Thesis (M.Ed.)--University of Hong Kong, 1988. / Includes bibliographical references (leaf 83-86). Also available in print.

Cette thèse présente une étude expérimentale de la transition de mouillage de l'hélium liquide sur la surface du césium métallique et une détermination de la ligne de prémouillage de ce liquide sur la surface du rubidium. On rappelle tout d'abord le critère établi par Cheng et al. pour le mouillage à température nulle d'une surface par l'hélium liquide, ainsi que son extension à température non-nulle pour déterminer la température de mouillage et la ligne de prémouillage. Une amélioration de ce modèle est présentée, qui prend en compte la variation avec la température de différents paramètres. Les méthodes expérimentales utilisées sont originales sur deux points: le mouillage d'une surface est détecté par une mesure de conductivité thermique, tirant parti du lien existant entre la superfluidité du film d'hélium à basse température et son épaisseur; Les surfaces étudiées sont celles du métal massif conservées dans des cellules scellées. La transition de mouillage a été observée sur le césium. Elle a été trouvée continue et hystérétique. La mise au point d'une technique de désaturation des films en cellule scellée a permis d'observer la transition de prémouillage à différentes températures sur une même surface de rubidium. La ligne de prémouillage a été déterminée expérimentalement entre 1 et 1.7 kelvins. Elle suggère que la surface du rubidium est faiblement mouillée à température nulle. Enfin, il a été montré que l'oxydation limitée d'une surface de rubidium la rendait moins mouillable par l'hélium. Pour la première fois, nous avons observé une surface de rubidium sèche à saturation, ainsi que la transition de démouillage sur cette surface.

Le pompage optique laser des mélanges gazeux d'hélium peut fournir, dans un champ magnétique faible, de fortes valeurs de polarisation nucléaire. Une technique thermique efficace qui permet la circulation des atomes d'hélium 3 dans la cellule expérimentale rend possible le transfert, en régime stationnaire, de la polarisation nucléaire du gaz à un échantillon liquide à basse température. La relaxation sur les parois de la cellule est efficacement réduite par l'utilisation de césium métallique. Dans de tels échantillons, fortement aimantés et anisotropes, c'est le champ dipolaire qui contrôle l'évolution de l'aimantation. Des modes amortis d'aimantation sont observés par une technique de RMN pulsée. L'analyse détaillée de leur fréquence et de leur taux d'amortissement donne des informations sur la densité d'aimantation et le coefficient de diffusion de spin dans des mélanges polarisés. Les expériences relatées ici sont réalisées à des températures supérieures à 0.2K dans des mélanges dont la concentration en hélium 3 est au moins de l'ordre de quelques pour cents. Quand survient la séparation de phase, la phase riche en hélium 3 renferme une forte densité d'aimantation. Des résultats préliminaires concernant l'observation de la séparation de phase et la mesure du potentiel chimique de l'hélium 4 sont également donnés dans cette thèse.

Cette thèse présente une étude expérimentale de la dynamique de mouillage d'un film d'4He liquide superfluide sur une surface de césium, lors de la traversée en température de la transition de prémouillage. Pour imager le film d'hélium, nous avons adapté aux basses températures une méthode de contraste interférentiel différentiel, qui donne accès selon une direction au gradient de couverture locale d'hélium. Les couches de césium ont été fabriquées par évaporation in situ à froid. Elles montrent des angles de contact de l'hélium équivalents à ceux des autres groupes utilisant le même procédé de fabrication. L'épaisseur de films d'hélium a été mesuré en fonction de la désaturation sur ces surfaces et est compatible avec les valeurs attendues. Suivant les surfaces de césium, deux types de comportement du film durant la transition de prémouillage ont été observés. Sur un premier type de surface, l'envahissement par le film s'effectue en passant par un état intermédiaire qui apparaît comme un mélange de surface sèche et mouillée à une échelle inférieure à la résolution. Nous n'avons eu accès alors qu'à la proportion de surface sèche à l'avant du film, moyennée sur la tache de diffraction. L'évolution de la couverture moyenne avec la température peut s'interpréter dans le cadre du modèle d'Ising à champ aléatoire à température nulle sans nucléation, avec un paramètre de désordre important. Sur un deuxième type de surface, les longueurs de corrélation du film sont bien plus importantes et l'invasion de la surface par le film s'effectue par une avancée observable du bord du film. Celle-ci se fait à température constante, le film progressant par avalanches successives activées thermiquement. La vitesse d'avancée du front du film a été mesuré en fonction de la température et de l'épaisseur du film.

