Bubble - particle interactions with hydrodynamics, XDLVO theory, and surface roughness for flotation in an agitated tank using CFD simulations

Abstract

The modeling of flotation in an agitated tank can be conducted by assuming the average turbulence or using more realistic turbulence at each point of the tank through computational fluid dynamics (CFD). Turbulence is used to calculate the probabilities of the collision, attachment, and stability of a particle with a bubble to ultimately predict the particle recovery. The extended Derjaguin - Landau - Verwey - Overbeek (XDLVO) theory helps to interpret the particle - bubble interaction and can be included in the calculation of the probabilities. Our study presents, for the first time, a theoretical work of flotation modeling in an agitated tank using CFD and the XDLVO theory with particle surface roughness (SR). In detail, we compared three approaches: one without XDLVO, one with XDLVO, and another with XDLVO including SR. We found that the approach without XDLVO predicts that the attachment probability decreases as the particle-bubble velocities increase, whereas the one with XDLVO predicts that it increases. The approach without XDLVO predicts higher particle stability against turbulence than the one with XDLVO. The modeling without XDLVO had fewer changes in particle fraction remaining in the tank as the contact angle increased than that in the one with XDLVO. Finally, the SR had a marked influence only on the attachment probability and, ultimately, did not significantly affect the particle fraction remaining in the tank.