In this thesis, quantum-mechanical calculations within density-functional theory on metallic systems are presented. The overarching goal has been to investigate effects of disorder. In particular, one of the properties investigated is the bindingenergy shifts for core electrons in binary alloys using different theoretical methods. These methods are compared with each other and with experimental results. One such method, the so-called Slater-Janak transition state method relies on the assumption that the single-particle eigenvalues within density-functional theory are linear functions of their respective occupation number. This assumption is investigated and it is found that while the eigenvalues to a first approximation show linear behavior, there are also nonlinearities which can influence the core-level binding energy shifts. Another area of investigation has been iron based alloys at pressures corresponding to those in the Earth’s inner core. This has been done for the hexagonal close packed and face entered cubic structures. The effects of alloying iron with magnesium and nickel on the equation of state as well on the elastic properties have been investigated. The calculations have shown that the hexagonal close packed structure in FeNi is more isotropic than the face-centered cubic structure, and that adding Mg to Fe has a large impact on the elastic properties. Finally, the effects of disorder due to thermal motion of the atoms have been investigated through ab-initio molecular dynamics simulations. Within the limits of this method and the setup, it is found that the face-centered cubic structure of molybdenum can be dynamically stabilized at high temperature, leading to a metastable structure, on the average. The dynamical stabilization of face-centered cubic molybdenum also rendered it possible to accurately calculate the lattice stability relative to the body-centered cubic phase. Inclusion of temperature effects for the lattice stability using ab-initio molecular dynamics simulations resolves the disagreement between ab-initio calculations and thermochemical methods.

The identity of dark matter is one of the key outstanding problems in both particle and astrophysics. In this thesis, I describe a number of complementary searches for particle dark matter. I discuss how the impact of dark matter on stars can constrain its interaction with nuclei, focussing on main sequence stars close to the Galactic Centre, and on the first stars as seen through the upcoming James Webb Space Telescope. The mass and annihilation cross-section of dark matter particles can be probed with searches for gamma rays produced in astronomical targets. Dwarf galaxies and ultracompact, primordially-produced dark matter minihalos turn out to be especially promising in this respect. I illustrate how the results of these searches can be combined with constraints from accelerators and cosmology to produce a single global fit to all available data. Global fits in supersymmetry turn out to be quite technically demanding, even with the simplest predictive models and the addition of complementary data from a bevy of astronomical and terrestrial experiments; I show how genetic algorithms can help in overcoming these challenges. / At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 5: Accepted. Paper 6: Submitted.

Ce document résume mes activités de recherche depuis 2005 portant sur le développement de méthodes pour calculer la structure électronique de systèmes moléculaires et s'organisant autour de deux thématiques : (1) Méthodes de Monte Carlo quantique (QMC) : j'ai développé plusieurs aspects des méthodes QMC pour pouvoir faire des calculs de référence en chimie quantique (méthodes d'optimisation des fonctions d'onde pour états fondamentaux et excités, nouvelles formes de fonctions d'onde explicitement corrélées et estimateurs statistiques améliorés pour les observables); (2) Théorie de la fonctionnelle de la densité (DFT) : afin d'améliorer la précision de la DFT actuelle, j'ai développé plusieurs méthodes qui combinent rigoureusement la DFT avec des calculs de fonctions d'onde corrélées utilisant une décomposition de l'interaction électron-électron (hybrides à séparation de portée, approximations "doubles hybrides" et hybrides multiconfigurationnels).

