Vous êtes sur la page 1sur 7

ACI MATERIALS JOURNAL

Title no. 106-M38

TECHNICAL PAPER

Critical Corrosion Amount to Cause Cracking of Reinforced Concrete Structures


by Byung Hwan Oh, Ki Hyun Kim, and Bong Seok Jang
The corrosion of a steel bar in concrete induces expansion pressure and thus causes tensile stresses in surrounding concrete. This may cause serious cracking in concrete cover, which greatly affects the serviceability and durability of concrete structures. Therefore, it is necessary to realistically determine the critical corrosion amount (CCA) that causes the initiation of cracking in concrete cover. The purpose of this study is to explore and determine the CCA that causes surface cracking of concrete cover. To this end, a comprehensive experimental study was conducted. Major test variables include concrete strength and cover thickness. The corrosion tests of steel bar in concrete have been conducted, and the strains on the surface of concrete cover have been measured according to various amounts of steel corrosion. The CCA, which causes the initiation of cracking on the surface of concrete cover, was determined from the present test results. It was seen that CCA increases greatly with an increase of cover thickness of concrete. The present study indicates that CCA increases approximately proportionally to the square of concrete cover thickness. The concrete strength also affects CCA. It can be seen from this study that CCA increases with an increase of compressive strength, especially in the case of normally used medium and large cover thicknesses, that is, for a cover thickness above 40 mm (1.6 in.) in actual concrete structures. It was found, however, that the effect of the compressive strength on CCA is not large for small cover thickness such as 20 mm (0.8 in.). The present study may provide a good base for future development of realistic durability design and service-life assessment of concrete structures.
Keywords: corrosion; cover; crack; durability; strength; test.

is assumed herein that the diffusion of corrosion products into porous concrete is not counted because it occurs internally around a reinforcing bar. A comprehensive experimental study was conducted. Major test variables include concrete strength and cover thickness. Several series of corrosion tests for a steel bar in concrete were conducted, and the strains on the surface of concrete cover were measured for various test series according to the different amounts of steel corrosion. The CCA that causes the initiation of surface cracking was determined from the present test results for various cover thicknesses and concrete strengths. RESEARCH SIGNIFICANCE During the last several decades, many concrete structures have been built in severe environments such as sea and marine environments. These structures are vulnerable to chloride attacks and may suffer seriously from corrosion of reinforcing bars. The corrosion of steel bars in concrete directly affects durability and reduces service life of concrete structures.8-10 It is well known that the corroded part of a steel bar increases in volume and this volume increase may induce pressure to the surrounding concrete due to the restraint of expansion. This corrosion-induced expansion pressure induces tensile stresses in concrete and may lead to serious cracking in concrete cover. The corrosion-induced cracking of concrete cover greatly affects not only durability but also the service life of concrete structures. Therefore, it is important to realistically determine the CCA that causes the initiation of cracking on the surface of concrete cover. The reasonable determination of CCA is important for the realistic prediction of the service life of concrete structures under corrosion environments because it is used to calculate the cracking time of concrete cover. The prediction of service life is one of the most important topics in recent years. The CCA may vary with cover thickness and concrete strength. Therefore, the effects of these parameters on the CCA are investigated in this study. CORROSION-INDUCED EXPANSION AND DEFORMATION Corrosion-induced expansion The corrosion of steel bar in concrete causes an increase of volume, and the amount of volume increase varies according to the types of oxidation products. Namely, the rates of volume increase due to corrosion are ; FeO = 1.7, Fe3O4 = 2,

INTRODUCTION The penetration of chlorides into concrete through a diffusion process will cause an accumulation of chlorides around a reinforcing bar1-2; this will cause corrosion of steel bars in concrete when the chloride content reaches the threshold values.3-7 The cracking of concrete cover due to steel corrosion is an important problem in concrete structures because it directly affects the durability and service life of such structures.8-12 The corrosion products of a reinforcing bar in concrete induce pressure to the surrounding concrete due to the expansion of corroded steel. This expansion pressure induces tensile stresses in concrete around the reinforcing bar, and the continuous increase of pressure eventually causes cracking through the concrete cover. The expansion pressure is in turn related to the amount of corrosion products in the reinforcing bar. It is very important to know in advance the critical corrosion amount (CCA) to predict the possible time to cracking for the concrete cover of real structures. Therefore, it is necessary to realistically determine the CCA that causes the initiation of cracking in concrete cover. The purpose of the present study is to explore the CCA that initiates the cracking on the surface of concrete cover. It ACI Materials Journal/July-August 2009

ACI Materials Journal, V. 106, No. 4, July-August 2009. MS No. M-2006-475.R3 received October 10, 2008, and reviewed under Institute publication policies. Copyright 2009, American Concrete Institute. All rights reserved, including the making of copies unless permission is obtained from the copyright proprietors. Pertinent discussion including authors closure, if any, will be published in the May-June 2010 ACI Materials Journal if the discussion is received by February 1, 2010.

