Since 1990, D. M. Eigler and E. K. Schweizer manipulated the xenon atoms on the Ni(110) surface to pattern the atomic scale structure. The researches about manipulation of atoms/ molecules become a major research in surface science. One of the ultimate goals of nanotechnology is manipulating or modifying the nanoclusters. In our experiment, first, deposit metal atoms on the Al2O3/NiAl(100) surface. These Co atoms will form a cluster structure because of the surface free energy. Second, use the STM manipulation technique to manipulate the clusters, and observe the differences by STM images. Closer the tip to Co clusters by reducing the bias voltage of STM. This action will induce a force to overcome the interaction between Co clusters and Al2O3 to remove the Co clusters. In the analysis of electronic spectroscopy, we can say that some of the clusters are not the Co clusters, those are formed by oxygen atom or other molecules by the difference of electronic spectroscopy curves. And we observe that the relation between Co cluster size and tunneling current gap is a linear relation. In the analysis of electronic spectroscopy, we can distinguish the cluster from the oxide surface by the difference of electronic spectroscopy curves. And we observe that the metal cluster size will be influenced the width of tunneling gap.

Thesis advisor: John T. Fourkas / One can extract a tremendous amount of information about the organizational and dynamic states of molecules, in situ and in real-time, through highly sensitive and noninvasive single particle optical probing. The highly efficient, multi-photon excited luminescence from stabilized metal nanoclusters renders these species useful as optical probes that can be used in detecting single particle and molecular dynamics. We generate stable, and monodisperse samples of Ag nanoclusters as small as 1 nm in diameter, and find that through substitution of various stabilizer molecules we can precisely tune the size of the clusters over a 1-6 nm range of diameters, ensuring monodispersity and stability at every stage. These clusters also exhibit highly efficient, polarized luminescence upon two photon excitation at 800 nm and remain highly photostable, not exhibiting the deleterious blinking that occurs with many single-molecule fluorophores. In order to demonstrate the utility of these clusters as single-molecule probes, we track their emission polarization over long periods in deeply supercooled liquids such as 4'(octahydro-4,7-methano-5H-inden-5-yliden) bisphenol dimethyl ether (ODE). Our results suggest that these clusters can detect nanoscale dynamics with high sensitivity. / Thesis (BS) — Boston College, 2004. / Submitted to: Boston College. College of Arts and Sciences. / Discipline: Chemistry. / Discipline: College Honors Program.

