The extraction of cobalt from secondary resources has become crucial, as cobalt has been identified as a strategically important and critical raw material due to the high risks of supply chain disruptions. In this work, selective sulfation roasting was investigated as a potential recycling strategy to extract cobalt and lithium from an industrial lithium cobalt oxide (LCO)-rich black mass. Additionally, the effect of graphite on metal extraction was studied. In the first set of experiments, the sieved black mass fraction containing both cathode and anode materials was directly roasted in a predetermined composition of gas mixtures of SO₂, O₂, and Ar for 1 h at 850 °C. The gas composition was determined from Kellogg’s diagram to allow for the selective sulfation of Co and Li. In another set of experiments, the carbon present in the black mass was first removed by roasting the material in Ar for 2 h and then in an Ar and O₂ gas mixture for five hours at 600 °C. Afterward, selective sulfation roasting was performed in mixtures of SO₂, O₂, and Ar gas similar to the previous set of experiments. For comparison, similar experiments were performed at 800 °C. The sulfation roasted black mass was leached in water to study the efficiency of Co extraction into the solution. Interestingly, the presence of carbon was found to be beneficial for Co extraction. The extraction efficiency for the first case (with carbon present in the raw material) was observed to be more than three times higher than in the second case (with carbon removed) for sulfation at 850 °C. The extraction efficiency and purity of the extracted Co were found to be better for higher temperature sulfation roasting conditions due to faster reaction kinetics. It was also found that almost all of the Li could be recovered while extracting Co. The maximum efficiency of the extraction was 99.51% Li and 61.21% Co for roasting under a gas flow of 10% SO₂-10% O₂-Ar at 850 °C for 60 min. These results suggest that Co and Li can be selectively extracted from the black mass by sulfation roasting pre-treatment followed by leaching in water. In holistic processing, the leach residue can then be further subjected to battery metal processing by state-of-the-art methods.

It is necessary to maintain the silicon content of hot metal in a stipulated range to improve the productivity and energy consumption of a blast furnace. Significant research therefore went into the development of data-driven models to predict hot metal silicon content in real-time. However, these models use only a small subset of blast furnace variables that are chosen using prior process knowledge. As each blast furnace is unique in its operation, using pre-selected variables would lead to sub-optimal models. To address this, a machine learning based ensemble feature selection and modeling approach is proposed. In this approach, all the available furnace variables are ranked using multiple feature selection techniques based on their impact on silicon content. The individual ranks are combined to obtain an ensemble ranking of variables and the top variables in the ranking are used to build data-driven silicon prediction models. This approach is applied to an industrial blast furnace wherein 374 variables are used to obtain the ensemble ranking. While some of the top 100 variables in the ensemble ranking matched those that are commonly used in silicon predictions models, several new variables have also been identified. Silicon prediction models trained using the top 100 variables resulted in a hot rate of ~90% demonstrating the efficacy of the proposed approach. Real-time predictions from the models will enable blast furnace operators to control the silicon content without having to wait for laboratory results.