333

Byung Hwan Oh, FACI, is a Professor in the Department of Civil Engineering at the Seoul National University, Seoul, Korea. He received his PhD from Northwestern University, Evanston, IL, in 1982. He received the ACI Wason Medal Award for Material Research in 2005 and the ACI/CANMET award in 2001 for his outstanding original contribution in concrete science and technology. His research interests include the corrosion behavior, durability, fracture, cracking, bond, fatigue, safety assessment, and nonlinear analysis/design of reinforced and prestressed concrete structures. Ki Hyun Kim is a Research Assistant in the Department of Civil Engineering at Seoul National University. His research interests include corrosion and durability of concrete structures. ACI member Bong Seok Jang is a Senior Researcher in the Dam Safety Research Institute of Korea Water Resources Corporation, Daejeon, Korea. His research interests include reinforcing bar corrosion, chloride penetration, and durability of concrete structures.

thickness along the reinforcing bar due to corrosion can be expressed as follows. W st 2 2 2 r b ( r b x p ) = ------- = w corr r b st (1)

where Wst is the mass of reinforcing bar consumed by corrosion, and st is the density of steel bar. Equation (1) can be rearranged to obtain the radius loss xp as follows. x p = r b ( 1 1 w corr ) (2)

Fe2O3 = 2.1, Fe(OH)2 = 3.6, Fe(OH)3 = 4.0, Fe(OH)3 3H2O = 6.15 , respectively.8,11 The expansive pressure due to the volume increase of a steel bar in concrete generally induces the tensile stresses and strains in the surrounding concrete. The tensile strains of surrounding concrete due to corrosion-induced expansion increase as the steel bar corrosion progresses. Further increase of tensile strain will cause cracking in the surrounding concrete, and the cracking will also occur at the surface of concrete cover during the expansion process. The tensile strains on the surface of concrete cover due to corrosion expansion are related to the corrosion amount wcorr of a steel bar in concrete, in which wcorr represents the ratio of weight loss due to corrosion to initial weight of a steel bar. Therefore, the deformation process around a steel bar due to corrosion is outlined in the following section to help understand how the corrosion-induced expansion is related to the corrosion amount wcorr that was measured in the present tests. Deformation around steel bar according to corrosion amount The deformation around a steel bar due to corrosion expansion can be described as shown in Fig.1, in which rb is the initial radius of the reinforcing bar, xp is the loss of radius of reinforcing bar due to corrosion, and wcorr is the ratio of weight loss due to corrosion to initial weight of the reinforcing bar, respectively. The loss of reinforcing bar volume per unit

The volume of corrosion product made by consumed steel is larger than the volume of consumed steel itself. Therefore, the radius of steel bar after corrosion will increase by rb under free expansion and the following relation holds. W rust W st 2 2 ( r b + r b ) r b = ----------- ------ rust st W st st = ------- -------------- 1 st rust W st = ------- ( v r 1 ) st = r b ( v r 1 )w corr where Wrust, rust are the mass and density of corrosion products, respectively, and vr = st /rust is the relative ratio of steel density to corrosion product density. The value of can be obtained from the fact that the mass of steel component (Fe) among corrosion products is the same as the mass of steel bar (Fe) consumed by corrosion, that is, Wst = Wrust. The value of is 0.523 for hydrated red rust (Fe(OH)3) and 0.622 for ferrous hydroxides (Fe(OH)2), respectively, because the atomic weights of Fe, O2, and H2 per one mol are 55.85 g, 32 g, and 2 g (0.1229 lb, 0.0704 lb, and 0.0044 lb), respectively.3 The value ranges from 0.523 to 0.622 depending on the types of corrosion products and the mean value being approximately 0.573. The ratio of steel density to corrosion product density, v r, normally ranges from 2 to 4. Now, Eq. (3) can be rearranged to obtain the radius increase rb due to corrosion as follows r b = r b ( 1 + ( v r 1 )w corr 1 ) (4)
2

