Machine learning-based classification of thermal stability in metal-organic frameworks

Abstract

As interest in Metal-Organic Frameworks (MOFs) increases, so does the need for research into their thermal stability. This study focused on predicting the thermal stability of MOFs based on various characteristics derived from their structure. We employed two main approaches: using the reactive force field (ReaxFF) to screen the thermal stability of multiple MOFs simultaneously and developing a machine learning model for large datasets to predict thermal stability and investigate the factors involved in it. Notably, the classification model developed in the second approach not only demonstrated high accuracy but also provided insights into the factors affecting the thermal stability of MOFs.