The aim of this review is to characterize the role of pressure solution creep in the ductility of the Earth's upper crust and to describe how this creep mechanism competes and interacts with other deformation mechanisms. Pressure solution creep is a major mechanism of ductile deformation of the upper crust, accommodating basin compaction, folding, shear zone development, and fault creep and interseismic healing. However, its kinetics is strongly dependent on the composition of the rocks (mainly the presence of phyllosilicates minerals that activate pressure solution) and on its interaction with fracturing and healing processes (that activate and slow down pressure solution, respectively).The present review combines three approaches: natural observations, theoretical developments, and laboratory experiments. Natural observations can be used to identify the pressure solution markers necessary to evaluate creep law parameters, such as the nature of the material, the temperature and stress conditions or the geometry of mass transfer domains. Theoretical developments help to investigate the thermodynamics and kinetics of the processes and to build theoretical creep laws.Laboratory experiments are implemented in order to test the models and to measure creep law parameters such as driving forces and kinetic coefficients. Finally, applications are discussed for the modelling of sedimentary basin compaction and fault creep. The sensitivity of the models to time is given particular attention: viscous versus plastic rheology during sediment compaction; steady state versus non-steady state behaviour of fault and shear zones. The conclusions discuss recent advances for modelling pressure solution creep and the main questions that remain to be solved. -IntroductionIn order to investigate the role of pressure solution creep in the ductility of the Earth's upper crust the various mechanical behaviour patterns of the upper crust must first be discussed. Two types of approach can be considered on this topic.1 -The mechanical behaviour of the upper crust is modelled using brittle theories, that include friction laws (Byerlee, 1978;Marone, 1998). This modelling approach is supported by two kinds of observations: (i) the maximum frequency of earthquakes is located within the first 15-20 km of the upper crust (Chen and Molnar, 1983;Sibson, 1982); (ii) laboratory experiments run at relatively fast strain-rates (faster than 10 -7 s -1 ) indicate a transition from frictional to plastic deformation at pressure and temperature conditions appropriate for a depth of 10-20 km (Kohlstedt et al., 1995;Paterson, 1978;Poirier, 1985).2 -Conversely, the behaviour of the upper crust is also modelled by ductile behaviour with creep laws (Wheeler, 1992). This modelling approach is supported by two kinds of observations: (i) geological structures exhumed from depth, such as compacted basin, folds and shear zones, and regional cleavage, indicate ductile behaviour throughout the upper crust (Argand, 1924;Hauck et al., 1998;Schmidt et al., 1996) (Fig. ...

No abstract

Stylolites are natural pressure-dissolution surfaces in sedimentary rocks. We present 3D high resolution measurements at laboratory scales of their complex roughness. The topography is shown to be described by a self-affine scaling invariance. At large scales, the Hurst exponent is ζ 1 ≈ 0.5 and very different from that at small scales where ζ 2 ≈ 1.2. A cross-over length scale at around L c = 1 mm is well characterized. Measurements are consistent with a Langevin equation that describes the growth of a stylolitic interface as a competition between stabilizing long range elastic interactions at large scales or local surface tension effects at small scales and a destabilizing quenched material disorder.

[1] Several models pertaining to earthquake cycles imply intermittent fluid flow through fault. During the interseismic period, increase in fluid pressure from hydrostatic to lithostatic values is a crucial parameter in mechanisms leading to earthquakes. To achieve such pressures, geodynamic processes (gouge compaction, fluid flow) and changes in permeability are required. Previous models have postulated that changes in permeability (by self-healing) are faster than the effects of geodynamic processes. We consider the different mechanisms and rates of crack sealing near active fault on the examples of uplifted Californian faults. We find that natural crack sealing is normally not achieved by a rapid self-healing process. Pressure solution, with mass transfer from solution cleavage to cracks, appears to be a more important mechanism for crack sealing and creep during postseismic deformation. The geometry of transfer path and experimental data have been used to model crack sealing rates by pressure solution which are estimated to be rather slow, similar to the recurrence time of some earthquakes. Such slow changes in permeability may be crucial factors in controlling the increase in fluid pressure and, consequently, the mechanism of critical failure in faults. Then, numerical modeling of fluid pressure and transfer around active faults has been performed integrating a slow change in permeability by crack sealing, gouge compaction, and fluid flow from depth. This modeling shows various location and evolution of fluid overpressure during the interseismic period depending on these processes and allows one to estimate the amount of fluid transferred from depth during interseismic periods.

