Investigating the influence of microstructural heterogeneity on fracturing behavior of the crystalline rock: an insight from the discrete fracture network simulation

Abstract

A grain-based model (GBM) for modeling crystalline rocks has been developed based on the discrete element method (DEM), in which the rock structure is tessellated using the Voronoi algorithm. In this study, an extended grain-based model (EGBM) is proposed by introducing discrete fracture network (DFN) into GBM for explicitly incorporating microstructural characteristics of crystalline rocks that cannot be captured by the conventional Voronoi tessellation. The grain boundary fabric is first extracted from the real rock image and then imported into the base model as a discrete fracture network (DFN). To avoid the influence of micromechanical heterogeneity rendered by variation in mineral compositions, the monomineralic Carrara marble was simulated. Three EGBMs with different rock fabrics (regular and irregular Voronoi structures, and a real rock structure) as well as a BPM were examined under uniaxial compression tests to investigate the influence of microstructural heterogeneity on fracturing behavior during brittle failure. The simulated results suggest the insufficiency of VGBM to account for the microstructural complexity of crystalline rocks that results in complex microcracking behavior and macroscopic fracturing mode.