Nanostructure systems possessing certain desirable features can arise from the self-organization of fundamental building blocks. In this thesis we explore two types of controlled self-assembly mechanisms in hetero-epitaxy: (a) classical assembly of atom vacancies into quasi one-dimensional line structures and (b) quantum-driven assembly of atoms into atomically-smooth two-dimensional thin films. In the classical assembly phenomenon, adatom vacancies, created via elastic strain-relaxation in compressively strained atom chains on a silicon substrate, self-organize into meandering vacancy lines. The average spacing between these line defects can be varied by adjusting the chemical potential μ of the adsorbed atoms. We implemented a lattice model that quantitatively connects density functional theory calculations for perfectly ordered structures to the fluctuating disorder seen in experiment and the experimental control parameter μ. The quantum-mechanical thin-film assembly explored in this thesis has an electronic origin. It is made possible by strong quantum size effects at the nanoscale and can be controlled experimentally by tuning the quantum mechanical boundary conditions and free carrier density of an ultrathin metal film. This is accomplished via atomic-scale template modification and chemical doping, respectively. Our investigations focused on the formation and structure-property relationship of these engineered quantum films, and specifically on the emergence of collective phenomena such as superconductivity and plasmon excitations. We succeeded in growing atomically-smooth Pb1-xGax (x = 0.06) alloy films on a Si(111)-7 × 7 substrate through quantum confinement, a remarkable observation because Pb and Ga are totally immiscible in the bulk. The resulting films exhibit large uniform-depth holes which turn out to be responsible for the exceptionally large critical current density in these films. Remarkably, the critical current density increases with temperature up to 3.25 K, a phenomenon that has not been seen before and that can be attributed to the unusual quantum-growth morphology of this material. The alloying experiments furthermore elucidate the likely origin of the Tc suppression generally observed in thin films. Finally, we demonstrate the existence of quantized plasmon modes in ultrathin metal films. Controlled self-organization experiments thus enable stabilization of novel nanophase materials, which in turn leads to discovery and understanding of novel collective properties.

There is by now compelling evidence that most of the matter in the Universe is in the form of dark matter, a form of matter quite different from the matter we experience in every day life. The gravitational effects of this dark matter have been observed in many different ways but its true nature is still unknown. In most models, dark matter particles can annihilate with each other into standard model particles; the direct or indirect observation of such annihilation products could give important clues for the dark matter puzzle. For signals from dark matter annihilations to be detectable, typically high dark matter densities are required. Massive objects, such as stars, can increase the local dark matter density both via scattering off nucleons and by pulling in dark matter gravitationally as a star forms. Annihilations within this kind of dark matter population gravitationally bound to a star, like the Sun, give rise to a gamma ray flux. For a star which has a planetary system, dark matter can become gravitationally bound also through gravitational interactions with the planets. The interplay between the different dark matter populations in the solar system is analyzed, shedding new light on dark matter annihilations inside celestial bodies and improving the predicted experimental reach. Dark matter annihilations inside a star would also deposit energy in the star which, if abundant enough, could alter the stellar evolution. This is investigated for the very first stars in the Universe. Finally, there is a possibility for abundant small scale dark matter overdensities to have formed in the early Universe. Prospects of detecting gamma rays from such minihalos, which have survived until the present day, are discussed. / Kosmologiska observationer har visat att större delen av materian i universum består av mörk materia, en form av materia med helt andra egenskaper än den vi upplever i vardagslivet. Effekterna av denna mörka materia har observerats gravitationellt på många olika sätt men vad den egentligen består av är fortfarande okänt. I de flesta modeller kan mörk materia-partiklar annihilera med varandra till standardmodellpartiklar. Att direkt eller indirekt observera sådana annihilationsprodukter kan ge viktiga ledtrådar om vad den mörka materian består av. För att kunna detektera sådana signaler fordras typiskt höga densiteter av mörk materia. Stjärnor kan lokalt öka densiteten av mörk materia, både via spridning mot atomkärnor i stjärnan och genom den ökande gravitationskraften i samband med att en stjärna föds. Annihilationer inom en sådan mörk materia-population gravitationellt bunden till en stjärna, till exempel solen, ger upphov till ett flöde av gammastrålning, som beräknas. För en stjärna som har ett planetsystem kan mörk materia även bli infångad genom gravitationell växelverkan med planeterna. Samspelet mellan de två mörk materia-populationerna i solsystemet analyseras, vilket ger nya insikter om mörk materia-annihilationer inuti himlakroppar och förbättrar de experimentella möjligheterna att detektera dem. Mörk materia-annihilationer inuti en stjärna utgör också en extra energikälla för stjärnan, vilket kan påverka stjärnans utveckling om mörk materia-densiteten blir tillräckligt stor. Denna effekt undersöks för de allra första stjärnorna i universum. Slutligen finns det också en möjlighet att det i det tidiga universum skapades mörk materia-ansamlingar som fortfarande finns kvar idag. Utsikterna att upptäcka dessa genom mätning av gammastrålning diskuteras. / QC 20120130

