Comparison of chloride penetration into surface-treated concrete in artificial and natural environments

Abstract

This study investigated the penetration of chloride into surface-treated high-performance concrete and normal concrete in natural and accelerated environments. Both high-performance concrete and normal concrete were applied in a real port. Concrete specimens that were cast together with the concrete port were transported to the laboratory and subjected to wetting and drying cycles with NaCl solution. The chloride contents of the specimens in the laboratory and the in situ components were tested. The chloride diffusion coefficients and surface chloride contents were calculated based on Fick’s second law. The results show that high-performance concrete and surface treatment clearly slow the chloride penetration into the concrete both in the laboratory and in situ. The chloride contents on the surface and in the concrete in the components of the concrete port are higher during the summer than during the winter. The chloride penetration performance in the concrete of real structures cannot be inferred from its performance in specimens under artificial environments in the laboratory.

Keywords

chloride diffusion coefficients chloride ion profiles in situ components surface chloride contents surface treatment

Introduction

Chloride-induced corrosion is a dominant cause of the deterioration of reinforced concrete structures (Hartt, 2012; Hobbs, 2001). Once the chloride content exceeds a threshold value, the passive film on the surface of the reinforcement will dissolve, and corrosion will occur in the presence of oxygen and moisture (Arya and Xu, 1995) and eventually cause the cracking and spalling of reinforced structures. To predict and improve the serviceability and durability of reinforced concrete structures, investigations of chloride penetration mechanisms and corresponding protective measures for structures are important.

The mechanisms of chloride penetration into concrete are widely thought to be diffusion due to a concentration gradient and convection under pressure (including capillary suction caused by surface tension; Kwon et al., 2009; Marchand and Samson, 2009; Muigai et al., 2012). Therefore, Fick’s second law of diffusion is widely applied to study chloride diffusion. However, most studies draw their conclusions based on specimens in artificial environments, such as wetting and drying cycles (Arya et al., 2014; Suryavanshi et al., 2002; Yang and Wang, 2004). Many studies have focused on chloride penetration into concrete in natural environments. Sadati et al. (2015) studied the performance of coated concrete samples that were naturally exposed to coastal soil, and Tang et al. (2014) studied the chloride profiles of concrete slabs that were exposed in Swedish marine environment over 20 years. Otieno et al. (2016) compared the corrosion performance of beams under cyclic wetting and drying to natural corrosion in a marine tidal zone. However, they did not study the components of the structures under loads. Nanukuttan et al. (2008) compared the chloride profiles of concrete bridge piers in different exposure zones in North East Scotland, and Arskøg et al. (2004) studied the chloride penetration of two beams above 3 m mean water level of a structure with a service life of 8 years in Norwegian. However, few studies were made on the comparison of components under different loads in the same natural environment.

An effective protective measure is to improve the concrete quality. High-performance concrete (HPC) has been widely applied in structures and has high strength and low permeability due to the use of mixtures of additives, such as fly ash, slag, silica fume, and superplasticizer, which reduce the water/binder ratio. However, HPC cannot completely prevent the ingression of harmful ions into concrete because HPC is an intrinsically porous material. Therefore, surface treatment can be used as an additional measure to improve the durability of structures by protecting the structures from the ingression of harmful ions. Different coatings utilize different protective mechanisms and have different protective efficacies. These surface coatings, including organic coatings and inorganic polymers (geopolymers), have been shown to have varying degrees of effectiveness in improving concrete durability using artificial wetting and drying cycles (Almusallam et al., 2003; Brenna et al., 2013; Li et al., 2015; Moon et al., 2007) or by being exposed to natural marine environments (Medeiros and Helene, 2009; Moradllo et al., 2012; Rodrigues et al., 2000; Shekarchi et al., 2009; Zhang et al., 2012). However, few reports have focused on the durability of in situ structures with surface coatings under loadings.

This study was conducted to investigate and compare the chloride penetration of surface-treated HPC and normal concrete (NC) in the laboratory and in situ. Surface-treated HPC and NC were applied in a real port. The chloride contents of the concrete from the port and concrete specimens with the same mixture that were subjected to wetting and drying cycles in the laboratory were measured and compared. The chloride diffusion coefficients and surface chloride contents of the specimens in the laboratory and the in situ components were compared. The results of this study can help to understand the chloride penetration behavior of different concretes and different components of a real concrete port and also provide a comparison between the chloride penetration behaviors of natural and artificial environments.