Il est bien connu que les propriétés optiques des nanoparticules de métaux nobles, en particulier d'or et d'argent, s'écartent fortement de celles de métaux macroscopiques. Pour les tailles comprises entre dix et quelques centaines de nanomètres, elles sont dominées par les plasmons de surface (SP) décrites par des modèles purement classiques. En revanche, les agrégats de quelques dizaines d’atomes se comportent comme des systèmes quantiques, ce qui induit des comportements optiques nouveaux. La structure des nanoparticules et l'environnement diélectrique peuvent affecter les propriétés optiques. Dans cette thèse, j'ai utilisé un microscope électronique à transmission à balayage (STEM) équipé d'un spectromètre à perte d'énergie des électrons (EELS) pour mesurer, en parallèle, les propriétés optiques et structurales de nanoparticules individuelles. Je présente comment des expériences complémentaires (STEM-EELS et absorption optique) sur de petites nanoparticules d'argent triées en taille et encapsulées dans une matrice de silice donnent au premier abord des résultats incohérents: tandis que, d’une part, l'absorption optique ne montre aucun effet de taille entre quelques atomes et environ 10 nm, un décalage en énergie est observé dans les mesures STEM-EELS. Notre interprétation quantitative, fondée sur un modèle mixte classique/quantique qui prend en compte tous les effets quantiques pertinents, a résolu les apparentes contradictions non seulement dans nos données expérimentales, mais également dans celles de la littérature. Notre modèle décrit comment l'environnement local est le paramètre crucial contrôlant la manifestation ou l'absence d'effets de taille quantique. En second lieu, je me suis intéressé à la région purement classique à travers des structureslithographiées de quelques centaines de nanomètres. Bien que les cavités plasmoniques triangulaires aient été largement étudiées dans la littérature, une classification en termes de modes de respiration et de bords plasmoniques manquait. Dans cette étude, les résultats expérimentaux de STEM-EELS, des modèles analytiques et des simulations classiques nous ont permis de décrire la nature des différents modes. / It is well known that the optical properties of nanoparticles of noble metals, in particular gold and silver, deviate strongly from those of macroscopic metals. For sizes between ten and a few hundred nanometers, they are dominated by surface plasmons (SPs) described by purely classical models. On the other hand, clusters of a few tens of atoms behave like quantum systems inducing new optical behaviors. The structure of the nanoparticles and the dielectric environment can affect the optical properties. In this thesis I used a scanning transmission electron microscope (STEM) fitted with an electron energy loss spectrometer (EELS) to measure, in parallel, the optical and structural properties of individual nanoparticles. I present how complementary experiments (STEM-EELS and optical absorption) on sizeselected small silver nanoparticles embedded in silica yield at first inconsistent results: while optical absorption shows no size-effect in the range between only a few atoms and ~10 nm, a clear spectral shift is observed in STEM-EELS technique. Our quantitative interpretation, based on a mixed classical/quantum model which takes into account all the relevant quantum effects, resolves the apparent contradictions, not only within our experimental data, but also in the literature. Our comprehensive model describes how the local environment is the crucial parameter controlling the manifestation or absence of quantum size effects. Secondly, I was interested in the purely classical region through lithographed structures of a few hundred nanometers. Although triangular plasmonic cavities have been widely studied in the literature, a classification in terms of plasmonic modes of breathing and edge was missing. In this study, experimental STEM-EELS results, analytical models and classical simulations enabled us to describe the nature of the different modes.

Yes / The fabrication of osmium nanoclusters from single atoms has been studied in real-time on B-doped and B-free graphitic surfaces. The dynamics of nucleation on both surfaces are identified, captured, and reported. The nucleation is ca. 2× faster on B-doped surface compared to the B-free surface (38 pm min−1versus 18 pm min−1), suggesting osmium–boron interactions within the nanomaterials.

The graphene moiré superstructure formed on Ru(0001) (g/Ru(0001)) has shown the potential as a template to self-assemble super-lattices of metal nanoparticles as model catalysts. To explore the possibility of rational catalyst design on g/Ru(0001), detailed density functional theory (DFT) calculations have been performed to investigate the adsorption and diffusion of Rh and Au adatoms on g/Ru(0001). The consequences of different hopping rates for cluster nucleation have been explored by performing Monte Carlo-based statistical analysis, which suggests that diffusing species other than adatoms need to be taken into account to develop an accurate description of cluster nucleation and growth on this surface. DFT calculations have also been carried out to investigate the adsorption and diffusion of 18 4d (Y-Ag) and 5d (La-Au) transition metal adatoms on g/Ru(0001). Given the necessity to study larger diffusing species than adatoms, DFT calculations have been performed to study the adsorption and diffusion of Rh and Au dimers and trimers on g/Ru(0001). It was shown that the mobility of Rh clusters decreases with the increase of cluster size, while for Au, dimers diffuse faster than monomers and trimers on the moiré surface. We then used a genetic algorithm combined with DFT calculations to predict the lowest energy structure of a Au8 cluster on g/Ru(0001). Our prediction leads us to propose that Au clusters aggregates through Oswald ripening with Au dimer being the major diffusing species. Finally, we examined the morphology of a Cu19 cluster on g/Cu(111) using MD simulations with COMB3 potential. We also studied the mobility of Cu clusters on g/Cu(111) at elevated temperatures. The analysis suggests that g/Cu(111) may not be a suitable substrate for the formation and growth of isolated Cu clusters. All these calculation results have provided us a better understanding and useful insights into the nucleation and growth mechanism of metal clusters on graphene moiré.