(3)

Fig. 1Deformation process due to corrosion-induced expansion (left side is free expansion of corroded part; right side is real expansion restrained by surrounding concrete). 334

Figure 1 shows the volume expansion of corrosion product layer caused by the radius loss of a steel bar. The expansion of corrosion product layer is restrained and compressed by surrounding concrete.12 Therefore, the freeexpanded displacement rb is now reduced to ur due to the restraint of surrounding concrete, as shown in Fig.1. Namely, the free-expanded thickness of corrosion product layer (xp + rb) decreases by (rb ur) to (xp + ur) due to pressure force caused by the restraint of surrounding concrete (refer to Fig. 1). This expansion displacement and, thus, pressure, will cause tensile stresses and strains in adjacent concrete. Furthermore, an increase of tensile strains will cause cracking in concrete cover. As can be seen from Eq. (2) and (4), ACI Materials Journal/July-August 2009

Table 1Mixture proportions of concretes


Test series S1 S2 S3
*

w/c 0.55 0.45 0.35

Water, kg/m3 (lb/yd3) Cement, kg/m3 (lb/yd3) Fine aggregate, kg/m3 (lb/yd3) Coarse aggregate, kg/m3 (lb/yd3) Admixture,* kg/m3 (lb/yd3) 168 (283) 170 (287) 172 (290) 305 (514) 378 (637) 491 (828) 792 (1355) 712 (1200) 608 (1025) 987 (1664) 1003 (1691) 1011 (1704) 0.30 (0.51) 0.76 (1.28) 2.95 (4.97)

High-range water-reducing admixture.

Table 2Measured material properties of concretes


Test series S1 S2 S3 w/c 0.55 0.45 0.35 Compressive strength, MPa (ksi) 27.5 (3.99) 40.3 (5.84) 44.3 (6.42) Tensile strength, MPa (ksi) 3.10 (0.449) 3.93 (0.569) 4.12 (0.597) Elastic modulus, MPa (ksi) 24,821 (3599) 30,019 (4353) 31,481 (4565) Poissons ratio 0.18 0.18 0.18 Cracking strain ( 103) 0.125 0.131 0.131

the expansion behavior due to corrosion is related to the corrosion amount (or weight loss ratio) wcorr of a steel bar. Therefore, the tensile strains occurring on concrete cover due to internal expansion pressure are also related to this corrosion amount (weight loss ratio) wcorr of a steel bar. Therefore, in this study, the tensile strains on the surface of concrete cover were measured for various corrosion amounts wcorr, and diagrams of strain versus corrosion amount (weight loss ratio) were obtained from those tests. These diagrams of strain versus corrosion amount enable one to determine the CCA that causes the initiation of cracking on concrete cover. This was done for various cover thicknesses and compressive strengths of concrete in this study. EXPERIMENTAL PROCEDURE Test outline and test variables In the present tests, the surface concrete strains were measured as the corrosion of steel bar in concrete progresses. This will allow the determination of CCA that induces the cracking on the surface of concrete cover as described in the previous section. To this end, a comprehensive experimental program was set up to execute the corrosion tests of steel bar in concrete. The major test variables are the cover thickness and compressive strength of concrete. The concrete cover thicknesses considered were 20 mm, 30 mm, 40 mm, and 50 mm (0.8 in., 1.2 in., 1.6 in., and 2.0 in.), respectively. To consider the effects of compressive strength fc, the watercement ratios (w/c) of concrete were varied from 0.55 to 0.35 (refer to Table 1). The test series were designated herein as test series S1 for w/c = 0.55 (fc = 27.5 MPa [3.99 ksi]), test series S2 for w/c = 0.45 (fc = 40.3 MPa [5.84 ksi]), and test series S3 for w/c = 0.35 (fc = 44.3 MPa [6.42 ksi]), respectively (refer to Tables 1 and 2 for test series identification). Therefore, the test specimen identification S1-C2 represents the specimen with fc = 27.5 MPa (3.99 ksi) and c = 20 mm (0.8 in.); S2-C3 represents the specimen with fc = 40.3 MPa (5.84 ksi) and c = 30 mm (1.2 in.); and S3-C4 represents the specimen with fc = 44.3 MPa (6.42 ksi) and c = 40 mm (1.6 in.), respectively. Test materials Type 1 ordinary portland cement and river sand, with a specific gravity of 2.55, were used. The specific gravity of crushed coarse aggregates was 2.6. The maximum aggregate size of concrete was 20 mm (0.8 in.). The slump value of ACI Materials Journal/July-August 2009 Fig. 2Configuration of test specimen and strain gauge attachment (200, 60, and 80 mm [8, 2.4, and 3.2 in.]). fresh concrete was controlled to be 150 mm (6 in.) by using high-range water-reducing admixture in the mixture; air content was 4.5%. Table 1 summarizes the mixture proportions for different series of test concretes. The measured material properties of concretes are summarized in Table 2. The compressive strengths of concrete for different test series ranged from approximately 27 to 45 MPa (3.91 to 6.52 ksi). The compressive strength of concrete for each test series in Table 2 was obtained from the mean value of three test cylinder specimens. The yield strength and tensile strength of steel bar were 392 and 480 MPa (56.8 and 69.6 ksi), respectively. All the tests for material properties were performed according to ASTM International standards. Fabrication of test specimens The test specimen cubes measured 200 x 200 x 200 mm (8 x 8 x 8 in.), with cover thicknesses of 20, 30, 40, and 50 mm (0.8, 1.2, 1.6, and 2.0 in.), respectively. A steel bar of 20 mm (0.8 in.) diameter was put in concrete specimen at the location of designated cover thickness. Therefore, the ratios of cover thickness to reinforcing bar diameter (c/d) are 1.0, 1.5, 2.0, and 2.5, respectively. The region for steel corrosion was limited to the central portion of the steel bar, as shown in Fig. 2. Both ends of steel bar were wrapped with polyvinyl chloride (PVC) pipes and sealed with epoxy. The cross-hatched region in Fig. 2 was exposed to corrosion. 335

