In this thesis, we first investigate the physical properties of some metal atoms, molecules and their clusters. We then study the interaction of these with silicon and graphene surfaces. The adsorption of NH3 and H2S molecules on Au3Pt3 is also studied. We calculate the equilibrium atomic structures of metal clusters using density functional theory (DFT) up to eight atoms. The electronic structures of these free and adsorbed clusters are also calculated in detail. We find that the adsorption generally modifies the structure of the Au3Pt3 cluster and the adsorbate (NH3 and H2S ). We also study the site-dependent shapes of the Au8 cluster, associated adsorption energies, band structures and the corresponding charge distribution for the S i(100) asymmetric surface. We show that the electronic properties of the cluster and the substrate complex change with the location of the cluster on the surface. We study the AunPtn clusters on graphene surface. We observe that graphene can be metallic or semiconducting depending on the number of Au and Pt atoms in the cluster and the charge transfer between the cluster and the graphene. We have studied bismuth both as an adsorbate and substitutional dopant in graphene. We have shown that bismuth causes a weak p-type doping for the adsorption case within generalized gradient approximation (GGA), but it n-dopes graphene when it is substitutional and for the adsorption case within local density approximation (LDA). Our results are in agreement with recent angle-resolved photoemission results for the weak adsorption.

One of the fundamental properties of bulk metals is the cancellation of electric fields. The free charges inside of a metal will move until they find an arrangement where the internal electric field is zero. This implies that the electric dipole moment of a metal particle should be exactly zero, because an electric dipole moment requires a net separation of charge and thus a nonzero internal electric field. This thesis is an experimental study to see if this property continues to hold for tiny sub- nanometer metal particles called clusters (2 - 200 atom, R < 1 nm). We have measured the electric dipole moments of metal clusters made from 15 pure elements using a molecular beam electric deflection technique. We find that the observed dipole moments vary a great deal across the periodic table. Alkali metals have zero dipole moments, while transition metals and lanthanides all have dipole moments which are highly size dependent. In most cases, the measured dipole moments are independent of temperature (T = 20 - 50 K), and when there is a strong temperature dependence this suggests that there is a new state of matter present. Our interpretation of these results are that those clusters which have a non- zero dipole moment are non-metallic, in the sense that their electrons must be localized and prevented from moving to screen the internal field associated with a permanent dipole moment. This interpretation gives insight to several related phenomena and applications. We briefly discuss an example cluster system RhN where the measured electric dipole moments appear to be correlated with a the N2O reactivity. Finally, we discuss a series of magnetic deflection experiments on lanthanide clusters (Pr, Ho, Tb, and Tm). The magnetic response of these clusters is very complex and highly sensitive to size and temperature. We find that PrN (which is non-magnetic in the bulk) becomes magnetic in clusters and TmN clusters have magnetic moments lower than the atomic value as well as the bulk saturation value implying that the magnetic order in the cluster involves non-collinear or antiferromagnetic order. HoN and TbN show very similar size dependent trends suggesting that these clusters have similar structures.

De nombreuses applications de calcul haute performance requièrent une part importante de ressources. L'utilisation pour leur exécution de support de type cluster ou grille est alors intéressante. Cette thèse vise l'étude et la mise en place d'applications parallèles et séquentielles sur des grappes d'ordinateurs en assurant une gestion fine des ressources afin de définir une qualité de service prédéfinie. Elle aborde la modélisation du support d'exécution, l'expression des besoins des applications et enfin le placement. Deux approches sont proposées : un mode déterministe où tout ce qui s'exécute sur la machine est supposé contrôlé et un mode stochastique où une part de la charge des machines est inconnue. Toute application appartient à une des quatre classes de services définies ce qui détermine le niveau de qualité de service demandé. Des algorithmes de placement sont proposés sur un support de type cluster avec garantie de la qualité de service. Ils sont basés sur des simulations événementielles et des équations différentielles déduites de modèles markoviens. Une extension de l'algorithme au niveau grille est présentée. Les applications sont modélisées par un graphe de tâches communicantes. Les algorithmes ont été validés par simulation et sont intégrés dans un gestionnaire de ressources (AROMA scAlable Resources Manager and wAtcher). Un modèle économique simple est proposé permettant la facturation des clients.

Reproduction de : Thèse de doctorat : Physique de la matière condensée : Lyon 1 : 2007. / Titre provenant de l'écran titre. 126 réf. bibliogr.

Thesis (M.S.)--Brigham Young University. School of Technology, 2003. / Includes bibliographical references (p. 93-96).

