Abstract:
Clusters of metal atoms have attracted considerable attention in recent years, owing to their ability to emulate the behaviour of individual atoms. These `superatoms' have a global shell structure, populated by the valence electrons of their constituent atoms. As one would expect the properties of these clusters are dependent on the atoms constituting them, leading to superatom research motivating the development of novel building blocks for nano-structured materials. Whilst a substantial body of work has been conducted on the potential uses and limitations of superatomic systems, the transition metals provide fertile ground for further exploration. In the work described in this thesis small transition metal clusters are explored using Density Functional Theory (DFT) methods to explore their utility in superatomic clusters, but also novel features which arise from the inclusion of d-orbitals into the superatomic context. Furthermore, this work aims to shed light on the limits of DFT in exploring these systems, with particular emphasis placed on the exploration of the self-interaction error (SIE) inherent to DFT. The first body of work outlined in this thesis shows how nickel atoms are one electron per atom superatomic species, which localise a single unpaired electron on each of the nickel centres, allowing for predictable determination of the magnetic moment of small nickel clusters. These clusters also demonstrate how individual nickel atoms, depending on how close a given cluster is to electronic shell closure, are able to change their valence electronic structure in order to maintain shell closure, consequently lowering the nett multiplicity of the cluster. The second body of work presented in this thesis expands upon this, by showing how the selective replacement of copper atoms for nickel can controllably alter the magnetic moment of clusters. This work cuts to the heart of superatomic investigation, in which it is illustrated how potential building blocks can be predictably altered to include dierent features. The third body of work tests the extent to which d-electrons can be considered delocalised by exploring group 3 and 4 transition metal clusters. This work introduces the concept of partial delocalisation and explores the interconnectedness states formed from d-electron density to classically superatomic states. The limits of DFT are also tested in this work, in which it is shown how the choice of exchange-correlation functional can have non-trivial inuences on the global electronic structure. The nal body of work presented is a broad based description of a range of seven atom clusters in which the importance of accounting for the SIE is tested, and what properties of dierent systems can be reliably extracted from DFT studies.