HR: 11:50h
AN: S32B-07    [Abstracts]
TI: Testing Friction Laws by Comparing Simulation Results With Experiments of Spontaneous Dynamic Rupture
AU: * Lu, X
EM: xiaol@caltech.edu
AF: Division of Engineering and Applied Science, California Institute of Technology, 1200 E. California Boulevard, Pasadena, CA 91030 United States
AU: Lapusta, N
EM: lapusta@caltech.edu
AF: Division of Geological and Planetary Sciences and Division of Engineering and Applied Science, California Institute of Technology, 1200 E. California Boulevard, Pasadena, CA 91030 United States
AU: Rosakis, A J
EM: rosakis@aero.caltech.edu
AF: Division of Engineering and Applied Science, California Institute of Technology, 1200 E. California Boulevard, Pasadena, CA 91030 United States
AB: Friction laws are typically introduced either based on theoretic ideas or by fitting laboratory experiments that reproduce only a small subset of possible behaviors. Hence it is important to validate the resulting laws by modeling experiments that produce spontaneous frictional behavior. Here we simulate experiments of spontaneous rupture transition from sub-Rayleigh to supershear done by Xia et al. (Science, 2004). In the experiments, two thin Homalite plates are pressed together along an inclined interface. Compressive load P is applied to the edges of the plates and the rupture is triggered by an explosion of a small wire. Xia et al. (2004) link the transition in their experiments to the Burridge-Andrews mechanism (Andrews, JGR, 1976) which involves initiation of a daughter crack in front of the main rupture. Xia et al. have measured transition lengths for different values of the load P and compared their results with numerical simulations of Andrews who used linear slip-weakening friction. They conclude that to obtain a good fit they need to assume that the critical slip of the slip-weakening law scales as P^-1/2, as proposed by Ohnaka (JGR, 2003). Hence our first goal is to verify whether the dependence of the critical slip on the compressive load P is indeed necessary for a good fit to experimental measurements. To test that, we conducted simulations of the experiments by using boundary integral methodology in its spectral formulation (Perrin et al., 1995; Geubelle and Rice, 1995). We approximately model the wire explosion by temporary normal stress decrease in the region of the interface comparable to the size of the exploding wire. The simulations show good agreement of the transition length with the experimental results for different values of the load P, even though we keep the critical slip constant. Hence the dependence of the critical slip on P is not necessary to fit the experimental measurements. The inconsistency between Andrews' numerical results and experimental measurements comes from different rupture initiation mechanisms. Andrews introduces an initial crack into his model, with the slip and stress distribution appropriate for the critical static crack, and then he increases the shear stress slightly to start the dynamic rupture. Since the critical crack size depends on the loading, Andrews effectively uses different initial crack lengths, in dimensional terms, for different loading. However, in the experiments, the size of the exploding wire and the explosion strength are the same for different values of compressive load P, and hence, for decreasing P, the size of the explosion becomes a progressively smaller fraction of the critical crack size. This feature is accounted for in our simulations. We plan to do more modeling and experiments to quantify the actual triggering mechanism. We will report on our current attempts to model the transition using experimentally and theoretically based Dieterich-Ruina rate and state friction law with Prakash-Clifton modification for variable normal stress and additional dynamic weakening due to flash heating. In the future, we plan to use these simulations to find experimentally realizable setups that, in simulations, produce different results for the enhanced rate and state friction law and the linear slip-weakening law.
DE: 7209 Earthquake dynamics (1242)
DE: 8020 Mechanics, theory, and modeling
DE: 8118 Dynamics and mechanics of faulting (8004)
DE: 8163 Rheology and friction of fault zones (8034)
SC: Seismology [S]
MN: Fall Meeting 2005