Experimentally validated DEM for large deformation powder compaction: mechanically-derived contact model and screening of non-physical contacts

Abstract

Despite widespread industrial reliance on powder compaction in manufacturing, a complete understanding of the underlying physical mechanisms that lead to pore structure, mechanical strength, and defects remains elusive, challenging ongoing efforts to optimize the process and improve product quality. The discrete element method (DEM) is a promising tool for studying powder compaction due to its algorithmic simplicity and particle-level insights, but its application is limited by the lack of accessible, physically justified contact models for large deformations. In this work, we help address this problem by extending the recently proposed mechanically-derived adhesive elastic-plastic contact model (Zunker and Kamrin, 2024a, 2024b) suitable for large deformation to the case of many-interacting particles. A topological penalty algorithm for the screening of non-physical contacts occurring through obstructing particles, a phenomenon unique to large deformation DEM, is also proposed. The extended version of the contact model and topological penalty algorithm are implemented into the open-source DEM software LAMMPS https://github.com/lammps/lammps and validated against the multi-particle finite element method (MPFEM). The contact model's unique ability to reconstruct deformed particle shapes is highlighted by comparison to FEM predictions. The industrially relevant problem of pharmaceutical tableting is simulated and comparisons to experimental data for the compaction of Avicel PH102 (microcrystalline cellulose) are made. Good agreement is observed between the experiment and numerical simulation for the axial and radial stress measurements as a function of relative density. Notably, the simulation is able to predict a similar residual radial stress after release of the axially confining pressure to that of the experiment.