The booming new energy industry has fueled a surge in global lithium demand, with the annual demand for lithium carbonate (Li₂CO₃) equivalent (LCE) projected to reach 11.2 million tons by 2050. As a key raw material for lithium extraction, spodumene generates approximately 10–15 tons of lithium slag per ton of lithium carbonate (Li₂CO₃) produced. However, the comprehensive utilization rate of lithium slag in China remains below 30%, and most of it is disposed of through landfilling, posing soil pollution risks. This review summarizes the main lithium extraction processes from spodumene: the sulfuric acid method (with a lithium recovery rate of over 96% but high acid consumption); alkali processes (achieving 96%–99% lithium recovery and featuring low equipment corrosion, yet with untested applicability to low-grade ores); salt roasting (simplifying purification processes but only achieving ~60% sulfate recovery); and chlorination roasting (with a lithium recovery rate of over 95% but requiring strict safety controls). Additionally, this review covers the resource utilization of lithium slag: 8–10 million tons of gypsum can be recovered annually (filling 16%–20% of China’s industrial by-product gypsum supply gap); the silica–alumina micro-powder can enhance concrete strength and reduce glass fiber production costs; and over 94% of tantalum (Ta) and niobium (Nb) can be recovered from fine tantalite concentrate slag. Key research gaps and future development directions are also identified to support the low-carbon development of the lithium industry.

Against the global drive for energy conservation and emission reduction, lithium-ion batteries (LIBs) have witnessed increasingly widespread adoption across diverse sectors, thereby generating an urgent imperative for the efficient recycling of electrodes from spent LIBs. This paper presents a comprehensive review of recent research progress in this domain. It begins by elucidating the fundamental architecture of LIBs and the key physicochemical characteristics of their electrode materials, followed by a critical analysis of the major technical bottlenecks inherent in current recycling processes. While conventional separation techniques are briefly discussed, the primary emphasis is placed on reviewing deep learning-based visual inspection systems, with particular attention to YOLOv8-based approaches reported in the literature. Finally, future development trajectories are delineated, and pivotal research directions including the optimization of multi-technology integration and the expansion of training dataset scale and diversity are proposed to facilitate the large-scale, standardized industrial implementation of electrode recycling technologies.