Effects of pressure solution and phyllosilicates on the slip and sompaction behaviour of crustal faults

This thesis aims to elucidate the effects of pressure solution and phyllosilicates on the compaction and shear behaviour of granular materials chosen to simulate fault gouge deforming under hydrothermal conditions. I implemented three different experimental programs. The first program consists of compaction experiments on quartz sand and phyllosilicate-bearing quartz sand. Clear evidence was found for compaction by IPS. The main conclusion is that IPS compaction under the conditions studied, is controlled by the rate of dissolution of quartz within the grain contacts. The presence of muscovite hinders compaction, through a chemical retardation effect, on the dissolution rate, of aluminium ions dissolved in the fluid from the muscovite. The second program consists of high strain rotary shear experiments performed on simulated fault gouges consisting of halite-muscovite mixtures. I observed velocity-strengthening behaviour up to a sliding velocity of ~0.3-1 μm/s, along with the development of a mylonitic microstructure consisting of an anastomosing foliation enveloping elongate halite clasts. The deformation mechanism was slip on the phyllosilicate foliation with accommodation by pressure solution of the intervening halite. At higher velocities, I observed a velocity weakening effect in the mixtures of halite and muscovite. This was accompanied by a cataclastic microstructure with no foliation but with significant porosity development. I explain this behaviour in terms of a granular flow mechanism involving competition between dilatation and pressure solution controlled compaction. The time-dependence of compaction leads to higher porosities at higher sliding velocities. Higher porosity implies a lower dilatancy angle for granular flow and thus faster shear produces a lower macroscopic shear strength. Slide-hold-slide experiments show that samples deformed in the low-velocity regime do not restrengthen on re-shear, while those deformed in the high velocity regime regain high strength. Healing rates, recorded for the latter samples, increase with sliding velocity. This is explained by a higher steady state porosity being maintained during steady state sliding, so that compaction is increased during hold periods. The increased compaction leads to an increased granular dilatancy angle, hence higher healing rates following rapid slip. The third program consists of rotary shear experiments on simulated quartz gouges under hydrothermal conditions. They show strain hardening up to a shear strain γ of 0.6-1.8, followed by weakening of up to 30% towards a steady state value at a strain γ of ~8-12. The steady state shear strength increases with decreasing grain size and temperature and increasing sliding velocity. The microstructure of the quartz gouge is characterised by the presence of a through-going boundary-parallel Y-shear. Deformation was largely by cataclastic processes, with most displacement being accommodated along the boundary-parallel Y-shear, causing the strong weakening observed. In Chapter 7, I present a microphysical model predicting the steady state velocity-weakening behaviourfor analogue samples deformed at high sliding velocities. The model is based on a quantitative description of the competition between dilatation and IPS-controlled compaction. The model results agree favourably with the experimental results. Extrapolation of the model to natural conditions shows a velocity weakening effect which is an order of magnitude larger than previously seen