Three-dimensional Voronoi analysis of monodisperse ellipsoids during triaxial shear

•DEM simulations of triaxial compression tests on assemblies of ellipsoidal particles were conducted.•Set Voronoi tessellation was performed on Voronoi cell construction of ellipsoid assemblies.•Effects of particle shape and shear-induced variation on Voronoi cell properties were investigated.•Shear-induced inhomogeneity and anisotropy of granular structure were analyzed in terms of Voronoi-based quantities.

The internal structure of a sheared assembly of monodisperse ellipsoids is investigated using three-dimensional Voronoi analysis. The discrete element method is employed to simulate isotropic and triaxial compression tests of ellipsoidal particles. A recently developed Voronoi tessellation technique, i.e., Set Voronoi tessellation, is applied to constructing Voronoi cells of assemblies at a series of shearing states. Several quantities are provided to quantify the properties of Voronoi cells, including local porosity, reduced surface area, sphericity, and a modified Minkowski tensor. We show that average local porosity, average reduced surface area, and average sphericity are functions of global porosity and mean coordination number during shearing, suggesting a relationship between void and particle networks. Moreover, local porosity and reduced surface area statistically comply with a modified lognormal distribution regardless of particle shape for a similar global porosity. However, the shape of a Voronoi cell is significantly dependent on the particle it encloses. Shear-induced inhomogeneity and anisotropy measured by Voronoi-based quantities are also examined. Increasing shear-induced entropy is observed, which means a more disordered void network during shearing. Furthermore, anisotropy of Voronoi cell orientations shows a comparable trend with anisotropy of contact normals. These findings are useful for developing a better understanding of void networks and their relationship with particle networks for non-spherical assemblies.

Graphical abstractDownload high-res image (239KB)Download full-size image