Using particle-based separation models to evaluate the selectivity of different collectors in chalcopyrite flotation

Particle-based separation models are a powerful tool for modelling and understanding mineral separation processes at the level of single particles. Latest developments in this field have enabled the incorporation of complete particle datasets from image-analysis based techniques, thus allowing for the full integration of material complexity into process models. So far, these models have mostly been applied to static processes, without variations in operating conditions. In this contribution, we used particle-based separation models to understand variations in the flotation behavior of a fine-grained and low-grade chalcopyrite-dominated copper ore using different collectors: PAX and kerosene. This approach highlights the influence of particle size and shape on the flotation of chalcopyrite-bearing particles. Moreover, it demonstrates that detailed information on mineral associations is critical to achieve a full description of the process behavior of single particles. Full association data should therefore be used instead of simplified ore mineral liberation data whenever possible. Finally, results indicate that higher selectivity against pyrite can be achieved when kerosene is used as a collector instead of PAX. In addition, ideas for improving separation (e.g., higher grade and recovery) are discussed based on detailed particle information.