University of Leeds
CatSD: structural database and high-throughout predictive workflows for homogeneous catalyst design
Abstract
dc:description.abstractIdentification of highly active catalysts is an important process across multiple industries including drug development, process chemistry and agrochemicals. The lack of understanding of ligand properties and catalytic pathways are limiting factors for the uptake of more sustainable and highly active catalysts. Herein we report a novel method for the identification of ligands and the prediction of their activity for homogeneous catalysts from the Cambridge Structural Database. We present CatSD, a structural database complete with catalytically relevant features to enable the mining of organometallic ligands from the CSD. We also present a high-throughput computational workflow for the prediction of activation energies and mechanistic exploration. This workflow is on a timescale similar to experimental high-throughput screening and provides energies with an accuracy of 3.9 kcal mol-1 . CatSD and the prediction workflow were applied to the Ullmann-Goldberg reaction to identify novel ligands for amine and amide coupling partners. Over 10,000 ligands were identified from the CSD for both coupling partners. The workflow showed excellent reliability for the generation of starting structures (99.7%) and good reliability for the optimisation of important intermediates (>84%) and transition states (TSOA: 33-61%, TSSig: 83-85%). Several ligands were validated experimentally identifying a previously unreported active ligand class. The effect of ligand properties was explored using machine learning to identify several key characteristics for both nucleophile coupling partners. Machine learning was also used to predict activation energies without the need to calculate the transition state. Models were optimised providing accuracy on par with the accuracy of the workflow calculations. It is our hope that the methodologies presented in this work will aid the discovery and design of ligands for homogeneous catalysts for the wider chemistry community as well as stimulate further research in this field.
Degree
thesis:*- Name dc:type.qualificationname
- Ph.D
- Level dc:type.qualificationlevel
- doctoral
- Grantor dc:publisher.institution
- University of Leeds
- Year dc:date.issued
- 2022
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Short, Marc Andrew Stephen
- Advisors dc:contributor.advisor
-
- Nguyen, Bao
- Willans, Charlotte
- Tovee, Clare
- ORCID person:identifier.orcid
Rights
- Language dc:language.dcterms_iso639_2
- eng
Identifiers
dc:identifier.*- Identifier
- uk.bl.ethos.874947
- Dc Identifier Grantnumber
- EP/L015285/1