Experimental and analytical study of the dynamic response of low‐porosity brittle rocks

Abstract

Motivated by an inadequate understanding of large differences observed in particle velocities and displacements in in situ large‐scale tests in wet and dry granite, a combined experimental and analytical study was undertaken to investigate the influence of porosity, crack population, pore water content and pressure, and confining pressure on the characteristics of high‐amplitude stress wave propagation in a low‐porosity (0.8–1.0%) brittle rock under well‐controlled laboratory conditions. A unique porosity enhancement technique, using a gas fracturing process, was developed to provide a controlled, homogeneous, isotropic increase in porosity to 2.4–2.9% (a factor of 2.5–3.5) in Sierra White granite. Combined with a technique to independently control confining and pore pressures, the experiments provided measured radial particle velocities, calculated displacements, and peak velocity and displacement attenuations. In the experiments, key parametric effects due to fracture‐induced porosity increase combined with pore water content were observed. In particular, a significant porosity (and accompanying fracture) increase combined with fully saturated, undrained condition (no effective stress) resulted in a substantial decrease in peak velocities and increased pulse durations, leading to greatly increased particle displacements. Reducing the pore pressure in enhanced‐porosity rock produces substantially smaller displacements because of narrower pulses. It was concluded that in the region where deviatoric response is important, anisotropic damage evolution through compression‐induced fracture formation rather than the resulting porosity increase itself appears to dominate material response. A phenomenological plastic damage model allowing damage accumulation under compression‐induced (splitting) cracking was used to numerically simulate the key experimental observations.