Process Systems Engineering for Spent Li-Ion Battery Recycling: Model-Based Diagnosis and Optimization of the Lithium Recovery Bottleneck

Abstract The efficient separation and recovery of lithium from spent lithium-ion batteries (LIBR) remains a critical challenge in hydrometallurgical recycling, primarily due to its significant and dispersed losses throughout the multistage purification process. This study addresses this bottleneck by developing a high-fidelity, steady-state mathematic model for an entire LIBR process. The model, which has been rigorously validated against plant data, serves as a digital twin to quantitatively diagnose the loss path of lithium. A full-process elemental flow analysis revealed that lithium recovery was the key limiting factor, with a process yield of only 61.08%, compared to >92% for nickel and cobalt. The analysis pinpointed the leaching and final precipitation units as primary loss sources, identifying soluble Li+ (81.94%) and insoluble LiF (17.94%) as the dominant loss forms. A systematic sensitivity analysis was then conducted to deconvolute the effects of key separation parameters: a reasonable pH of the leaching process is recommended to be 2.0 at the process temperature, and the operating pressure of the lithium precipitation unit should be higher than 1.6 bar to reduce entrainment losses. Implementing these optimized conditions led to an increase in the lithium carbonate precipitation ratio, thereby raising the final lithium recovery to 70.47%. A coupled techno-economic analysis confirmed the viability of the optimization strategy. Although the unit production cost increased by 2.3%, the enhanced lithium yield increased the unit net profit by 3.8%, which also shifted the revenue structure, making Li2CO3 the dominant product. Furthermore, the uncertainty analysis further reveals that the profitability of the LIBR process is most sensitive to the cost of raw materials and the prices of lithium and cobalt products. This work demonstrates a closed-loop methodology that integrates process simulation, bottleneck diagnosis, targeted optimization, and techno-economic verification to systematically improve the separation efficiency and economic potential of complex resource recovery processes.