Improving accuracy of polyoxometalate computational models

Malcolm, Daniel (2024) Improving accuracy of polyoxometalate computational models. PhD thesis, University of Glasgow.

Abstract

Given the growing materials and energy crises, where the Earth’s stock of precious, non-renewable metals and fuels being consumed by an ever-bigger population of human beings, it is more imperative than ever that we design smarter technology that is not only versatile, but also relevant to the problems we currently face; from being able to store renewable energy for long periods of time to finding alternative batteries to power our transportation, there is an ever greater need to develop electrochemically based methods of storing energy. Polyoxometalates (POMs) as a family of molecules are well-known for their ability to reversibly store large number of electrons per unit cage, solubility and structural durability under a wide range of environmental conditions, and studies show that alteration of the central heteroatom can further tune the oxidizing power of the species.¹

One of the main issues holding POMs back from being specially designed to exhibit the most desirable properties allowing for optimal deployment in key fields such as redox flow batteries is the lack of understanding surrounding their self-assembly mechanisms and, therefore, synthetic methods. Without this fundamental knowledge of how metal oxide reagents interact to yield unique structures under particular environmental conditions, we are left to rely on change discoveries to provide us with innovation. The first step towards building a greater wealth of knowledge is to refine our ability to simulate POM characteristics and properties using DFT, iteratively improving our calculations by benchmarking against empirical data; with this theory behind us, it should be easier to elucidate the mysteries hiding within the synthetic mixture.

With this investigation, we set out to establish how one can accurately model three species of POM: the well-known [X₂W₁₈Oₘ]ⁿ− Wells-Dawson (X = As, P, Se m = 60, 62), its hexalacunary variant [X₂W₁₂Oₘ]ⁿ− (X = As, P, Se m = 46, 48), and the wheel-shaped [X₈W₄₈Oₘ]ⁿ− framework these hexalacunaries can self-aggregate to form (X = As, P, Se m = 176, 184). After detailing the POM, computational chemistry, inverse design basics, and experimental details relevant to this work (Chapters 1, 2, and 3 respectively), we discuss our strategies and results with modelling the [X₈W₄₈O₁₈₄]ⁿ− POM wheel, highlighting how inclusion of only a few of the total number of countercations is sufficient for good empirical comparison (Chapter 5). Likewise, we repeat the same premise with the [X₂W₁₈O₆₂]ⁿ− Wells-Dawson and [X₂W₁₂O₄₈]ⁿ− hexalacunary species (Chapter 6), concluding that presence of several (but not all) countercations is essential for accurate replication of the POM structure but additional inclusion of protons within the framework will provide the best model for analysing regions of electron density and frontier orbital data. Finally, we review the available literature regarding computational modelling of POM UV-Vis spectra, arguing that the current level of theory is insufficient for accurate results and reviewing the available options (Chapter 7).

We also included results from a brief organic chemistry project we conducted over the course of this PhD thesis relating to the Ugi reaction which we, in the end, were not able to see through to completion (Chapter 8). Following this are the appendices for Chapters 10-13 (Appendix:1 – Appendix:4 respectively), collecting the vast number of tables, figures, and graphs produced by this work.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Subjects: Q Science > QD Chemistry
• Colleges/Schools: College of Science and Engineering > School of Chemistry
• Supervisor's Name: Vila-Nadal, Dr. Laia
• Date of Award: 2024
• Depositing User: Theses Team
• Unique ID: glathesis:2024-84592
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 27 Sep 2024 12:49
• Last Modified: 27 Sep 2024 12:50
• Thesis DOI: 10.5525/gla.thesis.84592
• URI: https://theses.gla.ac.uk/id/eprint/84592

Related URLs

• Book section
• Pre-print
• Article DOI