Leaching kinetics of valuable metals from waste Li-ion batteries using neural network approach

Abstract

The kinetic study of valuable metals recovery from waste lithium-ion batteries (LIBs) using the artificial neural network (ANN) was investigated. A multilayer feed-forward artificial neural network with back-propagation learning algorithm was used for kinetic analysis of Co, Mn, Ni, and Li leaching from waste LIBs using H2SO4 in the presence of H2O2. Required data for ANNs learning were generated using various random combinations of kinetic parameters in their extended ranges (i.e., activation energy and Arrhenius equation constant) as well as the most common solid–liquid reaction model. The predictive accuracy of the model was comparable with the correlation coefficient of the model fitting method (R2). Results showed that the model developed in this study can be a useful tool in accurately predicting the kinetics of leaching reactions. The activation energy of Co, Mn, Ni, and Li recovery from waste LIBs using the proposed method calculated to be 40.83, 43.35, 39.06, and 27.30 kJmol−1, respectively, and leaching reaction for metals found to follow the surface chemical reaction model.