Updating of chloride ingress models using core data measurements

Abstract

A large proportion of concrete bridges in North America are reaching the end of their service life. An important factor that accelerates deterioration is the amount of salts used during winter months. The chloride ions from the salt enter the concrete and destroy the passive-oxide layer that protects the reinforcing steel from corrosion. The TransChlor® software is used to simulate the time history of chloride ion ingress into a typical concrete bridge. A climate model is used to simulate the exposure of the structure to winter conditions. A probabilistic method is used to propagate the uncertainty on material properties in the model and to obtain the distribution of chloride ion concentrations. A procedure for updating the distribution is proposed using data from core samples at two depths (25/50 mm) and for the two types of exposure (splash/direct). The procedure is shown to be very efficient in reducing bias and variance from model predictions.