[1] Geological repositories subject to the injection of large amounts of anthropogenic carbon dioxide will undergo chemical and mechanical instabilities for which there are currently little experimental data. This study reports on experiments where low and high P co 2 (8 MPa) aqueous fluids were injected into natural rock samples. The experiments were performed in flow-through triaxial cells, where the vertical and confining stresses, temperature, and pressure and composition of the fluid were separately controlled and monitored. The axial vertical strains of two limestones and one sandstone were continuously measured during separate experiments for several months, with a strain rate resolution of 10 À11 s À1. Fluids exiting the triaxial cells were continuously collected and their compositions analyzed. The high P co 2 fluids induced an increase in strain rates of the limestones by up to a factor of 5, compared to the low P co 2 fluids. Injection of high P co 2 fluids into the sandstone resulted in deformation rates one order of magnitude smaller than the limestones. The creep accelerating effect of high P co 2 fluids with respect to the limestones was mainly due to the acidification of the injected fluids, resulting in a significant increase in solubility and reaction kinetics of calcite. Compared to the limestones, the much weaker response of the sandstone was due to the much lower solubility and reactivity of quartz in high P co 2 fluids. In general, all samples showed a positive correlation between fluid flow rate and strain rate. X-ray tomography results revealed significant increases in porosity at the inlet portion of each core; the porosity increases were dependent on the original lithological structure and composition. The overall deformation of the samples is interpreted in terms of simultaneous dissolution reactions in pore spaces and intergranular pressure solution creep.

[1] Indenter experiments have been performed on quartz crystals in order to establish a pressure solution creep law relevant at upper to middle crustal conditions. This deformation mechanism contributes to Earth's crust geodynamics, controlling processes as different as fault permeability, strength, and stress evolution during interseismic periods or mechanochemical differentiation during diagenesis and metamorphism. Indenter experiments have been performed at 350°C and 20-120 MPa during months with differential stress varying from 25 to 350 MPa. Several experimental parameters were varied: nature of quartz (synthetic or natural), nature of fluid, manner in which the solid/ solution/solid interface was filled, and orientation of the indented surfaces versus quartz crystallographic c axis. Significant strain rates could only be obtained when using high-solubility solutions (NaOH 1 mol L À1 ). Displacement rates of the indenter were found activated by differential stress, with exponential dependence, as theoretically predicted. The mean thickness of the trapped fluid phase below the indenter was estimated in the range 2-10 nm. Moreover, the development of this trapped fluid phase was relatively fast and allowed fluid penetration into previously dry contact regions by marginal dissolution. The indenter displacement rate was driven by differential stress, and its kinetics was controlled by diffusion along the trapped fluid and the development of a morphological roughness along the interface. Conversely, marginal strain energy driven dissolution was observed around the indenter, and its kinetics was controlled by freesurface reaction. These experimental results are applied to model the interactions between pressure solution and brittle processes in fault zones, providing characteristic time scales for postseismic transitory creep and sealing processes in quartz-rich rocks.

Previous studies show that pulverized rocks observed along large faults can be created by single high‐strain rate loadings in the laboratory, provided that the strain rate is higher than a certain pulverization threshold. Such loadings are analogous to large seismic events. In reality, pulverized rocks have been subject to numerous seismic events rather than one single event. Therefore, the effect of successive “milder” high‐strain rate loadings on the pulverization threshold is investigated by applying loading conditions below the initial pulverization threshold. Single and successive loading experiments were performed on quartz‐monzonite using a Split Hopkinson Pressure Bar apparatus. Damage‐dependent petrophysical properties and elastic moduli were monitored by applying incremental strains. Furthermore, it is shown that the pulverization threshold can be reduced by successive “milder” dynamic loadings from strain rates of ~180 s−1 to ~90 s−1. To do so, it is imperative that the rock experiences dynamic fracturing during the successive loadings prior to pulverization. Combined with loading conditions during an earthquake rupture event, the following generalized fault damage zone structure perpendicular to the fault will develop: furthest from the fault plane, there is a stationary outer boundary that bounds a zone of dynamically fractured rocks. Closer to the fault, a pulverization boundary delimits a band of pulverized rock. Consecutive seismic events will cause progressive broadening of the band of pulverized rocks, eventually creating a wider damage zone observed in mature faults.

Active faults in the upper crust can either slide steadily by aseismic creep, or abruptly causing earthquakes. Creep relaxes the stress and prevents large earthquakes from occurring. Identifying the mechanisms controlling creep, and their evolution with time and depth, represents a major challenge for predicting the behavior of active faults. Based on microstructural studies of rock samples collected from the San Andreas Fault Observatory at Depth (California), we propose that pressure solution creep, a pervasive deformation mechanism, can account for aseismic creep. Experimental data on minerals such as quartz and calcite are used to demonstrate that such creep mechanism can accommodate the documented 20 mm/yr aseismic displacement rate of the San Andreas fault creeping zone. We show how the interaction between fracturing and sealing controls the pressure solution rate, and discuss how such a stress-driven mass transfer process is localized along some segments of the fault