- © The Mineralogical Society Of America
This chapter addresses the mechanisms by which instabilities can develop in rocks in the nominally ductile regime under conditions that the authors believe are relevant to Earth. We will not address brittle failure or frictional sliding processes in detail, but much of our discussion will be couched in terms of knowledge gained in studies of these processes. As a consequence, we will begin by discussing why brittle failure and frictional sliding cannot be the mechanism by which rocks become unstable at depth, despite the strong evidence that they are the underlying mechanism of earthquakes at shallow depths (e.g., Scholz 1990, 2002).
Brittle shear failure in the normal sense is fundamentally a tensile process (tensile microcracks must first nucleate and self-organize via interaction of the stress concentrations at their tips). As a consequence, brittle failure and frictional sliding are strongly inhibited by pressure because work must be done against the pressure to open the Mode I cracks. The inhibition is so strong that the stress required to create a fault or initiate sliding on an existing fault becomes greater than the room-temperature flow stress of many rocks at pressures equivalent to only a few tens of kilometers in Earth. Increasing temperature has little effect on brittle processes, but in contrast, the ductile flow stress of rocks falls exponentially with temperature. Because both pressure and temperature increase with depth in Earth, earthquakes by unassisted brittle fracture mechanisms or frictional sliding can only occur at depths less than ∼30–50 km. Nevertheless, earthquakes occur to depths of almost 700 km in Earth and seismological evidence demonstrates unequivocally that they occur by displacement across a surface or narrow zone (i.e., they occur by faulting). This conundrum can be explained qualitatively by certain specific types of mineral reactions occurring under certain restricted conditions. The …