Large-scale granular packings of power-law-distributed polydisperse spheres*

Discrete element simulations of granular packings of size-dispersed spheres with largest-to-smallest diameter ratios >10 have remained elusive. Previous numerical studies of power-law-distributed particles have typically been constrained to work with relatively narrow spans of particle sizes, limiting the correspondence between simulation results and real-world granular materials of geophysical and industrial relevance. We employ a newly developed neighbor binning algorithm, implemented in LAMMPS, to efficiently generate multi-million particle granular packings of power-law-distributed spheres with largest-to-smallest diameter ratios of 50 and larger. By varying the exponent of the underlying power-law size distribution, along with interparticle coefficients of friction, our simulations reveal striking insights into the structure and properties of packings of both frictionless and frictional particles. Our work underscores the importance of balancing the relative abundance of large-large and small-small particle contacts to optimize packing properties. We contextualize our results with simulations of monodispersed packings and bidispersed packings with comparable size disparities.