Machine Learning Methods for the Prediction of Solution Synthesis Parameters

Despite rapid progress in predicting materials properties, the potential of using machine learning (ML) methods to predict optimal synthesis parameters is still untapped. In Luo et al. [1], we demonstrate how ML can be used for rationalization and acceleration of the discovery process of metal-organic frameworks (MOFs) by directly predicting the synthesis conditions of a MOF based on its crystal structure. Our approach is based on: i) establishing the first MOF synthesis database via automatic extraction of synthesis parameters from the literature, ii) training and optimizing ML models by employing the MOF database, and iii) predicting the synthesis conditions for new MOF structures. The ML models, even at an initial stage, exhibit a good prediction performance, outperforming human expert predictions, obtained through a synthesis survey. The automated synthesis prediction is available via a web tool on https://mof-synthesis.aimat.science. The methods presented in this work are not application-specific, and thus can be transferred to other materials classes.<br/> <br/>[1] Luo, Y., Bag, S., Zaremba, O., Cierpka, A., Andreo, J., Wuttke, S., Friederich, P. and Tsotsalas, M., 2022. MOF Synthesis Prediction Enabled by Automatic Data Mining and Machine Learning. Angewandte Chemie International Edition, 61(19), p.e202200242.