The use of electrically conductive metal-organic framework(MOFs) can open doors to electronic device applications, including electrodes, chemiresistive sensors, energy storage and more. However, the known number of intrinsically conductive MOFs is limited, and experimentation without computational guidance is arduous. Therefore, predicting MOFs that exhibit electrical conductivity via machine learning (ML) methods would save money, effort, and time. I ML models such as neural network were trained using 52,000 inorganic compounds from Open Quantum Materials Database (OQMD) and then used to screen and classify ~12,000 MOF entries within the Computation-Ready Experimental (CoRE) MOF database as either conductive or non-conductive. The majority of the classifying functions predicted 293 entries from the CoRE MOF database to be conductive, 87 of which have been mentioned in the literature.

Screening for conductive metal organic framework compounds by utilizing a neural network

ACS Spring Meeting 2021

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