The concrete surface strains and electric potentials were measured every 30 minutes by a data logging system, and these measurements were continued until the surface strain increased rapidly and reached sufficient values. The electric current can be obtained from the resistance value and is used to determine the corrosion amount of the reinforcing bar (Fig. 4). The procedure of determination of corrosion amount will be addressed in more detail in the following section. Determination of corrosion amount by Faradays Law The amount of corrosion can be calculated by Faradays Law. The amount of substance produced or consumed by the electrical quantity of one Faraday (F) is equal to the extracted substance of one chemical equivalent moved by one mol of electron. Equation (5) represents the amount of mol, X, extracted by electrolysis of substance with n electrons. It X = -----nF (5)

Fig. 3Arrangement for accelerated corrosion tests.

Fig. 4Electric current diagram and corrosion amount according to time obtained from present corrosion test. Test method and measurements Accelerated corrosion test by potentiostatic polarization The test specimens were immersed in NaCl 3% solution and the corrosion circuit was connected using a direct-current power supply as shown in Fig. 3. The reinforcing bar in the specimen plays as an anode and the stainless steel plate as a cathode. An electric resistance was installed in the corrosion circuit to measure the electric potential. The electrical potential difference between anode and cathode accelerates the penetration of chloride ions into concrete and thus accelerates the corrosion of steel bar. The potential used in the acceleration test for corrosion was 30 volts. 336

in which I is the electric current in ampere (A), n is the number of mol participating in production reaction (n = 2 was used here because the initial corrosion products are assumed mostly as n = 2, before surface cracking), t is the time (hours), and 1F is the electric quantity of one mol of electron = 96,500 C. An electric resistance was installed in the corrosion circuit to measure the electric potential, as shown in Fig. 3. The electric current was then obtained from the resistance value. The total electric change was also obtained by integrating the electric current values with respect to time and then the number of mol of corroded reinforcing bar was determined by using Faradays Law. Figure 4(a) shows the current values obtained from the present corrosion tests according to time for Specimen S2-C4, which is the specimen with fc = 40.3 MPa (5.84 ksi) and c = 40 mm (1.6 in.). Figure 4(b) depicts the increase of corrosion amount according to time that is obtained from both the current-time curve of Fig. 4(a) and Faradays Law. Measurement of strains on concrete surfaceThe strain values on the surface of the concrete specimen were measured according to time. This is to see the increase of tensile strain due to volume expansion of corroded reinforcing bar and to find the time at which the crack occurs on the concrete surface. For this purpose, concrete strain gauges were attached on the concrete surface near reinforcing bar, as shown in Fig. 2. The strain gauges were installed in the direction perpendicular to the anticipated crack direction. The strains were automatically measured and stored by a data logging system (refer to Fig. 3). From the measurement of concrete strains, the critical value of corrosion amount that causes the crack occurrence on the concrete surface has been determined. The CCA is the corrosion amount at which the crack occurs first on the surface of concrete cover. The procedure of determination of critical value will be addressed in the next section. ANALYSIS OF TEST RESULTS Variation of surface strains according to cover thickness The measured strains on the surface of concrete specimens were plotted. Figure 5 shows the concrete strains according to the amount of corrosion for various specimens with different cover thicknesses. It can be seen from Fig. 5 that the development of surface strains is larger and faster as ACI Materials Journal/July-August 2009

