Manual on Service Life of Corrosion-Damaged Reinforced Concrete Bridge Superstructure Elements (2006)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

21 Introduction There are two steps to conducting a rehabilitation of a rein- forced concrete element that has suffered corrosion-induced damage: (1) repair sections of the element that have suffered concrete failure in the form of cracking, delamination, and/or spalling and (2) provide a corrosion control system, if necessary, to prevent or minimize future deterioration. Performing a con- crete repair, such as sealing a crack or removing delaminated concrete and patching, restores the concrete element to its orig- inal geometric form and most often does not have any impact on the ongoing corrosion process.Consequently, either the con- crete repair can fail or concrete in areas not previously repaired can suffer from corrosion-induced damage. This requirement for periodic repair of failed sections of the concrete element can result in a significant increase in the resources required for maintenance. To minimize maintenance costs, a corrosion con- trol system along with the repairs is necessary. The discussion in this chapter is limited to selection of a corrosion control system for reinforced concrete elements suffering corrosion-induced damage. However, some types of repairs can also provide cor- rosion control, and they are discussed as appropriate. When a concrete patch is installed, fresh concrete is placed in the repair area. The new concrete’s chemical makeup is dif- ferent than that of the original concrete. This can, depending on the differences, result in startup or acceleration of ongoing corrosion. A corrosion cell is fueled by chemical or electrical imbalances. If, for example, a concrete patch is installed in a concrete element that has undergone exposure to chloride ions and/or had admixed chloride ions, the difference in chloride concentration between the original and the new concrete can result in the setup of a new corrosion cell or the acceleration of an existing cell. In a case such as this, corrosion on reinforcing bars traversing the old-new concrete boundary along the perimeter of the patch can initiate corrosion, or ongoing cor- rosion on those bars can accelerate, resulting in the failure of the patch along the repair perimeter. This failure mechanism, termed the “halo effect,” can result from differences in pH of the two types of concrete. Because the areas not adjacent to the concrete patch are not in any way impacted by the repair, cor- rosion in these areas can continue unabated, and failure of con- crete in those areas can occur after the concrete repairs have been completed. Corrosion control systems fall into two broad categories: local (i.e., those applied in a repair area) and global (i.e., those applied to the entire concrete element). The primary goal of the local corrosion control systems is to mitigate the halo effect and improve the service life of the patch repair, whereas global corrosion control systems mitigate the halo effect and provide corrosion control over the entire concrete element. The local systems are only applicable when the concrete ele- ment has suffered local exposure to chloride ions, such as exposure to contaminated water runoff in certain sections of the element. Local systems can be used to avoid or reduce the halo effect that is due to contaminated concrete adjacent to the repair. They are completely ineffective at controlling cor- rosion in areas that are not adjacent to the repair. Corrosion control systems can also be categorized by the mechanism by which they provide protection: mechanical (barrier) or electrochemical systems. The barrier systems control the transportation of chemical species such as chlo- ride ions, oxygen, and moisture to sites where the corrosion reactions are occurring or may occur. The electrochemical systems control the progression of corrosion by interfering with the chemical or electrical aspects of the corrosion process. Under each of these categories, many different types of corrosion control systems are available, and each has its own advantages and disadvantages. The effectiveness of a corrosion control system very much depends on the level of contamination of chloride ions in the concrete and the severity of chloride ion exposure. The barrier systems that are applied onto the surface of the concrete ele- ment are most effective when applied either to a repair or to C H A P T E R 4 Selection of Corrosion Mitigation Alternatives

existing concrete that is not sufficiently contaminated with chloride ions, thereby delaying or avoiding the accumulation of chloride ions in sufficient quantity at the steel depth and the consequent initiation of corrosion. If sufficient levels of chlo- ride ions are already present in the concrete and can diffuse to accumulate in sufficient quantity at the steel depth, a barrier system is not as effective. The effectiveness of electrochemical systems such as cathodic protection and electrochemical chlo- ride extraction generally do not depend on the quantity of chloride ions present in the concrete or on the severity of expo- sure because their level of protection can be adjusted to account for the corrosivity of the environment. The ability of other elec- trochemical systems such as corrosion inhibitors depends on the concentration of chloride ions and severity of exposure. The first step in selecting a corrosion control system is to identify if local systems will suffice. If not, appropriate global systems will need to be identified. To determine whether a local or global system is appropriate, the distribution of chlo- ride ions and the exposure conditions need to be determined. Subsequently, the selection of various alternatives from the local or global system type is also based on the distribution of chloride