Les simulations de Monte-Carlo, bien que consommatrices en temps de calcul, restent un outil puissant qui permet d'évaluer les méthodes de correction des effets physiques en imagerie médicale. Nous avons optimisé et validé une méthode de reconstruction baptisée F3DMC (Fully3D Monte CARLO) dans laquelle les effets physiques perturbant le processus de formation de l'image en tomographie d'émission monophotonique sont modélisés par des méthodes de Monte-Carlo et intégrés dans la matrice-système. Le logiciel de simulation de Monte-Carlo utilidé est GATE. Nous avons validé GATE en SPECT en modélisant la gamma-caméra (Philips AXIS) utilisé en routine clinique. Des techniques de seuillage, filtrage par analyse en composantes principales et de reconstruction ciblée (régions fonctionnelles, région hybrides) ont été testées pour améliorer la précision de la matrice-système et réduire le nombre de photons ainsi que le temps de calcul nécessaires. Les infrastructures de la grille EGEE ont été utilisées pour déployer les simulations GATE afin de réduire leur temps de calcul. Les résultats obtenus avec F3DMC sont comparés avec les méthodes de reconstruction (FBP,ML-EM,ML-EMC) pour un fantôme simulé et avec la méthode OSEM-C pour un fantôme réel. Les résultats de cette étude montrent que la méthode F3DMC ainsi que ses variantes permettent d'améliorer la restauration des rapports d'activité et le rapport signal sur bruit. L'utilisation de la grille de calcul EGEE a permis d'obtenir un gain de l'ordre de 300 en temps de calcul. Dans la suite, ces résultats doivent être confirmés par des études sur des fantômes complexes et des patients et ouvrent la voie vers une méthode de reconstruction unifiée, pouvant être appliquée aussi bien en SPECT qu'en PET.

A evolução continua dos computadores levou a várias modificações na maneira de fazer ciência, fazendo surgir uma infinidade de novas opções para resolvermos problemas científicos. A implementação computacional de métodos teóricos possibilitou a tratamento de sistemas grandes, complexos e em diferentes áreas da ciência.Uma área particularmente em destaque e a química de moléculas, atualmente podemos descrever sistemas moleculares relativamente complexos com extrema precisão. Os métodos teóricos em química molecular podem ser divididos basicamente em métodos clássicos ou quânticos, dependendo do que queremos estudar, podemos também utilizar métodos híbridos ou seqüenciais, contudo para uma descrição detalhada da estrutura eletrônica e de ligações químicas é necessária a utilização de métodos quânticos.Cálculos ab initio da estrutura eletrônica de átomos e especialmente moléculas foram realizados pela primeira vez utilizando-se o método de expansão de Roothaan na década de cinqüenta. Em 1986, uma versão integral das equações de Griffin-Wheele-Hartree-Fock (GW-HF) foi apresentada na literatura , inspirada no Método da Coordenada Geradora (MCG), introduzido por Griffin e Wheeler na década de cinqüenta. A versão integral das equações de Hartree-Fock foi denominada de Método da Coordenada Geradora Hartree-Fock (MCG-HF), e uma de suas primeiras aplicações foi na geração de bases atômicas universais . De fato, uma integração numérica cuidadosa das equações de GW-HF permite a geração de bases universais bem mais generalizadas do que as já publicadas na literatura.Recentemente, Barbosa e da Silva propuseram uma modificação na metodologia da discretização com o intuito de melhorar a obtenção de conjuntos de funções Gaussianas (GTF) por meio do MCG-HF, tornando possível a geração de GTF tão boas ou melhores quanto as até agora obtidas na literatura, porém mais compactas e precisas (acuradas). / The ongoing evolution of computers has led to several changes in the way of doing science, to create a multitude of new options for solving scientific problems. The implementation of computational methods allowed the theoretical treatment of large systems, complex and different areas of science. One area of particular focus and chemistry of molecules, now can describe relatively complex molecular systems with extreme precision. The theoretical methods in molecular chemistry can be basically divided into classical and quantum methods, depending on what we consider, we can also use sequential or hybrid methods, however for a detailed description of the electronic structure and chemical bonds is required the use of quantum methods. Ab initio calculations of the electronic structure of atoms and molecules have been specially made for the first time using the method of Roothaan expansion in the decade of the fifties . In 1986, a full version of the equations of Griffin-Wheeler-Hartree-Fock (HF-GW) was presented in the literature , inspired by the Method of Coordinate Generator (MCG), introduced by Griffin and Wheeler in the decade of fifty . The full equations of the Hartree-Fock method was called the Generator Coordinate Hartree-Fock (HF-MCG) and its first applications was the generation of atomic universal bases In fact, a careful numerical integration of equations of GW-HF allows the generation of universal bases and more widespread than those already published in the literature .Recently, Barbosa and Silva proposed a modification in the methodology of the discretization to improve the collection of sets of Gaussian functions (GTF) by means of MCG-HF, making possible the generation of GTF as good or better as the so far obtained in the literature, but more compact and precise (accurate). A polynomial expansion is proposed as a new way to discretize the Griffin-Wheeler-Hartree-Fock equations of the Generator Coordinate Hartree-Fock method. The implementation of the polynomial expansion in the Generator Coordinate Hartree-Fock method discretizes the Griffin-Wheeler-Hartree-Fock equations through a numerical mesh, which is not equally spaced. This procedure makes the optimization of Gaussian exponents in the Generator Coordinate Hartree-Fock method more flexible and more efficient. The results obtained with the polynomial expansion for atomic Hartree-Fock energies show this technique is very powerful when employed in the design of compact and high accurate Gaussian basis sets used in ab initio non-relativistic (Hartree-Fock) and relativistic (Dirac-Fock) calculations.

