Solvent extraction of cobalt from spent lithium-ion batteries: Dynamic optimization of the number of extraction stages using factorial design of experiments and response surface methodology

Abstract

The optimization of lithium-ion batteries (LiBs) recycling is crucial not only from a waste management perspective but also to decrease the dependence on imports of critical raw materials. In addition, the diversification of the recycling technologies is very important for better flexibility of the market. This study aims at investigating the recovery of Co from spent LiBs using solvent extraction from a real chloride-based solution obtained after the removal of Mn, which is very rarely reported. Cyanex 272 was used as the extractant and the effect of several variables on the extraction efficiency was considered to model and optimize the separation of Co and Ni. The number of extraction stages directly affects not only the process efficiency but also its cost. Thus, in this work, a novel approach was developed to assist in the selection of the number of extraction stages using a dynamic method based on the factorial design of experiments and response surface methodology combined with the Kremseŕs Equation. This method can assist the process design, decrease the overall cost of the operation, and optimize the separation of Co and Ni in a reduced number of extraction stages. The concentration of Co and Ni in the feed solutions is ∼ 8.3 g/L and 1.9 g/L, respectively. Based on the results, 98% extraction efficiency for Co can be achieved in 1 to 2 extraction stages with low co-extraction of Ni (<5%) when using 0.6–0.8 M Cyanex 272, O:A ratio below 1 and pH ∼ 5, but several combinations of conditions could provide similar results.