Experimental program

Materials and concrete mixtures

The concrete mixtures shown in Table 1 were exposed in a real port in wave-splashing condition. The two outside shelves (shelves S1 and S2 in Figure 1) were cast with HPC, while the other shelves were cast with NC. Additional details about the port are given in section “In situ concrete components and specimens.” The cement (Table 1) was ordinary Portland cement with a grade of 42.5 MPa complying with GB175-2007 and was produced by Jiaxing Dongjin Cement Limited Company. The fly ash was produced by the Dinghai power plant, and the slag was produced by Jiangsu Shagang Limited Company. The fine aggregate was medium natural river sand with a fineness modulus of 2.5, and the coarse aggregate was normal gravel in the size range of 5–25 mm. The 28-day compressive strengths of the HPC and NC, f_c, were 40.42 and 40.32 MPa, respectively, both of which have met the code for durability design of concrete structures of China (GB50476-2008, 2008). The strengths of both types of concrete were similar; however, their durabilities are different, which will be illustrated in the following sections.

Table 1.

Compositions of concrete mixtures (kg/m³).

	Water	Cement	Fly ash	Slag	Fine aggregate	Coarse aggregate	Superplasticizer	Corrosion inhibitor
HPC	139	180	180	90	779	1032	5.4	9.0
NC	180	400	0	0	607	1234	0	0

HPC: high-performance concrete; NC: normal concrete.

Figure 1.

The studied port in Zhoushan, China.

In situ concrete components and specimens

The studied concrete port is located in Zhejiang Province, China, and the East China Sea is located north of the port. The concrete port includes 13 shelves. Figure 2 shows shelves S1–S4. As shown in Figure 2, HPC was applied to the beams and columns in S1 and S2, while NC was applied to those of the other shelves. A beam and a column from each shelf that was located in the spray splash zone were studied. The tested beams, with 70 mm of concrete cover, had dimensions of 1200 mm × 900 mm × 12,000 mm, and the elevations ranged from 1.90 to 2.80 m. The elevation of the tested point was 2.05 m, which was 0.15 m above the bottom of the beam. The tested column, with 50 mm of concrete cover, was a cylinder with a diameter of 800 mm and a height of 1600 mm, and the elevations ranged from 0.30 to 1.90 m. The elevation of the tested point was 1.10 m, which is at the middle of the column.

Figure 2.

Diagram of the structure from S1 to S4: (a) top view of the structure, (b) section 1-1 of the structure, and (c) section 2-2 of the structure (dimensions are in millimeter for the structure size and in meter for the height above sea level).

A total of six 150 mm × 150 mm × 300 mm specimens of each type of concrete were used in the study. All of the specimens were cast in situ together with the concrete port using the same batch of HPC or NC. They were cured on the shore near the port and wetted once a day for 28 days. The studied in situ components and specimens are listed in Table 2.

Table 2.

Information about the studied specimens in the laboratory and the in situ components.

Label	Size (mm)	Concrete type	Surface condition	Environment	Numbers
LHT	150 × 150 × 300	HPC	All treated	In the laboratory	3
LHN	150 × 150 × 300	HPC	All treated except for top surface	In the laboratory	3
LNT	150 × 150 × 300	NC	All treated	In the laboratory	3
LNN	150 × 150 × 300	NC	All treated except for top surface	In the laboratory	3
SHB	1200 × 900 × 12,000	HPC	All treated	In situ	2
SHC	Diameter: 800; height: 1600	HPC	All treated	In situ	2
SNB	1200 × 900 × 12,000	NC	All treated	In situ	2
SNC	Diameter: 800; height: 1600	NC	All treated	In situ	2

HPC: high-performance concrete; NC: normal concrete.