<p>This thesis addresses the investigation of doubly excited <i>2l´nl</i> states in helium atoms and double core excitations in solid lithium compounds.</p><p>Measurements on <i>He</i> are made in field free environments and under the influence of electric and magnetic fields, using synchrotron based inelastic photon scattering. Cross sections for scattering to singly excited final states are directly determined and compared to theoretical results and are found to be in excellent agreement. Radiative and spin-orbit effects are quantified and are shown to play an important role in the overall characterization of highly excited <i>He </i>states below the <i>N =2</i> threshold. A dramatic electric field dependence is also observed in the flourecence yield already for relatively weak fields. This signal increase, induced by electric as well as magnetic fields, is interpreted in terms of mixing with states of higher fluorescence branching ratios.</p><p>Double core excitations at the lithium site in solid lithium compounds are investigated using resonant inelastic x-ray scattering (RIXS). The lithium halides <i>LiF, LiCl, LiBr</i> and <i>LiI </i>are studied as well as the molecular compounds <i>Li</i><i>2</i><i>O, Li</i><i>2</i><i>CO</i><i>3</i> and <i>LiBF</i><i>4</i>. States with one, as well as both, of the excited electrons localized at the site of the bare lithium nucleus are identified, and transitions which involve additional band excitations are observed. A strong influence of the chemical surrounding is found, and it is discussed in terms of the ionic character of the chemical bond.</p>

Polymer electrolytes are the most obvious candidates for safe "all-solid" Li-ion batteries and other electrochemical devices. However, they still have relatively poor ionic conductivities, which limits their wider adoption in commercial applications. It has earlier been the conventional wisdom that only amorphous phases of polymer electrolytes show usefully high ionic conduction, while crystalline forms are insulators. However, this has been challenged in the last decade by the discovery of highly organized, low-dimensional ion-conducting materials. Specifically, the crystalline phases of LiXF6.PEO6 exhibit higher ionic conductivities than their amorphous counterparts, with the Li-ion conduction taking place along the PEO channels. Polymer chain-length and chain-end registry has emerged as potentially significant in determining ionic conduction in these materials. Molecular Dynamics simulations have therefore been made of short-chain, monodisperse (Mw~1000), methoxy end-capped LiPF6.PEO6 to examine relationships between ion conduction and mode of chain-ordering. Studies of smectic and nematic arrangements of PEO chains have revealed that ion-transport mechanisms within the smectic planes formed by cooperative chain-end registry appear to be more suppressed by ion-pairing than in-channel conduction. Disorder phenomena in the chain-end regions emerge as a critical factor in promoting Li-ion migration across chain-gaps, as does the structural continuity of the PEO channels. Simulations incorporating ~1% aliovalent SiF62- dopants further suggest an increase in Li-ion conduction when the extra Li-ions reside within the PEO channels, with the anion influencing charge-carrier concentration through enhanced ion-pair formation. XRD techniques alone are shown to be inadequate in ascertaining the significance of the various short-chain models proposed; atomistic modelling is clearly a helpful complement in distinguishing more or less favourable situations for ion conduction. Though providing valuable insights, it must be concluded that this work has hardly brought us significantly closer to breakthroughs in polymer electrolyte design; the critical factors which will make this possible remain as yet obscure.

