A mechanistic approach For predicting the effect of various factors on partitioning between free and bound chlorides in concrete

The chloride-induced corrosion of reinforcing steel in concrete structures has become a widespread durability problem throughout the world. When concrete structures come in contact with chloride sources, the chloride ions will diffuse through the body of the concrete and ultimately reach the steel. Not all of the chloride ions which penetrate the concrete remain free in the pore solution. Some of the ions become bound to the hydration products in a chemical reaction to form calcium chloroaluminate hydrate (Friedel' salt). It is also well known that only the portion of the chloride ions that remains free is responsible for causing damage to the concrete structures by corroding steel rebar. Thus, the chloride binding capacity of the cementitious matrix plays a major role in controlling chlorides ingress and, consequently, the corrosion of steel reinforcement in concrete. The chloride binding capacity is affected by cement composition, environmental factors, and by the source of the chlorides ( vs. ). To quantify the durability of new and existing structures, a clear understanding of the mechanisms of chloride penetration into the concrete cover is required. Currently, most of the models available in the published literature for calculating free chloride ions in concrete use Fick’s law for chloride transport and chloride binding isotherms to account for bound chlorides. Binding isotherms are cement and environment specific. Thus, the existing models cannot be used for all types of cement and variable general environmental exposure conditions such as temperatures, pH levels, and chloride sources. A general mechanistic approach that can overcome those limitations is proposed in this thesis based on the concepts of ion-exchange theory for an accurate determination of chloride ingress in concrete under variable environmental conditions. Some of the model input parameters, such as exchange capacity and the equilibrium constant for the exchange reaction, were not easy to determine directly from experiments and were determined through an inverse modeling procedure. Verification experiments were carried out by varying different environmental parameters and making comparisons with the simulated results using the corresponding parameters. The experimental results showed that the proposed procedure is able to predict the amount of free chlorides in concrete, including predictions of chloride binding as a function of pH, temperature, chloride sources, and the presence of other ions such as carbonate. The proposed model was also used to clarify some unresolved issues such as the effect of chloride sources on binding and the effect of pH on the release of bound chlorides in the presence of carbonation