Stochastic model for the micromechanics of jammed granular materials: Experimental studies and numerical simulations

To describe statistical properties of complicated restructuring of the force network under isotropic compression, we measure the conditional probability distributions (CPDs) of changes of overlaps and gaps between neighboring particles by experiments and numerical simulations. We find that the CPDs obtained from experiments on a 2D granular material exhibit non-Gaussian behavior, which indicates strong correlations between the configurations of overlaps and gaps before and after deformation. We also observe the non-Gaussian CPDs by molecular dynamics simulations of frictional disks in two dimensions. In addition, the numerically calculated CPDs are well described by q-Gaussian distribution functions, where the q-indices agree well with those from experiments.