Compaction of mixtures of rigid and highly deformable particles: a micro-mechanical model

TL;DR

This paper develops a micro-mechanical model for the isotropic compaction of mixtures of rigid and deformable particles, predicting packing fraction, elastic modulus, and connectivity evolution under stress.

Contribution

It introduces a novel model linking micro-mechanical properties to macroscopic compaction behavior in mixed particle systems.

Findings

Packing fraction increases with stress and approaches a maximum value.

Bulk modulus diverges as packing fraction nears maximum.

Model accurately predicts maximum packing and elastic properties across densities.

Abstract

We analyze the isotropic compaction of mixtures composed of rigid and deformable incompressible particles by the non-smooth contact dynamics approach (NSCD). The deformable bodies are simulated using a hyper-elastic neo-Hookean constitutive law by means of classical finite elements. For mixtures that varied from totally rigid to totally deformable particles, we characterize the evolution of the packing fraction, the elastic modulus, and the connectivity as a function of the applied stresses when varying inter-particle coefficient of friction. We show first that the packing fraction increases and tends asymptotically to a maximum value $\phi_{max}$, which depends on both the mixture ratio and the inter-particle friction. The bulk modulus is also shown to increase with the packing fraction and to diverges as it approaches $\phi_{max}$. From the micro-mechanical expression of the granular stress tensor, we develop a model to describe the compaction behavior as a function of the applied pressure, the Young modulus of the deformable particles, and the mixture ratio. A bulk equation is also derived from the compaction equation. This model lays on the characterization of a single deformable particle under compression together with a power-law relation between connectivity and packing fraction. This compaction model, set by well-defined physical quantities, results in outstanding predictions from the jamming point up to very high densities and allows us to give a direct prediction of $\phi_{max}$ as a function of both the mixture ratio and the friction coefficient.