Stress-Dependent Flow, Mechanical, and Acoustic Properties for an Unconventional Oil Reservoir Rock

Abstract

Summary URTeC 1562443 Accurate estimation of rock properties has significant impact on reservoir simulation, geophysical interpretation, hydraulic fracture design, and geomechanical analysis. Because reservoirs are buried at depth, measurement of these properties should consider a s tress environment that is as close to in-situ as possible. Stress-dependence of permeability, pore volume compressibility, elastic moduli, and acoustic anisotropy has been widely studied in conventional reservoirs. H owever, due to challenges in sample preparation and laboratory setup, most measurements of unconventional rock are still limited to hydrostatic stress conditions. This paper investigates a nano-Darcy unconventional oil reservoir rock under triaxial stress conditions. Rock permeability and pore volume compressibility, strength and Young’s modulus, acoustic velocities, six stiffness coefficients, and Thomsen anisotropy parameters are measured under different levels of loading and confining stresses. The relationship of these properties to the differential stress between loading and confining stresses is identified and compared with conventional hydrostatic tests. The differential stress level reaches as high as 20,000 psi for rock mechanics tests and 10,000 psi for flow tests. Steady pressure and pulse decay methods have been applied to measure permeability with gas or oil fluid flows. Despite high rock strength (greater than 10,000 psi), the rock permeability and acoustic anisotropy strongly depend on the differential stress level. Following a typical slow permeability reduction with increasing stress, a distinguished stage of accelerated permeability reduction is observed when the differential stress exceeds 3,160 psi. Interestingly the initial pore volume compressibility of the strong rock is measured as high as 50×1/Mpsi. After slight decreases until the differential stress reaches a certain level, the compressibility reverses the trend and increases significantly. Both rock strength and Young’s modulus vary greatly with the angles to bedding planes. The stress-strain curves under different confining stresses challenge the traditional modulus-based brittleness. Thomsen anisotropy parameters reduce as much as 80% with the differential stress after observing an initial increase of anisotropy when the stress is low. Among the six stiffness coefficients, the two shear components along vertical and horizontal directions (C12 and C13) vary much more than the others. T his highlights the necessity of applying differential stress, rather than hydrostatic stress, in the laboratory. It also indicates that the studied nano-Darcy rock may be much more compressible and stress-dependent than its strength suggests.