Phase transitions in a number of models related to hightemperature superconductors are investigated, using scalingmethods and Monte Carlo simulations. This thesis considers twomain topics. The first topic is phase transitions, phase diagrams, andvortex motion in high temperature superconductors at finitetemperature, subject to magnetic fields and disorder. We studya vortex glass model at finite temperature, with stronguncorrelated vortex pinning and a magnetic field. We find thatthe vortex glass exists at finite temperature and calculate thecritical exponents of the transition. We also investigate hightemperature superconductors with columnar disorder in zero andapplied magnetic fields. Some of these studies are alsorelevant for the superfluid to Mott insulator transition ofbosons in two dimensions. We find that the unscreened Boseglass transition belongs to a new universality class. Wecalculate the critical exponents of the superconductingtransition with columnar defects in zero applied magneticfield. The transverse Meissner transition is studied, and wefind an exotic universality class with a correlation volumethat is infinitely anisotropic in all directions. The second topic is confinement-deconfinement transitions incompact Abelian Higgs models. We develop a new order parameter,related to a large Wilson loop for fractionalized charges, anduse it to study the concept of topological order. Thesetransitions may be relevant for strongly correlated electronsin two dimensions.

Observations from a broad range of astrophysical scales have forced us to the realization that the well understood matter comprising the stars and galaxies we see around us accounts for only a small fraction of the total mass of the Universe. An amount roughly five times larger exist in the form of dark matter about which we have virtually no direct evidence apart from its large scale gravitational effects. It is also known that the largest contribution to the energy density of the universe is the dark energy, a negative pressure form of energy which will not be dealt with here. I will present a candidate for the dark matter which is based completely in known physics and which presents several possible observational signatures. In this model the dark matter is composed of dense nuggets of baryonic matter and antimatter in a colour superconducting state. If these object are sufficiently massive their low number density will make them effectively dark in the sense that collisions with visible matter become infrequent. This work presents the basics of dark matter as a colour superconductor and then uses the physical properties of the quark nuggets to extract observational consequences.

The growth of vertically aligned arrays of single-walled carbon nanotubes (SWNT) has provided an efficient route toward production of highly aligned SWNT that can be grown ultra-long. Here, the growth of such aligned SWNTs is demonstrated from a thin e-beam deposited catalyst layer composed of Fe-Mo with ratio (5:1) respectively, supported by Al2O3. Using C2H2 decomposition in a hot filament chemical vapor deposition apparatus, experiments indicate that the order of deposition of the respective Fe and Mo catalyst components significantly affects growth characteristics, especially evident during growth at elevated reaction pressures under high carbon flux. The role of temperature and pressure on features of the nanotube arrays, such as height, alignment, quality, volumetric density, and diameter distribution are compared for each case of Fe/Mo and Mo/Fe considered. In order to better understand this effect, atomistic modeling using the Bozzolo-Ferrante-Smith (BFS) method for alloys is employed along with the Monte Carlo-Metropolis method. The dependence of the growth on the order of co-catalyst deposition is observed to be a structural effect that can be explained in a straight-forward interpretation of the BFS strain and chemical energy contributions toward the formation of Fe-Mo catalyst prior to growth. The competition between the formation of metastable inner Mo cores and clusters of surface-segregated Mo atoms in Fe-Mo catalyst particles influences catalyst formation, and the role of Mo concentration and catalyst particle size is found to also be a factor influencing this formation process. Finally, this modeling procedure is demonstrated as a general technique that can be employed to study the structural stability and formation of binary catalyst particles in a reducing environment-providing a potentially cheap and efficient method for catalyst design.

Single-walled carbon nanotubes (SWNTs) are fascinating materials to study one-dimensional photophysics. Their optical properties are strongly affected by strong Coulomb interactions and are determined by "excitons" which represent the quantum of polarization in non-metallic solids. In this thesis dissertation we have experimentally investigated both the structure and the dynamics of excitons in non-metallic SWNTs. In particular, we have performed micro-photoluminescence spectroscopy of individual semiconducting SWNTs at low temperature to study their intrinsic optical properties and investigate the excitonic fine structure. Using magnetic field parallel to the tube axis we were able to directly observe theoretically predicted dark states for the first time in SWNTs. In addition, we found that the inter-valley and exchange energy, which determines the energy separation between the dark and the bright state, to be very sensitive to the surrounding environment of the nanotube. We have also studied the temperature dependent lineshape of SWNT photoluminescence in order to gain insight into the dynamics of exciton-phonon interaction, finding evidence for acoustic phonon scattering. For the rest of this thesis dissertation, we have developed a model based on reaction-diffusion processes to theoretically explain the observation of photoluminescence saturation in SWNTs. Our model shows that efficient exciton-exciton annihilation under high pumping conditions can explain this observed behavior quantitatively.