 28, 56 and 90 d of
carbonation (E: AFt; F: FS; CH: calcium hydroxide; V: vaterite; CSH:
C-S-H gel; C: calcite)
357
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright © ICE Publishing, all rights reserved.
Magazine of Concrete Research Effect of carbonation on release of
Volume 68 Issue 7 bound chlorides in chloride-contaminated
concrete
Geng, Easterbrook, Liu and Li
5 10 15 20 25 30 35 40 45 50 55 60
According to the mass conservation of chlorides in a unit
2θ
volume of cement paste, the following mass conservation
Figure 6. XRD patterns of first, second and fourth layers after
equation can be established
56 d of carbonation (E: AFt; F: FS; CH: calcium hydroxide; V:
@C @B @ @C
vaterite; CSH: C-S-H gel; C: calcite) 1: þ ¼ D
@t @t @x @x
To simulate this interaction, a transport model that considers where C is the mass concentration of free chlorides, B is the
the transport of free chloride ions and carbon dioxide as well mass concentration of bound chlorides, t is the time, x is the
as the adsorption/desorption of bound chlorides induced by distance from the exposed surface and D is the diffusion coeffi-
carbonation is developed and presented in this section (see cient of free chloride ions in the cement paste. The mass con-
Figure 7). Since the specimen is exposed to the carbon dioxide centration of bound chlorides can be described using the
environment only on its one surface and all other surfaces are following first-order reaction equation (Wang, 2001)
sealed, the transport of free chloride ions and carbon dioxide
in the specimen can be treated as a one-dimensional transport @B
2: ¼ kb ðB Beq Þ
problem. @t
Cl Cl Liquid phase For non-carbonated concrete the bound chlorides at the equili-
Cl
Cl brium state can be expressed in terms of the free chlorides by
CO2
Cl using a chloride-binding isotherm such as a linear, Freundlish
Cl Cl
or Langmuir isotherm (Tang and Nilsson, 1993). If the
Langmuir isotherm is used (Li and Page, 2000), Beq can be
Cl Cl Cl Cl Solid phase
expressed as follows
Cl αC
Cl 3: Beq ¼
Cl 1 þ βC
Cl Cl
358
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright © ICE Publishing, all rights reserved.
Magazine of Concrete Research Effect of carbonation on release of
Volume 68 Issue 7 bound chlorides in chloride-contaminated
concrete
Geng, Easterbrook, Liu and Li
kc αC 0·4
4: Beq ¼
1 þ βC
14 d
0·2 28 d
where kc is the reduction factor due to carbonation, which 56 d
90 d
depends on the type of binders and mixture of the concrete
and the concentration of carbon dioxide. 0
0 0·5 1·0 1·5 2·0 2·5 3·0 3·5 4·0 4·5 5·0
Distance from exposed surface: cm
For the present one-dimensional accelerated carbonation
problem, the depth of carbonation can be approximated as Figure 8. Reduction function for carbonated concretes (Dco =
follows (Broomfield, 2006; Chávez-Ulloa et al., 2013; Villain 1·2 10−11 m2/s, kc = 0·01, n = 0·7, kt = 92·36 mm/year0·7, vertical
et al., 2007) dash lines are the depths of carbonation xco calculated from
Equation 5)
5: xco ¼ kt tn
where xco is the depth of carbonation, kt is the carbonation smooth increase of f (kc, xco) from kc to 1 in the transition
coefficient and n is a constant. Theoretically, Equation 3 is zone. Substituting Equations 5 and 7 into Equations 1 and 2
applied to the non-carbonated zone of x > xco and Equation 4 yields
is applied to the carbonated zone of x < xco. This means that
there will be a discontinuity at the junction between the carbo- @C @2C
¼D 2
nated and non-carbonated zones. In reality, however, there @t @x
8:
will be a transition zone between the carbonated and non- 1 kc x kt tn αC
þ kb B 1 erfc pffiffiffiffiffiffiffiffiffi
carbonated zones, in which the concentration of carbon 2 2 Dco t 1 þ βC
dioxide changes gradually and smoothly from the carbonated
zone to the non-carbonated zone. To take into account this
transition zone, the following modified reduction function is
proposed which can be applied to both carbonated and non-
@B 1 kc x kt tn αC
carbonated zones 9: ¼ kb B 1 erfc pffiffiffiffiffiffiffiffiffi
@t 2 2 Dco t 1 þ βC
1 kc x xco
6: f ðkc ; xco Þ ¼ 1 erfc pffiffiffiffiffiffiffiffiffi
2 2 Dco t
@Cð0; tÞ @CðL; tÞ
1 kc x xco αC 11: ¼0 and ¼0
7: Beq ¼ 1 erfc pffiffiffiffiffiffiffiffiffi @x @x
2 2 Dco t 1 þ βC
where Bo is the initial mass concentration of bound chlorides,
Co is the initial mass concentration of free chlorides and L is
Figure 8 graphically shows the variation of the reduction func- the length of the specimen. Equation 11 indicates that the free
tion with the distance at various different times. It can be seen chloride ions do not enter or exit the specimen at its two ends.
from the figure that f (kc, xco) = kc in the fully carbonated zone Equations 8 and 9 can be solved using numerical methods
and f (kc, xco) = 1 in the non-carbonated zone. There is a from which the mass concentrations of free and bound
359
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright © ICE Publishing, all rights reserved.
