Por: Aaron Aponte, Universidad Nacional Agraria La Molina; Elizbeth Champi, Universidad Nacional de San Agustín; Esmelin Ocón, Universidad Nacional de Cajamarca; Isabel Vicente, Universidad Nacional Daniel Alcides Carrión; Milton Cutipa, Universidad Nacional Jorge Basadre Grohmann, y Víctor Espinoza, Pontificia Universidad Católica del Perú.

Trabajo ganador de la IV Cantera de Talentos para la Minería del IIMP.

Abstract

This article proposes multidisciplinary strategies to mitigate over-foaming in Cu-Mo flotation, a frequent issue in porphyry-skarn deposits with high concentrations of hydrophobic phyllosilicates such as talc, serpentines, or chlorites, which generate unstable froths that affect recovery and concentrate quality. 

From the geological perspective, a predictive geometallurgical model is proposed based on the “total rock” concept and on both the physical and mineralogical characterization of phyllosilicates, enabling the classification and segregation of problematic blocks before they reach the plant. 

Mining engineering contributes through a blending strategy that combines materials with different mineralogies to stabilize the process. On the metallurgical side, the use of Jameson cells is reviewed, as they enhance separation kinetics and reduce entrainment. 

From an environmental standpoint, the benefits of applying the proposed strategies in Cu-Mo flotation are assessed, focusing on reduced water and energy consumption, as well as minimized use of reagents. 

Finally, a predictive model using Machine Learning (Random Forest) was developed. Fed with geological, operational, and environmental variables, it is able to anticipate over-foaming events with 85.7% accuracy, providing valuable support for process control.