Contact Details:

henry.thake@warwick.ac.uk

Department of Chemistry: G209

PhD student, September 2024-present

PhD Project

The aim of my project is to identify how certain local defects can be created in graphene networks and their general design principles for systematic, scaled growth. The ability of these defective networks to support novel types of metal catalyst materials and nanostructured magnetic materials will also be studied. Should such materials be found, I will look at how the properties of these materials are influenced by the underlying graphene structure and explore whether characteristics can be selected for based on the species used to grow the graphene sheets.

By considering known precursors to and defective graphitic materials, I will use DFT calculations to create training data for the development of machine learned interatomic potentials (MLIPs) that will enable me to predict the adsorption and lateral interactions of other precursors and products. This model will allow me to determine whether such precursor materials are viable for space-confined solution-phase (SCSP) type synthesis of defective graphene.

If secondary models can be developed that learn how precursor materials grow into defective graphene, it may become possible to predict how topological defects like flower-structures and 5-7 or 55-8 chains can be grown with predetermined tilt boundary angles and, most crucially, the precursor material(s) to do this.

As well as looking at the synthesis of graphitic structures, I intend to also examine how we can produce better (and cheaper) predictions of analytical data, focussing on near-edge adsorption fine structure spectroscopy (NEXAFS). The current level of simulation for such spectra using DFT scales poorly with the number of atoms in the molecule and doesn't yield correct absolute peak positions or intensities. One of my aims is, firstly, to improve the quality of the DFT simulations s.t. it can make more accurate predictions, and then use this, in combination with experimental data, to train a ML tool to accelerate the process.

Background

I completed a BA and MSci in Natural Sciences from the University of Cambridge. My MSci project, supervised by Prof. Steve Jenkins, used MD simulations to explore radical-mediated mechanisms for the fluorination of group 14 (100) surfaces.