HR: 0830h
AN: S41C-0084    [PDF]
TI: Effects of Variable Normal Stress on Shear Strength and Friction
AU: * Hong, T
EM: thong@geosc.psu.edu
AF: Dept. of Geosciences, Penn. State University, University Park, PA 16802 United States
AU: Marone, C
EM: cjm@geosc.psu.edu
AF: Dept. of Geosciences, Penn. State University, University Park, PA 16802 United States
AB: Previous work shows that dynamic and quasi-static variations in normal stress may play a key role in earthquake rupture dynamics and frictional stability. In order to improve our understanding of how normal stress changes influence frictional strength and stability we performed laboratory experiments using a servo-controlled, double-direct-shear testing apparatus. This assembly includes three rigid forcing blocks with two gouge layers sandwiched between grooved or smooth steel bounding surfaces. We sheared two 3-mm thick gouge layers of fine-grained quartz powder or smectite clay-quartz mixtures at constant load point velocity. Normal stress pulses and step experiments were carried out to investigate the shear strength response to an abrupt normal stress change. Normal stresses ranged from 10-45 MPa and we studied load point velocities in the range 1-1000 $\mu$m/s. Careful calibration of Poisson coupling within the testing apparatus and sample assembly was performed and these effects were removed from data prior to processing. Normal stress step tests involved a series of normal stress increases and decreases during steady sliding. For each step increase in normal stress, the parameter $\alpha$= ($\Delta$$\tau$/$\sigma$)/ln($\sigma$/$\sigma$$_{0}$) was calculated, where $\Delta$$\tau$ is the amplitude of the change in shear strength due to a normal stress step, and $\sigma$ and $\sigma$$_{0}$ are initial and final normal stress respectively. We find that $\alpha$ ranges from 0.28-0.35 for our range of experimental conditions, and that $\alpha$ does not change systematically with the variation of driving velocity. After exclusion of Poisson effects, there was generally no instantaneous change in shear strength in response to a step increase in normal stress. Our data show that step decreases in normal stress have the potential to destabilize steady sliding. For normal stresses of a few 10's of MPa, a normal stress reduction of $>= $1 MPa is sufficient to induce dynamic instability in our experiments. We compare our results to dynamic impact experiments on metals and other materials for which sliding velocity is of order 1-10 m/s. Those experiments consider only normal stress reductions and find that shear strength does not exhibit a sudden change in response to a step change in normal stress.
DE: 1744 Tectonophysics
SC: Seismology [S]
MN: 2003 Fall Meeting