Multivariate models are a useful tool when studying the effects of independent variables on one or more dependent variables, since this approach allows modeling of the dynamics of complex systems based on simple analytical models with considerable certainty. Due to the decrease in the copper oxide mineral grades, leaching of copper sulfide minerals (secondary sulfides) has positioned itself as a benchmark of operation for the Chilean mining industry. The present work proposes the study of the effects of sulfuric acid, chloride concentration and time on the extraction of copper from sulfuric minerals (chalcocite), considering an experimental design, the surface optimization methodology and the adjustment of a quadratic model. The experimental data were adjusted by multiple regression analysis and were statistically analyzed. A model was developed to represent the copper extraction from the Cu2S mineral as a function of the statistically significant variables (chloride concentration and time) that contribute to explain the variation of the response variable under the set of parameters sampled.

Lateritic ores constitute the main source of raw material for extraction of Ni and Co by the Caron process. Consumption of oil in the reduction furnace is one of the key indicators if the metallurgical process is economical. To date it has not been possible to replace the additive fuel oil that is used at commercial scales, therefore, the aim of this study was to partially replace the oil with bituminous coal on a pilot scale by using a mixture of 2 % coal and 1.25 % oil as the reducer additive. Phases of the reduced/leached ores were analyzed by powder X-ray diffraction, while the metallic state of the ore was determined by leaching the reacted samples with a bromine-ethanol solution followed by the atomic absorption spectrometry analysis. Extractions of Ni and Co were confirmed by leaching the reduced ore with ammoniacal-ammonium carbonate solutions. It was observed that the mixture used as a reducer additive can replace the fuel oil since it allows the adequate transformation of the main mineralogical phases of the laterite ore during the reduction process and the average extraction yields of Ni and Co for ~3 and ~8 %, respectively. Although the effect of bituminous coal particle size in the process was not analyzed, the reducing mixture ensured that the Caron process was more efficient.

To date, the grinding behavior of saprolite and lateritic overburden mixtures remains poorly understood. The Bond Work Index (BWI) is the principal indicator used to determine the specific energy consumption during the grinding process. To establish the F₈₀ and P₈₀ values, granulometric distribution models—Rosin–Rammler (RR), Gates–Gaudin–Schuhmann (GGS), and the Swebrec function (SWEF)—were evaluated. The mineral phases of the feed samples were analyzed by X-ray powder diffraction. This study provides evidence that the RR function is the most suitable for simulating the particle size distribution of the feed material, with residual errors below 6.30% and a coefficient of determination (R²) exceeding 97%. After the grinding equilibrium cycle is reached, the SWEF model proves to be the most appropriate, exhibiting residual errors under 3.50% and R² values above 98%. BWI reveals that saprolite is the most difficult ore to grind, with specific energy consumption increasing from 16.38 kWh/t to 25.50 kWh/t as the proportion of saprolite in the mixture rises. This reflects a clear upward trend, as confirmed by a fitted model with an R² of 98.54%. In contrast, the grindability index (Gbp) decreases, indicating that the material becomes increasingly resistant to grinding as the saprolite content increases. This may be attributed to inherent material properties, such as hardness, or to physical phenomena related to fragmentation. The declining Gbp further suggests that greater energy input is required to achieve additional particle size reduction. Overall, the findings demonstrate that saprolite is inherently difficult to grind and behaves according to its own grinding characteristics, regardless of whether it is processed alone or in combination with lateritic overburden.

La presente investigación se desarrolló en las instalaciones del Centro de Investigaciones del Níquel ¨Capitán Alberto Fernández Montes de Oca¨, teniendo como objetivo evaluar el efecto que tiene la sustitución del petróleo aditivo por el carbón bituminoso en el proceso de molienda de las lateritas. En ella se demostró que el índice de Bond se incrementa a medida que aumenta el contenido de carbón bituminoso en la mezcla. Al compararla con la mezcla de petróleo y laterita utilizada en la industria se puede afirmar que utilizando un porcentaje inferior a 4,5 % de carbón bituminoso no se incrementa el consumo energético específico en los molinos, mientras que al utilizar el carbón bituminoso en el rango de 5,0 a 5,5 % se incrementa en 3,0 kW-h/t. Las mezclas preparadas cumplieron con las especificaciones exigidas por el ensayo de Bond, presentando un comportamiento granulométrico con predominio de las partículas gruesas, antes y después de ser sometidas a dicho procedimiento.

Lateritic ore is currently the main raw material that allows extraction of Ni by the Caron process. To date, the kinetic behaviour of the roasting/selective reduction process of lateritic nickeli-ferous ores on a pilot scale is largely unknown. In the present study, the kinetic behaviour, the controlling stage and the mechanisms that describe this process with the use of a mixture of 2 wt.% bituminous coal and 1.25 wt.% fuel oil as a reducing additive were determined during the evaluation of the high and low thermal profiles, respectively. The phases of the reduced/le-ached minerals and the fed ore were analysed by X-ray powder diffraction. It was observed that the mixture used as a reducing additive guarantees an adequate transformation in both thermal profiles; the relationship between the residual Ni and residence time is described by a first-order reaction with the determination coefficients greater than 0.949. Although the influence of post-combustion air is not analysed, the controlling stage was diffusion through the ash layer with an activation energy of 14.4060 kJ mol-1. Thus, the most precise combination to describe the process is diffusion through the ash layer and the growth of the nuclei.