While a handful of intrinsically conducting metal-organic frameworks (MOFs) have been synthesized and characterized for electronic applications, there are currently limited efforts to expand this number without significant time, cost, and effort. Furthermore, though there are tens of thousands of reported MOF structures in existing databases, it is an arduous task to efficiently and effectively identify which are electronically conductive. Using machine learning (ML) techniques, such as neural networks, random forest, and support vector machines, we screened ~12,000 MOFs and identified 293 MOF crystal structures that are semiconductors based on their bandgaps and uncertainty quantification methods. We are currently in the process of expermentally verifying these MOFs predicted by ML. Our work demonstrates the potential power of ML to aid in down-selecting from large numbers of potential candidates while improving certainty and simulltaneously accelerate the discovery of novel conductive materials.

Accelerated discovery of conductive metal–organic frameworks by machine learning

ACS Spring Meeting 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.