La caractérisation de matériaux diffusants, tels les pigments, par leur indice optique complexe demeure actuellement un véritable challenge. Cet indice est la caractéristique intrinsèque au matériau, et est de ce fait, indispensable pour la simulation d'effets colorés et peut, dans le champ de la conservation-restauration, permettre la mise en place d'une méthode d'identification non-ambigüe des matériaux composant les œuvres d'art. L'objectif de cette recherche est d'élaborer une méthodologie permettant de déterminer l'indice optique complexe de ce type de matériaux. Elle constitue l'étape primordiale à l'élaboration d'une base de données d'indices optiques des pigments. Dans un premier temps, une étude théorique des phénomènes conjoint de diffusion et d'absorption résultant de l'interaction d'une onde électromagnétique avec un diffuseur, ou un ensemble de diffuseurs homogènes et sphériques, est développée. Elle consiste à définir l'influence des paramètres intrinsèques (des matériaux) et morphologiques (d'un milieu) sur les résultats obtenus lors de la résolution de l'équation du transfert radiatif (ETR) traité par la méthode 4-flux et de la théorie de Mie dans le sens direct. Dans un second temps, nous aborderons le problème inverse. À savoir, dans notre cas, remonter à l'indice complexe de particules diffusantes à partir des propriétés optiques de la couche. Nous illustrerons cette approche par l'étude d'un pigment rouge, le cinabre. Le modèle de calcul développé prend en compte les flux diffus et les flux spéculaires, mesurés sur les échantillons par spectrophotométrie couplée à une sphère d'intégration dans le domaine du visible

La croissance de couches minces d'oxydes sur les alliages métalliques et les métaux purs est un phénomène ayant fait l'objet d'un grand nombre d'études expérimentales. Les structures des couches d'oxydes sont bien connues, ainsi que les modèles mathématiques servant à modéliser les aspects macroscopiques de la croissance. Cependant, les détails des mécanismes de germination de la couche d'oxyde dans les premiers stades de la corrosion ainsi que sa croissance ultérieure demeurent peu ou mal connus. La simulation atomistique apparaît comme une alternative pour évaluer les différents mécanismes proposés et appréhender l'influence des différents paramètres physico-chimiques. Le développement d'un tel outil de simulation a démarré au LPCS avec la thèse de M. Legrand. En se basant sur l'exemple de l'alliage FeCr, un modèle informatique tridimensionnel dit "modèle de Legrand", permettant de simuler la dissolution sélective et la passivation des alliages binaires a été réalisé. L'évolution dynamique est basée sur une technique de type Monte Carlo classique. Le logiciel permet de simuler l'évolution d'un alliage quelconque, d'une composition et d'une structure cristallographique donnée. Il prend en compte la diffusion des atomes sur la surface, leur dissolution et le blocage de la dissolution par formation d'une couche de passivation. Cet outil était adapté pour la simulation des premiers stades de la corrosion. L'objectif de ce travail est d'améliorer ce modèle existant, afin de simuler l'évolution de la couche passive sur une échelle de temps plus longue. A l'issue de ce travail, de nombreux apports ont été effectués. Ainsi, l'introduction d'un champ de force MEAM (Modified Embedded Atom Method) pour le calcul des barrières de diffusion et de dissolution, a permis de remplacer les probabilités de diffusion empiriques par des probabilités calculées, et de mettre en évidence la diffusion préférentielle des Cr vers leurs semblables. L'introduction d'une dynamique de simulation Monte Carlo Cinétique (KMC) a permis une prise en compte réaliste de l'évolution cinétique du modèle avec des temps de simulation reliés au temps réel. Enfin Un second réseau cristallographique RVO (réseau virtuel d'oxyde) tridimensionnel, correspondant à celui de la couche passive (Cr2O3) a été implémenté, ainsi qu'une interface graphique pour un meilleur suivi de la simulation. Les résultats obtenus lors des simulations sont en accord avec les observations expérimentales: passivation totale à partir du Fe-16Cr, enrichissement en Cr de la couche passive, allures des courbes cinétiques, influence du champ électrique, mise en évidence l'apparition de cavités sous la couche passive.