Surface treatment

Two months after the casting of the port, a commercial surface treatment material was applied to the specimens and the port. The surface treatment contained three layers (Table 3). For each type of concrete, three specimens were painted on all of the surfaces, while the other three specimens were painted on the side and bottom surfaces to evaluate the effectiveness of the surface treatment (Figure 3). All of the surfaces of the port were surface treated as requested by the owner of the port.

Table 3.

Information about the surface treatment.

Protective layers	Main components	Amount (g/m²)	Dry film thickness (µm)
Primer	Low-molecular-weight epoxy resin, polyamide curing agent, and additive	100	40
Intermediate coat	Epoxy resin, pigments and fillers, and polyamide	333.3	300
Finish coat	Acrylate polyurethane	208.3	90

Figure 3.

Layout details of the concrete specimens in millimeter: (a) specimens LHN and LNN and (b) specimens LHT and LNT.

Chloride tests in the laboratory

After the surface treatment had dried for 2 months, all of the concrete specimens were transported to the laboratory at Zhejiang University and subjected to alternate wetting and drying cycles. Each cycle lasted for 7 days and consisted of a 3-day wetting period followed by a 4-day drying period. The specimens were submerged in NaCl solution during the wetting periods and were exposed to the air in the laboratory during the drying periods. To “accelerate” the chloride penetration, the NaCl concentration in the laboratory was about twice of the sea water (approximately 3%), that is, 6.15%, the same with the authors’ previous study (Zhao et al., 2010).

The chloride contents in the concrete specimens as a function of depth were measured after 2, 4, and 6 months of wetting and drying cycles. The concrete powders were collected at 5-mm intervals to a depth of 50 mm below the tested surface. At each position, the concrete powders that were collected from three holes were mixed to avoid the influence of the aggregates. The powder was oven-dried at 50°C ± 1°C for 24 h. At each position, 4.00 g of concrete powder was mixed with 40 mL of deionized water to conduct the chloride content measurements. A rapid chloride ion content determination method with automatic potentiometric titration (DY 2501-B) was used to precisely measure the free (water soluble) chloride content (concrete weight, %) in the concrete. After the tests, the holes were filled with epoxy resin to avoid the ingression of chloride.

In situ chloride tests

The in situ chloride contents of the beams and columns collected in section “In situ concrete components and specimens” were measured 11, 14, and 18 months (May, August, and December, respectively) after the surface treatment at the port. Similar to the concrete specimens, the concrete powders were collected at 5-mm intervals to a depth of 50 mm below from the exposed surface. The same test method as was described in section “Chloride tests in the laboratory” was applied to the powder from the beams and columns.

Results and discussion

Chloride ion profiles

Chloride ion profiles of the specimens in the laboratory

The chloride ion profiles of the specimens with and without surface treatment are shown in Figure 4. The chloride content is the average value of the free chloride content (concrete weight, %) from the three identical specimens. In Figure 4, 2M, 4M, and 6M represent specimens that were subjected to 2, 4, and 6 months of wetting and drying cycles, respectively.

Figure 4.

Chloride profiles of specimens (a) without surface treatment and (b) with surface treatment.

Figure 4 shows that the chloride content initially decreases and then reaches a constant low level (approximately 0.007%) along the concrete cover, which might come from the materials in the concrete, such as the water or aggregate, rather than the ingression of chloride from outside.

The chloride content also increases with longer periods of being subjected to alternate wetting and drying cycles. For the same period of time, the chloride content of HPC specimens without surface treatment (LHN) is lower than that of NC specimens without surface treatment (LNN; Figure 4(a)), and the chloride content of HPC specimens with surface treatment (LHT) is less than that of HPC specimens with surface treatment (LNT; Figure 4(b)). This can be attributed to the slower penetration of chloride ions through the HPC, which contains fly ash and slag (Thomas and Bamforth, 1999; Ye et al., 2016). The glassy SiO₂ and Al₂O₃ in the fly ash and slag could react with the Ca(OH)₂ that is generated by the hydration of cement, which would fill the pores in the concrete, improve crystal alignment, and increase the density of the interfacial transition zone (Thomas and Bamforth, 1999).

The chloride contents of the specimens without surface treatment are higher than those with surface treatment (Figure 4(a) and (b)). In addition, the increases of the chloride content in LHT and LNT with time are less than those of LHN and LNN, respectively. Therefore, the surface treatment clearly slows the ingression of chloride ions into the concrete.

