- Email: [email protected]
AgNO3-based colorimetric methods for measurement of chloride penetration in concrete
Construction and Building Materials 26 (2012) 1–8
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Review
AgNO3-based colorimetric methods for measurement of chloride penetration in concrete Fuqiang He a, Caijun Shi b,⇑, Qiang Yuan c, Changping Chen a, Keren Zheng c a
Department of Civil Engineering and Architecture, Xiamen University of Technology, Xiamen 361024, China College of Civil Engineering, Hunan University, Changsha 410082, China c School of Civil Engineering and Architecture, Central South University, Changsha 410075, China b
a r t i c l e
i n f o
Article history: Received 16 January 2011 Received in revised form 1 May 2011 Accepted 8 June 2011 Available online 14 July 2011 Keywords: AgNO3-based colorimetric methods Chloride penetration Concrete
a b s t r a c t This paper reviews reaction mechanisms of AgNO3, AgNO3 + K2CrO4 and AgNO3 + fluoresceine colorimetric methods, and their applications for measurement of chloride ion penetration in concrete. Among the three methods, AgNO3 method is most widely applied because it gives similar results but simpler and faster than the other two methods. AgNO3-based colorimetric methods can potentially measure two variables, chloride ion penetration depth and chloride concentration at the color change boundary. Reported chloride ion concentrations at the color change boundary (Cd) measured or calculated by some researchers vary over a broad range due to many factors, such as sampling procedure, chloride ion analysis method, alkalinity of concrete, sprayed volume and concentration of AgNO3 solution, pore solution volume and methods for measuring free chloride concentration in concrete, etc., in which sprayed volume and concentration of AgNO3 solution are main factors. The smaller the volume of sprayed AgNO3 solution with a certain concentration, the lower the Cd is. 0.1 mol/L AgNO3 solution has most clear color of boundary. To obtain a lower Cd, a proper sprayed volume of 0.1 mol/L AgNO3 solution should be determined. However, this important point was not mentioned at all times since the AgNO3 method was applied. Based on small range of Cd, AgNO3 method can be a useful method for rapidly evaluating chloride ion penetration in reinforced concrete structures in chloride environments. Ó 2011 Elsevier Ltd. All rights reserved.
Contents 1. 2.
3. 4.
5.
6.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AgNO3 based colorimetric methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. AgNO3 + fluoresceine method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. AgNO3 + K2CrO4 method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. AgNO3 method. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Comparison of the three methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Measurement of chloride penetration depth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Factors affecting chloride concentration at the color change boundary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Parameters of colorimetric reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Sampling methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Methods for free chloride measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Applications of AgNO3 colorimetric methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1. Measurement of chloride non-steady state diffusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1.1. Measurement of chloride penetration kinetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1.2. Measurement of chloride apparent diffusion coefficient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Measurement of non-steady electrical migration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3. Evaluation for corrosion risk of steel in concrete . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
⇑ Corresponding author. E-mail address: [email protected] (C. Shi). 0950-0618/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.conbuildmat.2011.06.003
2 2 2 2 2 2 3 5 5 5 6 6 6 6 6 6 7 7 7 7
2
F. He et al. / Construction and Building Materials 26 (2012) 1–8
1. Introduction Chloride-induced corrosion of steel reinforcement is a dominant factor on durability of concrete structures exposed to marine environments and/or de-icing salts. It is well known that high alkaline medium provided by cement matrix in concrete maintains steel reinforcements in a passive state and prevents them from corrosion. However, a local breakdown of the passivating layer occurs when a sufficient amount of chloride ions reaches the steel reinforcement; corrosion is then enabled. In the context of chlorideinduced corrosion, it is reasonable to consider the service life of reinforced concrete structures in two stages: the first stage is the time for a critical chloride concentration to reach the surface of steel reinforcement, and the second stage is the subsequent propagation extending onward to the time when the structure is damaged by the steel corrosion beyond acceptable limits. Due to the relatively fast development of the second stage and for the sake of safety, the duration of the first stage is sometimes regarded as the design service life of steel-reinforced concrete structures. The chloride ion penetration front, varying with time and environment, can be used to monitor the movement of chloride ions in concrete and for service life prediction [1]. Analysis of chloride ion content in concrete can be conducted by the so-called profile method, which requires cutting or drilling, grinding and chemical analysis of concrete samples. The profile method requires a series of special equipment and consumes much time. On the other hand, AgNO3-based colorimetric methods can measure chloride ion penetration depth in concrete quickly and simply. It can be a useful method for rapid measurement of chloride ion penetration in reinforced concrete structures in chloride environments. This paper reviews three AgNO3-based colorimetric methods, factors affecting the color change and chloride ion concentrations at the color change boundary, and their applications. 2. AgNO3 based colorimetric methods Since the 1970s, three AgNO3-based colorimetric methods, i.e., AgNO3 + fluoresceine, AgNO3 + K2CrO4 and AgNO3 method, have been proposed to measure the chloride ion penetration depth in concrete in the field and laboratory. Both AgNO3 + fluoresceine and AgNO3 + K2CrO4 methods were derived from the methods for analyzing free chloride ions in solutions [2], as described in Table 1. The three methods are described in detail in the following sections. 2.1. AgNO3 + fluoresceine method In the 1970s, Collepardi et al. [3,4] developed a colorimetric method to determine free chloride content in concrete, in which a fluoresceine solution (1 g/L in a 70% solution of ethyl alcohol in water) was first sprayed onto a cross-section of chloride-penetrated concrete and then 0.1 mol/L AgNO3 aqueous solution was sprayed onto the cross-section. Immediately after spraying the AgNO3 solution onto concrete surface, Ag2O and AgCl formed respectively by reactions between Ag+ and OH or Cl. Fluoresceine (HF1) is an organic weak acid, which dissociates into yellowish green ion F1 in a solution. When Cl is excessive, AgCl absorbs Cl and possesses negative charge. F1 in the solution is repelled by AgCl. In this case, the solution is yellowish. When excess AgNO3 is sprayed, Ag+ is absorbed by AgCl, AgCl has positive charge, which absorbs F1 in the solution and turns the solution into pink. When this method is applied for measuring chloride ion penetration in concrete, the chemical reactions occurring during this process are shown in Table 2. In most cases, the AgNO3 solution may be excessive. The chloride-containing zone appears to be dark pink due to the formation of pink AgClAgF1 on grey concrete. The ‘chloride-free’ zone turns to be dark brown due to the formation of brown Ag2O on grey concrete. This method has been standardized as Italian Standard 79-28 [5].
Table 1 Use of AgNO3 + fluoresceine and AgNO3 + K2CrO4 in silver measuring methods [2]. Method
AgNO3 + K2CrO4
AgNO3 + fluoresceine
Indicator Titration solution pH value of the solution Color change at titration endpoint Year of origin Developer
K2CrO4 AgNO3 solution 6.5–10.5 White ? brick red 1856 Mohr
Fluoresceine AgNO3 solution 7–8 Yellowish green ? pink 1924 Fajans
Once the potassium chromate aqueous solution is sprayed, the chloridecontaining zone remains yellow due to the formation of white AgCl and left yellow K2CrO4 solution. A small amount of white AgCl and ruby red Ag2CrO7, may also form, but their amounts are too small to affect the color significantly. Thus, the chloride-containing zone may remain yellowish in most cases. However, the chloride-free zone turns into a red-brown color due to the formation of brown Ag2O + brick red Ag2CrO4 and/or ruby red Ag2CrO7. A series of chemical reactions happening in both chloride-containing and chloride-free zones in this colorimetric method are summarized in Figs. 1 and 2. 2.3. AgNO3 method AgNO3 method involves spraying only an AgNO3 solution onto a freshly fractured concrete cross-section, which leads to the formation of white and brown zones with a clear color change boundary. Chemical reactions involved in this method are summarized in Table 3. Usually, the depth of the white zone is regarded as the chloride penetration depth, and brown zone corresponds to the chloride-free zone. In the 1990s, Otsuki et al. [13] investigated different chemical indicators, their concentrations at the color change boundary, and the accuracy of the AgNO3 method. They found that among silver nitrate, lead nitrate and thallium nitrate solutions, silver nitrate showed the clearest color change boundary. Four silver nitrate solutions with different concentrations, i.e. 0.05, 0.1, 0.2, 0.3, and 0.4 mol/L, were sprayed onto the freshly fractured surface of concrete samples containing chloride, respectively. Based on the brightness of color appearing on the concrete surface, they recommended the use of 0.1 mol/L silver nitrate solution. The amount of AgCl and Ag2O formed at the color change boundary from the AgNO3 method depends on the ratio of Cl to OH in the concrete pore solution. Yuan et al. [14] studied the color of AgCl, Ag2O and their mixtures, as shown in Fig. 4. Pure AgCl is silvery and pure Ag2O is dark brown. In the presence of a small amount of Ag2O, the color of the mixture could change very significantly from silvery to brown, which is very close to the color of concrete. This will produce a color change boundary. 2.4. Comparison of the three methods As described above, it can be found that two compounds with different colors form at the color change boundary of AgNO3-based colorimetric methods. The visible color of the color change boundary depends on the relative amount of the two compounds. AgNO3 method colorimetric mechanism is the simplest among the three methods Both AgNO3 + fluoresceine and AgNO3 + K2CrO4 methods need longer time to obtain a better coloring effect at the color change boundary. Thus, AgNO3 method can be conducted more simply and easily than the other two methods. Although in actual application, the color change boundary in the AgNO3 + K2CrO4 method could be more clearly visible than that in AgNO3 method, AgNO3 method can make clear color change boundary in the most cases [11]. Therefore, it is the most practical method for measuring chloride penetration in concrete. The typical colorimetric pictures from the three methods are shown in Fig. 3 [14]. The color change boundary between chloride-containing and so-called ‘chloride-free’ zones can be observed from the three methods. However, AgNO3 + fluoresceine method did not give a very clear boundary. Among the three colorimetric methods, color change depths measured by AgNO3 + fluoresceine
Table 2 Chemical reactions occurring during measurement of AgNO3 + fluoresceine method. Chemical reactions in zone with chloride
2.2. AgNO3 + K2CrO4 method
Ag+ + Cl ? AgCl (white) Ag+ + OH ? AgOH ? Ag2O
Since the 1980s, some researchers have used the AgNO3 + K2CrO4 method to measure chloride ion penetration in concrete [6–12]. In this method, a 0.1 mol/L AgNO3 solution with pH = 3–5 is sprayed onto a cross-section of concrete specimen first, then a K2CrO4 solution (5% by mass) is sprayed onto the cross-section after about 1 h of natural drying.
