Enhancement of scale up capability on AMIRA P9 flotation model by incorporating turbulence parameters

The success of any scale up process is dependent on how well the small scale test work represents the conditions of the larger scale and how well the predictive model captures the key contributors to the process. The AMIRA P9 flotation model was developed by the P9 team over a number of years (Gorain, 1997; Gorain et al., 2006). It has been widely used and generated significant value for many operations by assisting the decision making process through simulations of process changes and for flotation scale up predictions. This model assumes that the parameter P (ore floatability) remains constant over a wide range of surface area flux measurements (S_b) regardless of the amount of power introduced into the flotation cell. As a focus of this research, flotation test work was conducted at different hydrodynamic conditions in 5 and 60 L cells to challenge this assumption. Increasing the power input increased the P value especially in finer particle size classes (− 75 μm) in both cells demonstrating that P is not a true ore property because it is influenced by the cell hydrodynamics. To improve the accuracy of the AMIRA P9 flotation model in predicting the flotation rate constant (k) and to improve the consistency of the ore property in the model, measurable and appropriate turbulence parameters were sought to be incorporated into the model. Dimensionless turbulence parameters that could be obtained directly from measurements, ӕ (characterizing turbulence intensity, bubble size and viscosity) and EVF (characterizing the effective volume in flotation where majority of collision and attachment occurs) were formulated and introduced to the AMIRA P9 model. The consistency of P″ (the updated floatability component) as a more consistent ore property was improved which enhanced the flotation rate constant prediction for a variety of hydrodynamic conditions of the 60 L flotation cell. Further work is recommended to test the model for continuous rather than batch processing, in larger cells that are more similar to industrial cells and to further improve the accuracy and precision of predictions of the behaviour of coarse particles.