Chloride ion profiles of the in situ components

The in situ chloride profiles are shown in Figure 5. The chloride content is the average value of the free chloride content (concrete weight, %) from the two same components, and 11M, 14M, and 18M indicate the exposure time of the port of 11, 14, and 18 months, respectively.

Figure 5.

Chloride profiles of the in situ components: (a) HPC and (b) NC.

The chloride content of the in situ HPC is less than that of the in situ NC (Figure 5(a) and (b)), which indicates that the in situ HPC is denser and more resistant to chloride ingression than the in situ NC. The chloride contents at depths of 0–15 mm at month 14 are higher than those in months 11 and 18 (Figure 5). The corresponding months of the tests in situ are May, August, and December. The binding of chloride will decrease with the increase in temperature (Xu et al., 2016). More free chloride ions in concrete are released in summer, and it is on the contrary in winter. Besides, during the summer, the stronger wind and waves result in severe splashing of waves onto the port, and the higher temperatures accelerate the evaporation of water, which results in higher chloride contents near the surface of the concrete. During the winter, the lower temperatures slow the evaporation of water, which results in the lower chloride contents in the near-surface concrete. Therefore, the chloride content of the near-surface concrete in month 14 (i.e. August) is higher than those in months 11 and 18.

Figure 5(a) shows that the chloride content in the HPC beams (SHB) is slightly higher than that in the HPC columns (SHC). The column is under compressive loads, which could reduce the number of pores, change the distribution of the pore diameters, and reduce the average diameter of the micro-pores, which in turn reduces the water transport capacity and chloride transport capacity of the concrete (Wu et al., 2016). The beam is under a bending load, and the tested point is in the tensile region. The tensile loads will expand the micro-pores and connect some pores, which will accelerate the ingression of chloride ions (Xin, 2010). This phenomenon can also be found in the NC beams (SNB) and NC columns (SNC) except for the results of samples SNB-14M and SNC-14M. The authors believe that this disagreement might be due to testing errors.

Comparison of chloride ion profile in the laboratory and in situ

According to sections “Chloride ion profiles of the specimens in the laboratory” and “Chloride ion profiles of the in situ components,” surface treatment could significantly slow the chloride penetration into the concrete in both the laboratory and in situ. The increasing of chloride content in concrete during the test periods is slow according to Figures 4(b) and 5. Comparing Figures 4(b) and 5, the chloride contents of the in situ components are much higher than those of the specimens. We reckon that this is mainly due to the 2 months between the casting and the surface treatment. During these 2 months, the hydration reactions of the cementitious materials were insufficient, and the greater porosity and larger pore diameters allowed for the ingression of chloride ions (Arya et al., 2014). For the same reason, the chloride content of SNB-18M at 50 mm depth, that is, 0.046%, tends to reach the threshold value (0.05%–0.07% by weight of concrete depending on cement content of the mixes). Measures should be taken to avoid steel corrosion, although the concrete cover is 70 mm. The concrete specimens were cured, and their surfaces were treated before they were subjected to alternating wetting and drying cycles. In addition, due to the strong wind and waves, the longer exposure time may also result in higher chloride contents in the in situ components. The results indicate that concrete surfaces should be treated as soon as possible after casting of cast-in-place structures to significantly slow the ingression of chloride.

The chloride content increases with periods of being subjected to alternate wetting and drying cycles for the specimens in the laboratory, while it changes with seasons for the in situ components. Therefore, the chloride penetration performance of the concrete of the real structures cannot be inferred from the performance of specimens under wetting and drying cycles in the laboratory.