Table 3Critical corrosion amounts for various concrete strengths and concrete covers obtained from present tests
Test series S1-C2 S1-C4 S2-C2 S2-C3 S2-C4 S2-C5 S3-C2 S3-C4 Critical corrosion amount, % 0.83 3.60 0.88 1.92 4.00 5.59 0.95 5.20

the cover thickness becomes smaller. Namely, much larger strains occur for shallow-thickness specimens at the same corrosion amount. This is because expansive pressure due to the same corrosion amount causes much larger surface strains for thin-covered specimens. The surface strain increases slowly at a lower corrosion amount, but increase rapidly after a certain higher corrosion amount. The thincovered specimens show this rapid increase of strain at a relatively lower corrosion amount. The time of surface cracking, which is necessary for the determination of CCA, was reasonably defined herein as the moment at which the cover concrete strain reaches the cracking strain (the strain at the tensile strength of concrete). The thinner the cover thickness is, the lower the CCA. Variation of surface strains according to concrete strength Figure 6 shows the increase of surface strain with an increase of corrosion amount for various concrete strengths and cover thicknesses. It can be seen that in the case of cover thickness of 20 mm (0.8 in.), the surface strain versus corrosion amount curves are almost same irrespective of various concrete strengths. For a thicker concrete cover of 40 mm (1.6 in.), however, the surface strain profiles according to corrosion amount are much different among the different concrete strengths, that is, among the different test series of S1, S2, and S3, respectively. Here, the test series of S1, S2, and S3 mean three different concrete strengths, that is, fc = 27.5, 40.3, and 44.3 MPa (3.99, 5.84, and 6.42 ksi) for Series S1, S2, and S3, respectively (refer to Table 2). Figure 6 indicates that, as the strength of concrete increases, the amount of corrosion that induces the same value of surface strain increases. It is known that the concrete of higher strength has higher stiffness (that is, higher elastic modulus), as shown in Table 2. This means that higher stiffness of concrete needs higher expansive pressure (and thus higher corrosion amount) to develop the same surface strain. The strength of concrete, however, only slightly affects the corrosioninduced surface strains for the specimens with relatively thin cover (20 mm), as shown in Fig. 6. Determination of critical corrosion amount In general, it is difficult to determine surface cracking with the naked eye. Therefore, the time of surface cracking was reasonably defined herein as the moment at which cover concrete strain reached the cracking strain (the strain at the tensile strength of concrete). This cracking strain can be obtained by dividing the tensile strength by the elastic modulus of concrete. These cracking strains are summarized in Table 2. ACI Materials Journal/July-August 2009

Fig. 5Measured surface strains according to corrosion amount for various cover thicknesses (Series S2 concrete).

Fig. 6Measured surface strains according to corrosion amount for various-strength concretes (cover thickness c = 20 and 40 mm [0.8 and 1.6 in.]). Therefore, the CCAs that cause the initiation of surface cracking were determined from the strain versus corrosionamount curves (shown in Fig. 5 and 6) by employing the cracking strains for each test series. The CCAs for various concrete strengths and cover thicknesses are summarized in Table 3. From the present test results, the CCAs (wcorr,c) were found to be 0.83% and 3.60% for Series S1-C2 and S1-C4, respectively ( fc = 27.5 MPa [3.99 ksi] for test Series S1); 0.88%, 1.92%, 4.00%, and 5.59% for test Series S2-C2, S2-C3, S2-C4, S2-C5, respectively ( fc = 40.3 MPa [5.84 ksi] for test Series S2); and 0.95% and 5.20% for Series S3-C2 and S3-C4, respectively ( fc = 44.3 MPa [6.42 ksi] for test Series S3) (refer to Table 2 for concrete compressive strengths). Table 3 indicates that the CCA increases drastically with an increase in cover thickness and also increases with an increase in concrete strength. Figure 7 shows the relation between the CCA and cover thickness for Series S2 concrete ( fc = 40.3 MPa [5.84 ksi]) obtained from the present study (refer to Table 3). The quantity of each test sample in test procedure was 2 and the average value was used. Figure 7 indicates that the CCA increases approximately proportional to the square of cover thickness, as shown in Eq. (6) (also refer to Fig. 7). The correlation coefficient of Eq. (6) was found to be R2 = 0.9941 as shown in Fig. 7, which means that Eq. (6) correlates very well with the present test data. w corr, c = 0.0018c
2.07

