Дисертації з теми "Non-local correlation"

Kohn-Sham density functional theory (DFT) is a popular method used to investigate the properties of matter. Although exact in principle, DFT is in practice limited by a single approximation for the exchange-correlation functional - the quantity that describes the many-body interactions between electrons. This thesis is concerned with developing improved exchange-correlation functionals for use in practical DFT calculations. The standard functional currently used in solid state physics, and also popular in quantum chemistry, is the generalised gradient approximation (GGA), which requires only the local density, n(r), and the density gradient, ∆n(r), as input. A flexible semi-empirical GGA form, containing 15 free parameters that are fitted to near-exact molecular data is implemented within the plane-wave pseudopotential (PW-PP) Kohn-Sham scheme, to assess the possibility of employing semi-empirical GGAs in solid state applications. Self-consistent calculations performed for several bulk semiconductor properties using this GGA reveal that, overall, no improvement is attained over a conventional non-empirical GGA used in solid state physics. The remainder of the thesis focuses on a fully non-local functional known as the weighted density approximation (WDA), which utilises the global density of a system, n(r'), as input. An efficient computational algorithm is devised for use within the PW-PP formalism which enables fully self-consistent WDA calculations to be performed. Physical properties are shown to be intimately related to the particular form used for the pair-correlation function, Gwda/xc(r,r'), and by comparing with recent variational Monte Carlo (VMC) data, it is shown that the forms that provide a good description of exchange-correlation holes, nxc(r,r'), also lead to the most accurate bulk properties. For strongly inhomogeneous electron gas systems, the WDA provides close agreement with the VMC method for a variety of exchange-correlation quantities. The success of the fully non-local approach given by the WDA for other model electron gas systems studied suggests that the WDA is a very promising functional.

