Available online at www.sciencedirect.com Particuology 6 (2008) 445–454 DEM simulation of particle mixing in a sheared granular flow Li-Shin Lu a , Shu-San Hsiau b,∗ a Department of Mechanical Engineering, Technology and Science Institute of Northern Taiwan, Taipei 11202, Taiwan, China b Department of Mechanical Engineering, National Central University, Chung-Li 32001, Taiwan, China Received 24 March 2008; accepted 15 July 2008 Abstract Mixing behaviors of particles are simulated in a sheared granular flow using differently colored but otherwise identical glass spheres, with five different bottom wall velocities. By DEM simulation, the solid fractions, velocities, velocity fluctuations and granular temperatures are measured. The mixing layer thicknesses are compared with the calculations from a simple diffusion equation using the data of apparent self-diffusion coefficients obtained from the current simulation measurements. The calculations and simulation results showed good agreements, demonstrating that the mixing process of granular materials occurred through the diffusion mechanism. © 2008 Chinese Society of Particuology and Institute of Process Engineering, Chinese Academy of Sciences. Published by Elsevier B.V. All rights reserved. Keywords: DEM; Particle mixing; Sheared granular flow; Diffusion 1. Introduction Particle mixing is very important to industries ranging from food products to pharmaceuticals, ceramics and plas- tics. However, granular mixing is a complex and imperfectly understood process. In recent years, more and more stud- ies investigate fundamental mechanisms of granular mixing (Katayama, Aoyama, Yamamoto, & Sakaue, 2004; McCoy & Madras, 2005; Santomaso, Olivi, & Canu, 2004). In fact, the dominant mechanism affecting the flow behavior of granular materials is the random motions of particles resulting from the interactive collisions between particles (Campbell, 1990). The random motions of particles are quantified by granular tempera- ture, which is defined as the specific fluctuation kinetic energy of particles and serves as a key property in granular material flows. The energy of granular temperature is continually dissipated through the inelastic collisions between particles or between particles and boundaries. Thus the external energy must be con- tinually input into the system, such as shear cell, vibrating bed and rotating drum, to maintain the granular temperature. In recent years, sheared granular systems have received con- siderable attention as an important example of granular flow (Lätzel, Luding, & Herrmann, 2000; Losert, Bocquet, Lubensky, ∗ Corresponding author. Tel.: +886 3 426 7341; fax: +886 3 425 4501. E-mail address: sshsiau@cc.ncu.edu.tw (S.-S. Hsiau). & Gollub, 2000; Luding, 2008; Lun & Bent, 1994; Tardos, McNamara, & Talu, 2003). Diffusion, in particular, has been studied in both rapid (Campbell, 1997; Hsiau & Shieh, 1999; Radjai & Roux, 2002) and slow (Hazzard & Mair, 2003; Utter & Behringer, 2004) flow systems. Diffusivity is also a key quantity in describing velocity fluctuations in granular materials. Utter and Behringer (2004) found that self-diffusivities are propor- tional to the local shear rate with diffusivities along the direction of the mean flow approximately twice as large as those in the perpendicular direction. Campbell (1997) has reported measure- ments of the particle diffusive motion in an unbounded dense granular shear flow using computer simulations. Unexpectedly, at the solid fraction of ν = 0.56, the particles appeared to be trapped in a microstructure and prohibited from moving rela- tive to their neighbors, which indicates a transition to a stable ordered state. Campbell (1997) conjectured that a rapidly flow- ing granular material is a diffusive system except at large solid concentrations. In fact, at high solid fraction, shearing may cause crystallization, which ultimately results in the vanishing of dif- fusive mixing (Wang, Jackson, & Sundaresan, 1996). However, there was clear experimental evidence (Hsiau & Hunt, 1993) for the movements of the particles in the directions transverse to the bulk motion at even higher solid fractions than those examined in Campbell’s simulations. Besides, using numeri- cal simulations of a system with inelastic, rough, hard spheres of solid fraction ν = 0.565, Zamankhan, Tafreshi, Polashenski, Sarkomaa, and Hyndman (1998) found that a bounded rapid 1674-2001/$ – see inside back cover © 2008 Chinese Society of Particuology and Institute of Process Engineering, Chinese Academy of Sciences. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.partic.2008.07.006