Concrete surface chloride content and chloride diffusion coefficient

Previous studies applied Fick’s second law of diffusion to calculate the surface chloride content and the chloride diffusion coefficient of concretes under wetting and drying cycles and loading (Wu et al., 2016) or with surface coatings (Sadati et al., 2015). A similar fitting is performed in this study; however, it should be noted that Fick’s second law is suitable only in the condition that the constant chloride concentration penetrates into the concrete in single dimension. In this study, the chloride concentration is not constant because there is no chloride solution during the drying days or when the tide goes out. The chloride penetration could be thought as from single dimension since the surface treatment shows effective protection from chloride ingression (Zhao et al., 2010). Therefore, the surface chloride content and the chloride diffusion coefficient derived from the fitted line are the approximate values, and they are suitable only for the comparison. Besides, D_c in concrete with surface treatments, including the LHT and LNT specimens and the in situ beams and columns, can only reflect the chloride diffusion performance of the surface-treated concrete rather than only the concrete.

Fick’s second law of diffusion can be written as (Arya et al., 2014)

\frac{\partial C}{\partial t} = \frac{\partial}{\partial x} (D_{c} \frac{\partial C}{\partial x})

(1)

Assuming that concrete is homogenous material, surface chloride content and diffusion coefficient are constant values and bound between chloride ions and concrete is ignored, Equation (1) can be solved. Following initial and boundary conditions as

C |_{x = 0} = C_{s}, C |_{x > 0}^{t = 0} = C_{0}

The Equation (1) can be solved as

C (x, t) = C_{0} + (C_{s} - C_{0}) [1 - e r f (\frac{x}{2 \sqrt{D_{c} t}})]

(2)

where C(x, t) is the chloride content at a depth x and time t (% weight of concrete), C₀ is the initial chloride content of the concrete (% weight of concrete), and C₀ is taken as 0.007% because the chloride content of the concrete specimens deeper than the chloride penetration depth is approximately 0.007% (section “Chloride ion profiles of the specimens in the laboratory”). The values of D_c and C_s are calculated by iteration to produce the best fits, and chloride contents of <0.008% are ignored during the fitting.

Values of C_s and D_c of specimens in the laboratory

The surface chloride contents C_s and the chloride diffusion coefficient D_c of the concrete specimens in the laboratory are shown in Figures 6 and 7. It needs to be pointed out the values of C_s and D_c in this section and the following section are approximate values, and they are suitable only for the comparison as stated above. As expected, C_s increases with increasing numbers of wetting and drying cycles in all of the specimens. These results agree with those of several other authors (Arya et al., 2014; Kassir and Ghosn, 2002).

Figure 6.

Surface chloride contents (%, concrete weight) C_s of the specimens in the laboratory.

Figure 7.

Calculated results of the chloride diffusion coefficients D_c of the specimens.

The C_s value of LNN is significantly greater than that of LHN. The water-to-binder ratio of the NC is greater than that of the HPC, which results in greater porosity and attracts a higher surface chloride content (Bamforth et al., 1997). For the same surface conditions, there is no significant difference between the LHT and LNT. Therefore, the concrete type does not significantly influence the value of C_s for specimens with the same surface treatment.

The C_s values of the specimens without surface treatment are greater than those without surface treatment for the same type of concrete. In addition, the slope of the variation of C_s with time of the specimens with surface treatment is greater than that of the specimens without surface treatment. This indicates that the surface treatment can decrease the value of C_s and slow the increase of C_s with time.

According to Figure 7, the D_c values of the LHN and LHT are lower than those of LNN and LNT, respectively. This indicates that chloride diffusion is easier in NC than in HPC under the same surface conditions. This occurs because the fly ash and slag increase the density of the concrete and slow the penetration of chloride ions.

Figure 7 shows that the D_c values of the specimens without surface treatment (LHN and LNN) do not change significantly with the wetting and drying periods. However, D_c has been widely reported to decrease with exposure time because of several factors, such as ongoing cement hydration and pore blockage induced by the advancing chloride (Audenaert et al., 2010; Nokken et al., 2006). This difference might have occurred because the cement hydration was sufficient before the chloride content tests were conducted and did not continue during the 4 months of the tests.

In the specimens with surface treatment (LHT and LNT), the chloride diffusion coefficients D_c of all of the concretes and of the surface treatment decreased with the wetting and drying periods (Figure 7). This indicates that the surface treatment is effective at decreasing the diffusion of chloride ions. The chloride diffusion properties of the surface-treated concrete decrease with time, although the cement hydration of the concrete is sufficient. These results are consistent with previous studies (Moradllo et al., 2012; Sadati et al., 2015). The rate of chloride penetration (i.e. D_c) governs how fast chloride will enter and accumulate in the concrete. The decrease in D_c due to the surface treatment is expected to prevent the initiation of corrosion and increase the service life of the structure.

