Machine Learning and AI Empowering Metal-Organic Frameworks: Synthesis, Performance Prediction and Therapeutic Application

Abstract

Machine learning (ML), a subset of artificial intelligence(AI), has emerged as a powerful tool to address the bottlenecks in Metal-organic frameworks (MOF) research. With the exponential growth of structural libraries rendering traditional trial-and-error experimentation intractable, data-driven models enable efficient high-throughput screening and accurate prediction of MOF properties. In this review, we discussed the mechanism of ML and the database of MOFs. Then the latest applications of ML in the field of MOFs are highlighted, including the prediction of the synthesis route, crystal structure, ability of drug loading, ability to adsorb gases, and research progress of ML in diagnosing diseases. Finally, the challenges and limitations of ML in MOFs research are also discussed. Studying various ML algorithms to solve the performance prediction problems of MOFs in practical applications will help uncover the intrinsic connections between MOFs features and specific target properties, enhance the efficiency of materials science research, and promote the efficient application of ML in MOFs.