Forecasting chloride-induced reinforcement corrosion in concrete – effect of realistic reinforcement steel surface conditions

This paper presents experimental results on the effect of realistic steel surface conditions on the initiation of chloride-induced corrosion. Ten series of “as-received” reinforcement steel originating from different countries and having been subject to different exposure histories (storage, transport, handling, etc.) were studied. There were clear differences in visual appearance, which were reflected in electrochemical characterization (cyclic voltammetry). Metallographic analyses also indicated differences in terms of steel mi-crostructure. Neither the visual appearance nor the amount of initially present rust permitted any prognosis of resistance to chloride-induced corrosion. On the other hand, martensitic steels appeared to yield lower chlo-ride threshold values than ferritic-pearlitic microstructures. The results illustrate the difficulties that realistic conditions cause for the concept of the critical chloride content. Given the low degree of confidence with which corrosion initiation can be predicted, the feasibility of the concept of the critical chloride content is discussed.