Pendant ce travail de thèse, j'ai tout d'abord finalisé le développement d'un système expérimental unique dédié aux études de transport électronique de nanostructures multi-terminaux de faible impédance. Ce dispositif permet des mesures de conductance et de bruit à très basse température (30 mK), avec une résolution du pico-ampère en utilisant des SQUIDs comme amplificateurs de courant. Dans le chapitre 5, je fournis une description du fonctionnement de la mesure. De plus, je décris la calibration du dispositif et la manière de déduire des quantités physiques à partir des mesures.Au Chapitre 6, je décris des mesures de transport avec des jonctions diffusives à trois terminaux (trijonctions). Dans une géométrie, que l'on appelle T-shape, des électrodes supraconductrices d'Aluminium sont connectées entre-elles par une partie centrale métallique non-supraconductrice de Cuivre. Pour ces nanostructures, on observe des anomalies de conductance à basse tension qui n'ont jamais été observées expérimentalement. Ces anomalies de résistance/conductance ressemblant fortement à l'effet Josephson apparaissant lorsque deux des potentiels appliqués à la trijonction ont une somme nulle. Les anomalies sont présentes sur une large échelle de tension sans perte d'amplitude. De-même, elles montrent une grande robustesse en température. Des expériences sous champ magnétique appliqué montrent une forte suppression des anomalies pour un champ magnétique correspondant à flux magnétique dans la partie normale de l'ordre d'un quantum de flux. Ceci indique qu'un mécanisme cohérent de phase doit être à l'origine des anomalies. Dans la littérature, deux mécanismes sont proposés pour expliquer ces effets.Le premier, nommé "mode-locking", est un accrochage dynamique des courants Josephson ac, qui est induit par l'environnement expérimental (circuit). Cette situation a été étudiée dans les années soixante sur des microstructures Josephson couplées à base de liens faibles. Pour tester cette explication, nous avons mesuré un échantillon composé de deux jonctions Josephson spatialement séparées. Les anomalies n'apparaissent pas dans une telle géométrie, pas même avec une amplitude réduite. Ceci indique qu'une synchronisation par l'environnement expérimental ne peut pas être à l'origine des anomalies observées. Le deuxième mécanisme théorique évoqué est nommé "mode de quartet" et a été proposé récemment par Freyn et collaborateurs. L'une des électrodes supraconductrices distribue alors des doublets de paires de Cooper. Chacune de ces deux paires se scindent alors en deux quasiparticles se propageant chacune vers deux contacts supraconducteurs différents. Dans un tel mécanisme deux quasiparticules, issues de deux paires de Cooper différentes, arrivent sur chacun des deux contacts supraconducteurs. Lorsque les tensions appliquées entre le contact supraconducteurs émetteur et les deux autres contacts sont exactement opposés, les phases des fonctions d'ondes électroniques des quasiparticules arrivant sur un même contact supraconducteur sont telles que ces deux quasiparticules peuvent se recombiner pour former une paire de Cooper. Par ce mécanisme le doublet de paires de Copper émis se distribue de manière cohérente en deux paires de Cooper chacune dans un contact supraconducteur différent.Ce mécanisme est favorable, car il est robuste envers le désordre et peut ainsi exister sur une large échelle de tensions.Au cours de cette thèse, j'ai montré que ces anomalies sont effectivement présentes pour des tensions appliquées correspondant à des énergies bien supérieures à l'énergie de Thouless. A contrario, les effets cohérents responsables de l'effet Josephson ac doivent être fortement atténués sur cette même échelle d'énergie, ce qui rend peu probable le mécanisme de mode-locking
During this PhD, I have first finished the development of a unique experimental set-up, dedicated for studies of electronic transport of low impedance multi-terminal nanostructures. This set-up allows conductance and noise measurements at very low temperature (30 mK), with a resolution of a few pico-ampere by using SQUIDs as current amplifiers. In chapter 5, I give some explanation of the measurement working principle. Furthermore, I explain the calibration of the experimental set-up as well as how to extract physical quantities from the measurements.In chapter6, I explain transport measurements on diffusive tri-terminal junctions (tri-junction). In a T-shape called geometry, the superconducting Al-electrodes are connected via a common metallic, non-superconducting part of Copper. For these nanostructures, we observe features in the conductance at low voltage, which have been never observed yet experimentally. These features in conductance/resistance have a striking resemblance with a dc-Josephson effect, appearing when two applied potentials on the tri-junction compensate exactly each other.In literature, two mechanisms are proposed to explain this effect.The first mechanism, called "mode-locking", corresponds to a dynamic locking of ac-Josephson currents, which is induced by the experimental environment (circuit). This situation has been extensively studied in the 60's on coupled microstructures, based on weak links. In order to test this explanation, we have measured a junction, which is composed of two spatially separated Josephson junctions. The anomalies does not show up in such a geometry, even not with strongly reduced amplitude. This indicates, that synchronization via the experimental environment can't be the origin of the observed features. The second theoretical mechanism is named "quartet-mode" and has been recently proposed by Freyn and Co-workers. In this process, one superconducting electrode emits doublets of Cooper-pairs. Each of the two pairs splits into two quasi particles propagating toward different superconducting contacts. In such a mechanism, two quasi-particles originating of two different Cooper-pairs, arrive each in the two superconducting contacts. If the applied voltage between the emitting superconducting contact and the two other contacts is exactly opposite, the phase of the electronic wave functions of the arriving quasi-particles on the same superconducting contact are such, that these two quasi-particles can recombine by forming a Cooper-pair. Due to this mechanism, the emitted doublet of Cooper-pairs is coherently distributed as two Cooper-pairs, each of them in a different superconducting contact. This mechanism is favored, since it is robust with respect to disorder and can hence also exist over a large range of voltage. During this PhD, I have shown that these anomalies are indeed present for applied voltage corresponding to energies well above the Thouless energy. Argumentum a contrario, the coherent effects responsible for the ac Josephson-effect have to be strongly attenuated over the same range of energy, which makes low probable the effect of mode-locking

