Effects of normal stress vibrations on frictional healing

Abstract

Abstract. We conducted laboratory experiments to study frictional healing and the effects of normal stress vibrations on healing. The experiments were carried out using a servo-controlled double-direct shear apparatus on 10 cm x 10 cm blocks separated by a 3 rnm-thick gouge layer of fine-grained (grain size of 75-212 pro) quartz powder. We performed slide-hold-slide tests in which sliding surfaces were driven at a constant velocity, halted for a given interval, then restarted at the prior driving velocity. Healing varied systematically with cumulative displacement, and by conducting several sets of identical slide-hold-slides we calibrated and removed these effects. Forward modeling of the healing and relaxation curves using the rateand state-dependent friction laws shows that a displacement-dependent increase in the parameter b can account for our observations. To study the effects of vibration, we varied the mean normal stress of 25 MPa during holds by double amplitudes ranging from I to 13 MPa at a frequency of I Hz. Vibrations increased rates and magnitudes of frictional relaxation and healing, most likely due to increased gouge compaction. These effects increased with increasing amplitude of vibration. We performed normal stress step tests and used •he results to model the vibrational slide-hold-slide tests. Rate-and state-dependent constitutive laws cannot adequately describe the behavior we observed experimentally because they neglect gouge compaction. Mechanisms such as normal force oscillations may explain faster fault healing rates than would be predicted by standard laboratory measurements a• constant stress.

Rate and state dependent friction and the stability of sliding between elastically deformable solids

“…However, those experiments involved V of order 1 to 10 m=s. They lead to a di erent results than do experiments (Linker and Dieterich, 1992;Richardson and Marone, 1999) at slip rates of order 10 −6 slower, on rock materials. Studies of normal stress changes at those lower rates have led to the conclusion that is of the order of f=3.…”

“…However, experiments with variable normal stress histories (Linker and Dieterich, 1992;Prakash and Clifton, 1992;Prakash, 1998;Richardson and Marone, 1999;Bureau et al, 2000) show that there is also a memory dependence of variations in , and these e ects cannot be captured by the above formulation.…”

“…In fact, those e ects describe the actually gradual transition between what is often loosely called "static" and "kinetic" friction. Here, for some of our analyses, we want to use rate and state concepts to describe memory e ects in the response not only to variations in slip rate, but also to variations in normal stress, as studied in experiments by Linker and Dieterich (1992), Prakash and Clifton (1992), Prakash (1998), Richardson and Marone (1999), and Bureau et al (2000).…”

Application of rate‐and‐state friction laws to creep compaction of unconsolidated sand under hydrostatic loading conditions

“…[8] At the most basic level, state is associated with dilatancy and compaction of a gouge layer [e.g., Mair and Marone, 1999;Richardson and Marone, 1999]. Results from many laboratory friction experiments suggest that the state variable can be related to the effective contact area of asperities in the active shear band [e.g., Dieterich and Kilgore, 1994] or to the energy available to dilate the gouge layer [Beeler and Tullis, 1997].…”

“…Both are characterized by compaction strain and strain rates that decay with time. For compaction of quartz sand in friction experiments [e.g., Richardson and Marone, 1999], strain rate decays according to a power law function of porosity during a hold. In volumetric creep tests on unconsolidated sands from the upper terminal zone of the Wilmington field (and others from the Gulf of Mexico), strain rate decays according to a power law of time, under conditions of hydrostatic stress [Ostermeier, 1995;Chang et al, 1997;Hagin and Zoback, 2004].…”