(6) 337

Fig. 7Critical corrosion amount versus cover thickness relation for Series S2 concrete (fc = 40.3 MPa [5.84 ksi]) (cover thickness c = 20, 30, 40, and 50 mm [0.8, 1.2, 1.6, and 2.0 in.]).

Fig. 8Critical corrosion amount versus compressive strength relations for cover thickness c = 20 and 40 mm (0.8 and 1.6 in.), respectively (fc = 27.5, 40.3, and 44.3 MPa [3.99, 5.84, and 6.42 ksi], respectively). in which wcorr,c is the critical corrosion amount in percent of initial weight of steel bar, and c is the cover thickness in mm (1 mm = 0.04 in.). Figure 8 shows the relations between the CCA and concrete compressive strength for cover thicknesses of 20 mm and 40 mm (0.8 in. and 1.6 in.), respectively. It is interesting to note that the CCAs do not vary much depending on concrete strength when the cover thickness is rather small, such as c = 20 mm (0.8 in.), as shown in Fig. 8. This is because, when the cover thickness is small, the cracking occurs under a very small corrosion amount. The effect of concrete strength on the critical corrosion amount, however, becomes larger as the cover thickness increases (refer to Fig. 8). CONCLUSIONS The expansion pressure due to steel corrosion induces tensile stresses in concrete around the reinforcing bar and the continuous increase of pressure eventually causes cracking through the concrete cover. The cracking of concrete cover due to steel corrosion is an important and essential problem in concrete structures because it directly affects the safety and service life of such structures. The purpose of this study is to explore the CCA that causes the surface cracking of concrete cover. A comprehensive experimental study has 338

been conducted for this purpose. Major test variables include concrete strength and cover thickness of reinforcing bar. The following conclusions have been drawn from this study: 1. The CCAs, which initiate the cracking on the surface of concrete cover, were determined from the present corrosion tests for various concrete strengths and cover thicknesses; 2. From the present study, the CCAs (wcorr,c) are found to be 0.83% and 3.60% for test Series S1-C2 and S1-C4 ( f c = 27.5 MPa [3.99 ksi] and cover thickness c = 20 and 40 mm [0.8 and 1.6 in.]), respectively; and 0.88%, 1.92%, 4.00%, and 5.59% for test Series S2-C2, S2-C3, S2-C4, and S2-C5 ( fc = 40.3 MPa [5.84 ksi] and cover thickness c = 20, 30, 40, and 50 mm [0.8, 1.2, 1.6, and 2.0 in.]), respectively; and 0.95%, 5.20% for Series S3-C2 and S3-C4 (fc = 44.3 MPa [6.42 ksi] and cover thickness c = 20 and 40 mm [0.8 and 1.6 in.]), respectively; 3. It was found that the CCA increases greatly with an increase of cover thickness. The present study indicates that the CCA increases approximately proportional to the square of cover thickness of concrete; 4. The concrete strength also affects the CCA. The CCA increases with an increase of compressive strength of concrete, especially for the cases of medium and large cover thicknesses, that is, for the cover thicknesses greater than 40 mm (1.6 in.) in concrete structures. The present study, however, indicates that the CCA does not vary much depending on concrete compressive strength when the cover thickness is rather small, such as c = 20 mm (0.8 in.); and 5. Within the experimental data obtained from this study, it may be concluded that the present study may provide a good database, which is necessary for future development of advanced durability design and service life assessment of concrete structures. In the future, however, it may be necessary to study further the possible variability of corrosion in actual concrete structures to establish more realistic assessment method for the durability and service life of concrete structures. ACKNOWLEDGMENTS
The authors wish to express their gratitude and sincere appreciation to the Man-Made Disaster Prevention Program of Korea for financial support.