Excessive Cl: AgClCl + F1 (yellowish green) Excessive Ag+: AgClAg+ + F1 ? AgClAgF1 (pink)
Chemical reactions in chloridefree zone Ag+ + OH ? AgOH ? Ag2O (brown)
F. He et al. / Construction and Building Materials 26 (2012) 1–8
3
Fig. 1. Chemical reactions of AgNO3 + K2CrO4 method in chloride-containing zone.
Fig. 2. Chemical reactions of AgNO3 + K2CrO4 method in chloride-free zone.
Table 3 Chemical reactions in AgNO3 method. Chemical reactions chloride-containing zone
Chemical reactions in chloride-free zone
Ag+ + Cl ? AgCl (white)
Ag+ + OH ? AgOH ? Ag2O (brown)
Ag+ + OH ? AgOH ? Ag2O
[11,12,15,16]. Therefore, Xd measured by the AgNO3-based colorimetric methods should be called as color change depth. In fact, there exists steepness of chloride profile in concrete penetrated by chloride ions. Although electrical migration theoretically gives rectangular penetration profile, some factors such as coarse aggregates and leaking of solution may change the theoretical profile. AgNO3-based colorimetric methods cannot detect chloride penetration profile, but a color change boundary corresponding to a critical chloride ion concentration.
3. Measurement of chloride penetration depth method and AgNO3 method are similar. When AgNO3 + K2CrO4 method is applied, the color change depth was slightly larger than that derived from the AgNO3 method [11]. In AgNO3 + K2CrO4 method, the solution is acidified to avoid the influence of hydroxyl ions. This probably explains the larger penetration from the use of AgNO3 + K2CrO4 method. There are some similarities and differences among the three methods, as summarized in Table 4. The chloride-free zone is not really chloride-free. It may contain a low chloride content, which does not cause significant Cl-related reactions but results in different color from that of the chloride-containing zone. In the AgNO3-based colorimetric methods, chloride ion penetration depth Xd refers to the average distance between the chloride-exposed surface and the color change boundary. However, color change boundary is not really the boundary of area containing chloride and no chloride zone. It is only a curve near to the actual penetration front of chloride
NT Build 492 [17] recommends using slide caliper and a suitable ruler to measure the penetration depths from the centre to both edges at intervals of 10 mm after the chloride migration testing. Seven depths should be measured and an average of the seven measurements is used for calculating chloride migration coefficients. Special cares should be taken during the penetration depth measurement: (1) if the penetration front to be measured is obviously blocked by aggregate, move the measurement to the nearest front where there is no significant blocking by aggregate or, alternatively, ignore this depth if there are more than five valid depths.
4
F. He et al. / Construction and Building Materials 26 (2012) 1–8
Fig. 3. Typical colorimetric pictures of the three methods (a) and (b) from the Ref. [11], (c) was taken by the authors.
Fig. 4. Pictures of AgCl and Ag2O mixtures [14].
Table 4 Comparisons of color change boundary among the three colorimetric methods. Items
AgNO3 + fluoresceine method
AgNO3 + K2CrO4 method
Two colors
Pink and dark brown in the case of excessive Ag+ Low
Light yellow and brick red in the case of excessive K2CrO4 solution High
Visibility of color change boundary Factors influencing visibility of the color change boundary
yellowish green and dark brown in the case of excessive ClHigh
1. Interval between spraying the two solutions and sprayed amount of second solution affect visibility of the color change boundary, temperature and environmental humidity affect optimum spraying interval 2. Spraying too much K2CrO4 solution could wash precipitated Ag2CrO4 away and result in a dark yellow color (yellow + a little brick red) in chloride-free area, which affects the clearness of the boundary
(2) If there is a significant defect in the specimen which results in a penetration front much larger than the average, ignore this front as indicative of the penetration depth. (3) To obviate the edge effect due to a non-homogeneous degree of saturation or possible leakage, do not make any depth measurements in the zone within about 10 mm from the edge. Sometimes, due to the interference of aggregate, it is not easy to capture more than five valid penetration depths in practice. In the draft standard [9], it requires at least four measurements. Image analysis has also been used to measure the chloride penetration depths [11,12,18]. The average chloride penetration depth is calculated using the following equation:
Xd ¼
SCl L S
AgNO3 alone method Light red and brick red in the case of excessive AgNO3 solution Low
ð1Þ
where SCl is the area of the chloride-penetrated zone, S and L are the area and the thickness of the specimen respectively.
White and brown
Relatively high 1. Concrete with low alkalinity cause low visibility or disappearance of color change boundary 2. Concrete with dark grey color may produce a relatively low visible color change boundary
In the case of image analysis, the subjective feature in the measurement of the penetration front is reduced. However, there still exist some disadvantages for the image analysis as followings: (1) the image analysis measure average distance between penetrated surface to color change boundary, it cannot avoid effect of edge and coarse aggregates of concrete; (2) the image analysis needs calculation of area of chloride penetrated zone using certain software, which depends on comparison between colors at two sides of boundary. Different software will produce different error; (3) image analysis is influenced by roughness of fractured surface, to its better use, the fractured surface needs dry-cutting. This might help to avoid the errors associated with the coarse surface [11] and (4) image analysis may be influenced by color variation of concrete and the precipitated reaction products on the concrete surface. Although caliper and ruler method cannot avoid vision error of 0.5 mm, which maybe produce a little problem on application.
5
F. He et al. / Construction and Building Materials 26 (2012) 1–8
According to error evaluation [19,20], depth variation of 0.5 mm only caused a small variation of non-steady state migration coefficient. Therefore, caliper and ruler method can be proper method to measure penetration depth. 4. Factors affecting chloride concentration at the color change boundary Many researchers have measured the chloride concentration at the color change boundary, but obtained very controversial results. Experimental details and results from various publications are summarized in Table 5. As can be seen from Table 5, the chloride concentrations at the color change boundary is 0.01% by cement using AgNO3 + fluoresceine method [21], 0.1–0.4% by the mass of cement using AgNO3 + K2CrO4 method [11,12], and 0.28–1.69% by the mass of cement or 0.072–0.714 mol/L in pore solution using AgNO3 method [13,22–27]. The reasons why chloride concentration at color change boundary can be explained as followings. 4.1. Parameters of colorimetric reactions He [20] investigated chloride concentration at the color change boundary of AgNO3 method. They calculated the Cd according to Ag+–Cl–OH+–H2O system. When Ag+ can react with all OH and Cl, chloride concentration at color change boundary can be expressed as the following equation:
C critCl ¼ 1:6C OH
ð2Þ
+
When Ag can react with only partial OH and Cl , chloride concentration at color change boundary can be expressed as the following equation:
C critCl ¼ 0:00695C OH þ 0:608C Agþ V Agþ =V OH þCl
ð3Þ
where C Agþ and C OH are the mole concentrations of Ag+ and OH-.VAg+ is the volume of AgNO3 solution added to the NaOH + NaCl solution, V OH þCl is volume of NaOH + NaCl solution.
Calculated range of the C critCl is between 0.03–0.96 mol/L in pore solution (Cpd), or 0.011–2.27% by the mass of binder (Cbd) [20]. This is in agreement with results in publications, 0.01– 1.69% by the mass of binder and 0.072–0.714 mol/L in pore solution. As can be seen from Eqs. (2) and (3), concrete alkalinity, sprayed amount and concentration of AgNO3 solution and pore solution volume of concrete can influence chloride concentration at color change boundary. The smaller the volume of sprayed AgNO3 solution, the lower the C critCl is. To obtain a lower C critCl , AgNO3 solution should be sprayed as less as possible, provided that the surface of the cement-based materials is completely wetted by the AgNO3 solution. 0.1 mol/L AgNO3 solution gives the most clear boundary color, it was suggested to use for the indicator. Therefore, a proper sprayed volume of 0.1 mol/L AgNO3 solution should be determined. However, this important point was not mentioned at all times since the AgNO3 method was applied. 4.2. Sampling methods Since concrete is a heterogeneous material, some factors such as aggregates and defects, etc. will result in an irregular color change boundary, as shown in Fig. 5. Sampling at the color change boundary for chloride analysis may also be responsible for the variation in the chloride concentration. Even the boundary is a regular line, the method of sampling (e.g. cutting or drilling) and the thickness of sampling can also affect the results. Different sampling methods and sampling thickness have been used in the published literatures, which will result in the difference in reported chloride concentration at the color change boundary. From the angle of operation, drilling is better than cutting or grinding. This is because drilling is carried out in the central region of the specimens and does not consider effect of specimen edge. When considering homogeneity of sampling, especially at color change boundary, cutting or grinding is better than drilling. This is because drilling is just from one or several locations, which may not be very representative. This may be observed from Fig. 5. However, cutting or grinding is from the entire surface within central region, which is
Table 5 Summary of chloride concentrations at the color change boundary. No.