Le niveau de conversion des réacteurs CANDU et REP en cycle thorium a été étudié dans l'optique d'une utilisation en troisième et dernière strate de scénarios symbiotiques. Le plutonium du combustible REP usé serait par exemple utilisé en CANDU Th/Pu pour produire de l'233U, qui alimenterait ces réacteurs à eau et haute conversion. En cas d'augmentation importante de la production d'énergie à partir d'uranium, cette alternative basée sur des réacteurs existants pourrait suppléer une IVe génération trop tardive. Pour évaluer la compétitivité de tels scénarios, des calculs de cycles détaillés ont été effectués selon une méthodologie de simulation de coeur développée pour le CANDU-6 et adaptée au REP de type N4. Le CANDU Th/233U enrichi à 1.30 wt% est régénérateur, avec un burnup court de 7 GWj/t. Augmenter légèrement l'enrichissement allonge considérablement le cycle, au prix d'une sous-génération. Multirecycler conduit également à une perte de conversion, qui peut néanmoins être compensée par un chargement fissile hétérogène. La conversion à puissance standard est moins bonne en REP Th/233U qu'en CANDU (inventaire fissile réduit de moitié après 50 GWj/t) mais peut être améliorée par sous-modération. L'analyse neutronique montre que l'essentiel du gap de conversion entre CANDU et REP vient des conditions opératoires économes en neutrons du CANDU. Des scénarios ont été comparés du point de vue de l'économie d'uranium et de l'aval du cycle dans les deux cas, et ont confi rmé l'intérêt du CANDU. Deux pistes de recherche ont été identi fiées : l'évaluation de la sûreté des CANDUs au thorium par cinétique avec contre-réactions thermiques, et l'étude de coeurs fortement sous-modérés en cuve standard de REP.

Surface plasmon polaritons are non radiative modes which exist at the interface between a dielectric and a metal. They can confine light at sub-wavelength scales. However, their propagation is restricted by the intrinsic losses of the metal which imply a rapid absorption of the mode. The aim of this thesis is the study of the coupling of surface plasmons in metallo-dielectric planar structures. Obtaining the properties of the modes implies the extension of the solutions to the complex plane of propagation constants. The method used consists in determining the poles of the scattering matrix by means of Cauchy's integrals. The first solution to solve the problem of propagation of the surface plasmons consists in coupling these modes to one another. In a symmetric medium, when the thickness of the metallic film becomes thin enough, the coupling between the plasmon modes which exist on each side becomes possible. One of the coupled modes which is created, the so-called long range surface plasmon, has a bigger propagation length than the usual plasmon whereas the other coupled mode, named short range surface plasmon, has a smaller propagation length. We present a configuration which allows the excitation of the long range surface plasmon without the short range mode with a metallic layer deposited on a perfect electric conductor substrate. This excitation can be done in air and allows applications, such as the detection and the characterisation of molecules. Then, we present the coupling between dielectric waveguides, and, in particular, the coupled-mode theory in the case of the transverse magnetic polarisation. We consider also the case of PT symmetric structure. The last part of this work presents the demonstration of the strong coupling regime between a surface plasmon and a guided mode. We demonstrate an increase of the propagation length of the hybrid surface plasmon, which still has the confinement of a surface mode. A linear gain is added in the different layers of the structure. When the gain is added in the layer between both coupled modes allows an enhancement of the propagation lengths of the modes, and more precisely of the hybrid surface plasmon mode, which can propagate at the millimeter scale.

Le confinement magnétique dans les tokamaks est à l'heure actuelle la voie la plus avancée pour produire de l'énergie par fusion thermonucléaire. Des études théoriques et expérimentales ont montré que la génération de rotation permet d'en augmenter les performances par la réduction du transport turbulent à l'œuvre dans les plasmas de tokamaks. L'influence de la rotation sur les flux turbulents de chaleur et de particules ainsi que le transport du moment angulaire sont étudiés par simulation numérique dans le cadre du code gyro-cinétique, quasi-linéaire QuaLiKiz. A cette occasion, le code QuaLiKiz est modifié pour prendre en compte la rotation du plasma et calculer le flux de moment angulaire. Il est montré que le cadre de travail de QuaLiKiz permet de calculer le flux de moment angulaire y compris le stress résiduel induit par le cisaillement du champ électrique radial ainsi que l'effet de la rotation sur les flux de chaleur et de particules. Les approximations majeures du formalisme utilisé, en particulier la représentation de ballonnement à son ordre le plus bas et l'utilisation de fonctions propres analytiques calculées dans la limite hydrodynamiques, sont analysées en détail et leur validité vérifiée. La construction des flux quasi-linéaires est ensuite détaillée et le flux quasi-linéaire de moment angulaire dérivé. Les différentes contributions au flux turbulent de moment angulaire sont étudiées et comparées avec succès à la fois aux données de simulations gyro-cinétiques non-linéaires ainsi qu'aux données expérimentales.