pas de résumé

Ce document présente une étude des mécanismes de relaxation de fullerènes induits par photoexcitation. Nous avons exploré les processus d'ionisation et de fragmentation de ces systèmes modèles suite à une irradiation par différentes sources de lumière : impulsion femtoseconde, laser à électron libre (IR lointain) et Synchrotron (XUV). Afin d'étudier ces mécanismes, nous utilisons l'imagerie par cartographie de vitesse qui nous permet de suivre l'évolution de la distribution angulaire de vitesse des particules émises. Suivant la durée, l'intensité et la longueur d'onde de l'excitation utilisée, la nature des mécanismes change drastiquement (mécanismes statistiques, ionisation cohérente en champ fort, autoionization de plasmons, etc...). Plusieurs approches théoriques sont utilisées allant du modèle Weisskopf pour les processus statistiques à la résolution de l'équation de Schrödinger dépendante du temps pour les processus cohérents. Cette étude permet de relier les mécanismes de relaxation intervenant à différentes échelles de temps

We are developing a novel classical over-the barrier model for electron transfer from an in nitely thin conducting disc to a point charge projectile to model multiple electron capture in e.g. keV collisions of atomic ions with poly-cyclic aromatic hydrocarbons (PAHs). In its nal form, the present model will incorporate the polarization of the PAH molecules due to the active electron and the point charge projectile at a general angle of incidence. This will drastically improve the description of the potential barrier in comparisons with simpler versions of the model where the nite size and polarizability of the target molecule is neglected or treated in an averaged fashion. In this work we arrive at expressions for the electrostatic potential energy barrier experienced by the active electron in the two spatial orientations where the point charge projectile is located along the normal symmetry axis and in the tangent plane of the disc. Applied to coronene (C24H12) such barriers compare better with high level density functional theory (DFT) calculations than with the results from the simpler versions of the classical over-the-barrier models for atomic and spherical cluster targets. These results thus strongly supports the conducting disc approximation of PAHs. Finally we discuss the nal steps in the model development and possible extensions of the model to include less symmetric elliptical discs or spherical caps.

Laser-cooled atomic ions have led to an unprecedented amount of control over the quantum states of matter. The Coulombic interaction allows for information to be transferred between neighboring ions, and this interaction can be used to entangle qubits for logic operations in quantum information processors. The same procedure for logic operations can be used for high resolution atomic spectroscopy, and is the basis for the most accurate atomic optical clocks to date. This thesis describes how laser-cooled atomic ions can impact physical chemistry through the development of molecular ion spectroscopy techniques and the simulation of magnetic systems by ion trap quantum computers. A new technique developed for spectroscopy, Sympathetic Heating Spectroscopy (SHS), takes advantage of the Coulombic interaction between two trapped ions: the control ion and a spectroscopy ion. SHS uses the back action of the interrogating laser to map spectroscopy ion information onto the Doppler shift of the control ion for measurement. SHS only requires Doppler cooling of the ions and fluorescence measurement and represents a simplification of quantum logic spectroscopy. This technique is demonstrated on two individual isotopes of calcium: Ca-40(+) for cooling and Ca-44(+) as the spectroscopy ion. Having demonstrated SHS with atomic ions, the next step was to extend the technique by loading and characterizing molecular ions. The identification of an unknown molecular ion is necessary and can be achieved by monitoring the change in motion of the two ion crystal, which is dependent on the molecular ion mass. The motion of two trapped ions is described by their normal modes, which can be accurately measured by performing resolved sideband spectroscopy of the S(1/2)-D(5/2) transition of calcium. The resolved sidebands can be used to identify unknown ions (atomic and molecular) by calculating the mass based on the observed value in axial normal mode frequencies. Again, the trapped molecular ion is sympathetically cooled via the Coulombic interaction between the Ca-40(+) and the unknown molecular ion. The sensitivity of SHS could be improved by implementing sympathetic sideband cooling and determining the heating by measuring single quanta of motion. The ultimate limit of control would be the development of an ion trap quantum computer. Many theoretical quantum computing researchers have made bold claims of the exponential improvement a quantum computer would have over a classical computer for the simulation of physical systems such as molecules. These claims are true in principle for ideal systems, but given non-ideal components it is necessary to consider the scaling due to error correction. An estimate of the resource requirements, the total number of physical qubits and computational time, required to compute the ground state energy of a 1-D quantum Transverse Ising Model (TIM) of N spin-1/2 particles, as a function of the system size and the numerical precision, is presented. This estimate is based on analyzing the impact of fault-tolerant quantum error correction in the context of the quantum logic array architecture. The results show that a significant amount of error correction is required to implement the TIM problem due to the exponential scaling of the computational time with the desired precision of the energy. Comparison of this result to the resource requirements for a fault-tolerant implementation of Shor's quantum factoring algorithm reveals that the required logical qubit reliability is similar for both the TIM problem and the factoring problem.