Method of introducing chloride
Indicators
Sampling method
Method for measuring free chloride
Method for measuring total chloride
Free chloride
Total chloride
References
1
Diffusion, migration, and exposure in field
0.1 N AgNO3 and 0.1 N AgNO3 + K2CrO4 solution
Water extraction method
Acid extraction method
[11,12]
Immersing in NaCl solution
0.1 N AgNO3
Water extraction method
Fluorescent X-ray
0.1–0.4% by mass of cement 0.15% by cement
0.2–1% by mass of cement
2
Grinding layer by layer Splitting layer by layer
0.4–0.5% by cement
[13]
3
Mixing water with NaCl ASTM C1202
0.1 N AgNO3 + fluorescence 0.1 N AgNO3 + fluorescence solution 0.1 N AgNO3
4
0.01% by cement Not described
5
Immersing in NaCl solution
6
Immersing in NaCl solution
0.1 N AgNO3
Drilling in 2 mm steps
7
Wetting and drying ponding Electrical migration and nature diffusion
0.05-N AgNO3
Drilling at 2 mm steps Grinding layer by layer Grinding layer by layer
8
9
Electrical migration
0.1 N AgNO3 solution
0.1 and 0.035 N AgNO3 solution
Drilling in 2 mm steps
Acid extraction method Ion chromatography technique Ion chromatography technique
Water extraction method Water extracton method
UV spectrophotometer method Acid extraction
[21] 1.13 ± 1.4% by mass of cement
[22]
0.84–1.69% by mass of binder
[23]
0.28–1.41% by mass of binder
[24]
0.071– 0.714 mol/L 0.072– 0.142 mol/L
0.5–1.5% by mass of cement 0.019%–0.173% by mass of concrete
[25] [26]
[27]
6
F. He et al. / Construction and Building Materials 26 (2012) 1–8
2erf
1
C d qffiffiffiffiffiffiffiffiffi Dapp ¼ B 1 Cs
ð6Þ
Eq. (5) can be written as Eq. (7):
pffiffi Xd ¼ B t
Fig. 5. Pictures of the color change boundary of concrete.
more representative. As shown in Table 5, the chloride concentration range from grinding or cutting is much smaller than that from drilling. It seems that cutting or grinding is more suitable for sampling. On the other hand, when the same sampling method was used, Cd still changes in a large range. This means that sampling method cannot be mainly responsible for the broad range of Cd. 4.3. Methods for free chloride measurement Many methods, such as pore solution expression, water extraction method (water soluble chloride), alkaline solution extraction method and nuclear magnetic resonance (NMR), have been used to determine the free chloride concentration. Both ion chromatography and nuclear magnetic resonance techniques need special and costly equipment. Among these methods, the results from pore solution expression are close to the real free chloride concentration [28–30]. However, this method needs special expression apparatus, and is difficult to obtain enough pore solution from concrete with low water-to-cement ratio. The water extraction method has been often used for analyzing chloride content of the pore solution of harden concrete owing to its simple operation [28,31–33]. He [20] confirmed that the water extraction method can be used as an effective method to analyze the chloride content at the color change boundary. However, it is important to control experimental parameters. Because experimental results derived from different extraction parameters maybe produce large difference. Many researchers [11–13,26,27] used different water extraction parameters for measuring free chloride concentration at the color change boundary, which possibly partly responsible for the Cd with large range. 5. Applications of AgNO3 colorimetric methods 5.1. Measurement of chloride non-steady state diffusion Chloride non-steady state diffusion can be described by Fick’s second law, whose analytical solution is expressed as the following equation:
"
!# Xd Cðx; tÞ ¼ C s 1 erf pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 Dapp t
ð4Þ
Fick’s second law is the theoretical base of NT Build 443 [34]. In NT Build 443, CS and Dapp are obtained by fitting the experimental chloride profile to Eq. (2) by means of a non-linear regression analysis using the least square method. 5.1.1. Measurement of chloride penetration kinetics Eq. (4) can be written as follows:
X d ¼ 2erf
1
C d qffiffiffiffiffiffiffiffiffiffiffi Dapp t 1 Cs
ð5Þ
If CS, Cd and Dapp change in a small range, then Eq. (6) is obtained:
ð7Þ
Eq. (7) can be called as equation of chloride penetration kinetics. He et al. [11,12,20] found p that ffiffi there was relatively good linear correlation between Xd and t for certain concrete. This means that B may be almost a constant for certain concrete. For existing concrete structures, periodical spray tests can be easily performed for assessing Xd and its evolution vs. time, which can provide the kinetics of chloride penetration. These data can be regarded as useful monitoring parameters for the residual life prediction of existing concrete structures. 5.1.2. Measurement of chloride apparent diffusion coefficient Only if there is no chloride binding, or the bound and free chlorides follow a simple liner relationship, Dapp in Eq. (4) could be calculated according to the total chloride [19]. In fact, both free CS and Cd should be used in Eq. (4). Profile method is time-consuming and laborious. If an AgNO3based colorimetric method is conducted to measure the Cd and Xd, Eq. (4) can be rearranged as follows:
0
Dapp
12 X d pffiffiA ¼@ 1 2erf 1 CCds t
ð8Þ
Baroghel-Bouny et al. [12] compared the results between the apparent diffusion coefficients (DN) obtained from the profile method [34] and those (DA) obtained by Eq. (3), in which penetration depths were determined from the AgNO3 colorimetric method and Cd/CS = 0.14 was used based on a previous publication [35]. However, a good accord of DA with DN was not observed. Chiang and Yang [36] took chloride concentration in contact solution as CS (CS = 0.53 mol/L), associated with Cd = 0.07 mol/L [19], they found that there was a good linear relationship between DA and DN. He’s investigation results [20] maybe explain above contradiction. Based on proper AgNO3 sprayed amount and thus a small measured Cd, He [20] found that when chloride concentrations in contact solution and first 2 mm layer were close, DA (CS is chloride concentration in contact solution) and DN are very close. Variation of DA increased with Cd. Therefore, smaller Cd can measure more accurate DA. This means that depending on determination of CS, AgNO3 colorimetric method can rapidly and relatively accurately measure chloride apparent coefficient. Compared with AgNO3 colorimetric method, profile method gives more reliable and convincing results owing to the variation of CS and Cd used in the colorimetric methods. However, if chloride color change depth could be measured by AgNO3-based colorimetric methods before profile grinding, it will be helpful for determining thickness and total depth of sampling. Once the chloride color change depth is measured, the DA can be quickly calculated, which can be a quick estimation of DN from the profile method. This is very useful for field measurement of Dapp and estimation of remained service life of steel reinforced concrete. 5.2. Measurement of non-steady electrical migration NT Build 492 [17] is one of the most widely used methods for evaluating the resistance of concrete to chloride penetration. In this method, a 0.1 mol/L AgNO3 is used to measure the chloride penetration depth after the non-steady-state migration testing, and then the non-steady-state chloride migration coefficient (Dnssm) is calculated according to two equations which include Cd and Xd as parameters. However, when Cd is taken as 0.07 mol/L,
F. He et al. / Construction and Building Materials 26 (2012) 1–8
the measured Cd varying in a large range (calculated range of 0.03– 0.96 mol/L) may result in large error. According to error evaluation [20], when Xd < 10 mm, possible error of Dnssm caused by Cd can reach above 40% and when Xd < 5 mm, the possible error can reach above 100%. This is a bad news for high performance concrete with low color change depth after non-steady electrical migration. When Cd is effectively controlled in small range by reducing amount of sprayed AgNO3 solution, error of Dnssm caused by Cd can be less than 10%, in spite of Xd. Because CS is relatively constant after electrical migration [19], when Cd is effectively controlled within small range by controlling sprayed amount of AgNO3 solution, the use of Cd = 0.07 mol/L causes a little error of Dnssm [20]. Tang’s [19] investigation result indicated that Dnssm was almost a constant when depth was greater than a certain value. He [20] found that the slight variation of Xd measured by AgNO3 colorimetric method had a little effect on Dnssm. 5.3. Evaluation for corrosion risk of steel in concrete Evaluation for corrosion risk of steel in concrete is based on comparison between threshold chloride concentration for steel corrosion (Ccrit) and chloride concentration at color change boundary. The reported threshold chloride concentration with respect to corrosion risk varies greatly [37,38]. The Building Research Establishment [39] has proposed a classification for assessing the risk of corrosion in terms of acid-soluble (or total) chloride contents by the mass of cement: low, less than 0.4%; medium, 0.4–1.0%; and high, greater than 1%. Based on measured range of Cd, Meck et al. [12,24] suggested that owing to range of Cd is generally lower than Ccrit, evaluation for corrosion risk of steel in concrete using AgNO3 colorimetric method can be operated. As above discussed, concrete alkalinity, sprayed amount and concentration of AgNO3 solution and pore solution volume of concrete can influence chloride concentration at color change boundary, measured and calculated Cd change in a large range. Therefore, it is necessary to point out again that a proper sprayed amount of AgNO3 solution with a certain concentration should be determined to obtain a small range of Cd. A smaller Cd will produce more accurate applications.
