We have been studying use of 3D molecular orbitals of products and reactants as input to conditional generative adversarial networks (cGAN) to model organic chlorinations. Our previous work with electron densities has been extended to include nuclear density. The resulting cGAN then takes a 3D computed density and generates an image of the electronic structure and nuclear positions of the chlorinated product. The current status, successes, and limitations will be discussed.

Use of conditional generative adversarial networks (cGAN) with molecular orbital fields to study chlorination sites

ACS Fall 2020 Virtual Meeting

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