ions in areas that have not yet suffered damage. In addition, other information—such as half-cell potential data, corrosion rate information, clear concrete cover distribution, electrical continuity of the reinforcing steel, carbonation, and the presence or absence of other concrete deterioration such as freeze-thaw and ASR—is used in the decision making. To provide for a logical framework to analyze the chloride dis- tribution and exposure conditions, the SI was proposed in Chapter 2 and defined in Chapter 3. The discussion of how the SI can be used follows. The two most important pieces of information normally obtained during condition evaluation are the quantity of damage and the distribution of chloride ions in sound (i.e., undamaged) areas. The quantity of damage is obtained dur- ing the visual and the delamination surveys. The distribution of chloride ions in sound areas is obtained by collecting pow- dered concrete samples or cores to determine the concentra- tion of chloride ions at various depths, including the depth at which the reinforcing steel is located. The distribution of chloride ions reveals whether the chloride ions are admixed or whether they are from an external source. If they are uni- formly distributed as a function of depth (in quantities in excess of the background chloride level), then most likely they were admixed. If the chloride ions show a diffusion distribu- tion, it can be safely assumed that the chlorides are coming from the outside. It is also possible to have a combination of admixed chlorides and a diffusion of chlorides coming from the environment. In addition, an analysis of the distribution allows one to determine if the chloride exposure is localized or common to the entire concrete element. Standard practice would be to analyze the distribution of the chloride ions and determine the level of threat they pose in terms of corrosion- induced damage in the future. If the chloride ions are close to or above the threshold, a more aggressive corrosion control system such as cathodic protection or electrochemical chlo- ride extraction may be required. However, if the distribution of chloride ions makes it insufficient to initiate corrosion at the time of the evaluation or to reach the threshold (at the steel depth) in the near future due to diffusion of the existing chloride ions in the concrete, a barrier system in the form of sealers, membranes, and overlays may be appropriate. To assist in the process of analysis of the distribution of the chloride ions at the steel depth, the SI—a measure of the average distribution of the chloride ions with respect to the threshold—was developed.When the distribution of chlo- ride ions is such that the concentration at all locations is below the threshold, the SI will vary from 0 to 10 (0 represents a con- dition in which the chloride concentration at all locations at the steel depth is equal to the threshold and 10 represents the condition in which the chloride concentration at all locations at the steel depth is close to 0). However, in practice there are situations in which the chloride concentration at the steel depth at some of the locations has exceeded the threshold and corrosion has initiated. As the number of locations increase where the chloride ion concentration at the steel depth exceeds the threshold, the SI will approach 0. When the numbers of locations with concentrations in excess of the threshold have a larger moment with respect to the threshold compared with the locations that have chloride concentration below the threshold, the SI will become negative. A negative SI would signify that corrosion has initiated at the majority of the loca- tions and that corrosion-induced damage can be expected in the near future in the sound areas. For the SI to be successful, the threshold value has to be properly ascertained. As discussed in Chapter 3, the threshold can vary over a significant range from one concrete mix to another and even in the same concrete. Therefore, the service life model can be used to select an appropriate threshold value. Service life modeling is performed at various threshold values, and the threshold value that provides the best estimate of dam- age, as compared with actual damage on the structure, is used as the threshold value for the SI. This threshold value is termed the “apparent threshold” because it represents the sum average of all variations in the threshold value required to initiate cor- rosion. The service life model can be used to determine the distribution of chloride ions at the steel depth at a given age. The apparent threshold value and the distribution of chloride ions can then be used to calculate the SI. For structures or con- crete mixtures for which the threshold value is known with a reasonable degree of confidence, this threshold value can be used along with the actual chloride distributions without depending on the service life model to provide the apparent threshold value. Figure 2 presents the distribution of chloride 22

