Granular Flow - Application
Size segregation by a vibrating screen
Separation of granular materials according to their physical
properties is a very important process. Vibration induced segregation can
be predicted by the DEM model. The figures below shows the initial state
and state after 78 s of vibration of a box containing a binary mixture of
8 cm particles (initially at the bottom) and 2 cm particles (initially on
top). The initial hexagonal microstructure of the large particles is very
stable making this a good test.
 |
 |
initial state |
after shaking for 78 s |
As the shaking occurs, short lived gaps are created in the
microstructure, allowing small particles to fall to lower levels. The
intrusion of smaller particles prevents the larger particles from
returning to their previous packed state. This allows further small
particles to fall into the now long-lived gaps that have been opened
between the large ones. Eventually the small particles reach the bottom.
In this example, the small particles behave as invaders, slowly eating
away at the microstructure of large particles. As each larger particle is
separated it rises quickly to the surface. By t = 78 s the large particles
clearly occupy the upper part, demonstrating that size segregation can be
predicted by DEM modelling. Further segregation is prevented by the
large-scale convective motions generated by such vibration [1].
The figures below show the initial and final state of a binary mixture
of 3 cm particles. The grey particles have a density of 1600 kg/m3,
while the pink particles are heavier with a density of 3600 kg/m3.
After shaking for 55 s, there is a clear concentration of heavy particles
towards the bottom and lighter particles towards the top demonstrating
that density segregation can be predicted by the DEM model. The mechanism
responsible for density segregation is presently not well understood.
 |
 |
initial state |
after shaking for 55 s |
Rates of segregation can be predicted using the methodology described
in [2] involving the calculation of the coefficient of variation of the
distribution of local average diameter or density. This allows optimal
vibrational modes to be identified that give the best separation rates for
given types of materials.
| Download animation: |
AVI (320x240 pixels;19.8
MB) |
|
QuickTime (320x240 pixels;
34.5MB) |
References
[1] K. Liffman, G. Metcalfe, P.W. Cleary, Particle segregation and
convective transport due to horizontal shaking, Physical Review
Letters, 79, 4574-4576 (1997).
[2] P.W. Cleary, Discrete element modelling of industrial granular
flow applications, TASK Quarterly Journal, 2,
385-415 (1998).
|
To
top