Segregation phenomena in gravity separators: A combined numerical and experimental study

Abstract

In this paper we formulate a computational framework for characterizing and optimizing the performance of a destoner, an example of a density-based separation technique. The numerical framework combines Computational Fluid Dynamics (CFD) simulations with a Discrete Element Method (DEM), implemented in an open-source computation package (OpenFOAM). This framework is validated first by comparing the simulations with experiments for a standardized test case and further with our experimental study of a pilot-scale destoner. We evaluate the combined effects of process conditions, such as separator deck inclination, vibration speed and fluidization velocities on destoner performance. Our simulations show how the heavy product fraction in the discards stream increases over time with a corresponding accumulation of the ‘valuable’ light product at the base of the deck, indicating segregation between the stones (heavy product) and the grains (light product). We also find that these separation profiles are highly sensitive to changes in deck surface air velocities, with the gradual development of segregation zones at velocities close to the minimum fluidization velocity of the heavier component. Optimal separation is seen at a deck inclination of 4° and a fluidization velocity of between 1.75 and 2m/s. Our simulation results also agree well with the experimental findings indicating the usability of the proposed framework for the design and optimization of gravity separators.