Corrosion Cell Formation on a Bar Embedded in Concrete Exposed to Chlorides

Abstract

This thesis investigated corrosion of a reinforcing steel bar embedded in concrete and the effect of corrosion coupling on the bar/concrete interface induced by the variation of corrosion potentials with concrete depth. Two separate numerical models were used to simulate the corrosion process which included: a two-dimensional finite element model for mass transport of oxygen and chloride to the bar/concrete interface; and, a one-dimensional model for the corrosion current flow through the electrolyte induced by corrosion potential differences on the bar/concrete interface. A novel approach to corrosion modeling in reinforced concrete that had not been identified in the literature was used. This new approach, incorporated: variable solution conductivity developed from concentrations within the pore solution; anodic and cathodic areas modified to maximize corrosion current through the electrolyte; and, kinetics of corrosion set by the pore solution chemistry. Various reinforcing configurations and moisture conditions were evaluated within the simulation to obtain insight into the effect these variables have on corrosion potentials measured on the concrete surface and the corresponding corrosion currents generated on the bar/concrete interface. Variables related to bar diameter, concrete cover, and bar spacing where all shown to affect corrosion potentials and current densities on the bar/concrete interface and the concrete surface where field measurements are obtained. Moisture conditions were found to have the largest impact on corrosion potentials and current density’s on the bar/concrete interface. When relative humidity’s of 90% or higher were used, simulated corrosion potentials on the concrete surface under high chloride conditions were found to reach values identified in ASTM C876 and Alberta Transportations Deck Testing Guidelines that indicate active corrosion. However, when moisture conditions were reduced to below 90% relative humidity, simulated corrosion potentials on the concrete surface for high chloride concentrations did not achieve values that indicate a high probability of corrosion. This result suggests a secondary mechanism must be present on the bar/concrete interface that changes the chemical composition within the pore solution to shift the kinetics of corrosion to an environment that will produce the negative corrosion potentials recognized as indicating a high probability of corrosion. Therefore, a new mechanism is proposed that outlines the process necessary for the pore solution on the bar/concrete interface to transition the kinetics of corrosion to an actively corroding state at low relative humidity. This mechanism requires local acidification of the pore solution along portions of the bar where anodic processes are increased due to the presence of chloride and reduced oxygen availability. Reaching this environment requires free OH- to be consumed without replenishment from the surrounding environment by either diffusion from high pH areas or dissolution of the hardened portions of the pore structure. The proposed mechanism begins with corrosion by-products formed when Fe2+ reacts with free OH-, precipitates from the pore solution onto the pore structure as Fe(OH)2. Once precipitated, the contact area between pore solution and hardened portions of the pore structure are reduced which restricts the dissolution process for restoring OH- removed from the electrolyte. Additionally, precipitation of Fe(OH)2 reduces the flow of OH- from the surrounding high pH zones as the pore structure is restricted. Both mechanisms result in a pH gradient being formed with acidified zones created on the bar/concrete interface in the anodic regions. These acidified zones cause the kinetics of corrosion to transition from a passivated state, towards an environment similar to carbonation.