Wang, Yuchen2024-04-052024-04-052024https://hdl.handle.net/2097/44172Gold and silver nanoparticles can exhibit unique physical and chemical properties such as plasmonic resonance characters. Those types of plasmonic nanoparticles usually exist in large sizes, which leads to difficulty in computing their optical properties using quantum chemistry methods. Facing these challenges, this dissertation will focus on employing different computational approaches to investigate the physical and chemistry properties of noble metal nanoclusters. The two central techniques in this dissertation will be DFT (density functional theory) and TDDFT (time-dependent density functional theory), which are two quantum chemistry approaches that have been widely used for ground-state and excited-state calculations, respectively. In addition, we also applied other techniques such as molecular dynamics (both Born-Oppenheimer molecular dynamics and non-adiabatic molecular dynamics) and a machine learning model to assist with the study of noble metal nanoclusters. In the first part of the dissertation, our goal is to employ a reasonable model system to examine the plasmon-induced catalytic process involving nanocluster/nanowire systems for small molecule activation/dissociation. In chapter 3, we examine the dissociation of a nitrogen molecule on atomically-thin silver nanowires with different lengths and on a Ag₁₉ nanorod. We also applied different field strengths to the system and examine field strength effects on the N₂ dissociation properties. In chapter 4, we study how a transition metal dopant affects the system's electronic structure and its plasmon-enhanced N₂ dissociation properties. In chapter 5, to study the effects of different electric fields, we apply both static electric fields and continuous wave fields to gold/silver triangle nanoclusters@H₂ systems and examine the system evolution as various external fields are applied. For these three projects, linear-response TDDFT provides the absorption spectra information, real-time TDDFT calculations show the electronic transitions and population of electrons, and Ehrenfest dynamics provides information about the motion of nuclei during the dynamics process. In addition, it is also possible to develop new computational methods with lower computational costs in order to accelerate the calculations. In the second part of the dissertation, we develop new methods to examine the properties of large metal nanoclusters. In chapter 6, we develop a new method called TDDFT-aas, which is an approximate method to TDDFT. In this method, instead of calculating the exact two electron integrals in the K coupling matrix when solving the Casida equations, we approximate the integrals in the K coupling matrix, which therefore reduces the computational cost. In chapter 7, we apply the DFT method to examine the gold thiolate-protected nanoparticle growth mechanism in a diglyme solvent environment. DFT approaches can be successfully applied to calculate relatively small metal nanocluster systems, but larger systems are difficult to study, including the dimers of interest in this work. Therefore, in chapter 8, we develop a type of machine learning force field that shows success in predicting energies for gold thiolate-protected nanoclusters not only in the training database but also outside the database. Overall, in this dissertation, we both apply theoretical approaches and develop new computational methods to study different aspects of noble metal nanoclusters.en-US© the author. This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).http://rightsstatements.org/vocab/InC/1.0/Metal nanoclusterQuantum chemistryMachine learningDissertation