We are constantly struggling to understand how nature works, trying to identify recurrent events and looking for analogies and relations between objects or individuals. Knowing patterns of behavior is powerful and fundamental for survival of any species. In this thesis, datasets of diverse systems related to transportation, economics, sexual and social contacts, are characterized by using the formalisms of time series and network theory. Part of the results consists on the collection and analyzes of original network data, the rest focuses on the simulation of dynamical processes on these networks and to study how they are affected by the particular structures. The majority of the thesis is about temporal networks, i.e. networks whose structure changes in time. The new temporal dimension reveals structural dynamical properties that help to understand the feedback mechanisms responsible to make the network structure to adapt and to understand the emergence and inhibition of diverse phenomena in dynamic systems, as epidemics in sexual and contact networks. / Vi är ständigt kämpar för att förstå hur naturen fungerar, försöker identifier återkommande evenemang och söker analogier och relationer mellan objekt eller individer. Veta beteendemönster är kraftfull och grundläggande för överlevnad av arter. I denna avhandling, dataset av olika system i samband med transporter är ekonomi, sexuella och sociala kontakter, som kännetecknas av att använda formalismer av tidsserier och nätverk teori. En del av resultatet utgörs av insamling och analys av ursprungliga nätdata, fokuserar resten på simulering av dynamiska processer i dessa nätverk och att studera hur de påverkas av de särskilda strukturer. Huvuddelen av avhandlingen handlar om tidsmässiga nät, i.e. nät vars struktur förändringar i tid. Den nya tidsdimensionen avslöjar strukturella dynamiska egenskaper som hjälper till att förstå den feedback mekanismer som ansvarar för att göra nätverksstruktur att anpassa sig och förstå uppkomsten och hämning av olika företeelser i dynamiska system, epidemier i sexuella och kontaktnät. / Constantemente nos esforçamos para entender como a natureza funciona, tentando identificar eventos recorrentes e procurando por analogias e relações entre objetos ou indivíduos. Conhecer padrões de comportamento é algo poderoso e fundamental para a sobrevivência de qualquer espécie. Nesta tese, dados de sistemas diversos, relacionados a transporte, economia, contatos sexuais e sociais, são caracterizados usando o formalismo de séries temporais e teoria de redes. Uma parte dos resultados consiste na coleta e análise de dados de redes originais, a outra parte concentra-se na simulação de processos dinâmicos nessas redes e no estudo de como esses processos são afetados por determinadas estruturas. A maior parte da tese é sobre redes temporais, ou seja, redes cuja estrutura varia no tempo. A nova dimensão temporal revela propriedades estruturais dinâmicas que contribuem para o entendimento dos mecanismos de resposta responsáveis pela adaptação da rede, e para o entendimento da emergência e inibição de fenômenos diversos em sistemas dinâmicos, como epidemias em redes sexuais e de contato pessoal.

We investigate the manifestation of strong electronic correlations in the quasiparticle interference (QPI), arising from the scattering of conduction electrons from defects and impurities in an otherwise translationally-invariant host. The QPI may be measured experimentally as the Fourier transform of the spatial modulations in the host surface density of states that result, which are mapped using a scanning tunnelling microscope. We calculate the QPI for a range of physically relevant models, demonstrating the effect of strong local electronic correlations arising in systems of magnetic impurities adsorbed on the surface of non-interacting host systems. In the first instance the effect of these magnetic impurities is modelled via the single Anderson impurity model, treated via numerical renormalization group (NRG) calculations. The scattering of conduction electrons, and hence the QPI, demonstrate an array of characteristic signatures of the many-body state formed by the impurity, for example due to the Kondo effect. The effect of multiple impurities on the QPI is also investigated, with a numerically-exact treatment of the system of two Anderson impurities via state-of-the-art NRG calculations. Inter-impurity interactions are found to result in additional scattering channels and additional features in the QPI. The QPI is then investigated for the layered transition metal oxide Sr2RuO4, for which strong interactions in the host conduction electrons give rise to an unconventional triplet superconducting state at T_c ∼ 1.5K. The detailed mechanism for this superconductivity is still unknown, but electron-electron or electron-phonon interactions are believed to play a central role. We simulate the QPI in Sr₂RuO₄, employing an effective parametrized model consisting of three conduction bands derived from the Ru 4d t2g orbitals that takes into account spin orbit coupling and the anisotropy of the Ru t2g orbitals. Signatures of such interactions in the normal state are investigated by comparing these model calculations to experimental results. We also calculate the QPI in the superconducting state, and propose how experimental measurements may provide direct evidence of the anisotropy and symmetry of the superconducting gap, and thus offer insight into the pairing mechanism and the superconducting state.

