Method to evaluate and prove-the-concept of magnetic separation and/or classification of particles

Abstract

When designing new magnetic separators and/or classifiers or optimizing existing ones, it is usual to face several obstacles: the high cost of a proof of concept full laboratorial setup (including preliminary optimization procedures and/or feasibility demonstrations), time-consuming experiments, lack of flexibility of the assembled laboratorial apparatus, feed complexity, among others. In this work a method and corresponding methodology are proposed to apply in such cases, representing a low-cost, flexible and robust alternative to overcome the mentioned obstacles, from which working parameters of a laboratorial or even larger version of the device may be extrapolated. This represents a powerful tool when designing magnetic separators. In the proposed methodology by determining in one experiment the magnetic force required to separate/classify a particle in a certain point, it may be derived immediately the change in magnitude and shape of the magnetic force index (Bâ B) that must exist to separate other particles (with the same or different magnetic susceptibility) in another point, and it is possible to estimate, for example, the optimum, maximum or minimum value of other variables affecting the competing forces (e.g. radius of the particles, fluid density, rotation velocity), and also determine the critical limits of separation by extrapolating and obtaining the magnetic force required in those limits. It represents an open field allowing determining freely the values of the main variables. This methodology and associated method also allow repeating quickly and easily the experiments with different sets of geometrical design and positions. A case study was analyzed and tested for both processes: magnetic separation and magnetic classification, with good results, that allowed to conclude about the feasibility of the system for both processes, and to determine the best configuration geometry. The main objective of the present study was to demonstrate a cheap method and corresponding methodology that may be applied when designing magnetic separators and/or classifiers. Â(c) 2016 Elsevier B.V.