Prediction of chloride ingress and binding in cement paste

Abstract

This paper summarizes recent work on an analytical model for predicting the ingress rate of chlorides in cement-based materials. An integral part of this is a thermodynamic model for predicting the phase equilibria in hydrated Portland cement. The model’s ability to predict chloride binding in Portland cement pastes at any content of chloride, alkalis, sulfates and carbonate was verified experimentally and found to be equally valid when applied to other data in the literature. The thermodynamic model for predicting the phase equilibria in hydrated Portland cement was introduced into an existing Finite Difference Model for the ingress of chlorides into concrete which takes into account its multi-component nature. The “composite theory” was then used to predict the diffusivity of each ion based on the phase assemblage present in the hydrated Portland cement paste. Agreement was found between profiles for the Cl/Ca ratio predicted by the model and those determined experimentally on 0.45 water/powder ratio Portland cement pastes exposed to 650 mM NaCl for 70 days. This confirms the assumption of essentially instantaneous binding where quasi-equilibrium is established locally. This does not imply steady state diffusion however. It simply implies that incremental increases in the concentration of diffusing ions in the pore solution will rapidly re-equilibrate with the hydrates present locally, where, the greater the ratio of bound to free ions, the greater the buffering effect which slows down the rate of ingress. In the case of chlorides, this buffering effect is greatest at high contents of AFm (AFm is aluminium ferrite compounds with a single (mono) formula unit CaX.) phase and low alkali metal contents.