The corrosion performance over a nearly 30-year service period of epoxy-coated rebar (ECR) in Florida marine bridges is presented. Severe ECR corrosion was noted earlier in several of those bridges, built with relatively high permeability concrete. Corrosion development took place later in some other structures with less permeable concrete, with instances where the corrosion was associated with preexisting cracks. Corrosion was found to be associated with coating imperfections and coating disbondment. Laboratory experiments and modeling indicated that macrocell coupling with remote cathodes was an aggravating factor. Quantitative damage functions relating the observed deterioration with service time of the affected bridges are presented. A predictive corrosion initiation-propagation model was developed and the model output is compared with the field results to identify suitable parameters for forecasting future rehabilitation needs. The prognosis for future corrosion development is discussed, with attention to corrosion of ECR in otherwise low-permeability concrete with preexisting cracks or local deficiencies.

Severe corrosion recently documented in Florida post-tensioned bridges were related to grout segregation that created deficient grout with characteristics of having low chloride content, high sulfate concentration, high pore water pH, and high moisture content. The role of elevated sulfate concentrations in the development of steel corrosion in deficient grout has not been elucidated. As first approach to determine the corrosion mechanism of steel strand in deficient grout, the objective of the research, described here, was to evaluate the role of sulfate ions on the corrosion of steel in alkaline solutions. Steel coupons were exposed to two different alkaline solutions (SPS) simulating pore solutions with pH 12.6 (SPS1) or 13.3 (SPS2) with varying sodium sulfate levels (early fixed content and later increments). Electrochemical experiments included the steady-state condition and potentiodynamic polarization tests. Solution pH was found as an important parameter in the corrosion development of steel in sulfate SPS solutions. In the highly SPS2 solution (pH 13), sulfate ions did not impair passive film growth or stability. In the moderate SPS1 solution (pH 12.6), early fixed sulfate presence could be aggressive by impairing passive film development; however, even high levels of later increments of sulfates cannot depassivate steel. The instability of the passive film in alkaline sulfate solutions resulted in pitting corrosion of steel.

The chloride corrosion threshold of steel in concrete depends not only on concrete and steel properties but also on the potential of the steel while it is still in the passive condition. An innovative model to predict corrosion in reinforced concrete structures is presented, incorporating potential dependence and corrosion macrocell effects. The novel approach integrates the initiation and the propagation stages of corrosion in a single predictive model. The model is applied to a generic, partially submerged, reinforced concrete marine pile. The purpose of this paper was to establish and illustrate the concept of a functional approach to implement potential-dependent chloride threshold in corrosion damage projections, plus articulation of the initiation and propagation stages, in one predictive model. The long-term damage projection results are markedly lower contrasted with those of an identical system with a time-invariant chloride threshold. The results show that implementing chloride threshold dependence on potential can reveal otherwise ignored damage progression trends that may impact structural design decisions to control corrosion.