Mesostructure-induced uncertainty of chloride transport in concrete

Chloride-induced corrosion is a primary cause of performance degradation and lifespan attenuation in reinforced concrete (RC) structures in marine and de-icing salt environments. This paper focuses on identifying and modeling the uncertainty of chloride transport induced by the mesoscopic structure of concrete. Utilizing mesoscopic statistics conducted on random concrete specimens with different volume fractions, this investigation elucidates the probabilistic characteristics and the spatial and temporal variation of chloride concentration and diffusion coefficients. Building on the quantitative relationships and Spearman correlations between aggregate characteristic parameters and chloride diffusion coefficients derived from mesoscopic statistics, a prediction model for chloride diffusion coefficients incorporating aggregate volume fractions and particle sizes was developed and validated. On this basis, a mesoscopic probabilistic framework for chloride diffusion coefficients was proposed. The findings reveal the spatial and temporal variability in chloride concentration and diffusion coefficient. The chloride concentration approximates a normal or log-normal distribution, whereas the diffusion coefficient approximates a log-normal distribution. The chloride diffusion coefficient generally shows a noticeable positive Spearman correlation with the maximum and average aggregate sizes and a negative correlation with the aggregate particle number. The research is beneficial for probabilistic chloride transport and lifespan estimation in RC structures.