A simple model for rock deformation by pressure solution, assuming grain boundary diffusive mass transfer to be deformation rate controlling, is presented. The model leads to a constitutive flow law which is of the same form as that for Coble creep. It is argued that the presence of a fluid film in stressed grain boundaries leads to enhanced diffusivity of solute particles in the grain boundary. Some simple experiments are described, which demonstrate rapid diffusion in solutions in pores, much slower diffusion in stressed interfaces and deformation by pressure solution. By using the theoretical model, and by assuming that the pressure of the interfacial solution is equal to the applied normal stress, so that available experimental data on the effect of pressure on mineral solubility could be used, rates of deformation by pressure solution have been calculated. These are compared with rates of deformation by crystal plastic and high temperature diffusive flow processes, by using deformation mechanism maps. Predicted transition conditions between various deformation mechanisms are found to be consistent with those inferred from the study of textures of naturally deformed rocks.

Some of the features of water enhanced deformation of rocks by diffusive mass transfer (pressure solution) in nature which are pertinent to the rate controlling mechanism of the deformation are reviewed, and it is inferred that (a) the diffusion of matter in an aqueous intergranular film which can support shear stress is an essential part of the process, and (b) the diffusion is driven by stress induced chemical potential gradients, together with gradients due to local chemical reactions. The theoretical approach to the derivation of constitutive flow laws for creep by diffusive mass transfer is outlined, and a simplified flow law proposed. Crucial to the absolute rate of deformation predicted by the flow law is the estimation of the phenomenological coefficient which links diffusive flux to chemical potential gradient. It is argued that this should be several orders of magnitude less in thin, stressed aqueous films than for solutions of ions in large water volumes. Some simple experiments are described to address the question of (a) the existence of thin intergranular aqueous films which can support shear stress, and(b) the magnitude of the above phenomenological coefficient. The results obtained are consistent with the inferences made from the study of microstructures in naturally deformed rocks, and this is illustrated by means of extrapolation of theoreticallv derived relationships to conditions of natural rock deformation and sediment compaction by pressure solution.