Predictive Geometallurgical Modeling for Flotation Performance in Mixed Copper Ores Using Discriminatory Methods

This study presents a predictive geometallurgical model for the Ouansimi copper deposit in Morocco, focusing on the flotation of mixed copper ores. The deposit was divided into two geometallurgical domains (GMD) with distinct oxidation rates using K-means clustering. The principal component analysis-partial least squares regression method was applied to establish the relationship between mineralogical features, chemical assay, location factors, and flotation performance, enabling the identification of critical factors influencing flotation responses and offering opportunities for improvement. The developed models for copper and silver in GMD (I) outperformed those in GMD (II). Specifically, the negative impact of copper oxide-bearing minerals’ response to flotation xanthate collectors and refractory copper minerals in the gangue on copper flotation performance was identified. Additionally, the performance of silver, likely occurring as an inclusion within chalcocite, was influenced by the interlocking and association between copper-bearing minerals and dolomite. While valuable insights were gained, knowledge gaps related to the presence of copper oxide minerals in the hypogene stage and the genesis of the Ouansimi deposit were identified. Further investigation is warranted to comprehend better the mode of occurrence and potential upgrading of precious metals in low-grade by-products in the final concentrate. In conclusion, our integrated approach, incorporating geological, mineralogical, and metallurgical data, holds promise for enhancing efficiency and profitability in the mining industry through optimized flotation performance in mixed copper ores. The findings contribute to a deeper understanding of mineral processing challenges and offer valuable insights for future exploration and ore processing strategies.