Based on a review of the literature

Based on a review of the literature data and an analysis of the model developed by Shanti et al., we propose here that the densification of ice templated ceramics is governed both by the influence of the osmotic pressure during the solidification stage and by the morphology of the macropores, and more specifically the radius of curvature of the solid/vapour interface. The choice of a solvent for the process, such as water, camphene, dioxane or glycerol, will have a partial influence through the variation of the surface tension, which affects local particles packing and therefore the balance between macroporosity and microporosity, but this alone cannot explain the magnitude of porosity variations observed experimentally. The Shanti derivation neglects the effect of the solvent on the osmotic pressure.Instead of Shanti derivation one should, for instance, use the more general equation of the osmotic pressure derived by White. Variations of the osmotic pressure when changing the solvent can yield significantly different behaviour during the freezing stage and therefore significantly affect the densification behaviour, an effect not apparent when using the Shanti model. In addition,and equally important, a modification of the pores morphology is induced by the evolution of the solvent crystals morphology when changing the solvent. Highly dendritic and small crystals will facilitate the densification of the green body and yield porous materials with a lower total porosity. Determination of the osmotic pressure in the typical ice-templating suspensions is required to validate these propositions.