Modelling of chemical evolution of low pH cements at long term

International audienceIn the context of the underground radioactive waste repository, low-pH cements were developed to reduce interactions between concrete and clay barrier. These cements contain high proportions of mineral additions like silica fume, fly ash or blast furnace slag for example. The high ratio of cement replacement by pozzolanic additions allows to reduce the pH by a global reduction of Ca/Si ratio of the hydrates (according to the one observed on CEM I pastes). In order to predict the short term development of the hydration for each component of this cement, a multiphasic hydration model, previously developed, is used. The model predicts the evolution of hydration degree of each anhydrous phase and consequently the quantity of each hydrate in paste (CH, aluminates, CSH with different Ca/Si ratios). However, this model is not suitable to determine the long term mineralogical and chemical evolution of the material, due to the internal change induced by chemical imbalance between initial hydrates. In order to evaluate the chemical characteristics of low pH cement based materials, and thus assess its chemical stability in the context of radioactive waste storage, a complementary model of chemical evolution at long term is proposed. This original model is based on "solid-solution" principles. It assumes that the microdiffusion of calcium plays a major role to explain how the different Ca/Si ratio of initial C-S-H tends together toward a medium stabilized value. The main mechanisms and full development of the model equations are presented first. Next, a comparison of the model with experimental data issue from EDS analysis on low pH cement allows to test the model