This thesis investigates the synthesis and activation of highly monodisperse Au25(SR)18 - clusters and bimetallic clusters (AuAg and AuPd) protected with various stabilizers for reduction and hydrogenation catalytic reactions. The first chapter is the introduction chapter, which summarizes the literature involving monolayer protected Au clusters, atomically precise Au clusters, bimetallic clusters, X-ray absorption spectroscopy, research objectives, and organization and scope. The second chapter describes the synthesis of Au25(SR)18 - clusters protected with various thiolate stabilizers for nitrophenol reduction catalysis using NaBH4 as a reducing agent. This chapter also describes the stability of these clusters under reaction conditions using UV-Vis spectroscopy and MALDI mass spectrometry. The third chapter details the synthesis of carboxylic acid-protected Au25 clusters using a NaBH4 purification strategy. Here, the knowledge obtained in the second chapter regarding the exceptional stability of Au25(SR)18 - clusters in the presence of NaBH4 was used to isolate carboxylic acid protected Au25 clusters from a polydisperse reaction mixture. The fourth chapter describes the synthesis and activation of mesoporous carbon supported Au25(SR)18 - clusters for nitrophenol reduction catalysis. Here, thermal removal of thiolate stabilizers led to the enhancement in the catalytic activity at low calcination temperatures; however, at higher calcination temperatures activity dropped as particle sintering was observed. Activation of these clusters on mesoporous carbon support was followed by TEM and X-ray absorption spectroscopy. The fifth chapter describes the thermal and chemical removal of thiolate stabilizers from supported Au25(SC8H9)18 - clusters. Here, the removal of thiolate stabilizers and subsequent growth of Au25 clusters was followed by TEM and EXAFS spectroscopy. The sixth and seventh chapters describe the synthesis of AuPd and AuAg bimetallic clusters using Au25(SR)18 - clusters as precursors and their characterization using UV-Vis spectroscopy, transmission electron microscopy, and X-ray absorption spectroscopy. Here, AuPd bimetallic clusters were thermally and chemically treated, which resulted in the formation of AuPd bimetallic nanoparticles with segregated Pd atoms on the surface. AuPd bimetallic nanoparticles were used for the selective hydrogenation catalysis of allyl alcohol. The last chapter of this thesis includes final conclusions and possible avenues for future work.

Galaxy clusters are the most massive bound systems known in the Universe and are believed to have formed through large scale structure formation. They host relativistic cosmic-ray (CR) populations and are gravitationally bound by large amounts of Dark Matter (DM), both providing conditions in which high-energy gamma rays may be produced either via CR interactions with the intracluster medium or through the annihilation or decay of DM particles. Prior to the launch of the Fermi satellite, predictions were optimistic that these sources would be established as γ-ray-bright objects by observations through its prime instrument, the Large Area Telescope (LAT). Yet, despite numerous efforts, even a single firm cluster detection is still pending. This thesis presents a number of studies based on data taken by the LAT over its now seven year mission aiming to discover these γ rays. Using a joint likelihood technique, we study the γ-ray spectra of a sample of nearby clusters searching for a CR-induced signal due to hadronic interactions in the intracluster medium. While we find excesses in some individual targets, we attribute none to the cluster. Hence, we constrain the maximum injection efficiency of hadrons being accelerated in structure formation shocks and the fraction of CR-to-thermal pressure. We also perform a refined search targeting the Coma cluster specifically due to its large variety of existing observations in other wavebands. In the latter case we find weak indications of an excess which however falls below the detection threshold. Because the cluster emission we consider is inherently extended, we need to take into account the imperfect modeling of the foreground emission, which may be particularly difficult such as is the case with the Virgo cluster. Here, we assess the systematics associated with the foreground uncertainties and derive limits based on an improved background model of the region. For the first time we derive limits on the γ-ray flux from CR and DM-interactions in which we take into account the dynamical state of the system. For DM we also include the contribution from substructure. The DM domain is further explored by searching for line-like features as they arise from the annihilation of DM into two photons in a large sample of clusters, including Virgo and Coma. Finding no evidence for γ-ray lines, we derive limits on the DM annihilation cross section that are roughly a factor 10 (100) above that derived from observations of the galactic center assuming an optimistic (conservative) scenario regarding the boost due to DM substructure. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 3: Submitted. Paper 4: Submitted.</p>