ions at various SI levels based on an apparent chloride thresh- old value of 800 ppm. All corrosion control alternatives that may be most applicable for that particular distribution of chlo- ride ions have been identified, and the types of corrosion con- trol systems that are most likely to provide optimal protection for a particular SI are shown in Figure 3. The following simplifications were made to prepare Figure 3 and apply primarily to global systems: • All surface-applied coatings that are capable of controlling the flow of moisture into the concrete without hampering the outflow of moisture out of the concrete (i.e., they are breathable) have been lumped together as sealers. The term “sealers” is most often used to refer to silane/siloxane-based material that is applied to the surface of the concrete to reduce the inflow of water vapor but that is not as effective when water is ponded on the surface of the sealer. These 23 Chloride Ion Distribution for Susceptibility Index 9 0 0.005 0.01 0.015 0.02 0.025 0.03 0 100 200 300 400 500 600 700 800 Chloride Ion Concentration (PPM) Fr eq ue nc y Chloride Ion Distribution for Susceptibility Index 10 0 0.02 0.04 0.06 0.08 0.1 0.12 0 50 100 150 200 250 300 350 400 450 Chloride Ion Concentration (PPM) Fr eq ue nc y Chloride Ion Distribution for Susceptibility Index 8 0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.01 0 200 400 600 800 1000 1200 Chloride Ion Concentration (PPM) Fr eq ue nc y Chloride Ion Distribution for Susceptibility Index 7 0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0 200 400 600 800 1000 1200 Chloride Ion Concentration (PPM) Fr eq ue nc y Chloride Ion Distribution for Susceptibility Index 6 0 0.0002 0.0004 0.0006 0.0008 0.001 0.0012 0.0014 0.0016 0.0018 0.002 0 200 400 600 800 1000 1200 Chloride Ion Concentration (PPM) Fr eq ue nc y Chloride Ion Distribution for Susceptibility Index 5 0 0.0002 0.0004 0.0006 0.0008 0.001 0.0012 0.0014 0.0016 0.0018 0.002 0 200 400 600 800 1000 1200 Chloride Ion Concentration (PPM) Fr eq ue nc y Chloride Ion Distribution for Susceptibility Index 4 0 0.0002 0.0004 0.0006 0.0008 0.001 0.0012 0.0014 0.0016 0.0018 0 200 400 600 800 1000 Chloride Ion Concentration (PPM) Fr eq ue nc y Chloride Ion Distribution for Susceptibility Index 3 0 0.0002 0.0004 0.0006 0.0008 0.001 0.0012 0.0014 0.0016 0.0018 0 200 400 600 800 1000 Chloride Ion Concentration (PPM) Fr eq ue nc y Figure 2. Chloride distribution at various values of SI.

sealers provide varying degrees of barrier to the transport of chloride ions into the concrete. • The term “membrane” is used to denote surface-applied systems that do not allow transport of moisture in either direction and are considered waterproofing material. These systems, as long as their integrity is maintained, do not allow moisture or chloride ions to enter the concrete. In this category, the asphalt overlay with the waterproofing membrane is also included. • Overlays include all kinds of concrete overlays, and they are installed to decrease the permeability of the top layer of concrete and to increase the depth that the chloride ions have to travel to reach the steel. In addition to providing corrosion benefits, the concrete overlays serve as a wearing surface on bridge decks. • The corrosion inhibitor category includes all materials that have chemicals that can interfere with the corrosion process or parts of the corrosion process (such as the cathodic reaction or the anodic reaction) and reduce or stop the corrosion. The corrosion inhibitors can be surface applied or admixed with repair concrete. Some manufac- turers market a combination of corrosion inhibitors, pore pluggers, and sealers. These materials are also included in the corrosion inhibitor category. • The cathodic protection systems include galvanic systems, impressed current systems, and re-alkalization (which is considered a part of electrochemical chloride extraction). Formulation of a Repair and Corrosion Control Strategy Several different factors are considered in selecting the repair and the corrosion control strategy: 1. Type of superstructure element (i.e., deck, beams, and girders). 2. Type of reinforcement (conventional, prestressed, post- tensioned, black, or epoxy coated). 3. Quantity and distribution of damage over the concrete element. 4. Susceptibility of sound concrete to corrosion-induced damage in the future. 5. Severity of exposure. 6. Presence of other concrete deterioration processes, such as freeze-thaw and ASR. 7. Structural deficiencies. 8. Conformance with newer codes and the need to upgrade. 9. Structure close-down and phasing requirements. Only Items 1 to 6 are discussed in this chapter; the other items and prestressed concrete elements are beyond the scope of this manual. The type of superstructure element plays a significant role in the selection of the repair and corrosion control system. In the deicing salt environment, the deck generally experiences the most severe exposure condition. The other superstructure ele- ments, such as the beams and the girders,are sheltered under the deck and are not directly exposed to deicing salts or the weather. They are most often only exposed to the corrosive environment in areas where the expansion joints in the deck have failed and chloride-contaminated water is running off onto the beams and girders. Therefore, these superstructure elements most often only require repairs and a local corrosion control system. It is possible that the beams and girders may have admixed chloride ions. However, because they are generally precast, they are less likely to contain admixed chloride ions. The beams and the girders can suffer from other concrete deteri- oration processes, such as freeze-thaw and ASR, and may experience cracking, which would accelerate the diffusion of chloride ions into the concrete and require a less severe expo- sure to chloride ions to result in corrosion-induced damage. Both freeze-thaw and ASR require direct exposure to mois- ture to generate cracking. Because two exposed surfaces of the beams and the girders are vertical and the third one is the underside, the surfaces do not experience a severe enough exposure to moisture to suffer sufficient cracking necessary to accelerate the diffusion of chloride ions. In a marine environ- ment, when the elevation of the beams and the girders are as low as, or lower than, 15 feet from the high water mark, then they are exposed to a sufficiently corrosive environment and 24 CATHODIC PROTECTION, ELECTROCHEMICAL EXTRACTION SEALERS 4 MEMBRANES OVERLAYS & OVERLAYS + MEMBRANES CORROSION INHIBITORS 6 Do Nothing 5 7 9 10SI ≤ 0 1 2 3 Figure 3. Optimal solutions for various values of Susceptibility Index.

