Investigation on the Flotation Recovery of Mixed Copper Ore Through an Integrated Mineralogical Approach

The flotation behavior of mixed copper ore during selective copper flotation was investigated using a geometallurgical approach. To this end, samples were taken from a production block at the Ouansimi copper deposit and subjected to laboratory batch kinetics tests and quantitative mineral analyses. Using a range of ore samples from different location in the same deposit, the influence of mineralogy on flotation performance was studied. The automated quantitative mineralogy technique was utilized to characterize the samples, revealing that bornite, chalcopyrite, and chalcocite are the copper-bearing sulfide minerals, whereas malachite is the major copper-bearing carbonate mineral. In order to construct a robust model with the highest quality data (Q2) and excellent goodness of fit (R2), the deposit was divided into two homogeneous geometallurgical domains using a K-means clustering algorithm based on mineralogical and metallurgical information. Copper recovery of samples by geometallurgical domain was correlated to the feed mineralogy through statistical algorithms for principal component analysis (PCA) and partial least squares regression (PLS) regression to look for relationships between mineralogical features and flotation recovery. For the domain 1, a significant negative correlation was found between copper recovery and the middling and locked copper association (iron oxide with a trace of copper, copper associated with gangue like biotite, barite, kaolinite, chalcocite locked and delafossite). By contrast, in domain 2, a negative effect on copper recovery was established due to the presence of iron oxide with a trace of copper and malachite and bornite.