Poro-acoustoelastic constants based on Padé approximation

Abstract

Stress-induced velocity variations for porous rocks are generally characterized by a strong nonlinear dependence on stress associated with complex deformations under loading. The classical theory of poro-acoustoelasticity with high-order elastic constants is based on the Taylor expansion of the strain energy function, encountering problems of divergence and limitless elastic wave velocities in describing stress-associated velocity variations, especially for high effective stresses. The extension of the theory beyond the high-order elastic constants based on the Padé approximation to the strain energy function is addressed in this article. The resultant acoustoelastic constants are characteristics of a reasonable theoretical limit in elastic wave velocities with increasing effective stresses, avoiding some of the problems associated with high-order elastic constants such as decreasing moduli with increasing effective pressure at high effective pressure, possibly implying the microstructural dependence of elastic constants. That is, the loading stress increases strain energy and wave velocity, but also induces frame-related attenuation, which in turn reduces stiffness and elastic constants. The Padé nonlinear constants can be reduced for low effective stresses to the conventional acoustoelastic constants based on the Taylor expansion. Theoretical results are compared with ultrasonic measurements for a perfectly elastic crystal, topaz (Al₂SiO₄F₂), and a porous rock, demonstrating that the Padé-approximation-based acoustoelasticity gives a more accurate description of stress-associated velocity variations, especially for higher effective stresses.