6. Summary (1) AgNO3-based colorimetric methods are fast and easy to perform. They have been widely applied to measure the chloride penetration depth in field and laboratory. Among the three methods, AgNO3 method can be conducted more rapidly, simply and easily than AgNO3 + K2CrO4 or AgNO3 + fluoresceine method, the three methods give similar measured results. (2) The color change boundary shown by the colorimetric method is not a real boundary between chloride-free zone and chloride-penetrated zone. The boundary has a certain chloride concentration owing to OH ions in pore solution of concrete. The measurement of the chloride ion penetration depth can be carried out on specimens by caliper and ruler method and a ruler or by means of analyzing the digitized image of the specimen. Caliper and ruler method can be proper method to measure penetration depth. (3) Cutting or grinding is more suitable for sampling purpose than drilling since the sample obtained by cutting or grinding is more representative. The water extraction method can be used as an effective method to analyze the chloride content at the color change boundary. However, it is important to control experimental parameters.
7
(4) Analysis of published results indicate that the chloride concentrations at the color change boundary is 0.01% by the mass of cement for AgNO3 + fluoresceine method, and 0.1– 0.4% by the mass of cement for AgNO3 + K2CrO4 method, 0.28–1.69% by the mass of cement or 0.072–0.714 mol/L for the AgNO3 method. Many factors such as concrete alkalinity, sprayed volume and concentration of AgNO3 solution, pore solution volume of concrete, sampling method and methods used for measuring free chloride in concrete can be responsible for the high variability. Among all factors, sprayed volume and concentration of AgNO3 solution are main ones. (5) The smaller the volume of sprayed AgNO3 solution with a certain concentration, the lower the Cd is. 0.1 mol/L AgNO3 solution has most clear color of boundary. To obtain a lower Cd, a proper sprayed volume of 0.1 mol/L AgNO3 solution should be determined. However, this important point was not mentioned at all times since the AgNO3 method was applied. Based on small range of Cd, the method can be potentially to be used as a useful tool within the framework of evaluating chloride penetration in reinforced concrete structure in chloride environments.
Acknowledgments Financial supports from the National Natural Science Foundation of China (51072050) and Xiamen Department of Science and Technology (3502Z20103042) are greatly appreciated. References [1] Stanish K. The migration of chloride ions in concrete. Ph.D. dissertation. Canada: Department of Civil Engineering, University of Toronto; 2002. [2] Li KA. Analytical chemistry. Peking University Press; 2005. [3] Collepardi M, Marcialis A, Turriziani R. Kinetics of penetration of chloride ions into concrete. Il Cemento 1970;67:157–64 [in Italian]. [4] Collepardi M, Marcialis A, Turriziani R. Penetration of chloride ions in cement pastes and in concretes. J Am Ceram Soc 1972;55(10):534–5. [5] Italian Standard 79–28. Determination of the chloride, Rome; 1978. [6] Maultzsch M. Concrete related effects on chloride diffusion (in German). Contribution to Int Coll Chloride Corrosion, Vienna, BAMAmts- und. Mitteilungsblatt 1983;13(3):387–9. [7] Maultzsch M. Effects on cement pastes and concrete of chloride solution impact. Mater Tech (CH) 1984;12(3):83–90 [in German]. [8] Baroghel-Bouny V, Chaussadent T, Henry D. Testing and modeling chloride penetration into concrete – first round robin test. RILEM Technical Committee 178 Report, LCPC; 1999. p. 15. [9] Measurement of the colorimetric front. RILEM Technical Committee 178 Procedure. Preliminary draft; 2003. [10] Frederiksen JM. Testing chloride in structures – an essential part of investigating the performance of reinforced concrete structures. In: Proceedings of COST 521 workshop. Belfast, UK; 2000. [11] Baroghel-Bouny V, Belin P, Maultzsch M, Henry D. AgNO3 spray tests: advantages, weaknesses, and various applications to quantify chloride ingress into concrete. Part I: Non-steady-state diffusion tests and exposure to natural conditions. Mater Struct J 2007;40(8):759–81. [12] Baroghel-Bouny V, Belin P, Maultzsch M, Henry D. AgNO3 spray tests: advantages, weaknesses, and various applications to quantify chloride ingress into concrete. Part II: Non-steady-state migration tests and chloride diffusion coefficients. Mater Struct J 2007;40(8):783–99. [13] Otsuki N, Nagataki S, Nakashita K. Evaluation of AgNO3 solution spray method for measurement of chloride penetration into hardened cementitious matrix materials. ACI Mater J 1992;89(6):587–92. [14] Yuan Q, Shi C, He F, et al. Effect of hydroxyl ions on chloride penetration depth measurement using the colorimetric method. Cem Concr Res 2008;38(10):1177–80. [15] He F, Shi C, Yuan Q, et al. A study on factors influencing chloride concentration at color change boundary of concrete using AgNO3 colorimetric method. J Chin Ceram Soc 2008;36(7):890–5 [in Chinese]. [16] Chen G, Li W, Wang P. Penetration depth and concentration distraction of chloride ions into cement mortar. J Tong Ji Univ 1996;24(3):19–24 [in Chinese]. [17] NT Build 492. Nordtest method, concrete, mortar and cement-based repair materials: Chloride migration coefficient from non-steady-state migration experiments. Espoo, Finland; 1999.
8
F. He et al. / Construction and Building Materials 26 (2012) 1–8
[18] Baroghel-Bouny V, Arnaud S, Henry D, Carcasse‘s M, Que´nard D. Ageing of concretes in natural environments: experimentation for the 21st century. III – Durability properties of concretes measured on laboratory specimens, vol. 241. Bul Lab Ponts Chaus; 2002. p. 13–59. [19] Tang L. Chloride transport in concrete – measurement and prediction. Ph.D. dissertation. Goteborg, Sweden: Department of Building Materials, Chalmers University of Technology; 1996. [20] He F. Measurement of chloride migration in cement-based materials using AgNO3 colorimetric method. Ph.D. dissertation. Changsha, China: School of Civil Engineering and Architecture, Central South University; 2010. [21] Collepardi M. Quick method to determine free and bound chlorides in concrete. In: Nilsson LO, Ollivier JP, editors. Proceedings of the international RILEM workshop on chloride penetration into concrete, Paris; 1995. [22] Andrade C, Castellote M, Alonso C, Gonzalez C. Relation between colorimetric chloride penetration depth and charge passed in migration tests of the type of standard ASTM C1202-91. Cem Concr Res 1999;29(3):417–21. [23] Sirivivatnanon V, Khatri R. Chloride penetration resistance of concrete, presented to concrete. In: Institute of Australia conference getting a lifetime out of concrete structures. Brisbane, Australia, October 1998. [24] Meck E, Sirivivatnanon V. Field indicator of chloride penetration depth. Cem Concr Res 1999;33(8):1113–7. [25] McPolin D, Basheer PAM, Long AE, Grattan KTV, Sun T. Obtaining progressive chloride profiles in cementitious materials. Constr Build Mater 2005;19(9):666–73. [26] Yuan Q. Fundamental studies on test methods for transport of chloride ions in cementitious materials. Ph.D. dissertation. Belgium: Department of Structural Engineering, Gent University; 2009. [27] He F, Shi C, Yuan Q, et al. Factors influencing chloride concentration at the color change boundary using AgNO3 colorimetric method. J Chin Ceram Soc 2008;36(7):890–5 [in Chinese].
[28] Arya C, Newman JB. An assessment of four methods of determining the free chloride content of concrete. Mater Struct J 1990;23:319–30. [29] Barneyback RS, Diamond S. Expression and analysis of pore fluids from hardened cement pastes and mortars. Cem Concr Res 1986;16(5):760–70. [30] Barneyback RS, Diamond S. Expression and analysis of pore fluids from hardened cement pastes and mortars. Cem Concr Res 1981;11(2):279–85. [31] Arya C, Buenfeld NR, Newman JB. Assessment of simple methods of determining the free chloride ion content of cement paste. Cem Concr Res 1987;17(6):907–18. [32] Glass GK, Wang Y, Buenfeld NR. An investigation of experimental methods used to determine free and total chloride contents. Cem Concr Res 1996;26(9):1443–9. [33] Haque MN, Kayyali OA. Free and water soluble chloride in concrete. Cem Concr Res 1995;25(3):531–42. [34] NT Build 443. Nordtest method. Accelerated chloride penetration. Espoo, Finland; 1995. [35] Tang L. Electrically accelerated methods for determining chloride diffusivity in concrete-current development. Mag Concr Res 1996;48(176):173–9. [36] Chiang CT, Yang CC. Relation between the diffusion characteristic of concrete from salt ponding test and accelerated chloride migration test. Mater Chem Phys 2007;106(2–3):240–6. [37] Thomas MDA, Matthews JD. Performance of PFA concrete in a marine environment – 10-year results. Cem Concr Compos 2004;26(1):5–20. [38] Alonso C, Andrade C, Castellote M, Castro P. Chloride threshold values to depassivate reinforcing bars embedded in a standardized OPC mortar. Cem Concr Res 2000;30:1047–55. [39] Everett LH, Treadaway KWJ. Deterioration due to corrosion in reinforced steel concrete. Garston (United Kingdom): Building Research Establishment; 1980. Information Paper IP 12/80.
