Lithium ion batteries (LIBs) are inescapable in everyday life. From electric vehicles to portable devices, LIBs are in high demand. With more knowledge of the effect unrecycled battery components have on the environment and the economy, researchers from the Beijing Institute of Technology have compiled a review of the current recycling approaches for spent cathode materials as well as proposing a closed-loop recycling system powered by artificial intelligence. This process uses AI to match the specific failure state of the cathode materials to the most effective strategy for recycling. By developing this approach, Lai Chen and his team hope to relieve the load on the environment and the supply chain by reusing materials that still have life in them, along with altering the paradigm we operate under to shift from “traditional, rough processing” to “precise recycling”.
The results of the review were published in Environmental Chemistry and Safety on July 20th, 2026.
Future estimates on waste produced by LIBs for electric vehicles alone are projected to be up to 8 million tons by 2040: a staggering number for only one area of LIB use. Not only is this waste detrimental to the environment, but many of the metal resources within the cathode material of the battery might still have use available for other applications. Recycling will decrease the pressure on the supply chain for metals like lithium, manganese, cobalt and nickel.
There are recycling methods in use that vary in their effectiveness in reducing greenhouse gas emissions, namely pyrometallurgical (17% reduction), hydrometallurgical (51%), and direct recycling/upcycling (61%). The potential greenhouse gas emission reduction by these methods could be up to 16.3% in 2060 using the Lithium Cycle Computable General Equilibrium (LCCGE) model. However, these methods use rough sorting, which can damage the materials and affect the overall amount recovered.
The use of artificial intelligence can change the way metal cathode material recycling occurs by using precision recycling, a method that would require highly trained AI models and robust sets of data on batteries, their materials, and various capacities and degradation metrics.
“ The application of artificial intelligence can enable adaptive, data-informed decision-making throughout the battery life cycle, effectively overcoming the critical barrier caused by the lack of reliable data regarding the state of retired batteries, ” said Lai Chen, author of the review and associate professor at Beijing Institute of Technology.
AI use in this field would highlight the optimal recycling route using non-destructive methods based on the failure features of the cathode materials. Such failure features that would be used to determine recycling strategy include lithium loss, surface/interface deterioration, transition-metal dissolution, and structural disorder and phase reconstruction. The digital intelligence technologies and sorting/recycling procedures would be integrated and work together in one facility to promote the closed-loop system proposed.
Chen suggests a call to action for future researchers to systematically document cathode degradation metrics to begin establishing comprehensive datasets for AI models to learn. The information needed would include cathode degradation metrics such as lithium inventory loss ratios, structural phase transition percentages and transition-metal dissolution concentrations.
This type of information is indispensable to developing the closed-loop design and will allow for evaluations of state of charge, state of health and remaining useful life to be determined. Additionally, the standardization of battery health diagnostics, which has yet to be established, is needed to develop this system.
Rui Tang of the School of Materials Science and Engineering at the Beijing Institute of Technology, along with Hong Liu, Lai Chen, Jinyang Dong, Yi Jin, Xianglei Meng, Yiling Ren, Yuchen Wei, Huimin Yang, Yun Lu, Qing Huang and Yuefeng Su also of the Beijing Institute of Technology Chongqing Innovation Center, and Yibiao Guan of the China Electric Power Research Institute contributed to this research.
The National Natural Science Foundation of China, The High-Level Talent Introduction Project of Yibin and the China Postdoctoral Science Foundation supported this study.
D OI Link:
https://doi.org/10.26599/ECS.2026.9600053
Environmental chemistry and safety
Precision Recycling of Spent Cathode Materials: From Degradation Mechanisms to Recycling Strategies
20-Jul-2026