can experience corrosion-induced damage that will require repair and a global corrosion control system. The type of reinforcement plays a significant role in the selection of the repair type and corrosion control system. The majority of the corrosion control systems are targeted toward the black reinforcing steel type. If epoxy-coated rebars are present, a compatible repair and corrosion control system will be needed. In addition to repairing concrete, some form of protection will need to be provided to the damaged epoxy coating in the repair area. The report of the NCHRP 10-37C study documents the various repair and corrosion control strategies that are compatible with epoxy-coated rebars [35]. The quantity of damage plays a significant role in the selec- tion of the repair type and the corrosion control system because it impacts both the structural integrity of the element and the cost of various options of repair. If the element is in question or the element is considered to have failed, then repair is needed to restore the service capac- ity of the element. When the level of damage is not sufficient to impact the integrity of the element, then the cost of vari- ous repair options govern the selection of the repair. For example, on a bridge deck, the quantity of damage may be such that partial-depth replacement along with full-depth repair using a single concrete pour may be more cost-effective than patching each damaged area individually. This par- ticular option of partial-depth replacement is not only a repair, but also a corrosion control system. In this approach, the contaminated top layer of concrete is removed by stan- dard or hydrodemolition techniques and replaced with fresh concrete. If the depth of replacement has been selected to ensure that the majority of the chloride contamination has been removed, then the deck is as good as new. Otherwise, corrosion of the bottom mat of reinforcing steel will result from the formation of a large macrocell between the top mat, which is in fresh concrete, and the bottom mat, which is in contaminated concrete. The soffit repairs are more expensive to perform because they require access from the bottom of the structure. As concrete fails on the bottom surface, chunks of concrete can fall and cause injury to commuters below or cause damage to property. The diffusion model or the chlo- ride profiles from concrete cores can be used to determine the depth to which the concrete has been contaminated with chloride ions and the depth of concrete removal. Another option is to place a monolithic layer of concrete to repair damaged areas and install an overlay to provide a new wear- ing surface and increase the clear concrete cover over rein- forcing steel. This approach increases the time to corrosion initiation and can result in a significant extension in service life if chloride ions are not left in the original deck concrete in sufficient quantities to continue to initiate and support the corrosion reaction. Otherwise, corrosion may continue to occur and result in damage to the original concrete. The dam- age occurring in the original concrete may not be reflected immediately upon formation and may become noticeable only after significant damage has occurred under the overlay. The selection of an overlay as the sole corrosion control sys- tem depends on the SI. Figure 3 shows that such layers may be used only when the SI is equal to or greater than 4. In conjunction with the selection of the repair type, the type of corrosion control system must also be identified. Two types of repairs that also served as the corrosion control system were discussed above. However, these types of repairs are effective only when the susceptibility of the sound concrete to corrosion is low and the SI is equal to or greater than 4. When the con- centration of chloride ions in the concrete is sufficient to result in a SI of 4 or less, more aggressive corrosion control systems such as corrosion inhibitors, cathodic protection, and electro- chemical chloride extraction must be considered. Numerous products and systems are available. Because each product has its own limitations and applications, and some are more effec- tive than others in a given application, selection must be made with care. A brief discussion of some of the more popular cor- rosion control systems is provided in Chapter 5. The quantity of damage and the SI provide a good defini- tion of the condition of the concrete element and provide suf- ficient information on the types of repairs and the types of corrosion control systems that may be effective for that par- ticular structure. The alternative strategies so identified must be further refined by considering the exposure condition. Although the SI may suggest a category of corrosion control system, some of the products in that category may not be as robust and reliable in the exposure condition that the struc- ture experiences. For example, arc-sprayed zinc galvanic cathodic systems fall in the category of cathodic protection system, but they do not work well in dry conditions. Similarly, certain corrosion inhibitors may not be applicable to certain environments. A sealer system may be more appropriate on a beam or a girder but less appropriate on a bridge deck because of possible deterioration under traffic. Finally,other deterioration processes have a significant impact on the selection of the repair and corrosion control system. When freeze-thaw damage is occurring on the concrete element, some form of waterproofing will be required to reduce the direct exposure of the concrete surface to moisture. Cathodic protec- tion systems and electrochemical chloride extraction technolo- gies can, under certain circumstances, accelerate the ASR and generate more cracking damage on the concrete. When selecting the type of repair and corrosion control system, one must consider all of the factors discussed above. Sufficient information must be collected on the structure, including during the condition evaluation, to allow consider- ation of all the above factors. 25