L'objectif de ce travail est de prédire le gonflement de fûts d'enrobés bitumineux contenant des déchets radioactifs en condition d'entreposage, de transport et de stockage. L'incorporation de radionucléides dans du bitume provoque en effet sa radiolyse et du dihydrogène est ainsi généré dans le matériau. Ce gaz est solubilisé dans la matrice jusqu'à ce que sa concentration atteigne une valeur limite, la saturation. Au-delà de celle-ci, un mécanisme de germination a lieu : des bulles sont formées. Elles vont par la suite croître grâce à l'apport de gaz par radiolyse et migrer sous l'effet de la poussée d'Archimède. Le gonflement du matériau résulte alors de la compétition entre création et évacuation de gaz. Un modèle a été construit pour décrire l'évolution d'une population de bulles. Sa résolution analytique n'étant pas possible, un code numérique a été développé. Les résultats montrent qu'une augmentation de la viscosité du fluide, du terme source de gaz apporté par radiolyse ou de la hauteur du récipient induisent une augmentation du gonflement et qu'une augmentation du coefficient de diffusion du gaz dans la matrice diminue le gonflement. Lorsque le fluide présente un seuil d'écoulement, le comportement de la population de bulles est modifié et l'évacuation de gaz par migration des bulles se fait par à-coups, contrairement au cas d'un fluide newtonien où une évacuation stationnaire est atteinte. Cette dernière a enfin servi à l'optimisation des conditions opératoires. / The aim of this PhD thesis is to predict the swelling of bitumen barrels in which radioactive salts are mixed. The bitumen exposed to radioactivity undergoes a chemical reaction: the radiolys. This implies a generation of dihydrogen. The created is solubilized until the concentration reaches a limit value which is called saturation. Over this limit nucleation of bubbles is observed. Then they will grow thank to the contribution of the gas generated by radiolys and they will be submitted to Archimede's principle so that they will rise in the fluid. The swelling is the result of the competition between generation and evacuation of gas. A model has been built to describe theevolution of a bubble population. Because of it is not possible to solve it analitically, a numerical program was developed. The results show that an increase of the fluid viscosity, the gas generation or the container height lead to an increase of the swelling and that an increase of the diffusioncoefficient contributes to a decrease of the swelling. In the particular case of a yield stress fluid, the behavior of the bubble population is modified and the evacuation of gas is done with several shots, at the opposite of the case of a Newtonian fluid for which a stationary evacuation is reached.

We study a stochastic particle system with a logarithmically-singular inter-particle interaction potential which allows for inelastic particle collisions. We relate the squared Bessel process to the evolution of localized clusters of particles, and develop a numerical method capable of detecting collisions of many point particles without the use of pairwise computations, or very refined adaptive timestepping. We show that when the system is in an appropriate parameter regime, the hydrodynamic limit of the empirical mass density of the system is a solution to a nonlinear Fokker-Planck equation, such as the Patlak-Keller-Segel (PKS) model, or its multispecies variant. We then show that the presented numerical method is well-suited for the simulation of the formation of finite-time singularities in the PKS, as well as PKS pre- and post-blow-up dynamics. Additionally, we present numerical evidence that blow-up with an increasing total second moment in the two species Keller-Segel system occurs with a linearly increasing second moment in one component, and a linearly decreasing second moment in the other component.

