Image-informed fracture model for complex-shaped particles in discrete element simulations

Particle breakage is a complex physical process that immensely affects the behavior of granular systems. Achieving accurate simulations of real crushable materials is challenging since modeling irregular particles is computationally intensive and hinders scaling up to large granular assemblies while predicting the evolution of particle morphology after each breakage event. The factors affecting it still need to be explored. We propose a novel particle breakage algorithm for irregular particles in the Discrete Element Method (DEM). Unlike standard practices in the literature, the proposed algorithm intrinsically preserves material bulk properties during the fragmentation of complex-shaped particles. This is achieved via the concurrent consideration of imaging data in the form of surface meshes and multi-sphere particle representations. The particle meshes are used to provide morphological information and to generate multi-spheres for each new generation of fragments, while the multi-spheres are used to determine the contact forces acting on the particles in the DEM simulations. We employ a combined Mohr-Coulomb-Weibull failure criterion to determine when a fracture occurs. The validity of the proposed algorithm is demonstrated via comparison with available experimental tests, where the model is found to result in statistically reliable results.