(1) Thermodynamics provides a general framework to the formulation of constitutive equations of unsaturated materials. The "simple" system analyzed here is a multi-phase system, constituted of s olid skeleton particles, bulk phases and liquid-liquid and solid-liquid interfaces which can sustain constant surface tensions (only temperature-dependent). A thermodynamic analysis of this system, at the macroscopic scale, can provide the conditions for which an effective stress could be assumed. Thanks to the concept of the Lagrangian saturation degree, we can show that a Bishop-like stress tensor accounting for surface tensions at the interfaces, can be derived in the elastic range provided that pores deform identically whenever subjected to a uniform pressure. This model was applied to a clay used in engineering barriers and compared to the elastic variations of the void ratio along different loading paths. The agreement between the model and experiment implies that the isodeformation hypothesis seems valid for elastic response of this material. The model can be extended to plasticity. If isodeformation of pores is also assumed for the plastic components, the same Bishop-like effective stress tensor is derived as work conjugate to the plastic deformations. The general approach shows however that in addition to mechanical hardening, there is also a saturation (or capillary) induced hardening hence putting the existence of effective stress into perspectives. A model as simple as possible was developed as a generalization of the Cam-clay model for which the capillary hardening can be calibrated on the basis of the Basic Barcelona Model. This model is applied to clay along different loading paths. The computed void ratio changes are compared to those of experiments.

 

(2) A model is proposed that accounts for the isothermal drying process of hardened cement-based materials. It constitutes a further development of previous French works. The equations of isothermal drying are derived (i) from mass balance equations written for the liquid water phase, water vapor and dry air, (ii) from the Fick’s law governing the relative diffusion process of water vapor and dry air to the gaseous mixture, and (iii) from the Darcy’s law describing the transport of wet air and liquid water. Intrinsic liquid water (Kl) and gas (Kg) permeabilities are distinguished, since the concept of intrinsic permeability, which is independent of the fluid nature, is not relevant for a cementitious material. New laws for gas transfers are introduced according to measurements on concrete specimens. Thus, a semi-empirical law gives the effective diffusion coefficient of water vapor vs. porosity and degree of liquid water saturation. In the same way, a new function, expressing the relative permeability to gas with respect to this degree of saturation, is proposed on the basis of experimental results. In order to describe the global movement of gas, viscous and slip flows are taken into account according to the Klinkenberg’s concept. A numerical study shows, on the one hand, that a gas depression (below the atmospheric gas pressure) can be observed and, on the other hand, that transfers of water in the gas phase may significantly contribute to the drying of cementitious materials in addition to liquid water transport by capillarity movements. Simplified approaches and their range of application are presented. Moreover, this presentation provides an insight into the impact of carbonation on the microstructure of OPC materials. Complementary techniques of microstructure investigation have been used and coupled with thermogravimetric analysis results. We propose an assessment of the evolution of porosity in relation to CSH carbonation. The changes in intrinsic permeability related to carbonation have been studied on a large panel of concretes by using two different methods: an inverse analysis of drying kinetics and the Katz-Thompson methods.