Precipitation from supersaturated aqueous fluids in
rocks is the principal mode of mineralization in nature. In modelling natural crystal growth processes experimentally it is necessary to estimate the degree of this supersaturation. Conventional wisdom suggests that the supersaturation is 'not very high', based on the assumption that in a material as heterogeneous as a natural rock, nucleation should not be a problem. However, fluids in natural rocks migrate along grain boundaries and other porous features - the fluid is not 'free' in the sense of allowing convection to be the principal cause of mass transport. Such small volumes of fluid are subject to departures from equilibrium according to the universal relationship between supersaturation and volume:
Supersaturation = A-B lnv
where A and B are constants and v is the fluid volume. Fluids constrained in porous media also do not have uniform supersaturation in space and time. The existence of supersaturation gradients which develop if mass transport is diffusion controlled requires a kinetic definition of the supersaturation concept, distinct from the thermodynamic definition of critical supersaturation, which assumes free solutions with a uniform composition and an infinite time-scale. This kinetic concept, termed the threshold supersaturation, bth, is related to the supersaturation rate Rb (i.e. the rate of change of supersaturation with distance or time) by the expression
Rb = K(bth)m
where K and m are constants.
The validity of this expression can be demonstrated in experiments when supersaturation gradients are established in space by diffusion controlled crystal growth in a medium with restricted porosity (silica gel), as well as when supersaturation gradients exist in time (by crystallisation as a function of cooling rate). Emphasis in this paper is on experimental crystallisation of BaSO4, CaSO4, SrCO3 and BaCO3 by counter diffusion of cations and anions in a microporous silica gel transport system. The results indicate that crystals of highly insoluble salts grow at very high supersaturations when mass transport is diffusion controlled. The range
of supersaturation experimentally available is extended
by doping the fluid in the gel with organic inhibitors. The development of the morphology can be studied in the pure and doped systems and show many features similar to that in natural crystallization. Other evidence such as estimates of the volume of fluid available for crystallization and the distribution of mineralisation in space and time also support the conclusion that natural crystal growth takes place at much higher supersaturations than previously assumed.
The structure of a crystal surface depends on the supersaturation of the fluid with which it is in contact. The consequences of high supersaturation are an increase in the surface roughness and a change in the mechanism and kinetics of crystal growth. The role of kinetics is particularly important when considering the partitioning of elements between the fluid phase and the crystal surface. Under non-equilibrium conditions the expected partitioning between fluid and crystal in a two component system can be surface dependent leading to sector zoning or completely reversed, leading to oscillatory zoning. Examples of the experimental generation of such zoning patterns will be given for Mg-calcites and Ba,Sr sulphates. The role of supersaturation on the morphological development will also be described in these systems.
Fernández-Díaz, L., Putnis, A., Prieto, M. & Putnis, C.V., Journ. Sed. Res. (1996, in press).
Putnis, A., Prieto, M. & Fernández-Díaz, L., Geol. Mag. 132, 1-13 (1995).