Stress partition and micro-structure in size-segregating granular flows

Lydie H Staron, Jeremy C Phillips

Research output: Contribution to journalArticle (Academic Journal)peer-review

31 Citations (Scopus)
367 Downloads (Pure)


When a granular mixture involving grains of different sizes is shaken, sheared, mixed, or left to flow, grains tend to separate by sizes in a process known as size segregation. In this study, we explore the size segregation mechanism in granular chute flows in terms of pressure distribution and granular micro-structure. Therefore, 2D discrete numerical simulations of bi-disperse granular chute flows are systematically analysed. Based on the theoretical models by Gray and Thornton 2005 and Hill and Tan 2014, we explore the stress partition in the phases of small and large grains, discriminating between contact stresses and kinetic stresses. Our results support both gravity-induced and shear-gradient-induced segregation mechanisms. However, we show that the contact stress partition is extremely sensitive to the definition of the partial stress tensors, and more specifically, to the way mixed contacts (i.e. involving a small grain and a large grain) are handled, making conclusions on gravity-induced segregation uncertain. By contrast, the computation of the partial kinetic stress tensors is robust. Kinetic pressure partition exhibits a deviation from continuum mixture theory of a significantly larger amplitude than contact pressure, and display a clear dependence on the flow dynamics. Finally, using a simple approximation for the contact partial stress tensors, we investigate how contact stress partition relates with the flow micro-structure, and suggest that the latter may provide an interesting proxy for studying gravity-induced segregation.
Original languageEnglish
Article number022210
Number of pages8
JournalPhysical Review E: Statistical, Nonlinear, and Soft Matter Physics
Publication statusPublished - 24 Aug 2015


Dive into the research topics of 'Stress partition and micro-structure in size-segregating granular flows'. Together they form a unique fingerprint.

Cite this