On appelle nanoalliages les nanoparticules d'alliages, leur composition chimique et leur taille en modifient les propriétés physico-chimiques. La relation entre structure et propriétés est intéressante pour la différence existant entre les propriétés du volume et celles du nanoalliage. Comme les alliages volumiques, caractérisés par un diagramme de phase en température et concentration, les nanoalliages sont étudiés pour déterminer un diagramme de phase en fonction de la concentration, la température, la taille et la forme. Les nanoparticules de CoPt ont un intérêt pour le développement du stockage magnétique à haute densité nécessitant une forte anisotropie magnétique telle qu'existant dans la phase volumique L10. L'expérience a montré que pour des nanoparticules de CoPt de 2-3nm la phase L10 existe. Des études théoriques, tenant compte des relaxations du réseau, ont confirmé l'existence d'agrégat de type polyèdre de Wulff présentant un tel ordre sur un réseau cfc.Nous proposons une étude théorique de l'alliage CoPt, utilisant un modèle énergétique simple, le modèle d'Ising en liaisons fortes, reproduisant l'ordre en volume et les trois moteurs de la ségrégation de surface (effets d'alliage, de surface et de relaxation atomique). Les analyses statistiques conduites à partir de simulations Monte Carlo dans les ensembles canonique et semi-grand canonique caractérisent un diagramme de phase de volume et de nanoalliage. Dans l'ensemble semi-grand canonique on s'intéresse à la synergie entre l'ordre au coeur et la ségrégation de surface du système CoPt. L'ordre à courte et longue distance a été caractérisé au cœur de l'agrégat et nous en avons étudié la frustration. / Nano-alloys are commonly the name given to nano-particles of alloys. The size and chemical composition of nano-alloys affect physical and chemical properties. Relation between structure and properties is interesting for the difference of bulk and nano-alloys properties. As bulk alloys which are characterised by phase diagram (in concentration and temperature), nano-alloys can be studie to determine a phase diagram in function of concentration, temperature, size and shape. CoPt nano-particles are interesting for the development of ultra-high density magnetic memories which require strong magnetic anisotropy as it exists in the L10 bulk phase. It has been shown experimentally the CoPt nano-alloys order with the L10 structure exist for the small size (2-3nm). Theoretical studies, using model taking into account the lattice's relaxations have confirmed the ordering of the fcc nano-particles with a Wulff polyhedron (WP) shape. We propose a theoretical study focus on CoPt WP, use a simple energetic model the Tight Binding Ising Model which reproduce the bulk ordering and the three driving surface segregation force (alloying, surface and atomic relaxation effect). Statistical analysis are conducted using a Monte Carlo simulation in the canonical and semi-grand canonical ensemble to characterize a phase diagram for bulk and nano-particles. In the semi-grand canonical ensemble we are interested on the synergy or competition between cluster core ordering and surface segregation of CoPt system. We characterised the order (short and long distance) at the core of the cluster and we studied frustration of the ordering.

Neuromorphic computing applications could be made faster and more power efficient by emulating the function of a biological synapse. Non-conventional spintronic devices have been proposed that demonstrate synaptic behavior through domain wall (DW) driving. In this work, current induced domain wall motion has been studied through micromagnetic simulations. We investigate the synaptic behavior of a head to head domain wall driven by a spin polarized current in permalloy (Py) nanostrips with shape anisotropy, where triangular notches have been modeled to account for edge roughness and provide pinning sites for the domain wall. We seek optimal material parameters to keep the critical current density for driving the domain wall at order 1011 A/m2.