Particle Trajectories and Segregation in a Two-Dimensional Rotating Drum

Abstract

The size segregation phenomenon of granular materials in a two-dimensional rotating drum is simulated very efficiently using a simplified deterministic model. We demonstrate ergodicity in the sense that the statistical properties of a particle trajectory reflect the behaviour of the whole assembly. Segregation is shown to occur in a bidisperse system even if the radii of the particles are very similar (ratio as big as 0.9). Since Oyama's pioneering experiments in 1939 [l] it has been well known that particles of different sizes tend to accumulate in different regions in a partially fdled cylinder rotating around its horizontal symmetry axis, Recent experiments [2-41 show that radial segregation occurs also in two dimensions. That is, the smaller particles assemble into an inner core of the filling. Clbment, Duran and Rajchenbach[3] recorded the trajectory of a single particle of different size in an otherwise monodisperse granular medium. The experimental set-up for tracing the position of a coloured or illuminating particle by a videocamera, and evaluating the record by digital image processing is easily accessible, and highly automationable. They showed that a single small particle will more likely be found in the middle of the packing than at the container wall, demonstrating that segregation is not a collective effect requiring a certain density of small particles.Many parameters are involved in this process, such as size, shape, mass, friction forces, rotating velocity, filling of the drum, etc. Some of them should be more important than others, thus the question naturally arises: what is the minimal model, which captures the essential physics? Moreover, one would like to know to what extent the trajectory of a tagged particle reflects the behaviour of the whole assembly quantitatively.

Discrete particle simulations predicting mixing behavior of solid substrate particles in a rotating drum fermenter

“…Various discrete particle simulation methods have been shown to provide a convenient way to study granular¯ow. Among these are Monte Carlo simulations (Rosato, 1986), cellular automata models (Ktitarev and Wolf, 1998), and bottom-to-top restructuring (Baumann et al, 1994). However, the most widely used simulation method is based on molecular dynamics simulations in which interaction forces are modeled on individual grain particles.…”

Simulation of rotating drum experiments using non-circular particles

“…Nevertheless under certain conditions this model is able to reproduce the experimental behavior correctly. With a model which is based on the same method Baumann et al found an interesting phenomenon which is also measured experimentally [10][11][12]: Two sorts of circular particles, originally perfectly mixed, will decompose after only one or two rotations of the drum. The smaller disks gather themselves in the center of the drum.…”

Characterising porosity of multi-component mixtures in rotary mills

“…Particularly useful to this current study work is the proof by Baumann et al, (1994 Baumann et al, we can be confident that the assumption of ergodicity can be applied to the study of rotary mills using the PEPT technique.…”

“…A significant amount of work has also been done on binary systems (Arntz et al, 2008) for the purpose of studying phenomena such as size segregation and mixing (Baumann et al, 1994;Dury and Ristow, 1999). Some work has also been done on packing density.…”