Cette thèse a pour objectif le développement d'une stratégie d'identification des paramètres de plasticité et d'endommagement jusqu'à amorçage, pour des métaux ductiles.Un formalisme logarithmique est utilisé pour simuler les grandes déformations subies par les éprouvettes et une formulation non-locale multi-champ permet de simuler l'adoucissement indépendamment du maillage utilisé et d'éviter le verrouillage volumique.La Corrélation d'Images Digitales est utilisée pour obtenir des mesures hétérogènes plein champ à partir d'éprouvettes entaillées.La stratégie proposée s'appuie sur des observations microscopiques et sur une approche d'identification par recalage de modèle éléments finis (FEMU), visant à minimiser l'écart entre une mesure et son pendant simulé. L'écart est exprimé en termes de force et déplacement grâce à une normalisation appropriée. L'application de la FEMU est guidée par des analyses de sensibilité.La robustesse de la comparaison essai-calcul est assurée par l'application de conditions au bord mesurées. L'effet négatif de l’incertitude de mesure est mis en évidence et une solution de filtrage innovante est proposée.La stratégie est appliquée pour l'identification des paramètres de l'alliage Inconel625. Elle permet de reproduire l'amorçage pour des éprouvettes planes, en termes de réponse macroscopique et de localisation des sites d'amorçage
This thesis proposes an identification strategy for plastic behaviour and damage up to the onset of fracture, for an application to ductile metals.A logarithmic finite strain formulation is used to simulate the large deformations undergone by the specimens, while a locking-free non-local formulation allows a mesh independent simulation of the softening behaviour.Digital Image Correlation is used to obtain heterogeneous full-field measurements from tensile tests on notched specimens.The identification strategy is based both on microscopic observations and on a Finite Element Model Updating (FEMU) technique, according to which the parameters are identified by minimizing the discrepancy between experiment and simulation. The discrepancy is quantified both in terms of displacement and force thanks to an appropriate normalization. The application of FEMU is guided using sensitivity analysis.The robustness of the comparison between simulation and measurement is ensured by prescribing measured displacements as boundary conditions for the simulation. The negative effect of the measurement uncertainty is underlined, and an innovative filtering approach is proposed.The proposed strategy is used to identify the materials' parameters of alloy Inconel625. It allows to reproduce the onset of fracture for flat specimens, both in terms of macroscopic response and crack initiation location

