Corrosion of metal reinforcements within concrete and localisation of supporting reactions under natural conditions

Corrosion in concrete prevents in-situ observation, necessitating models to provide insight into the local reaction currents. We present a computational method for predicting corrosion rates of reinforcements within concrete under natural conditions, i.e. requiring the corrosion current to be supported by equal cathodic currents. In contrast to typical corrosion models, where these two counteracting currents are required to be co-located, we allow these currents to be separated such that pitting corrosion can be supported by cathodic reactions over a much larger area. Pitting corrosion is investigated, elucidating the effects of the concrete porosity and water saturation, the presence of dissolved oxygen, and chlorine concentration within the pore solution. The presented model is capable of capturing the dynamic growth of acidic regions around corrosion pits, showing the limited region over which the hydrogen evolution reaction occurs and how this region evolves over time. The ability of oxygen to diffuse towards the metal surface due to increased porosity is seen to have a major effect on the corrosion rate, whereas changes in the chlorine concentration (and thus changes in the conductivity of the pore solution) play a secondary role. Furthermore, external oxygen is seen to enhance corrosion but is not required to initialise and sustain acidic corrosion pits. • We present a model for self-sustained corrosion of concrete reinforcements. • The location of cathodic reactions, metal potential and relevant currents are a natural output of the model. • The hydrogen reaction attains high reaction currents, localised around corrosion pits. • Oxygen initially present in concrete can sustain corrosion for over a month. • Estimates of corrosion based on concrete porosity and pore fluid salinity are given, agreeing with experiments.