Values of C_s and D_c of the in situ components

The surface chloride contents C_s of the in situ components are shown in Figure 8. The variation of C_s with time is similar to the chloride profiles in section “Chloride ion profiles of the in situ components; that is, C_s is higher in the summer and decreases in the winter. The result agrees with authors’ previous study (Zhao et al., 2013). This is because that the stronger wind and waves in the summer result in severe splashing of waves onto the port, and the higher temperatures accelerate the evaporation of water, which leaves a higher chloride content on the surface concrete. The opposite situation occurs during the winter.

Figure 8.

Surface chloride contents (%, concrete weight) of the in situ components.

There is no significant difference between the C_s values of the beams and columns (Figure 9). In addition, the concrete type has little influence on the surface chloride content of the specimens and the columns and beams with surface treatment. This might be because the surface conditions are similar in all of the components.

Figure 9.

Calculated results of chloride diffusion coefficients D_c of the in situ components.

Figure 9 shows the chloride diffusion coefficients D_c of the in situ components. The D_c values of the HPC are smaller than those of the NC for the same components.

The D_c values are smaller in the summer (month 14) and increase in the winter. This could be because the higher chloride contents on the surface and in the near-surface concrete steepen the chloride profile, which results in a lower fitted value of D_c in the summer. However, previous studies have indicated that higher temperatures can accelerate the movement of chloride ions (Dousti et al., 2013) and decrease the amount of bound chloride (Larsson, 1995; Roberts, 1962). As a result, D_c increases with increasing temperature (Dousti et al., 2013; Nguyen et al., 2009). Therefore, it is not reasonable to determine the chloride diffusion coefficient only based on the results from the summer when fitting the chloride profiles of components.

The D_c values of the beams are greater than those of the columns for the same type of concrete except for the results of SNB and SNC in month 14, which might be caused by testing errors. This result agrees with many previous studies (Wu et al., 2016; Xin, 2010). The D_c values of concrete increase under tensile stresses and decrease under compressive stresses.

Comparison of C_s and D_c in the laboratory and in situ

The discussion in sections “Values of C_s and D_c of specimens in the laboratory” and “Values of C_s and D_c of the in situ components” showed that there is no significant difference between the surface chloride contents C_s of HPC and NC because the surfaces of both types of concrete were treated with the same materials. However, C_s increases with time for the specimens in the laboratory but changes with the seasons in the in situ components.

The chloride diffusion coefficient D_c of HPC is smaller than that of NC. However, there is larger difference in the time-varying chloride diffusion property between the specimens in the laboratory and the in situ components. The chloride diffusion coefficients of the surface-treated specimens in the laboratory decrease with time, and they decrease in the summer for the in situ components.

Therefore, conclusions that are obtained from specimens that are subjected to artificially accelerated chloride ingression in the laboratory cannot be directly applied to structures in natural environments. The results of the chloride profiles from different seasons should be included when determining the chloride diffusion coefficients of structures.

Conclusion

HPC and NC were applied in a real port and were surface treated. Concrete specimens that were cast together with the concrete port were transported to the laboratory to accelerate the ingression of chloride ions. The chloride contents of the specimens in the laboratory and the in situ components were tested, and the chloride diffusion coefficients and surface chloride contents were fitted based on Fick’s second law. The following conclusions can be drawn from the results of this study:

The application of both HPC and surface treatment significantly slows the chloride penetration into the concrete in both the laboratory and in situ.

The chloride contents on the concrete surface and in the near-surface concrete are higher in the summer and lower in the winter.

The chloride penetration is more severe in situ than in specimens in designed artificial environment because of the severe ingression of chloride on insufficient hydration concrete in situ before the port was surface treated.

The chloride penetration properties of concretes in natural structures cannot be inferred from the performance of the concrete in the laboratory under wetting and drying cycles. The chloride content changes with seasons in natural structures, while they increase with time in the laboratory.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was financially supported by the National Natural Science Foundation of China (no. 51278460) and the Science and Technology Project of the Department of Communications of Zhejiang Province (no. 2012H032).

