Instability-induced clustering and segregation in high-shear Couette flows of model granular materials

Controlling unwanted segregation of components within a particle mixture is a longstanding goal in particle processing technologies. We investigate flow-induced clustering and consequent segregation of cohesionless particulate mixtures flowing rapidly in high-shear Couette geometries, comparing results from particle-dynamic (PD) simulations, kinetic theory and experiments. Using spheres and disks and a simplified plug instability as a surrogate for the variety of coherent flow structures, we find that density, velocity, and granular temperature gradients, and possibly, initiation of vorticity, influence the onset and nature of segregation. For equal density particles, simulations and experiments show that there exists a critical solids fraction at which the direction of segregation is reversed and streamwise diffusivity drops significantly, which appears to correspond to a point where the thermal diffusion flux no longer dominates over other terms. Results compare favorably to those from binary kinetic theories with one exception. We find that the 1D steady-state theoretical solutions do not capture the flux reversal observed in PD simulations of the equal density mixtures. Finally, we illustrate how local density variations can severely affect particle size distribution measurements.