Limit of validity of the moisture diffusivity for the study of moisture transfer inside terracotta

In order to know the limit of validity of the concept of hydrical diffusivity, D_W, we tried to validate the Boltzmann's method used to identify it. We show that for a medium strictly satisfying the equation of diffusion, the method gives quite good results with nevertheless some inaccuracies for the extreme values of water contents. For the real studied material, a backed clay, its identification becomes more difficult because of the dispersion of the wetting phenomenon observed for different specimens of the same material. We are showing that the mean behaviour of one sample is near to be the same as the more probable behaviour of all the samples. Otherwise the values of D_W determined for a wetting cannot be used to describe a drying process. It practical interest is then considerably reduced. In order to test the influence of the gas phase pressure effects on the behaviour of the material, we have determined all the parameters necessary to describe, numerically, wetting and drying processes without to resort to D_W. We noticed an overpressure of the gas phase inside the material due to drying. If the gas pressure effects are, numerically, switched off, we cannot notice any influence on water content profiles for this type of material.