NOTATION
c d fc I n rb ur vr = st /rust Wrust Wst wcorr wcorr,c X xp rb rust rust st = = = = = = = = = = = = = = = = = = = cover thickness of concrete, mm (1 mm = 0.04 in.) diameter of reinforcing bar, mm (1 mm = 0.04 in.) compressive strength of concrete electric current in ampere (A) number of mol participating in production reaction initial radius of reinforcing bar, mm (1 mm = 0.04 in.) displacement of reinforcing bar due to corrosion, mm (1 mm = 0.04 in.) relative ratio of steel density to corrosion product density mass of corrosion products, g (1 g = 0.0022 lb) mass of reinforcing bar consumed by corrosion, g (1 g = 0.0022 lb) ratio of weight loss due to corrosion to initial weight of reinforcing bar critical corrosion amount in percent of initial weight of steel bar amount of mol extracted by electrolysis loss of radius of reinforcing bar due to corrosion, mm (1 mm = 0.04 in.) Wst /Wrust increase of reinforcing bar radius due to corrosion, mm (1 mm = 0.04 in.) strain of corrosion product layer density of corrosion products, g/mm3 (1 g/mm3 = 36.13 lb/in.3) density of steel bar, g/mm3 (1 g/mm3 = 36.13 lb/in.3)

ACI Materials Journal/July-August 2009

REFERENCES
1. Oh, B. H., and Jang, B. S., Chloride Diffusion Analysis of Concrete Structures Considering the Effects of Reinforcements, ACI Materials Journal, V. 100, No. 2, Mar.-Apr. 2003, pp. 143-149. 2. Oh, B. H., and Jang, S. Y., Prediction of Diffusivity of Concrete Based on Simple Analytic Equations, Cement and Concrete Research, V. 34, No. 3, 2004, pp. 463-480. 3. Oh, B. H., and Jang, S. Y., Experimental Investigation of the Threshold Chloride Concentration for Corrosion Initiation in Reinforced Concrete Structures, Magazine of Concrete Research, V. 55, No. 2, Apr. 2003, pp. 117-124. 4. Alonso, C.; Andrade, C.; Castellote, M.; and Castro, P., Chloride Threshold Values to Depassivate Reinforcing Bars Embedded in a Standardized OPC Mortar, Cement and Concrete Research, V. 30, 2000, pp. 1047-1055. 5. Hussain, S. E.; Al-Gahtani, A. S.; and Rasheeduzzafar, Chloride Threshold for Corrosion of Reinforcement in Concrete, ACI Materials Journal, V. 93, No. 6, Nov.-Dec. 1996, pp. 534-538. 6. Hope, B. B., and Ip, A. K. C., Chloride Corrosion Threshold in Concrete, ACI Materials Journal, V. 84, No. 4, July-Aug. 1987, pp. 306-314.

7. Thomas, M., Chloride Thresholds in Marine Concrete, Cement and Concrete Research, V. 26, No. 4, 1996, pp. 513-519. 8. Liu, Y., and Weyers, R. E., Modeling the Time-to-Corrosion Cracking in Chloride Contaminated Reinforced Concrete Structures, ACI Materials Journal, V. 95, No. 6, Nov.-Dec. 1998, pp. 675-681. 9. Francois, R.; Castel, A.; Vidal, T.; and Vu, N.-A., Long-Term Corrosion Behavior of Reinforced Concrete Structures in Chloride Environment, Journal de Physique, Archives, EDP Sciences, Dec. 22, 2006. (published online) 10. Tang, L., and Nilsson, L. O., Service Life Prediction for Concrete Structures under Seawater by a Numerical Approach, DBMC Component 7, E&FN Spon, 1996, pp. 97-106. 11. Broomfield, J. P., Corrosion of Steel in Concrete: Understanding, Investigation and Repair, E&FN Spon, 1997, pp. 20-25. 12. Lundgren, K., Modeling Bond between Corroded Reinforcement and Concrete, Fracture Mechanics of Concrete Structures, V. 1, Proceedings of the Fourth International Conference on Fracture Mechanics of Concrete and Concrete Structures, Cachan, France, May 28 -June 1, 2001, pp. 247-254.

ACI Materials Journal/July-August 2009

339