Dependency of C–S–H carbonation rate on CO2 pressure to explain transition from accelerated tests to natural carbonation

The use of normalized accelerated carbonation tests is currently limited to the classification of concretes in terms of carbonation resistance and the results are not easily transposable to forecasting concrete carbonation in natural conditions. Common models assume that the kinetics of the carbonation front ingress in concrete is a square root function of the CO2 pressure but observations in the field generally invalidate this assumption. Based on an experimental program including carbonation tests at several CO2 pressures, this paper shows that the amount of carbonated product depends largely on the CO2 pressure. Several experimental analyses of carbonated concrete under different pressures are confronted, to finally propose a new analytical model able to predict carbonation ingress in natural conditions using the results of accelerated tests. The model takes both the cement chemical composition and its amount in concrete into account. The carbonation kinetics dependence on CO2 pressure is considered through two underlying functions including, for the first, the dependence of the CSH carbonation rate on the pressure and, for the second, the effect of this additional carbonation on the reduction of the CO2 diffusion coefficient.