Magazine of Concrete Research Effect of carbonation on release of
Volume 68 Issue 7 bound chlorides in chloride-contaminated
concrete
Geng, Easterbrook, Liu and Li
chlorides can be obtained. Numerical schemes for solving a set of Cs(t) is shown in Figure 9, from which a reasonably good
of partial differential equations can be found in many math- agreement between them is observed.
ematical textbooks and thus are not provided or discussed
further here. After Bo, Co, D, Dco, kt, kc and kb are defined, Equations 8
and 9 can be solved numerically. Figure 10 and Figure 11
show the predicted mass concentration distribution profiles of
Numerical results bound and free chlorides in the specimen at three different
Before Equations 8 and 9 can be solved numerically, one has times (0, 56 and 90 d of carbonation). To demonstrate the
to know the values of all parameters involved in the model.
The initial mass concentrations of bound and free chlorides 6
can be obtained from the experimental data shown in
0
kc αCs kc αCs 0 0·005 0·010 0·015 0·020 0·025 0·030 0·035 0·040 0·045
13: BðtÞ ¼ þ Bso ekb t
1 þ βCs 1 þ βCs Distance from exposed surface: m
360
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright © ICE Publishing, all rights reserved.
Magazine of Concrete Research Effect of carbonation on release of
Volume 68 Issue 7 bound chlorides in chloride-contaminated
concrete
Geng, Easterbrook, Liu and Li
7
carbonation chamber. The experimental data showed that the
carbonation of cement paste results in a release of bound
6
chlorides, which is related to not only the decomposition of
5 Friedel’s salt but also the decomposition of C-S-H gel. The
release of bound chlorides leads to an increase of free chlorides
4 and the corresponding diffusion of free chlorides from a region
of high concentration to a region of low concentration.
3
361
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright © ICE Publishing, all rights reserved.
Magazine of Concrete Research Effect of carbonation on release of
Volume 68 Issue 7 bound chlorides in chloride-contaminated
concrete
Geng, Easterbrook, Liu and Li
Chang CF and Chen JW (2006) The experimental investigation Ngala VT and Page CL (1997) Effects of carbonation on
of concrete carbonation depth. Cement and Concrete pore structure and diffusion properties of hydrated
Research 36(9): 1760–1767. cement pastes. Cement and Concrete Research
Chávez-Ulloa E, Camacho-Chab R, Sosa-Baz M, Castro-Borges P 27(7): 995–1007.
and Pérez-López T (2013) Corrosion process of reinforced Page CL and Page MM (2007) Durability of Concrete and
concrete by carbonation in a natural environment and an Cement Composite. Woodhead Publishing in Materials,
accelerated test chamber. International Journal of Oxford, UK.
Electrochemical Science 8(7): 9015–9029. Page CL and Venneland O (1983) Pore solution composition
Delagrave A, Marchand J, Olliver JP, Julien S and Hazrati K and chloride binding capacity of silica-fume cement pastes.
(1997) Chloride binding capacity of various hydrated Materials and Structures 16(1): 19–25.
cement systems. Advanced Cement Based Materials 6(1): Papadakis VG, Vayenas CG and Fardis MN (1991)
28–35. Fundamental modeling and experimental investigation
Dias WPS (2000) Reduction of concrete sorptivity with age of concrete carbonation. ACI Material Journal
through carbonation. Cement and Concrete Research 30(8): 88(4): 363–373.
1255–1261. Reddy B, Glass GK, Lim PJ and Buenfeld NR (2002) On the
Fabbri A, Corvisier J, Schubnel A et al. (2009) Effect of corrosion risk presented by chloride bound in concrete.
carbonation on the hydro-mechanical properties of Cement and Concrete Composites 24(1): 1–5.
Portland cements. Cement and Concrete Research 39(12): Ryou JS and Ann KY (2008) Variation in the chloride
1156–1163. threshold level for steel corrosion in concrete arising from
Florea MVA and Brouwers HJH (2012) Chloride binding related different chloride sources. Magazine of Concrete Research
to hydration products: part I: ordinary Portland cement. 60(3): 177–187, http://dx.doi.org/10.1680/macr.2008.60.
Cement and Concrete Research 42(2): 282–290. 3.177.