References

Almusallam

Khan

Dulaijan

et al . (2003) Effectiveness of surface coatings in improving concrete durability. Cement and Concrete Composites 25(4): 473–481.

Arskøg

Ferreira

Gjørv

(2004) Durability and performance of Norwegian concrete harbor structures. The fourth international conference on concrete Under severe conditions – environment and loading, Seoul, pp. 133–140. Seoul National University and Korea Concrete Institute.

Arya

(1995) Effect of cement type on chloride binding and corrosion of steel in concrete. Cement and Concrete Research 25(7): 893–902.

Arya

Vassie

Bioubakhsh

(2014) Modelling chloride penetration in concrete subjected to cyclic wetting and drying. Magazine of Concrete Research 66(7): 364–376.

Audenaert

Yuan

De Schutter

(2010) On the time dependency of the chloride diffusion coefficient in concrete. Construction and Building Materials 24(3): 396–402.

Bamforth

Price

Emerson

(1997) An International Review of Chloride Ingress into Structural Concrete, vol. 359. London: Thomas Telford Ltd.

Brenna

Bolzoni

Beretta

et al . (2013) Long-term chloride-induced corrosion monitoring of reinforced concrete coated with commercial polymer-modified mortar and polymeric coatings. Construction and Building Materials 48: 734–744.

Dousti

Rashetnia

Ahmadi

et al . (2013) Influence of exposure temperature on chloride diffusion in concretes incorporating silica fume or natural zeolite. Construction and Building Materials 49: 393–399.

GB50476-2008 (2008) Ministry of housing and urban-rural development of the People’s Republic of China, code for durability design of concrete structures of China.

10.

Hartt

(2012) Service life projection for chloride-exposed concrete reinforced with black and corrosion-resistant bars. Corrosion 68: 754–761.

11.

Hobbs

(2001) Concrete deterioration: causes, diagnosis, and minimising risk. International Materials Reviews 46(3): 117–144.

12.

Kassir

Ghosn

(2002) Chloride-induced corrosion of reinforced concrete bridge deck. Cement and Concrete Research 32(1): 139–143.

13.

Kwon

Park

et al . (2009) Service life prediction of concrete wharves with early-aged crack: probabilistic approach for chloride diffusion. Structural Safety 31(1): 75–83.

14.

Larsson

(1995) The enrichment of chlorides in expressed concrete pore solution submerged in saline solution. Proceedings of the Nordic seminar on field studies of chloride initiated reinforcement corrosion in concrete, Lund University of Technology, Report TVBM-3064, February, pp. 171–176.

15.

Yang

Guo

et al . (2015) Time dependence and service life prediction of chloride resistance of concrete coatings. Construction and Building Materials 83: 19–25.

16.

Marchand

Samson

(2009) Predicting the service-life of concrete structures—limitations of simplified models. Cement and Concrete Composites 31(8): 515–521.

17.

Medeiros

Helene

(2009) Surface treatment of reinforced concrete in marine environment: influence on chloride diffusion coefficient and capillary water absorption. Construction and Building Materials 23(3): 1476–1484.

18.

Moon

Shin

Choi

(2007) Evaluation of the durability of mortar and concrete applied with inorganic coating material and surface treatment system. Construction and Building Materials 21(2): 362–369.

19.

Moradllo

Shekarchi

Hoseini

(2012) Time-dependent performance of concrete surface coatings in tidal zone of marine environment. Construction and Building Materials 30: 198–205.

20.

Muigai

Moyo

Alexander

(2012) Durability design of reinforced concrete structures: a comparison of the use of durability indexes in the deemed-to-satisfy approach and the full-probabilistic approach. Materials and Structures 45(8): 1233–1244.

21.

Nanukuttan

Basheer

McCarter

et al . (2008) Full-scale marine exposure tests on treated and untreated concretes—initial 7-year results. ACI Materials Journal 105(1): 81–87.

22.

Nguyen

Lorente

Carcasses

(2009) Effect of the environment temperature on the chloride diffusion through CEM-I and CEM-V mortars: an experimental study. Construction and Building Materials 23(2): 795–803.

