A standard triaxial apparatus was set up to perform sliding tests. Slide-hold-slide (SHS) tests, with hold times from 10 to 3000s and at invariant effective confining pressures of 1MPa, incremented to 3MPa and then 5MPa, were conducted on both dry and wet split limestone samples to study the evolution of fracture friction under holding and sliding processes. Limestone fracture shows time-dependent behaviors in SHS tests, friction drops (creep) gradually during hold periods, and recovers (healing) after re-slide. Frictional healing and creep increment are proportional with hold times, while are inversely proportional with effective confining pressures. Moreover, the wet sample displays heavier healing than the dry one, whereas creep increment of the wet sample is lighter than the dry one. We conclude that the evolution of limestone fracture friction is significantly controlled by mechanisms of fluid-rock interactions.


The evolution of the mechanical properties of fractures and faults exerts a critical influence on the behavior of fault zones and on the earthquake cycle. In addition to natural processes, this behavior has broad implications in induced seismicity and influences oil and gas production, the sequestration of carbon dioxide and the disposal of nuclear wastes.

Fault friction is typically measured by slide-hold-slide experiments, where fault gouge is deformed at low velocities (<0.1 mms−1) and over small shear offsets (<30 mm) (Dieterich, 1979, 1981; Mair et al., 1999; Marone, 1998, 2004) or alternatively with rotary-shear devices, that enable higher slip velocities over very large shear offsets (Hirose & Shimamoto, 2005; Mizoguchi et al., 2006). Typical experiments (Dieterich, 1972; Marone, 1998; Olsen et al., 1998; Bos & Spiers, 2002; Tenthorey et al., 2003) show significant creep during the holding period and frictional healing (strength gain) upon reloading. Fault creep and healing are often time-dependent with longer hold times typically resulting in greater magnitudes of creep and in strength gains (Tenthorey et al., 2003; Yasuhara et al., 2005; Carpenter et al., 2012). Frictional healing may arise from growth of grain contacts (potentially induced by pressure solution) (Rabinowicz, 1951; Dieterich & Kilgore, 1994; Yasuhara et al., 2005), strengthening of the contacts (correlates with cementation) (Hickman & Evans, 1992; Beeler et al., 1994; Karner et al., 1997; Olsen et al., 1998; Muhuri et al., 2003; Tenthorey et al., 2003; Li et al., 2011) or via other changes in state including changes in porosity and rearrangement of force chains. Experimental evidence (Dieterich, 1979, 1981; Johnson, 1981; Marone & Scholz, 1988; Marone, 1998; Saffer & Marone, 2003; Moore & Lockner, 2007; Ikari et al., 2007; Ikari et al., 2011; Carpenter et al., 2012) quantifies both the mode and magnitude of second-order changes in the velocity-dependent (including velocity- strengthening and weakening) evolution of frictional strength.

This content is only available via PDF.
You can access this article if you purchase or spend a download.