The power of parallel supercomputing resources has progressed to the point where first-principles calculations involving systems up to 10³ atoms are feasible, allowing ab initio exploration of increasingly complex systems such as amorphous networks, nanostructures, and large defect clusters. Expansion of our fundamental understanding of modified Si-based materials is paramount, as these materials will likely flourish in the foreseeable cost-driven future in diverse micro- and nanotechnologies. Here, density-functional theory calculations within the generalized gradient approximation are applied to refine configurations of Si-based materials generated from Metropolis Monte Carlo simulations and study their resultant structural properties. Particular emphasis is given to the contributions of strain and disorder on the mechanical, optical, and electronic properties of modified Si-based materials in which aspects of compositional variation, phase, strain scheme, morphology, native defect incorporation, and quantum confinement are considered. The simulation strategies discussed are easily extendable to other semiconductor systems. / text

Stellar bars are the most important internal drivers of the evolution of disk galaxies because they efficiently redistribute mass and angular momentum in the baryonic and dark matter components of galaxies. Mounting evidence suggests that mechanisms other than major mergers of galaxies, such as minor mergers, gas accretion, and bar-driven secular processes, play an important role in galaxy evolution since a redshift z~2. In order to characterize the evolution of barred disks, this thesis presents one of the most comprehensive studies of barred galaxies in the field at low redshifts, and also as a function of environment across galaxy clusters of different densities. This work improves significantly on earlier studies by using quantitative methods to characterize bars, analyzing high-quality data from some of the largest disk galaxy samples to date, and using results across a range of Hubble types and environments to test different theoretical models for the evolution of disk galaxies. Our main results are summarized below: (1) Studies done as a part of this thesis have quantitatively shown for the first time that the optical bar fraction in z~ 0 field galaxies is a sensitive and non-monotonic function of host galaxy properties, such as the luminosity, stellar mass, and bulge-to-disk ratio. We find that at z~0, the bar fraction increases significantly from galaxies of intermediate mass and Hubble types (Sb) toward those of lower mass and late Hubble types (Sd-Sm). The behavior from intermediate to early Hubble types is more uncertain. These results, which have been subsequently confirmed by independent studies, set constraints for theoretical models and in particular underline the importance for bar growth of angular momentum exchange between the bar, disk, bulge, and dark matter halo, as well as the possible triggering of bars by external satellites and interactions with the dark matter. Furthermore, our results at optical and near-infrared wavelengths on the fraction and sizes of bars at z~0 provide the zero-redshift anchor point for studies of bars at higher redshifts with current and future space missions (e.g., ACS, WFC3, JWST), and allow us to assess the systematic effects in such studies. (2) Although cluster environments are unique laboratories for investigating the evolution of barred disks, only sparse and disparate results have emerged from early studies. In this thesis, we study barred disks in clusters using high-quality data from the Hubble Space Telescope Advanced Camera for Surveys for the moderately-rich cluster Abell 901/902 (characterized by a galaxy number density n~1,000 gal Mpc⁻³) at z~0.165, and of the Coma cluster at z~0.02, the densest cluster (n~10,000 gal Mpc⁻³) in the nearby Universe. We find that the optical bar fraction for bright, early Hubble type disk galaxies does not show a statistically significant variation (within the error bars of ± 10 to 12%) as a function of galaxy environment within the Abell 901/902 cluster, as well as between the Abell 901/902 cluster and the field. Similarly, the optical bar fraction for bright S0 galaxies shows no statistically significant variation (within the error bars of ±10%) between the Virgo, Abell 901/902, and core of the Coma clusters, even though these environments span over an order of magnitude in galaxy number density (n~300 to 10,000 gal Mpc⁻³). We suggest that the S0 bar fraction is not greatly enhanced in denser environments, such as the core of Coma, due to the predominance of high speed encounters over slow ones, the tidal heating of S0 disks, and the low gas content of S0s in rich clusters. / text

Traditional soil classification methods invoke physical differences based on particle size to group soils into textural classes. Resulting groupings are used to make predictions about soil attributes and processes of interest including hydrologic response. My hypothesis is that more useful classification schemes will be created by starting with response and applying an inverse approach to generate soil groupings. I propose an alternative classification scheme based on these hypotheses, using techniques of cluster analysis. The resulting system has high predictive capacity with simplicity comparable to the U.S. Dept. of Agriculture soil textural triangle or other similar classification diagrams. I conclude that: classification is most appropriate when carried out on process and objective specific bases; there is a physical meaning to cluster-based groupings, which allows for more appropriate segregation of response as compared to textural groupings; using clusters, a small number of samples can be used to characterize the range of response.