Computational Studies of Structure and Surface Reactivity in Metal Nanoclusters

Metal nanoparticles exhibit physical and chemical properties unique to their length scale that have the potential to shape next generation technologies. The large structural and chemical space that determines nanoparticle properties requires feedback between computational and experimental studies to drive the discovery of both new architectures and target properties of these systems. This dissertation describes the application and development of theoretical methods to study metal nanoparticle electronic structure, developing new structure-property relationships and new concepts that govern nanoparticle behavior while connecting theoretical insight with laboratory observations. In Chapter 1, the dissertation is introduced by detailing how the connection between geometry and electronic structure has shaped the way we teach and understand chemistry across length scale, and projects these concepts onto the 1-3 nm length scale where traditional descriptors of electronic structure break down. In Chapter 2, the optoelectronic impact of alloying Cu with a Au nanocluster is studied, revealing how atomic descriptors and position of the heteroatom determine optical absorption in [Au25(SR)18] , providing an easily accessible experimental readout of electronic structure. Building on hypotheses tested in Chapter 2, Chapter 3 explores the size dependence of Cu/Au alloying. Here, the atom concentration and composition architecture are key parameters predicted to drive the emergence of plasmonic behavior in Au144-xCux(SR)60 nanocluster. In addition to the composition dependence of nanoparticle properties, the surface chemistry is known to dominate overall nanocluster electronic structure. In Chapter 4, both the type and specific molecular descriptors of the ligand were shown to impact the total magnetic moment of Co13 and Co55 model nanoclusters. Chapter 5 extends these concepts of surface chemistry to address the size dependent evolution of the ligand mediated magnetic properties in CoN nanoclusters, demonstrating how energy level alignment and orbital symmetry contribute to size dependent trends. Finally, Chapter 6 describes a reduced scaling computational method that improves the approximation of mean field excited state energy predictions, increasing the size and complexity of systems that can be treated with high accuracy.