Journalof StructuralGeology, Vol. 8, No. 8, pp. 845 to 856, 1986 0191-8141/86 $03.00 + 0.00 Printed in Great Britain Pergamon Journals Ltd. Experimental pressure solution-deposition on quartz grains: the crucial effect of the nature of the fluid J. P. GRATIER and R. GUIGUET IRIGM, University of Grenoble, B.P. 68, 38402 Saint-Martin-d'H6res, France (Received 12 June 1985; accepted in revisedform 3 January 1986) Abstract--Experimental deformation by pressure solution was performed on an aggregate of small grains subjected to deviatoric stress (50 MPa) for a long time (several weeks or months) at relatively high temperature and pressure in contact with various fluids (air, water, 0.1 to 1 N NaOH for quartz, water and 5% NH4C1 for calcite). The change in shape of the grains by solution-deposition depended; on the duration of the experiment (with the same fluid) and on the concentration of the solid in solution (with the same duration but various fluids). Significant shape changes were obtained for quartz grains, but only with both long duration and very good solvents (1 N NaOH). By comparison with previously obtained results on the change of shape of fluid inclusions (where the kinetics of dissolution was the rate controlling process), the limiting process of the deformation of the quartz grains was inferred to be the rate of diffusion along grain boundaries saturated with trapped fluid. INTRODUCTION FOLLOWINGPatterson (1976), the mechanisms of defor- mation that are possibly of importance in rocks can be conveniently grouped into three categories: cataclastic flow (with microfracturing and slipping), crystal plastic- ity (with glide motion of dislocations), and diffusional flow (with mass transfer). The deformation of rocks by pressure solution-deposition belongs to this last category (Elliott 1973). In comparison with the other mechanisms, this one needs an intergranular solution phase, it occurs at relatively low temperature (25- 400°C), low differential stress (1-100 MPa), and very slow strain rate (10-tl-10 -16 s -1) (Rutter 1976). This process is one of the most common mechanisms of ductile deformation for wet rocks in the upper crust. The mass transfer from solution zones (such as stylolites or solution cleavage) to zones with redeposition (such as veins or pores) leads to a change in shape (or a change in density) of rocks. This deformation may be analysed as a creep mechanism by establishing the relation between the strain rate and the various parameters, including driving force for mass transfer, nature of the solid and the fluid, temperature, pressure and geometry of the structure. THEORETICAL APPROACH The driving force for the solution-deposition process is often linked to the difference in normal stresses between the solid-fluid interface at the site of dissolution and that at the site of crystallization, but may be imposed by other forces such as the difference in elastic strain energy, in plastic strain energy or in surface energy between the two interfaces. All these forces lead to a difference in chemical potential between the dissolution zone and the crystallization zone (Paterson 1973, Robin 1978). Theoretically, the effect of a difference in normal stresses is higher than the others. To impose such a difference in normal stresses around a solid, it is neces- sary to put this solid in contact with another solid. Several authors have discussed the behaviour of a fluid phase trapped in an interface between two solids under compressive stresses (Weyl 1959, Bathurst 1975, Rutter 1976, Robin 1978). The rate of mass transfer by diffusion along this interface could be drastically reduced if the fluid phase was locally expelled from the interface. To avoid this problem, Pharr & Ashby (1981) proposed that the mass transfer could occur in the free fluid phase around the solid, where there is no effect of the differ- ence of normal stresses. Here, the driving force is the difference of surface energy linked to a difference of curvature of the solid boundaries. Gratier & Jenatton (1984) observed that a difference of curvature correlates with the amount of mass transfer around cigar-shaped microcavities (fluid inclusions) in heated synthetic quartz and calcite. It remains to make a careful investiga- tion of the pressure solution process along solid to solid contact with a trapped solvent, between two stressed grains. The preliminary results of an experimental approach, allowing observation of significant shape changes of small grains, are given in this paper. In all the cases (change in shape of cavities or grains), the rate of mass transfer is dependent on three successive processes: (i) the kinetics of dissolution at the solid/fluid interface; (ii) the rate of displacement of matter (by diffusion or infiltration) and (iii) the kinetics of the deposition process. If one of these three successive processes is much slower than the others, the overall rate of mass transfer is dominated by the kinetics of the slowest process (Raj 1982). It is generally assumed that the rate of displace- ment is the slowest process, but it has been shown experimentally that this is not always the case: the kinetics of interface reaction may also be the limiting process (Raj 1982, Gratier & Jenatton 1984). 845