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Review
AgNO3-based colorimetric methods for measurement of chloride penetration in concrete Fuqiang He a, Caijun Shi b,⇑, Qiang Yuan c, Changping Chen a, Keren Zheng c a
Department of Civil Engineering and Architecture, Xiamen University of Technology, Xiamen 361024, China College of Civil Engineering, Hunan University, Changsha 410082, China c School of Civil Engineering and Architecture, Central South University, Changsha 410075, China b
a r t i c l e
i n f o
Article history: Received 16 January 2011 Received in revised form 1 May 2011 Accepted 8 June 2011 Available online 14 July 2011 Keywords: AgNO3-based colorimetric methods Chloride penetration Concrete
a b s t r a c t This paper reviews reaction mechanisms of AgNO3, AgNO3 + K2CrO4 and AgNO3 + fluoresceine colorimetric methods, and their applications for measurement of chloride ion penetration in concrete. Among the three methods, AgNO3 method is most widely applied because it gives similar results but simpler and faster than the other two methods. AgNO3-based colorimetric methods can potentially measure two variables, chloride ion penetration depth and chloride concentration at the color change boundary. Reported chloride ion concentrations at the color change boundary (Cd) measured or calculated by some researchers vary over a broad range due to many factors, such as sampling procedure, chloride ion analysis method, alkalinity of concrete, sprayed volume and concentration of AgNO3 solution, pore solution volume and methods for measuring free chloride concentration in concrete, etc., in which sprayed volume and concentration of AgNO3 solution are main factors. The smaller the volume of sprayed AgNO3 solution with a certain concentration, the lower the Cd is. 0.1 mol/L AgNO3 solution has most clear color of boundary. To obtain a lower Cd, a proper sprayed volume of 0.1 mol/L AgNO3 solution should be determined. However, this important point was not mentioned at all times since the AgNO3 method was applied. Based on small range of Cd, AgNO3 method can be a useful method for rapidly evaluating chloride ion penetration in reinforced concrete structures in chloride environments. Ó 2011 Elsevier Ltd. All rights reserved.
Contents 1. 2.
3. 4.
5.
6.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AgNO3 based colorimetric methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. AgNO3 + fluoresceine method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. AgNO3 + K2CrO4 method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. AgNO3 method. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Comparison of the three methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Measurement of chloride penetration depth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Factors affecting chloride concentration at the color change boundary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Parameters of colorimetric reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Sampling methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Methods for free chloride measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Applications of AgNO3 colorimetric methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1. Measurement of chloride non-steady state diffusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1.1. Measurement of chloride penetration kinetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1.2. Measurement of chloride apparent diffusion coefficient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Measurement of non-steady electrical migration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3. Evaluation for corrosion risk of steel in concrete . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
⇑ Corresponding author. E-mail address: [email protected] (C. Shi). 0950-0618/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.conbuildmat.2011.06.003
2 2 2 2 2 2 3 5 5 5 6 6 6 6 6 6 7 7 7 7
2
F. He et al. / Construction and Building Materials 26 (2012) 1–8
1. Introduction Chloride-induced corrosion of steel reinforcement is a dominant factor on durability of concrete structures exposed to marine environments and/or de-icing salts. It is well known that high alkaline medium provided by cement matrix in concrete maintains steel reinforcements in a passive state and prevents them from corrosion. However, a local breakdown of the passivating layer occurs when a sufficient amount of chloride ions reaches the steel reinforcement; corrosion is then enabled. In the context of chlorideinduced corrosion, it is reasonable to consider the service life of reinforced concrete structures in two stages: the first stage is the time for a critical chloride concentration to reach the surface of steel reinforcement, and the second stage is the subsequent propagation extending onward to the time when the structure is damaged by the steel corrosion beyond acceptable limits. Due to the relatively fast development of the second stage and for the sake of safety, the duration of the first stage is sometimes regarded as the design service life of steel-reinforced concrete structures. The chloride ion penetration front, varying with time and environment, can be used to monitor the movement of chloride ions in concrete and for service life prediction [1]. Analysis of chloride ion content in concrete can be conducted by the so-called profile method, which requires cutting or drilling, grinding and chemical analysis of concrete samples. The profile method requires a series of special equipment and consumes much time. On the other hand, AgNO3-based colorimetric methods can measure chloride ion penetration depth in concrete quickly and simply. It can be a useful method for rapid measurement of chloride ion penetration in reinforced concrete structures in chloride environments. This paper reviews three AgNO3-based colorimetric methods, factors affecting the color change and chloride ion concentrations at the color change boundary, and their applications. 2. AgNO3 based colorimetric methods Since the 1970s, three AgNO3-based colorimetric methods, i.e., AgNO3 + fluoresceine, AgNO3 + K2CrO4 and AgNO3 method, have been proposed to measure the chloride ion penetration depth in concrete in the field and laboratory. Both AgNO3 + fluoresceine and AgNO3 + K2CrO4 methods were derived from the methods for analyzing free chloride ions in solutions [2], as described in Table 1. The three methods are described in detail in the following sections. 2.1. AgNO3 + fluoresceine method In the 1970s, Collepardi et al. [3,4] developed a colorimetric method to determine free chloride content in concrete, in which a fluoresceine solution (1 g/L in a 70% solution of ethyl alcohol in water) was first sprayed onto a cross-section of chloride-penetrated concrete and then 0.1 mol/L AgNO3 aqueous solution was sprayed onto the cross-section. Immediately after spraying the AgNO3 solution onto concrete surface, Ag2O and AgCl formed respectively by reactions between Ag+ and OH or Cl. Fluoresceine (HF1) is an organic weak acid, which dissociates into yellowish green ion F1 in a solution. When Cl is excessive, AgCl absorbs Cl and possesses negative charge. F1 in the solution is repelled by AgCl. In this case, the solution is yellowish. When excess AgNO3 is sprayed, Ag+ is absorbed by AgCl, AgCl has positive charge, which absorbs F1 in the solution and turns the solution into pink. When this method is applied for measuring chloride ion penetration in concrete, the chemical reactions occurring during this process are shown in Table 2. In most cases, the AgNO3 solution may be excessive. The chloride-containing zone appears to be dark pink due to the formation of pink AgClAgF1 on grey concrete. The ‘chloride-free’ zone turns to be dark brown due to the formation of brown Ag2O on grey concrete. This method has been standardized as Italian Standard 79-28 [5].
Table 1 Use of AgNO3 + fluoresceine and AgNO3 + K2CrO4 in silver measuring methods [2]. Method
AgNO3 + K2CrO4
AgNO3 + fluoresceine
Indicator Titration solution pH value of the solution Color change at titration endpoint Year of origin Developer
K2CrO4 AgNO3 solution 6.5–10.5 White ? brick red 1856 Mohr
Fluoresceine AgNO3 solution 7–8 Yellowish green ? pink 1924 Fajans
Once the potassium chromate aqueous solution is sprayed, the chloridecontaining zone remains yellow due to the formation of white AgCl and left yellow K2CrO4 solution. A small amount of white AgCl and ruby red Ag2CrO7, may also form, but their amounts are too small to affect the color significantly. Thus, the chloride-containing zone may remain yellowish in most cases. However, the chloride-free zone turns into a red-brown color due to the formation of brown Ag2O + brick red Ag2CrO4 and/or ruby red Ag2CrO7. A series of chemical reactions happening in both chloride-containing and chloride-free zones in this colorimetric method are summarized in Figs. 1 and 2. 2.3. AgNO3 method AgNO3 method involves spraying only an AgNO3 solution onto a freshly fractured concrete cross-section, which leads to the formation of white and brown zones with a clear color change boundary. Chemical reactions involved in this method are summarized in Table 3. Usually, the depth of the white zone is regarded as the chloride penetration depth, and brown zone corresponds to the chloride-free zone. In the 1990s, Otsuki et al. [13] investigated different chemical indicators, their concentrations at the color change boundary, and the accuracy of the AgNO3 method. They found that among silver nitrate, lead nitrate and thallium nitrate solutions, silver nitrate showed the clearest color change boundary. Four silver nitrate solutions with different concentrations, i.e. 0.05, 0.1, 0.2, 0.3, and 0.4 mol/L, were sprayed onto the freshly fractured surface of concrete samples containing chloride, respectively. Based on the brightness of color appearing on the concrete surface, they recommended the use of 0.1 mol/L silver nitrate solution. The amount of AgCl and Ag2O formed at the color change boundary from the AgNO3 method depends on the ratio of Cl to OH in the concrete pore solution. Yuan et al. [14] studied the color of AgCl, Ag2O and their mixtures, as shown in Fig. 4. Pure AgCl is silvery and pure Ag2O is dark brown. In the presence of a small amount of Ag2O, the color of the mixture could change very significantly from silvery to brown, which is very close to the color of concrete. This will produce a color change boundary. 2.4. Comparison of the three methods As described above, it can be found that two compounds with different colors form at the color change boundary of AgNO3-based colorimetric methods. The visible color of the color change boundary depends on the relative amount of the two compounds. AgNO3 method colorimetric mechanism is the simplest among the three methods Both AgNO3 + fluoresceine and AgNO3 + K2CrO4 methods need longer time to obtain a better coloring effect at the color change boundary. Thus, AgNO3 method can be conducted more simply and easily than the other two methods. Although in actual application, the color change boundary in the AgNO3 + K2CrO4 method could be more clearly visible than that in AgNO3 method, AgNO3 method can make clear color change boundary in the most cases [11]. Therefore, it is the most practical method for measuring chloride penetration in concrete. The typical colorimetric pictures from the three methods are shown in Fig. 3 [14]. The color change boundary between chloride-containing and so-called ‘chloride-free’ zones can be observed from the three methods. However, AgNO3 + fluoresceine method did not give a very clear boundary. Among the three colorimetric methods, color change depths measured by AgNO3 + fluoresceine
Table 2 Chemical reactions occurring during measurement of AgNO3 + fluoresceine method. Chemical reactions in zone with chloride
2.2. AgNO3 + K2CrO4 method
Ag+ + Cl ? AgCl (white) Ag+ + OH ? AgOH ? Ag2O
Since the 1980s, some researchers have used the AgNO3 + K2CrO4 method to measure chloride ion penetration in concrete [6–12]. In this method, a 0.1 mol/L AgNO3 solution with pH = 3–5 is sprayed onto a cross-section of concrete specimen first, then a K2CrO4 solution (5% by mass) is sprayed onto the cross-section after about 1 h of natural drying.
