US Geological Survey


Date of this Version



Published in JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 100, NO. B7, PAGES 13,113-13,132, JULY I0, 1995


Pressure solution is widely regarded as a potentially important deformation
mechanism along crustal faults and during a diagenesis, yet the mechanisms and kinetics of this process remain highly controversial. To better understand the fundamental factors controlling the rates of pressure solution at the grain-to-grain scale, we conducted experiments in which convex halite lenses were pressed against fiats of fused silica in brine. Fluid pressures were maintained at 0.1 MPa; temperatures and mean contact normal stresses ranged from 8.3 ° to 90.2 °C and 0.5 to 13.5 MPa, respectively. The geometry and growth rate of the contact spot between the two lenses and the rate at which the lenses approached one another (convergence) were monitored using reflected light interferometry and transmitted light photomicrography. Convergence occurred when halite and silica lenses were pressed together in brine (halite/silica experiments). No undercutting was observed, and dry control experiments
indicated negligibled islocation creep. Convergence rates in experiments at 50.2 °C ranged from 0.01 to 0.05 μm/d, depending on mean normal stress and contact spot radius. The data are consistent with intergranular pressure solution (IPS) rate-limited by diffusion through an intergranular film with a very high diffusion coefficient (~10-5-10-7cm2/s). The data further suggests that the diffusion coefficient and/or thickness of this film increases with decreasing normal stresses, at least for normal stresses less than about 4 MPa. As no island-channel boundary structures were observed, we propose that this film consists of a continuous layer of strongly adsorbed (i.e., structured) water that is maintained between the halite and silica lenses during deformation. Convergence rates in similar experiments conducted at a constant load of 0.11 N but at 8.3 °, 50.2 °C were approximately constant at a given normal stress and contact spot size. The cause of this temperature insensitivity is unknown but might result from changes in interphase boundary structure or thickness with increasing temperature that are sufficient to offset the expected thermal activation. An experiment was also conducted in which a halite lens was pressed against a fused silica fiat coated with an 0.8- μm-thick film of Na-montmorillonite in brine. This clay film produced an approximately fivefold increase in convergence rates over those observed in a halite/ silica experiment conducted without clay at the same load and temperature. The strong sensitivity of IPS rates both to contact spot radius and to the presence of second phases along grain boundaries suggests that fine-grained, clay-rich fault gouges and multiphase granular aggregates should be particularly susceptible to pressure solution creep in the middle to upper crust.