A critical view on service life predictions based on chloride induced corrosion

A critical look is taken on the parameters used for service life predictions of reinforced concrete when assuming the limit state is initiation of chloride induced corrosion and the chloride ingress mechanism is diffusion. The fib model code and Life 365 are compared in this respect. Consequences of letting the aging factor work for the whole service life (as in the fib model code) versus letting it be a discontinuous function where it works for a certain time and thereafter being constant (as in Life 365) are discussed. The concept of a discontinuous aging factor is recommended. The surface concentration and apparent diffusion coefficient is coupled when fitting the total chloride content vs. distance from the concrete surface. Also, the total chloride surface concentration measured for an old concrete in marine tidal zone is much higher than the value predicted when assuming the porosity saturated with seawater caused by physical and chemical binding of chlorides. Furthermore, measured values of total surface chloride content are still much higher than predicted from laboratory measured chloride binding isotherms. This indicates that laboratory measurements do not mimic the field by the procedures commonly applied for chloride binding isotherms.