Quantum Information Theory (QIT) studies how information can be processed and transmitted when encoded on quantum states. Practically, it can be understood as the effort to generalize Classical Information Theory to the quantum world. Interestingly, the fact that very-small scale Physics differs considerably from that of macroscopic objects offers a richer structure to the new theory. Among other phenomena, entanglement is at the heart of many quantum information protocols. It is the most spectacular and counter-intuitive manifestation of quantum mechanics: it signifies the existence of non-local correlations. Although intrinsically non-intuitive, these strange effects have been shown to lead to intriguing applications with no classical analogue. The main scope of this thesis is to establish qualitative and quantitative connections among the different quantum and classical information resources. Among the many weird effects that quantum systems present, the non-additivity concept plays an important role. In the quantum realm, the joint processing of two quantum resources is often better than the sum of the two resources. Activation is the strongest manifestation of non-additivity. It can be understood as the capability of two objects to achieve a given task that is impossible for each of them when considered individually. From a classical point of view, it is unknown whether such a process can hold. Here we focus on the classical secret-key rate. We provide two probability distributions conjectured to have bound information, hence from which it is conjectured that no secret key can be extracted when taken individually, but that lead to a positive secret-key rate when combined. For that, we exploit the close connection between the information-theoretic key agreement and the quantum entanglement scenario. Successively, we move to the multipartite scenario showing a one-to-one correspondence between bound information and bound entanglement. We provide an example of multipartite bound information which shares the same features of its quantum analogue, the Smolin state. Later, we move to prove a deep connection between privacy and non-locality. We do it by showing that all private states violate the Bell-CHSH inequality. Private states are those entangled states from which a perfectly secure cryptographic key can be extracted. An example of those is the maximally entangled state. But still, there are other private states that are not maximally entangled. While a maximally entangled state violates a Bell's inequality, this is not known a priori for the whole set. We give a general proof valid for any dimension and any number of parties. Private states, then, not only represent the unit of quantum privacy, but also allow two distant parties to establish a different quantum resource, namely non-local correlations. Lastly, we tackle the connection between non-locality and genuine randomness. Non-locality and genuine intrinsic randomness have been the subject of active interest since the early days of quantum physics. Initially, this interest was mainly derived from their foundational and fundamental implications but recently it also has acquired a practical aspect. Recent development in device independent scenario have heightened the need to quantify both the randomness and non-locality inherent in quantum systems. While some works try to deepen this relation, we provide a simple method to detect Bell tests that allow the certification of maximal randomness. These arguments exploit the symmetries of Bell inequalities and assume the uniqueness of the quantum probability distribution maximally violating it. We show how these arguments can be applied to intuit the randomness intrinsic in a probability distribution without resorting to numerical calculations.
La Teoría de la Información Cuántica (QIT) estudia como la información puede ser procesada y transmitida al codificarse en estados cuánticos. Prácticamente, se puede pensar como la generalización de la Teoría de Información Clásica al mundo cuántico. El hecho que la física a esta escala difiera considerablemente de aquella de los objetos macroscópicos ofrece una mayor riqueza a la estructura de la nueva teoría. Entre otros fenómenos, el entrelazamiento está a la base de muchos protocolos cuánticos. Es la más espectacular y anti-intuitiva manifestación de la mecánica cuántica observada en sistemas cuánticos compuestos: implica la existencia de correlaciones no-locales. No obstante la extrañeza de estos efectos, se han demostrado distintas aplicaciones sin ningún análogo clásico. El objetivo de esta tesis es establecer conexiones cualitativas y cuantitativas entre los diferentes recursos descritos por la teoría cuántica y clásica. Entre los efectos raros que los sistemas cuánticos muestran, la no-aditividad desempeña un papel muy importante. En el mundo cuántico, el uso de dos recursos cuánticos puede ser más ventajoso que la suma de los dos, considerados individualmente. La activación es la mas fuerte manifestación del fenómeno de no-aditividad. Este proceso se puede entender como la capacidad de dos objetos juntos de lograr una tarea que sería imposible por cada uno de ellos singularmente. Desde un punto de vista clásico, es desconocido si existen procesos o cantidades que no respetan la aditividad. Aquí, nos centramos en la tasa de clave secreta. Presentamos aquí dos distribuciones de probabilidad que conjeturamos contener bound information, o sea a partir de la cuales es imposible destilar bits secretos que dan bits secretos cuando utilizadas conjuntamente. Para probar este resultado, utilizamos la conexión existente entre entrelazamiento y el proceso de establecimiento de seguridad. Sucesivamente desplazándonos al caso multipartito, probamos una correspondencia uno a uno entre la bound information y el entrelazamiento no-destilable. Presentamos un ejemplo de bound information multipartita que comparte las mismas propiedades de su análogo cuántico, el estado de Smolin. Luego profundizamos la relación entre privacidad y no-localidad. Probamos que todos los estados que pertenecen al conjunto de estados privados violan una desigualdad de Bell, conocida como CHSH. Los estados privados son aquellos estados entrelazados de los cuales es posible extraer una clave secreta. Un ejemplo de estos estados es el estado máximamente entrelazado, pero hay otros que son privados aunque no máximamente entrelazados. Es conocido que un estado máximamente entrelazado puede violar una desigualdad de Bell, pero lo que se desconoce es si esto pasa para todos los estados privados. Nuestro resultado es general ya que nuestra prueba es válida para cualquier número de partes y cualquier dimensión del espacio local de cada una. Los estados privados, entonces, no solo permiten destilar una clave de forma segura sino que también presentan una propiedad tan fuerte como la no-localidad. Finalmente, investigamos la relación entre los conceptos de no-localidad y de aleatoriedad. Desde los orígenes de la teoría cuántica, los conceptos de no-localidad y de aleatoriedad fueron objeto de gran interés. A principio este interés se debía más a razones relacionadas con los fundamentos de la teoría, pero recientes resultados han empujado la comunidad científica a investigar ulteriormente y sobre todo a cuantificar la no-localidad y la aleatoriedad presente en los estados cuánticos. Aunque algunos autores se hayan movido en esta direccion, muchas preguntas han quedado sin respuestas. Aquí presentamos un simple método que permite detectar aquellas desigualdades de Bell que pueden certificar la presencia de máxima aleatoriedad. Nuestros resultados prueban como simples argumentos pueden dar complejas respuestas sin la necesidad de recorrer a computaciones numéricas.

In this thesis we theoretically examine spin-polarized non-local transport in two junctions built up by materials with superconducting and magnetic properties. The first consists of a Zeeman-split superconductor placed between two normal metals. We find that such systems can generate a spin-polarized non-local current even when the injected current is unpolarized. The effect is however mostly geometric and take place also when the superconductor is switched with a normal metal. The second system is built up by two ferromagnetic wires coupled by a superconductor via spin-active interfaces. We investigate the circumstances which allow for an inverse spin-valve effect. For certain parameter ranges we find that a crossover between positive and negative magnetoresistance is possible as a function of temperature.

