THEORETICAL AND SINGLE-WIRE STUDIES OF VORTEX MAGNETIC SEPARATION

This paper investigates the mechanisms of a new technique, Vortex Magnetic Separation (VMS) which can not only greatly reduce the serious mechanical entrainment problem in HGMS, as normally applied at low Reynolds Number Re(=2-rho-V(o)a/eta where rho is the density of the fluid, V(o) is the velocity of the fluid, a is the radius of the wire and eta is the viscosity of the fluid), but also provide much higher output because the high slurry velocity V(o) is used. This method will have wide application to the beneficiation of complex ores and to other industrial processes. In this process the particles are captured by entraining them in the standing vortices which appear in the flow behind the matrix wire at sufficiently high Reynolds Number. Based on the observations and experiments carried out on VMS, using a cryogenic single-wire cell, a theoretical model of VMS has been constructed. The fit to the data indicates that in order for a particle to be captured in the vortex process, the particle must remain substantially within a distance of 0.303-delta from the wire surface where delta is the thickness of the boundary layer around the wire. It assumed that there are fluctuations in the flow within the boundary layer which drive the particles into the vortex region. The theoretical model of the VMS process predicts great improvements in the selectivity, especially in cases where mechanical entrainment is a problem. Further, the particle size compared to the boundary layer thickness is an important parameter, it allows the separator to distinguish between small strongly magnetic particles and larger, but more weakly magnetic materials which is a difficult problem for conventional HGMS. This model also predicts that the total capture cross-section is almost directly proportional to the applied magnetic field, so superconducting magnets may be advantageous for large scale applications of this process.