Abstract: Morphology is an important grain scale property that substantially influences the macroscopic behavior of granular materials. Particle-based methods such as discrete element method are suitable to study the morphology effect because the particle geometry can be systematically controlled. However, coupling realistic particle morphology with simulation involves significant computational costs. Therefore, considerably simplified particle morphology has been typically adopted in the particle-based simulations to keep the computational costs manageable. This paper aims to systematically address how modeling error results from the morphology simplification and investigates how the modeled fidelity of particle morphology influences the simulated granular material behavior. To this end, a spectrum of particle morphology is developed by gradually simplifying a polyhedral particle model that represents a realistic particle morphology. The modeled fidelity is then quantified in term of Shape Factor and Angularity Factor to characterize the global form and the local angularity of the particle morphology. A set of 3D printed particles is developed from the numerical particle models and used in laboratory testing for systematic parametric study. A series of triaxial compression tests is performed on the synthetic specimens with the 3D printed particles. The result indicates the relative importance of high-fidelity modeling of (1) local angularity under low-confining pressure and (2) global form under high-confining pressure. This study also demonstrates the potential of leveraging 3D printed synthetic particles to study the morphology-dependent mechanical behavior of granular materials. Graphic abstract: [Figure not available: see fulltext.].
All Science Journal Classification (ASJC) codes
- Materials Science(all)
- Mechanics of Materials
- Physics and Astronomy(all)