Glass GK and Buenfeld NR (2000) The influence of chloride Saetta AV and Vitaliani RV (2004) Experimental investigation
binding on the chloride-induced corrosion risk in and numerical modeling of carbonation process in
reinforced concrete. Corrosion Science 42(2): 329–344. reinforced concrete structures. Part I: theoretical
Goñi S and Guerrero A (2003) Accelerated carbonation of formulation. Cement and Concrete Research 34(4):
Friedel’s salt in calcium aluminate cement paste. Cement 571–579.
and Concrete Research 33(1): 21–26. Saetta AV and Vitaliani RV (2005) Experimental investigation
Hassan Z (2001) Binding of External Chloride by Cement and numerical modeling of carbonation process in
Pastes. PhD thesis, Department of Building Materials, reinforced concrete structures. Part II: practical
University of Toronto, Toronto, Ontario, Canada. applications. Cement and Concrete Research 35(35):
Ioannou S, Paine K, Reig L and Quillin K (2015) Performance 958–967.
characteristics of concrete based on a ternary calcium Saetta AV, Schrefler BA and Vitaliani RV (1993) The carbonation
sulfoaluminate–anhydrite–fly ash cement. Cement and of concrete and the mechanism of moisture, heat, carbon
Concrete Composites 55: 196–204. dioxide flow through porous materials. Cement and
Isgor OB and Razaqpur AG (2004) Finite element modeling of Concrete Research 23(4): 761–772.
coupled heat transfer, moisture transport and carbonation Saillio M, Baroghel-Bouny V and Barberon F (2014) Chloride
processes in concrete structures. Cement and Concrete binding in sound and carbonated cementitious materials
Composites 26(1): 57–73. with various types of binder. Construction and Building
Jirickova M and Cerny R (2006) Chloride binding in building Materials 68: 82–91.
materials. Journal of Building Physics 29(3): 189–200. Suryavanshi AK and Swamy RN (1996) Stability of Friedel’s salt
Kobayashi K, Suzuki K and Uno Y (1994) Carbonation of in carbonated concrete structural elements. Cement and
concrete structures and decomposition of CSH. Cement Concrete Research 26(5): 729–741.
and Concrete Research 24(1): 55–61. Suryavanshi AK, Scantlebury JD and Lyon SB (1996)
Li LY and Page CL (2000) Finite element modelling of chloride Mechanism of Friedel’s salt formation in cements rich in
removal from concrete by electrochemical method. tri-calcium aluminate. Cement and Concrete Research
Corrosion Science 42(12): 2145–2165. 26(5): 717–727.
Mangat PS and Molloy BT (1995) Chloride binding in concrete Tang L and Nilsson LO (1993) Chloride binding capacity and
containing PFA, gbs or silica fume under sea water binding isotherms of OPC pastes and mortars. Cement and
exposure. Magazine of Concrete Research 47(171): Concrete Research 23(2): 247–253.
129–141, http://dx.doi.org/10.1680/macr.1995.47.171.129. Thomas MDA and Jones MR (1996) A critical review of service
Mesbah A, Cau-dit-Coumes C, Renaudin G, Frizon F and Leroux F life modeling of concretes exposed to chlorides. In
(2012) Uptake of chloride and carbonate ions by calcium Concrete in Service of Mankind: Radical Concrete
monosulfoaluminate hydrate. Cement and Concrete Technology (Dhir RK and Hewlett PC (eds)). E & F N
Research 42(8): 1157–1165. Spon Ltd, London, UK, pp. 723–736.
362
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright © ICE Publishing, all rights reserved.
Magazine of Concrete Research Effect of carbonation on release of
Volume 68 Issue 7 bound chlorides in chloride-contaminated
concrete
Geng, Easterbrook, Liu and Li
Villain G, Thiery M and Platret G (2007) Measurement of concrete. Atmospheric Environment 41(34):
methods of carbonation profiles in concrete: 7274–7285.
thermogravimetry, chemical analysis and Yu SW, Sergi G and Page CL (1993) Ionic diffusion across an
gammadensimetry. Cement and Concrete Research interface between chloride-free and chloride-containing
37(8): 1182–1192. cementitious materials. Magazine of Concrete Research
Wang Y (2001) Mathematical Modelling of Chloride Ingress 45(165): 257–261, http://dx.doi.org/10.1680/macr.1993.45.
into Concrete and Electro-chemical Chloride Removal 165.257.
from Concrete. PhD thesis, Aston University, Yuan Q, Shi C, De Schutter G, Audenaert K and Deng D (2009)
Birmingham, UK. Chloride binding of cement-based materials subjected to
Yoon IS, Copuroglu O and Park KB (2007) Effect of external chloride environment – a review. Construction and
global climate change on carbonation progress Building Materials 23(1): 1–13.
363
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright © ICE Publishing, all rights reserved.