Excessive Cl: AgClCl + F1 (yellowish green) Excessive Ag+: AgClAg+ + F1 ? AgClAgF1 (pink)
Chemical reactions in chloridefree zone Ag+ + OH ? AgOH ? Ag2O (brown)
F. He et al. / Construction and Building Materials 26 (2012) 1–8
3
Fig. 1. Chemical reactions of AgNO3 + K2CrO4 method in chloride-containing zone.
Fig. 2. Chemical reactions of AgNO3 + K2CrO4 method in chloride-free zone.
Table 3 Chemical reactions in AgNO3 method. Chemical reactions chloride-containing zone
Chemical reactions in chloride-free zone
Ag+ + Cl ? AgCl (white)
Ag+ + OH ? AgOH ? Ag2O (brown)
Ag+ + OH ? AgOH ? Ag2O
[11,12,15,16]. Therefore, Xd measured by the AgNO3-based colorimetric methods should be called as color change depth. In fact, there exists steepness of chloride profile in concrete penetrated by chloride ions. Although electrical migration theoretically gives rectangular penetration profile, some factors such as coarse aggregates and leaking of solution may change the theoretical profile. AgNO3-based colorimetric methods cannot detect chloride penetration profile, but a color change boundary corresponding to a critical chloride ion concentration.
3. Measurement of chloride penetration depth method and AgNO3 method are similar. When AgNO3 + K2CrO4 method is applied, the color change depth was slightly larger than that derived from the AgNO3 method [11]. In AgNO3 + K2CrO4 method, the solution is acidified to avoid the influence of hydroxyl ions. This probably explains the larger penetration from the use of AgNO3 + K2CrO4 method. There are some similarities and differences among the three methods, as summarized in Table 4. The chloride-free zone is not really chloride-free. It may contain a low chloride content, which does not cause significant Cl-related reactions but results in different color from that of the chloride-containing zone. In the AgNO3-based colorimetric methods, chloride ion penetration depth Xd refers to the average distance between the chloride-exposed surface and the color change boundary. However, color change boundary is not really the boundary of area containing chloride and no chloride zone. It is only a curve near to the actual penetration front of chloride
NT Build 492 [17] recommends using slide caliper and a suitable ruler to measure the penetration depths from the centre to both edges at intervals of 10 mm after the chloride migration testing. Seven depths should be measured and an average of the seven measurements is used for calculating chloride migration coefficients. Special cares should be taken during the penetration depth measurement: (1) if the penetration front to be measured is obviously blocked by aggregate, move the measurement to the nearest front where there is no significant blocking by aggregate or, alternatively, ignore this depth if there are more than five valid depths.
4
F. He et al. / Construction and Building Materials 26 (2012) 1–8
Fig. 3. Typical colorimetric pictures of the three methods (a) and (b) from the Ref. [11], (c) was taken by the authors.
Fig. 4. Pictures of AgCl and Ag2O mixtures [14].
Table 4 Comparisons of color change boundary among the three colorimetric methods. Items
AgNO3 + fluoresceine method
AgNO3 + K2CrO4 method
Two colors
Pink and dark brown in the case of excessive Ag+ Low
Light yellow and brick red in the case of excessive K2CrO4 solution High
Visibility of color change boundary Factors influencing visibility of the color change boundary
yellowish green and dark brown in the case of excessive ClHigh
1. Interval between spraying the two solutions and sprayed amount of second solution affect visibility of the color change boundary, temperature and environmental humidity affect optimum spraying interval 2. Spraying too much K2CrO4 solution could wash precipitated Ag2CrO4 away and result in a dark yellow color (yellow + a little brick red) in chloride-free area, which affects the clearness of the boundary
(2) If there is a significant defect in the specimen which results in a penetration front much larger than the average, ignore this front as indicative of the penetration depth. (3) To obviate the edge effect due to a non-homogeneous degree of saturation or possible leakage, do not make any depth measurements in the zone within about 10 mm from the edge. Sometimes, due to the interference of aggregate, it is not easy to capture more than five valid penetration depths in practice. In the draft standard [9], it requires at least four measurements. Image analysis has also been used to measure the chloride penetration depths [11,12,18]. The average chloride penetration depth is calculated using the following equation:
Xd ¼
SCl L S
AgNO3 alone method Light red and brick red in the case of excessive AgNO3 solution Low
ð1Þ
where SCl is the area of the chloride-penetrated zone, S and L are the area and the thickness of the specimen respectively.
White and brown
Relatively high 1. Concrete with low alkalinity cause low visibility or disappearance of color change boundary 2. Concrete with dark grey color may produce a relatively low visible color change boundary
In the case of image analysis, the subjective feature in the measurement of the penetration front is reduced. However, there still exist some disadvantages for the image analysis as followings: (1) the image analysis measure average distance between penetrated surface to color change boundary, it cannot avoid effect of edge and coarse aggregates of concrete; (2) the image analysis needs calculation of area of chloride penetrated zone using certain software, which depends on comparison between colors at two sides of boundary. Different software will produce different error; (3) image analysis is influenced by roughness of fractured surface, to its better use, the fractured surface needs dry-cutting. This might help to avoid the errors associated with the coarse surface [11] and (4) image analysis may be influenced by color variation of concrete and the precipitated reaction products on the concrete surface. Although caliper and ruler method cannot avoid vision error of 0.5 mm, which maybe produce a little problem on application.
5
F. He et al. / Construction and Building Materials 26 (2012) 1–8
According to error evaluation [19,20], depth variation of 0.5 mm only caused a small variation of non-steady state migration coefficient. Therefore, caliper and ruler method can be proper method to measure penetration depth. 4. Factors affecting chloride concentration at the color change boundary Many researchers have measured the chloride concentration at the color change boundary, but obtained very controversial results. Experimental details and results from various publications are summarized in Table 5. As can be seen from Table 5, the chloride concentrations at the color change boundary is 0.01% by cement using AgNO3 + fluoresceine method [21], 0.1–0.4% by the mass of cement using AgNO3 + K2CrO4 method [11,12], and 0.28–1.69% by the mass of cement or 0.072–0.714 mol/L in pore solution using AgNO3 method [13,22–27]. The reasons why chloride concentration at color change boundary can be explained as followings. 4.1. Parameters of colorimetric reactions He [20] investigated chloride concentration at the color change boundary of AgNO3 method. They calculated the Cd according to Ag+–Cl–OH+–H2O system. When Ag+ can react with all OH and Cl, chloride concentration at color change boundary can be expressed as the following equation:
C critCl ¼ 1:6C OH
ð2Þ
+
When Ag can react with only partial OH and Cl , chloride concentration at color change boundary can be expressed as the following equation:
C critCl ¼ 0:00695C OH þ 0:608C Agþ V Agþ =V OH þCl
ð3Þ
where C Agþ and C OH are the mole concentrations of Ag+ and OH-.VAg+ is the volume of AgNO3 solution added to the NaOH + NaCl solution, V OH þCl is volume of NaOH + NaCl solution.
Calculated range of the C critCl is between 0.03–0.96 mol/L in pore solution (Cpd), or 0.011–2.27% by the mass of binder (Cbd) [20]. This is in agreement with results in publications, 0.01– 1.69% by the mass of binder and 0.072–0.714 mol/L in pore solution. As can be seen from Eqs. (2) and (3), concrete alkalinity, sprayed amount and concentration of AgNO3 solution and pore solution volume of concrete can influence chloride concentration at color change boundary. The smaller the volume of sprayed AgNO3 solution, the lower the C critCl is. To obtain a lower C critCl , AgNO3 solution should be sprayed as less as possible, provided that the surface of the cement-based materials is completely wetted by the AgNO3 solution. 0.1 mol/L AgNO3 solution gives the most clear boundary color, it was suggested to use for the indicator. Therefore, a proper sprayed volume of 0.1 mol/L AgNO3 solution should be determined. However, this important point was not mentioned at all times since the AgNO3 method was applied. 4.2. Sampling methods Since concrete is a heterogeneous material, some factors such as aggregates and defects, etc. will result in an irregular color change boundary, as shown in Fig. 5. Sampling at the color change boundary for chloride analysis may also be responsible for the variation in the chloride concentration. Even the boundary is a regular line, the method of sampling (e.g. cutting or drilling) and the thickness of sampling can also affect the results. Different sampling methods and sampling thickness have been used in the published literatures, which will result in the difference in reported chloride concentration at the color change boundary. From the angle of operation, drilling is better than cutting or grinding. This is because drilling is carried out in the central region of the specimens and does not consider effect of specimen edge. When considering homogeneity of sampling, especially at color change boundary, cutting or grinding is better than drilling. This is because drilling is just from one or several locations, which may not be very representative. This may be observed from Fig. 5. However, cutting or grinding is from the entire surface within central region, which is
Table 5 Summary of chloride concentrations at the color change boundary. No.
