Toward a probabilistic design of reinforced concrete durability: application to a marine environment

This work aims to propose a complete design methodology for concrete durability problems based on a probabilistic method including an original chloride diffusion modelling compatible with the new performance-based approaches. This is illustrated by the study of a concrete immersed in sea water. The physical and chemical processes that lead to the corrosion of the concrete reinforcement bars are presented first. A chloride penetration model based on Fick's second law is proposed. Next, the durability modelling parameters chosen among the durability indicators (Baroghel-Bouny, Concrete design for structures with predefined service life-durability control with respect to reinforcement corrosion and alkali-silica reaction, 2004; Alexander et al., Mater Struct 41:921-936, 2008) are assessed. They depend on the concrete formulation and the chemical composition of the cement. These indicators are characterized by their statistical distributions, which are realistically specified from a wide literature review. The whole probabilistic modelling is included in a Bayesian network so that it can be easily updated to include new experimental data. The evaluation of the time dependant corrosion risk is estimated for two types of cement: CEM I and CEM I with silica fume. The result shows the effect on the Lind-Hasofer reliability index of the type of cement, the concrete quality and the design options. The quality is integrated through the mean value and the standard deviation of the modelling parameters. The method could be used either directly for cover design or for semi-probabilistic design code calibration.