23.

Nokken

Boddy

Hooton

et al . (2006) Time dependent diffusion in concrete—three laboratory studies. Cement and Concrete Research 36(1): 200–207.

24.

Otieno

Beushausen

Alexander

(2016) Chloride-induced corrosion of steel in cracked concrete. Part I: experimental studies under accelerated and natural marine environments. Cement and Concrete Research 79: 373–385.

25.

Roberts

. (1962) Effect of calcium chloride on the durability of pretensioned wire in prestressed concrete. Magazine of Concrete Research 14(42): 143–154.

26.

Rodrigues

MPMC

Costa

MRN

Mendes

et al . (2000) Effectiveness of surface coatings to protect reinforced concrete in marine environments. Materials and Structures 33(10): 618–626.

27.

Sadati

Arezoumandi

Shekarchi

(2015) Long-term performance of concrete surface coatings in soil exposure of marine environments. Construction and Building Materials 94: 656–663.

28.

Shekarchi

Rafiee

Layssi

(2009) Long-term chloride diffusion in silica fume concrete in harsh marine climates. Cement and Concrete Composites 31(10): 769–775.

29.

Suryavanshi

Swamy

Cardew

(2002) Estimation of diffusion coefficients for chloride ion penetration into structural concrete. Materials Journal 99(5): 441–449.

30.

Tang

Boubitsas

Utgenannt

(2014) Chloride ingress in concrete—results from over 20 years’ field exposure in Swedish marine environment. In: Proceedings of XXII Nordic concrete research symposium, Reykjavik, 13–15 August, pp. 377–380. Gothenburg: Chalmers Publication.

31.

Thomas

MDA

Bamforth

(1999) Modelling chloride diffusion in concrete: effect of fly ash and slag. Cement and Concrete Research 29(4): 487–495.

32.

Wang

et al . (2016) Transport model of chloride ions in concrete under loads and drying-wetting cycles. Construction and Building Materials 112: 733–738.

33.

Xin

(2010) Study on Uniaxial Tensile Performances and Durability Damage of HPC under Cyclic Load. Qingdao, China: Qingdao Technological University.

34.

Song

Jiang

et al . (2016) Influence of elevated temperature on release of bound chlorides from chloride-admixed plain and blended cement pastes. Construction and Building Materials 104: 9–15.

35.

Yang

Wang

(2004) The diffusion characteristic of concrete with mineral admixtures between salt ponding test and accelerated chloride migration test. Materials Chemistry and Physics 85(2): 266–272.

36.

Jin

et al . (2016) Chloride penetration in concrete exposed to cyclic drying-wetting and carbonation. Construction and Building Materials 112: 457–463.

37.

Zhang

Yao

Wang

(2012) Potential application of geopolymers as protection coatings for marine concrete III. Field experiment. Applied Clay Science 67: 57–60.

38.

Zhao

Chen

Gao

et al . (2013) Seasonal variation of surface chloride ion content and chloride diffusion coefficient in a concrete dock. Advances in Structural Engineering 16(2): 395–403.

39.

Zhao

Jin

(2010) Evaluation of the performance of surface treatments on concrete durability. Journal of Zhejiang University Science A 11(5): 349–355.

Comparison of chloride penetration into surface-treated concrete in artificial and natural environments

Abstract

Keywords

Introduction

Experimental program

Materials and concrete mixtures

In situ concrete components and specimens

Surface treatment

Chloride tests in the laboratory

In situ chloride tests

Results and discussion

Chloride ion profiles

Chloride ion profiles of the specimens in the laboratory

Chloride ion profiles of the in situ components

Comparison of chloride ion profile in the laboratory and in situ

Concrete surface chloride content and chloride diffusion coefficient

Values of Cs and Dc of specimens in the laboratory

Values of Cs and Dc of the in situ components

Comparison of Cs and Dc in the laboratory and in situ

Conclusion

Footnotes

Declaration of Conflicting Interests

Funding

References

Values of C_s and D_c of specimens in the laboratory

Values of C_s and D_c of the in situ components

Comparison of C_s and D_c in the laboratory and in situ