Method of introducing chloride
Indicators
Sampling method
Method for measuring free chloride
Method for measuring total chloride
Free chloride
Total chloride
References
1
Diffusion, migration, and exposure in field
0.1 N AgNO3 and 0.1 N AgNO3 + K2CrO4 solution
Water extraction method
Acid extraction method
[11,12]
Immersing in NaCl solution
0.1 N AgNO3
Water extraction method
Fluorescent X-ray
0.1–0.4% by mass of cement 0.15% by cement
0.2–1% by mass of cement
2
Grinding layer by layer Splitting layer by layer
0.4–0.5% by cement
[13]
3
Mixing water with NaCl ASTM C1202
0.1 N AgNO3 + fluorescence 0.1 N AgNO3 + fluorescence solution 0.1 N AgNO3
4
0.01% by cement Not described
5
Immersing in NaCl solution
6
Immersing in NaCl solution
0.1 N AgNO3
Drilling in 2 mm steps
7
Wetting and drying ponding Electrical migration and nature diffusion
0.05-N AgNO3
Drilling at 2 mm steps Grinding layer by layer Grinding layer by layer
8
9
Electrical migration
0.1 N AgNO3 solution
0.1 and 0.035 N AgNO3 solution
Drilling in 2 mm steps
Acid extraction method Ion chromatography technique Ion chromatography technique
Water extraction method Water extracton method
UV spectrophotometer method Acid extraction
[21] 1.13 ± 1.4% by mass of cement
[22]
0.84–1.69% by mass of binder
[23]
0.28–1.41% by mass of binder
[24]
0.071– 0.714 mol/L 0.072– 0.142 mol/L
0.5–1.5% by mass of cement 0.019%–0.173% by mass of concrete
[25] [26]
[27]
6
F. He et al. / Construction and Building Materials 26 (2012) 1–8
2erf
1
C d qffiffiffiffiffiffiffiffiffi Dapp ¼ B 1 Cs
ð6Þ
Eq. (5) can be written as Eq. (7):
pffiffi Xd ¼ B t
Fig. 5. Pictures of the color change boundary of concrete.
more representative. As shown in Table 5, the chloride concentration range from grinding or cutting is much smaller than that from drilling. It seems that cutting or grinding is more suitable for sampling. On the other hand, when the same sampling method was used, Cd still changes in a large range. This means that sampling method cannot be mainly responsible for the broad range of Cd. 4.3. Methods for free chloride measurement Many methods, such as pore solution expression, water extraction method (water soluble chloride), alkaline solution extraction method and nuclear magnetic resonance (NMR), have been used to determine the free chloride concentration. Both ion chromatography and nuclear magnetic resonance techniques need special and costly equipment. Among these methods, the results from pore solution expression are close to the real free chloride concentration [28–30]. However, this method needs special expression apparatus, and is difficult to obtain enough pore solution from concrete with low water-to-cement ratio. The water extraction method has been often used for analyzing chloride content of the pore solution of harden concrete owing to its simple operation [28,31–33]. He [20] confirmed that the water extraction method can be used as an effective method to analyze the chloride content at the color change boundary. However, it is important to control experimental parameters. Because experimental results derived from different extraction parameters maybe produce large difference. Many researchers [11–13,26,27] used different water extraction parameters for measuring free chloride concentration at the color change boundary, which possibly partly responsible for the Cd with large range. 5. Applications of AgNO3 colorimetric methods 5.1. Measurement of chloride non-steady state diffusion Chloride non-steady state diffusion can be described by Fick’s second law, whose analytical solution is expressed as the following equation:
"
!# Xd Cðx; tÞ ¼ C s 1 erf pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 Dapp t
ð4Þ
Fick’s second law is the theoretical base of NT Build 443 [34]. In NT Build 443, CS and Dapp are obtained by fitting the experimental chloride profile to Eq. (2) by means of a non-linear regression analysis using the least square method. 5.1.1. Measurement of chloride penetration kinetics Eq. (4) can be written as follows:
X d ¼ 2erf
1
C d qffiffiffiffiffiffiffiffiffiffiffi Dapp t 1 Cs
ð5Þ
If CS, Cd and Dapp change in a small range, then Eq. (6) is obtained:
ð7Þ
Eq. (7) can be called as equation of chloride penetration kinetics. He et al. [11,12,20] found p that ffiffi there was relatively good linear correlation between Xd and t for certain concrete. This means that B may be almost a constant for certain concrete. For existing concrete structures, periodical spray tests can be easily performed for assessing Xd and its evolution vs. time, which can provide the kinetics of chloride penetration. These data can be regarded as useful monitoring parameters for the residual life prediction of existing concrete structures. 5.1.2. Measurement of chloride apparent diffusion coefficient Only if there is no chloride binding, or the bound and free chlorides follow a simple liner relationship, Dapp in Eq. (4) could be calculated according to the total chloride [19]. In fact, both free CS and Cd should be used in Eq. (4). Profile method is time-consuming and laborious. If an AgNO3based colorimetric method is conducted to measure the Cd and Xd, Eq. (4) can be rearranged as follows:
0
Dapp
12 X d pffiffiA ¼@ 1 2erf 1 CCds t
ð8Þ
Baroghel-Bouny et al. [12] compared the results between the apparent diffusion coefficients (DN) obtained from the profile method [34] and those (DA) obtained by Eq. (3), in which penetration depths were determined from the AgNO3 colorimetric method and Cd/CS = 0.14 was used based on a previous publication [35]. However, a good accord of DA with DN was not observed. Chiang and Yang [36] took chloride concentration in contact solution as CS (CS = 0.53 mol/L), associated with Cd = 0.07 mol/L [19], they found that there was a good linear relationship between DA and DN. He’s investigation results [20] maybe explain above contradiction. Based on proper AgNO3 sprayed amount and thus a small measured Cd, He [20] found that when chloride concentrations in contact solution and first 2 mm layer were close, DA (CS is chloride concentration in contact solution) and DN are very close. Variation of DA increased with Cd. Therefore, smaller Cd can measure more accurate DA. This means that depending on determination of CS, AgNO3 colorimetric method can rapidly and relatively accurately measure chloride apparent coefficient. Compared with AgNO3 colorimetric method, profile method gives more reliable and convincing results owing to the variation of CS and Cd used in the colorimetric methods. However, if chloride color change depth could be measured by AgNO3-based colorimetric methods before profile grinding, it will be helpful for determining thickness and total depth of sampling. Once the chloride color change depth is measured, the DA can be quickly calculated, which can be a quick estimation of DN from the profile method. This is very useful for field measurement of Dapp and estimation of remained service life of steel reinforced concrete. 5.2. Measurement of non-steady electrical migration NT Build 492 [17] is one of the most widely used methods for evaluating the resistance of concrete to chloride penetration. In this method, a 0.1 mol/L AgNO3 is used to measure the chloride penetration depth after the non-steady-state migration testing, and then the non-steady-state chloride migration coefficient (Dnssm) is calculated according to two equations which include Cd and Xd as parameters. However, when Cd is taken as 0.07 mol/L,
F. He et al. / Construction and Building Materials 26 (2012) 1–8
the measured Cd varying in a large range (calculated range of 0.03– 0.96 mol/L) may result in large error. According to error evaluation [20], when Xd < 10 mm, possible error of Dnssm caused by Cd can reach above 40% and when Xd < 5 mm, the possible error can reach above 100%. This is a bad news for high performance concrete with low color change depth after non-steady electrical migration. When Cd is effectively controlled in small range by reducing amount of sprayed AgNO3 solution, error of Dnssm caused by Cd can be less than 10%, in spite of Xd. Because CS is relatively constant after electrical migration [19], when Cd is effectively controlled within small range by controlling sprayed amount of AgNO3 solution, the use of Cd = 0.07 mol/L causes a little error of Dnssm [20]. Tang’s [19] investigation result indicated that Dnssm was almost a constant when depth was greater than a certain value. He [20] found that the slight variation of Xd measured by AgNO3 colorimetric method had a little effect on Dnssm. 5.3. Evaluation for corrosion risk of steel in concrete Evaluation for corrosion risk of steel in concrete is based on comparison between threshold chloride concentration for steel corrosion (Ccrit) and chloride concentration at color change boundary. The reported threshold chloride concentration with respect to corrosion risk varies greatly [37,38]. The Building Research Establishment [39] has proposed a classification for assessing the risk of corrosion in terms of acid-soluble (or total) chloride contents by the mass of cement: low, less than 0.4%; medium, 0.4–1.0%; and high, greater than 1%. Based on measured range of Cd, Meck et al. [12,24] suggested that owing to range of Cd is generally lower than Ccrit, evaluation for corrosion risk of steel in concrete using AgNO3 colorimetric method can be operated. As above discussed, concrete alkalinity, sprayed amount and concentration of AgNO3 solution and pore solution volume of concrete can influence chloride concentration at color change boundary, measured and calculated Cd change in a large range. Therefore, it is necessary to point out again that a proper sprayed amount of AgNO3 solution with a certain concentration should be determined to obtain a small range of Cd. A smaller Cd will produce more accurate applications.
