Concrete structures suffer from corrosion if chloride ions in the environment penetrate to the depth of reinforcing steel; surfaces periodically wetted with chloride solution are at most risk, when chloride ions can enter concrete by combined diffusion and absorption. BS8500-1 recommended minimum cover requirements primarily assume that chloride ingress is diffusion controlled, and absorption is only partly accounted for, so cover values may be inadequate. This paper investigates factors that influence absorption of chloride ions into concrete and how this affects distribution of chloride at different depths from the surface. Concrete cubes (CEM I cement; water/cement ratio 0 . 45) are subjected to repeated wetting (with 50% saturated salt solution) and drying cycles, during which their mass is monitored. The effective porosity, weight and distance sorptivity and chloride penetration depths are calculated for a number of curing, conditioning and drying conditions. Samples are extracted from the cubes at a range of depths and analysed for chloride content. Results show that the quantity of chloride entering the concrete, and in particular surface chloride content, is very sensitive to effective porosity/drying conditions immediately before wetting; also as much as 31% of the protection provided by concrete cover can be lost after exposure to just one wet/dry cycle, thereby significantly reducing time to corrosion of concrete structures.

The objective of the work is to investigate the effect on the distribution of chloride at different depths of some factors concerning concrete pore structure, namely water/cementitious material ratio, cement type and percentage ground granulated blast furnace slag (GGBS). Concrete cubes are subjected to repeated wetting and drying cycles with various concentrations of sodium chloride solution, during which their mass is monitored. The effective porosity, weight and distance sorptivity and chloride solution penetration depths are calculated. Samples are extracted from the cubes at a range of depths and analysed for chloride content. The results show that the most significant effect of sorptivity on long-term chloride ingress to concrete is its effect on surface chloride content.The value of this parameter is a way of taking account of absorption when modelling chloride ingress under cyclic wetting and drying conditions. The values of cover depth recommended in BS8500-1 that are needed to ensure a time to corrosion of 100 year are calculated assuming the surface chloride content is 0 . 14% for GGBS and pulverised fuel ash (PFA) concretes. Our cyclic wetting and drying experiments produce higher surface chloride contents (0 . 29-0 . 62%) that would lower the time to corrosion using the cover depths recommended in the code.

Corrosion of steel in concrete structures remains a major problem worldwide, and surfaces periodically wetted with chloride solution are particularly vulnerable. BS 8500-1 cover requirements have been calculated using the error function solution to Fick's second law of diffusion. However, the values of surface chloride content (Cs) assessed via the diffusion coefficient are rather low, suggesting the cover thicknesses may have been underestimated, which could account for the high incidence of reinforcement corrosion. This paper investigates the influence of a number of factors on Cs and the diffusion coefficient and considers the implications for modelling chloride ingress. Concrete cubes made from 100% CEM I, 70% CEM I + 30% pfa (pulverised fuel ash) and 50% CEM I + 50% GGBS (ground granulated blastfurnace slag) cements were subjected to either one-week (1 d wetting with 3%, 10% or 50% sodium chloride solutions and 6 d drying at 20°C) or two-week (2 d wetting with 50% sodium chloride solution and 12 d drying at 20°C, 30°C or 40°C) cycles, in both cases for 24 weeks during which time their mass was monitored. Identical specimens were totally immersed in 3%, 10% and 50% sodium chloride solutions, also for 24 weeks. The distribution of chloride with depth was measured. The specimens subjected to wet–dry cycling initially appeared to have high diffusion coefficients that quickly reduced to values comparable with the fully submerged specimens, suggesting that the former values are predominantly a function of moisture content and the latter a function of pore structure. The results further show that Cs increases with an increasing number of wet–dry cycles, the final value being a function of the diffusion coefficient, effective porosity and concentration of the chloride solution. Cs exerts a strong influence on the depth of chloride penetration and current estimates of Cs do not appear to be suitable for prediction purposes.