Smart recovery of rare earth elements from acid mine drainage: Current methods, machine learning integration, and bibliometric analysis

Acid mine drainage (AMD) presents a persistent environmental threat due to its high acidity and metal content. Rare earth elements (REEs), which are essential for contemporary technologies, can also be found in it as a valuable secondary source. This review critically examines the current state of REEs recovery from AMD, focusing on both physicochemical and biological techniques. Key methods discussed include precipitation, adsorption, ion exchange, membrane separation, and solvent extraction. Strategies for recovering REEs and developing hybrid systems are also presented. With an emphasis on predictive modelling, process Optimisation, material discovery, and real-time AMD composition monitoring, the use of machine learning (ML) and artificial intelligence (AI) into REE recovery procedures is investigated. A bibliometric analysis of global research production from 1998 to 2025 identified important figures, new trends, and areas of knowledge that need attention. Findings reveal that hybrid approaches, particularly those integrating functionalized materials with ML-driven process Optimisation, offer the most promise in enhancing selectivity and recovery efficiency while minimizing environmental impact. However, challenges remain in scaling up these technologies, managing co-contaminants, and accurately modelling complex AMD compositions. The development of zero-waste mining techniques and sustainable resource recovery through transdisciplinary innovation are suggested as future research objectives.