CAREER: A continuum model for simultaneous prediction of granular flow and size-segregation in general geometries

Brown University, Providence RI

Investigators

Abstract

CBET - 1552556 PI: Henann, David L. The goal of this CAREER project is to develop a theoretical model that describes the behavior of flowing granular materials. Granular materials are ubiquitous - they include naturally-occurring materials such as sand, soil, and snow, as well as industrial and agricultural materials such as pharmaceutical pills and capsules, food grains, and catalyst particles. Predicting the behavior of these materials during flow poses special challenges, especially when the granular material consists of particles of different sizes. In this case, the flow causes the particles to segregate according to size, which, in turn, affects the flow of the material as a whole. The theory developed in this project will use nonlocal modeling to describe a representative granular material consisting of particles of two different sizes - called bidisperse. The resulting model will predict the flow behavior of this bidisperse system in several prototype flow geometries, and the results will be compared with corresponding experiments conducted as part of the project. A variety of educational activities are also planned, including the development of new modules illustrating computational methods in solid mechanics. The project will enlist undergraduates to participate in the research, and workshops will be conducted to disseminate results and engage broad technical communities. A predictive continuum theory, based on nonlocal modeling, will be developed that simultaneously predicts size segregation and flow in bidisperse dense granular systems composed of spherical particles. Preliminary tests have demonstrated that this approach produces a model that can predict coupled flow fields and segregation patterns. A systematic set of discrete-element calculations will be performed to isolate shear-strain-rate-driven and gravity-driven mechanisms of segregation, which will clarify the two-way coupling between flow and segregation. The model will then be generalized to three-dimensional flows using a continuum thermomechanics approach, and a finite-element implementation of the non-local, coupled theory will be developed. Predictions of the model will be validated by comparing with experiments of bidisperse granular systems in representative flow configurations.