This thesis describes the work undertaken over a 3 year period under a Department of the Environment, Transport and the Regions (DETR) Partners in Technology programme in collaboration with four industrial partners. The aim of the work was to investigate the two-phase flow heat transfer and fluid dynamic performance of the non-ozone depleting refrigerants and lubricating oil mixtures in a horizontal tube. A single tube test facility was designed and built to determine local heat transfer coefficients and overall pressure drop for evaporation and condensation. A circulating pump system pressurised through an accumulator charged with nitrogen was selected to enable accurate setting of the evaporation and condensation conditions. The system was validated using R22. Tests were undertaken for R404A, R407C, Isceon 59 with a Polyol-ester lubricating oil. The experimental data consist of two-phase local heat transfer coefficients and overall pressure drops inside a smooth copper tube. Numerous data were obtained for evaporation and condensation, and constituted a large database of two-phase heat transfer coefficients. The experimental results were compared with several existing correlations traditionally used in heat exchanger design, to assess their suitability with the new refrigerants. As a result of these comparisons offering a poor agreement, a model characterising the variation of the local heat transfer coefficient was developed. The evaporation model consists in a modification of the Kattan et al. model (1998a, b & c). The new model predicts the experimental results well, with a standard deviation of 6.1% for refrigerant R407C and Isceon 59. The effects of lubricating oils on two-phase flow heat transfer are also discussed qualitatively. This work provides a design tool when dealing with the non-ozone depleting refrigerants, and recommendations are made on the use of several correlations.

Diploma thesis is about image sensor security. Goal of the thesis was study data integrity gained from the image sensors. Proposed method is about source camera identification from noise characteristics in image sensors. Research was about influence of denoising algorithms applied to digital images, which was acquired from 15 different image sensors. Finally the statistical evaluation had been done from computed results.

En vue d'obtenir des informations sur l'importance structurale de h, experiences de diffusion dee neutrons sur des echantillons obrtenus par pulverisation cathodique. Calcul des trois fonctions de correlation de paire si-si, si-h et h-h en utilisant la substitution de h et d. Mesures precises des premiers voisins si par experiences exafs sur des couches minces. Obtention d'informations sur l'orientation relative des tetraedres si

Density functional theory (DFT) is a rapidly growing approach to quantum theory that employs the three dimensional density for energy calculations in contrast to the many dimensional wavefunction method of traditional quantum mechanics. It also introduces the concept of F (n), a universal density functional which is independent of any external potential, that exists in theory but that in practice needs to be approximated. This is accomplished by separating F (n) into different energy functional components and then approximating these functionals. The purpose of the following research is to approximate the correlation-energy component, $\rm E\sb{c}\lbrack n\rbrack,$ by trying to satisfy certain theoretical constraints Uniform and non-uniform coordinate scaling requirements for a correlation-energy density functional led initially to the development of the Wilson-Levy (WL) functional. The numerical results of this functional, which is Wigner-like, are compared to results of both recent and traditional functionals. Values calculated from the correlation-energy functionals of both the original Wigner and the VWN are also tabulated. Parameterization of the WL functional is explained. The advantages and disadvantages of the WL and other functionals are discussed and the functional derivative of both the closed-shell and open-shell forms of the WL is presented More recent developments including non-uniform coordinate scaling requirements lead to the development of a new correlation-energy Wigner-like functional. Coordinate scaling requirements along with other constraints are reviewed and investigated for the functionals of Perdew, Becke, Lee-Yang-Parr, Wilson-Levy, Perdew-Wang 91 and the newest functional. Numerous other equations that do not comply with these constraints are also evaluated. The importance of $$\rm\biggl\vert\left({\partial E\sb{c}\lbrack n\sb\lambda\rbrack \over\partial\lambda}\right)\sb{\lambda=1}\biggr\vert$$is assessed and used in a new method for selecting a spin-function for open-shell species. The correlation between coordinate constraint obedience and a functional's efficiency is finally evaluated and suggestions for an improved functional are proposed
acase@tulane.edu