This dissertation focuses on computational study of the geometry and energetics small molecules and nanoclusters involving transition metals (TM). These clusters may be used for various industrial applications including catalysis and photonics. Specifically, in this work we have studied hydrides and carbides of 3d-transition metal systems (Sc through Cu), small nickel and gold clusters. Qualitatively correct description of the bond dissociation is ensured by allowing the spatial and spin symmetry to break. We have tested applicability of new exchange-correlation functional and alternative theoretical descriptions (spin-contamination correction in broken symmetry DFT and ensemble Kohn-Sham (EKS)) as well. We studies TM hydrides and carbides systems to understand the importance of underlying phenomenon of bond breaking in catalytic processes. We have tested several exchange-correlation functionals including explicit dependence on kinetic energy density for the description of hydrides (both neutral and cationic) and carbides formed by 3d-transition metals. We find M05-2x and BMK dissociation energies are in better agreement with experiment (where available) than those obtained with high level wavefunction theory methods, published previously. This agreement with experiment deteriorates quickly for other functionals when the fraction of the Hartree-Fock exchange in DFT functional is decreased. Higher fraction of HF exchange is also essential in EKS formalism, but it does not help when spin-adapted unrestricted approach is employed. We analyze the electron spin densities using Natural Bond Orbital population analysis and find that simple description of 3d electrons as non-bonding in character is rarely correct. Unrestricted formalism results in appreciable spin-contamination for some of the systems at equilibrium, which motivated us to investigate it further in details. In order to correct the spin contamination effect on the energies, we propose a new scheme to correct for spin contamination arising in broken-symmetry DFT approach. Unlike conventional schemes, our spin correction is introduced for each spin-polarized electron pair individually and therefore is expected to yield more accurate energy values. We derive an expression to extract the energy of the pure singlet state from the energy of the broken-symmetry DFT description of the low spin state and the energies of the high spin states (pentuplet and two spin-contaminated triplets in the case of two spin-polarized electron pairs). We validate our spin-contamination correction approach by a simple example of H2 and applied to more complex MnH system. Ensemble KS formalism is also applied to investigate the dissociation of C2 molecule. We find that high fraction of HF exchange is essential to reproduce the results of EKS treatment with exact exchange-correlation functional. We analyze the geometry and energetics of small nickel clusters (Ni2-Ni5) for several lowest energy isomers. We also study all possible spin states of small nickel cluster isomers and report observed trends in energetics. Finally we determine the geometry and energetics of ten lowest energy isomers of four small gold clusters (Au2, Au4, Au6, and Au8). We have also investigated the influence of cluster geometry, ligation, solvation and relativistic effects on electronic structure of these gold clusters. The effect of one-by-one ligand attachment in vacuum and solvent environment is also studied. Performance of five DFT functionals are tested as well; Local Spin Density Approximation (SVWN5), Generalized Gradient Approximation (PBE), kinetic energy density-dependent functional (TPSS), hybrid DFT (B3LYP), and CAM-B3LYP which accounts for long-range exchange effects believed to be important in the analysis of metal bonding in gold complexes and clusters. Our results exhibit the ligand induced stability enhancement of otherwise less stable isomers of Au4, Au6 and Au8. Ligands are found to play a crucial role in determining the 2D to 3D transition realized in small gold clusters. In order to select an appropriate theory level to use in this study, we investigate the effect of attachment of four different ligands (NH3, NMe3, PH3, PMe3) on cluster geometry and energetics of Au2 and Au4 in vacuum and in solution. Our results benchmark the applicability of DFT functional model and polarization functions in the basis set for calculations of ligated gold cluster systems. We employ five different basis sets with increasing amount of polarization and diffuse functions; LANL2DZ, LANL2DZ-P, def2-SVP, def2-TZVP, and def2-QZVP. We obtain NMe3 = NH3 > PH3 > PMe3 order of ligand binding energies and observe shallow potential energy surfaces in all molecules. Our results suggest appropriate quantum-chemical methodologies to model small noble metal clusters in realistic ligand environment to provide reliable theoretical analysis in order to complement experiments.