Impossible experiments: Using the discrete element method to explore drag forces in granular segregation and frictional hysteresis

Abstract

Granular materials are of vital importance in both geophysical and industrial settings. As well as obeying a non-Newtonian rheology, granular materials exhibit a wide range of interesting phenomena, including non-locality, hysteresis, and size segregation. The discrete element method, or DEM, offers a way to study granular flows by modelling grains as elements of a specified shape which interact through contact forces. Using DEM it is possible to construct simulations which would be impossible, or at least highly impractical, to recreate physically, but nonetheless can be used to extract useful information about how granular materials behave. In this thesis, we use DEM simulations to investigate two processes: granular size segregation, which is a process by which mixtures of small and large grains can spontaneously separate, and frictional hysteresis, which is a process which allows certain inclinations to support both static and flowing layers of granular material. In granular segregation, there is a drag force which resists the forces driving differing grains apart. We make use of DEM simulations to measure this drag force directly. Our findings indicate that the drag force is primarily responsible for determining the rate at which segregation takes place, and how bulk flow properties influence the segregation rate. We also examine these findings in the context of a free surface avalanche. When examining hysteresis we consider depth-averaged models for granular flows on inclined planes. Depth-averaged models can include hysteretic effects through the inclusion of an empirical basal- friction function. For static and steady flowing material, it is possible to infer the functional form of the basal friction from experimental data. However, there is also an intermediate regime where steady flows cannot be observed. We use DEM simulations of inclined-plane flows in periodic domains to infer the basal friction in this intermediate regime by investigating whether specific initial conditions come to rest or accelerate to a steady flow. This is complicated by the fact that the rate at which flows come to rest is stochastic, and is dependent on the size of the periodic domain used.

Date of Award	31 Dec 2022
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Nico Gray (Supervisor) & Christopher Johnson (Supervisor)