A model for flash weakening by asperity melting during high‐speed earthquake slip

Abstract

[1] Recent results from laboratory experiments on a broad range of mineral systems exhibit dramatic drops in the effective friction coefficient m once the slip rate exceeds a critical level V w , which is typically O(0.1) m/s. This ''flash weakening'' has been attributed to the effects of localized heating at highly stressed microscopic asperities. We extend previous phenomenological treatments to assess whether melting at asperity contacts can explain the observed changes in strength. Using physical parameters obtained from the literature on the phase behavior and mechanical properties of quartz, albite, dolomite, gabbro, Westerly granite, and serpentinite, the predictions of our simplified model are in reasonable agreement with available experimental data. We derive approximate analytical expressions that suggest that strength changes are insensitive to the melt viscosity under conditions that likely include those during earthquake slip along major fault systems. Instead, the primary controls on m are the ratio of slip rate V to V w and the Stefan number S, which is defined as the ratio of the latent heat of fusion to the sensible heat required to raise the temperature from ambient levels. The phase behavior during the short lifetimes and at the high confining pressures of asperity contacts is a significant source of uncertainty in the parameter choices, as are the presence and availability of water. Nevertheless, our results are encouraging for further efforts to incorporate the microphysics of fault zone processes into earthquake simulations.Citation: Rempel, A. W., and S. L. Weaver (2008), A model for flash weakening by asperity melting during high-speed earthquake slip,

Flash weakening of serpentinite at near-seismic slip rates

“…A lower bound of ∼500°C is provided by experimental studies on the conditions of antigorite dehydration at pressures between 2 and 4 GPa [ Perrillat et al , 2005]. We choose this pressure range to model the conditions at the asperity contact [e.g., Rempel and Weaver , 2008], where we estimate τ c = 2 to 4 GPa. The overlap in the estimated values of τ c and T w shown in Figure 10 suggests an asperity size D ≈ 3–10 μ m, which is qualitatively in agreement with the observed contact size in SEM micrographs (Figure 8d).…”

On the transient behavior of frictional melt during seismic slip

“…It is observed in high‐velocity experiments that before bulk melting occurs, shear stress may oscillate but remains on average at a relatively high level (peak stress or τ p ) in the same range as the static Coulomb friction for rocks (0.6 σ n < τ p < 0.7 σ n ). Our analysis of frictional melt applies to fault evolution after bulk melting has occurred, while premelt friction may be controlled by other processes such as flash heating at the asperity contacts [ Rice , 2006; Rempel and Weaver , 2008; Beeler et al , 2008], selective melting of minerals [ Shand , 1916; Spray , 1992], or occurrence of melt at localized patches only [ Hirose and Shimamoto , 2005; Del Gaudio et al , 2009], which are well beyond the scope of this study. In addition, under normal stresses greater than a few tens of megapascals, experimental results indicate that the premelt oscillations tend to disappear and bulk melting occurs after an extremely short time lapse [ Nielsen et al , 2008; Niemeijer et al , 2009].…”

“…In several such mechanisms, the weakening is related to some form of thermally activated process and is triggered by frictional heating. Some well‐known examples are fluid pressurization [ Bizzarri and Cocco , 2006], decarbonation [ Han et al , 2007], silica gel formation [ Goldsby and Tullis , 2002; Di Toro et al , 2004], hydrodynamic lubrication [ Brodsky and Kanamori , 2001], acoustic fluidization [ Melosh , 1996], and flash heating [ Rice , 2006; Beeler et al , 2008; Rempel and Weaver , 2008]. The present study is dedicated essentially to the case of melting, observed both in laboratory experiments [ Spray , 1987; Tsutsumi and Shimamoto , 1997; Hirose and Shimamoto , 2005; Spray , 2005; Di Toro et al , 2006b; Del Gaudio et al , 2009] and on natural faults (e.g., ancient faults now found at the surface but originally active at depths of several km in the seimogenic Earth crust or upper mantle; see Sibson [1975], Swanson [1992], Di Toro and Pennacchioni [2004], Spray [2005], Di Toro et al [2006a], Ueda et al [2008] and, for a review, Snoke et al [1998]).…”

Dislocation Motion and the Microphysics of Flash Heating and Weakening of Faults during Earthquakes

“…Theoretical and experimental investigations [5,[7][8][9][10][11][12][13][14][15] on silicate-bearing rocks (gabbro, granite, peridotite, etc.) suggested that such frictional heat initiates at highly stressed microscopic asperities, a process termed flash heating.…”