Abstract
Most theories of magnetic separation have been developed for monosize particles. These are often referred to as monodisperse systems in contrast to polydisperse systems which contain more than one size and often, in practice, a wide particle size distribution. Further, the separation of at least two mineral species is usually required, both of which can be assumed to have a wide distribution of particle size and, in many cases, variable magnetic properties. When a monodisperse suspension is passed through a separator, the theory provides a method to exactly calculate the outlet concentration, but when a polydisperse mixture is passed through, the outlet concentration cannot easily be predicted because of additional factors. In this paper a number of cases will be discussed, requiring additional factors to those normally considered in monodisperse theory, which produces results very different from those predicted by monodisperse theory. This paper examines some of these additional factors, both theoretically and experimentally, and the treatment of polydisperse mixed-mineral systems is considered. In particular it will be shown that for the capture of a polydisperse, single mineral system by a separator of length L, the values predicted by the monodisperse theory, are often completely at variance with experiment due to a failure to consider the mechanical trapping of particles.