6. Summary (1) AgNO3-based colorimetric methods are fast and easy to perform. They have been widely applied to measure the chloride penetration depth in field and laboratory. Among the three methods, AgNO3 method can be conducted more rapidly, simply and easily than AgNO3 + K2CrO4 or AgNO3 + fluoresceine method, the three methods give similar measured results. (2) The color change boundary shown by the colorimetric method is not a real boundary between chloride-free zone and chloride-penetrated zone. The boundary has a certain chloride concentration owing to OH ions in pore solution of concrete. The measurement of the chloride ion penetration depth can be carried out on specimens by caliper and ruler method and a ruler or by means of analyzing the digitized image of the specimen. Caliper and ruler method can be proper method to measure penetration depth. (3) Cutting or grinding is more suitable for sampling purpose than drilling since the sample obtained by cutting or grinding is more representative. The water extraction method can be used as an effective method to analyze the chloride content at the color change boundary. However, it is important to control experimental parameters.
7
(4) Analysis of published results indicate that the chloride concentrations at the color change boundary is 0.01% by the mass of cement for AgNO3 + fluoresceine method, and 0.1– 0.4% by the mass of cement for AgNO3 + K2CrO4 method, 0.28–1.69% by the mass of cement or 0.072–0.714 mol/L for the AgNO3 method. Many factors such as concrete alkalinity, sprayed volume and concentration of AgNO3 solution, pore solution volume of concrete, sampling method and methods used for measuring free chloride in concrete can be responsible for the high variability. Among all factors, sprayed volume and concentration of AgNO3 solution are main ones. (5) The smaller the volume of sprayed AgNO3 solution with a certain concentration, the lower the Cd is. 0.1 mol/L AgNO3 solution has most clear color of boundary. To obtain a lower Cd, a proper sprayed volume of 0.1 mol/L AgNO3 solution should be determined. However, this important point was not mentioned at all times since the AgNO3 method was applied. Based on small range of Cd, the method can be potentially to be used as a useful tool within the framework of evaluating chloride penetration in reinforced concrete structure in chloride environments.
Acknowledgments Financial supports from the National Natural Science Foundation of China (51072050) and Xiamen Department of Science and Technology (3502Z20103042) are greatly appreciated. References [1] Stanish K. The migration of chloride ions in concrete. Ph.D. dissertation. Canada: Department of Civil Engineering, University of Toronto; 2002. [2] Li KA. Analytical chemistry. Peking University Press; 2005. [3] Collepardi M, Marcialis A, Turriziani R. Kinetics of penetration of chloride ions into concrete. Il Cemento 1970;67:157–64 [in Italian]. [4] Collepardi M, Marcialis A, Turriziani R. Penetration of chloride ions in cement pastes and in concretes. J Am Ceram Soc 1972;55(10):534–5. [5] Italian Standard 79–28. Determination of the chloride, Rome; 1978. [6] Maultzsch M. Concrete related effects on chloride diffusion (in German). Contribution to Int Coll Chloride Corrosion, Vienna, BAMAmts- und. Mitteilungsblatt 1983;13(3):387–9. [7] Maultzsch M. Effects on cement pastes and concrete of chloride solution impact. Mater Tech (CH) 1984;12(3):83–90 [in German]. [8] Baroghel-Bouny V, Chaussadent T, Henry D. Testing and modeling chloride penetration into concrete – first round robin test. RILEM Technical Committee 178 Report, LCPC; 1999. p. 15. [9] Measurement of the colorimetric front. RILEM Technical Committee 178 Procedure. Preliminary draft; 2003. [10] Frederiksen JM. Testing chloride in structures – an essential part of investigating the performance of reinforced concrete structures. In: Proceedings of COST 521 workshop. Belfast, UK; 2000. [11] Baroghel-Bouny V, Belin P, Maultzsch M, Henry D. AgNO3 spray tests: advantages, weaknesses, and various applications to quantify chloride ingress into concrete. Part I: Non-steady-state diffusion tests and exposure to natural conditions. Mater Struct J 2007;40(8):759–81. [12] Baroghel-Bouny V, Belin P, Maultzsch M, Henry D. AgNO3 spray tests: advantages, weaknesses, and various applications to quantify chloride ingress into concrete. Part II: Non-steady-state migration tests and chloride diffusion coefficients. Mater Struct J 2007;40(8):783–99. [13] Otsuki N, Nagataki S, Nakashita K. Evaluation of AgNO3 solution spray method for measurement of chloride penetration into hardened cementitious matrix materials. ACI Mater J 1992;89(6):587–92. [14] Yuan Q, Shi C, He F, et al. Effect of hydroxyl ions on chloride penetration depth measurement using the colorimetric method. Cem Concr Res 2008;38(10):1177–80. [15] He F, Shi C, Yuan Q, et al. A study on factors influencing chloride concentration at color change boundary of concrete using AgNO3 colorimetric method. J Chin Ceram Soc 2008;36(7):890–5 [in Chinese]. [16] Chen G, Li W, Wang P. Penetration depth and concentration distraction of chloride ions into cement mortar. J Tong Ji Univ 1996;24(3):19–24 [in Chinese]. [17] NT Build 492. Nordtest method, concrete, mortar and cement-based repair materials: Chloride migration coefficient from non-steady-state migration experiments. Espoo, Finland; 1999.
8
F. He et al. / Construction and Building Materials 26 (2012) 1–8
[18] Baroghel-Bouny V, Arnaud S, Henry D, Carcasse‘s M, Que´nard D. Ageing of concretes in natural environments: experimentation for the 21st century. III – Durability properties of concretes measured on laboratory specimens, vol. 241. Bul Lab Ponts Chaus; 2002. p. 13–59. [19] Tang L. Chloride transport in concrete – measurement and prediction. Ph.D. dissertation. Goteborg, Sweden: Department of Building Materials, Chalmers University of Technology; 1996. [20] He F. Measurement of chloride migration in cement-based materials using AgNO3 colorimetric method. Ph.D. dissertation. Changsha, China: School of Civil Engineering and Architecture, Central South University; 2010. [21] Collepardi M. Quick method to determine free and bound chlorides in concrete. In: Nilsson LO, Ollivier JP, editors. Proceedings of the international RILEM workshop on chloride penetration into concrete, Paris; 1995. [22] Andrade C, Castellote M, Alonso C, Gonzalez C. Relation between colorimetric chloride penetration depth and charge passed in migration tests of the type of standard ASTM C1202-91. Cem Concr Res 1999;29(3):417–21. [23] Sirivivatnanon V, Khatri R. Chloride penetration resistance of concrete, presented to concrete. In: Institute of Australia conference getting a lifetime out of concrete structures. Brisbane, Australia, October 1998. [24] Meck E, Sirivivatnanon V. Field indicator of chloride penetration depth. Cem Concr Res 1999;33(8):1113–7. [25] McPolin D, Basheer PAM, Long AE, Grattan KTV, Sun T. Obtaining progressive chloride profiles in cementitious materials. Constr Build Mater 2005;19(9):666–73. [26] Yuan Q. Fundamental studies on test methods for transport of chloride ions in cementitious materials. Ph.D. dissertation. Belgium: Department of Structural Engineering, Gent University; 2009. [27] He F, Shi C, Yuan Q, et al. Factors influencing chloride concentration at the color change boundary using AgNO3 colorimetric method. J Chin Ceram Soc 2008;36(7):890–5 [in Chinese].
[28] Arya C, Newman JB. An assessment of four methods of determining the free chloride content of concrete. Mater Struct J 1990;23:319–30. [29] Barneyback RS, Diamond S. Expression and analysis of pore fluids from hardened cement pastes and mortars. Cem Concr Res 1986;16(5):760–70. [30] Barneyback RS, Diamond S. Expression and analysis of pore fluids from hardened cement pastes and mortars. Cem Concr Res 1981;11(2):279–85. [31] Arya C, Buenfeld NR, Newman JB. Assessment of simple methods of determining the free chloride ion content of cement paste. Cem Concr Res 1987;17(6):907–18. [32] Glass GK, Wang Y, Buenfeld NR. An investigation of experimental methods used to determine free and total chloride contents. Cem Concr Res 1996;26(9):1443–9. [33] Haque MN, Kayyali OA. Free and water soluble chloride in concrete. Cem Concr Res 1995;25(3):531–42. [34] NT Build 443. Nordtest method. Accelerated chloride penetration. Espoo, Finland; 1995. [35] Tang L. Electrically accelerated methods for determining chloride diffusivity in concrete-current development. Mag Concr Res 1996;48(176):173–9. [36] Chiang CT, Yang CC. Relation between the diffusion characteristic of concrete from salt ponding test and accelerated chloride migration test. Mater Chem Phys 2007;106(2–3):240–6. [37] Thomas MDA, Matthews JD. Performance of PFA concrete in a marine environment – 10-year results. Cem Concr Compos 2004;26(1):5–20. [38] Alonso C, Andrade C, Castellote M, Castro P. Chloride threshold values to depassivate reinforcing bars embedded in a standardized OPC mortar. Cem Concr Res 2000;30:1047–55. [39] Everett LH, Treadaway KWJ. Deterioration due to corrosion in reinforced steel concrete. Garston (United Kingdom): Building Research Establishment; 1980. Information Paper IP 12/80.