PBM of an industrial-scale vertical wet stirred media mill (HIGMill): Assessment of back-calculation and hybrid methods

An industrial-scale vertical wet stirred media mill (HIGMill) was simulated via a population balance model (PBM). Feed and product particle size distributions (PSDs) of copper ore were measured at various steady-state conditions. The PBM incorporated 1-large–2-small-tanks model and power-law specific breakage rate function S_i. Parameters of S_i and three non-normalized cumulative breakage functions B_ij were estimated by a back-calculation method with global optimization. In a hybrid method, breakage index t_n-based B_ij, obtained from laboratory drop-weight tests, were input to the optimizer to estimate S_i. Results suggest that higher rotor speed, lower solids loading, and use of steel rotor vs. rubber rotor and coarser beads led to faster breakage. The back-calculation with the modified Tavares B_ij achieved the best fits, whereas the hybrid method ranked third. The predictions by all models were reasonably good. We demonstrated pros/cons of the PBM approaches and their parameters' plausibility via a microhydrodynamic scaling analysis.