Assessing synthetic difficulty in computational organic materials discovery

Abstract

This thesis presents a study of computational organic materials discovery, focused on the generation and assessment of small molecule aromatic fused ring systems. This was accomplished through the use of MolBuilder, an evolutionary algorithm built to efficiently navigate chemical space by guiding molecular generation according to calculated properties; various methods to incorporate the assessment of synthetic feasibility in generated molecules are demonstratedjavascript:void(0);, in particular the use of computational tools which estimate synthetic difficulty from molecular structure.

Several molecular generation campaigns were conducted with MolBuilder, aiming to optimize physical properties and incorporate a bias towards synthetically feasible candidates for applications in organic semiconducting materials, by constructing fitness functions which optimize one or more objectives as part of the molecular generation process. Top performing species sampled in this manner exhibit promising calculated reorganisation energy values while maintaining low predicted synthetic complexity, and could be suggested as targets for experimental work.

More information

Identifiers

ORCID for Graeme Day: orcid.org/0000-0001-8396-2771

ORCID for Russell Minns: orcid.org/0000-0001-6775-2977

Catalogue record

Export record