Detection and Characterization of Precursors to Shear Failure

Purdue University, West Lafayette IN

Investigators

Abstract

Rocks are difficult materials to characterize and monitor because of their highly heterogeneous nature that leads to complex time- and scale-dependent behavior. Some of the scale-dependent behavior arises from "defects" or "mechanical discontinuities" that exist in the rock. These discontinuities range from the atomic scale, where atom lattices may have irregularities; to the grain scale, where cracks are pervasive at the grain contacts, and up to the regional or continental scale where discontinuities such joints occur from the millimeter scale to faults that range over hundreds of kilometers, such as the North Anatolian fault or the San Andreas fault. One of the most important limitations in research and engineering practice on rocks is our inability to detect accurately the presence of defects and discontinuities inside the rock and to determine the evolution of discontinuities with stress from external agents. Current knowledge of rock damage strongly relies on experiments in the laboratory where direct observations of the surfaces of the rock specimens are made to observe new crack formation. The objective of the research is to establish the relationship between mechanical and geophysical responses of rock under loading to provide a basis for seismic monitoring techniques of rock systems, and to enable the development of analysis tools to evaluate the condition of a rock mass during and after construction. The promise of using elastic wave propagation through fractured rock as an effective monitoring technique is rooted in previous findings by the two Co-PIs who discovered the presence of precursors prior to shear failure of frictional discontinuities, in the form of either distinct maxima or minima in transmitted and reflected waves, respectively. Advances in our ability to detect impending new damage in the form of new cracks or slip along pre-existing discontinuities will prevent failures during construction, ensure long term containment for waste storage, or avoid earthquakes induced by anthropogenic activities such as mining, underground fluid storage or deep wastewater injection. A series of laboratory experiments will be performed to test the hypotheses that these precursors (1) result from a reduction in local fracture specific stiffness in the region monitored by the seismic transducers, and (2) are caused by slip and damage of the asperities, and the hypothesis that failure along the entire discontinuity is the result of progressive slip along the discontinuity. The experiments will consist of subjecting rock joints to biaxial compression loading, and more specifically to direct shear. Digital Image Correlation and geophysical methods will be used to monitor slip and slip failure of the joints, which include wave propagation measurements (both transmitted and reflected waves, as well as other converted modes). The tests will be conducted on granite, limestone or sandstone and shale, each rock having different grain size and anisotropy, and on joints that may be tight or filled with a soil deposit. Both dry and saturated rock specimens will be investigated to determine the geophysical signatures associated with changes in rock saturation, and to identify and quantify the modes that are best-suited for monitoring the potential shear failure of a discontinuity.