Les mousses sont des matériaux très largement utilisés dans l'industrie par exemple dans l'agroalimentaire ou la cosmétique. Ce sont des matériaux instables qui vieillissent au cours du temps à cause du mûrissement dû à la diffusion du gaz entre les bulles, de la coalescence qui est la fusion de deux bulles après rupture du film les séparant et du drainage dû à l'écoulement de liquide entre les bulles par gravité. Ces trois processus sont en compétition dans les mousses. Au cours de ma thèse, j'ai réalisé différentes expériences dans le but d'éliminer le drainage et d'étudier le mûrissement, la coalescence et la compétition entre les deux. Après un rappel sur la structure, le vieillissement des mousses et la description du dispositif expérimentale que j'ai mis au point pour contrôler la quantité de liquide dans la mousse, je me suis intéressée au mûrissement et à la coalescence de mousses 2D horizontales. J'ai montré tout d'abord grâce à l'analyse d'image qu'il est possible de différencier les mousses évoluant principalement par mûrissement de celles évoluant par coalescence en introduisant un paramètre de convexité sur les bulles composant la mousse. Les mousses ayant mûri sont alors utilisées pour étudier l'impact de l'humidité des mousses sur la perméabilité des films. La perméabilité est calculée à l'aide d'une nouvelle théorie prenant en compte l'humidité de la mousse. Les mousses qui coalescent, quant à elles, sont utilisées pour étudier l'impact de la fraction liquide, de la pression capillaire et de la taille des bulles sur la coalescence dans les mousses 2D. Cette étude permet de montrer qu'une analyse de ces paramètres à l'échelle de la mousse n'est pas suffisante pour conclure sur leur réel impact sur la coalescence et qu'il est nécessaire d'analyser ces paramètres à l'échelle de la bulle afin de différencier leurs effets. Pour ce faire, j'ai mis au point un programme qui permet d'isoler chaque évènement de coalescence. Ainsi, il existe une fraction liquide critique à partir de laquelle il n'y a plus de coalescence car les films entre les bulles n'existent plus. A 3D, j'ai commencé par établir un lien entre les fractions liquides surface et volumique donnant une nouvelle manière de de mesurer la fraction liquide volumique à partir d'une image de la surface. Ensuite, j'ai étudié le mûrissement de mousses humides 3D. Un des buts était de mettre en évidence la transition entre le régime sec où le rayon moyen croit comme le temps avec un exposant β =1/2 et le régime très humide où β = 1/3. J'ai réalisé des expériences de diffusion multiple de la lumière, sur des mousses confinées et trouvé de valeurs de β voisine de 1/2. Néanmoins, la mesure β dans les mousses très humides n'a pas été possible. Pour parer à cela, le projet "Soft Matter Dynamics" vise à étudier le mûrissement et le réarrangement des bulles dans les mousses humides en microgravité dans la station spatiale internationales (ISS). Mon implication dans ce projet a consisté à réaliser des protocoles de remplissage et de nettoyage de cellules en collaboration avec Airbus, à créer un programme d'automatisation de la mesure de la taille des bulles et à participer à une campagne de test du système de génération des mousses. / Foams are materials widely used in the industry, like in food industry or cosmetics industry for example. These are unstable materials that age over time by coarsening which is gas diffusion between bubbles, by coalescence which is the fusion of two bubbles after rupture of the film between them or by drainage which is the liquid flow between the bubbles due to gravity. These three processes are in competition in foams. During my thesis, I carried out various experiments in order to eliminate drainage and study coarsening, coalescence and the competition between them. After a quick reminder on foam structure and aging and a description of the experimental set-up that I made allowing to control the liquid content of the foam, I present coarsening and coalescence experiments of horizontal 2D foams. First, I show that it is possible to differentiate foams which evolved mainly by coarsening from those which evolved by coalescence. This is done using a convexity parameter that compares the convexity of the bubbles in the foams. The foams which coarsen are then used to study the impact of foam humidity on the films permeability. The permeability is calculated with the help of a new theory which takes into account the humidity of the foam. The coalescing foams are used to study the impact of the liquid fraction, the capillary pressure and the bubbles' size on coalescence in 2D foams. This study shows that an analysis of these parameters at the scale of the foam does not allow to conclude on their real impact on coalescence and that it is necessary to analyze these parameters at the scale of each bubble in order to differentiate their effects. To this end, I have developed a program that follow-up these bubbles and which isolate each event of coalescence. There is a critical liquid fraction above which there is no more coalescence because the films between the bubbles no longer exist. In 3D, I began by establishing link between surface and volume liquid fractions giving a new method to measure the volume liquid fraction from an image of the surface. Then, I studied coarsening of 3D wet foams. One of the goals was to show the transition between the dry regime where le mean radius increase with time as a power law, with an exponent β in the dry regime β =1/2 and the very wet regime where β = 1/3. I carried out multiple light scattering experiments on confined foams and found the values of $eta$ to be near $1/2. Nevertheless, the measurement of β in very wet foams wasn't possible. To counter this, the "Soft Matter Dynamics" project aims to study the coarsening and bubbles' rearrangement of wet foams in microgravity in the international space station (ISS). My involvement in this project consisted in the realisation of cell filling and cleaning protocols in collaboration with Airbus, the creation of a program for the automatisation of the bubbles' size measurement and the participation in a test campaign on the foam generation system.

Les mousses d’émulsions sont des systèmes complexes de bulles d’air et de gouttes d’huile dans l’eau. Ils sont omniprésents dans la vie de tous les jours mais leur stabilité n’est encore pas parfaitement comprise. L’objet de cette thèse expérimentale est de caractériser les propriétés rhéologiques et de comprendre les mécanismes de vieillissement de tels matériaux. Dans une première partie, j’étudie l’élasticité d’émulsions attractives dans lesquelles ont vient inclure des bulles d’air. Mes résultats montrent que si les bulles d’air sont plus rigides que l’émulsion (haute pression de Laplace), les inclusions ont pour effet le renforcement de la matrice élastique et ce phénomène peut être décrit par des modèles micromécanique. Toutefois, si les bulles sont trop petites, les prédictions ne fonctionnent plus et la restructuration des bulles et des gouttes a pour effet d’augmenter davantage le module élastique. Dans une seconde partie, j’étudie la déstabilisation de ce même système modèle par la gravité menant des études systématiques du crémage de l’émulsion et du drainage dans la mousse. J’ai monté une expérience permettant de quantifier temporellement les fractions liquides des trois fluides que sont l’huile, l’eau et l’air et décelé trois régimes dans le drainage des mousses d’émulsions. Dans un troisième temps, j’étudie le mûrissement dans les fluides visqueux dans un cas de figure modèle où la perméabilité des films et la diffusion du gaz sont négligeables afin de mettre en évidence à la fois théoriquement et expérimentalement des temps caractéristiques de relaxation des bulles en fonction de la rhéologie de la phase continue. Ces travaux de thèses permettraient de pouvoir fabriquer des mousses très stables ou des émulsions moussées plus légères et plus rigides. / Foamed emulsions are complex systems composed of air bubbles and oil droplets in water. They are ubiquitous in our daily life but their stability is still poorly understood. The aim of this experimental thesis is to characterize the rheological properties and to understand the mechanisms of ageing of such materials. In the first part, I study the elasticity of attractive emulsions in which air bubbles are included. Our results show that if the air bubbles are slightly stiffer than the emulsion (high Laplace pressure), one can observe a volume fraction dependent reinforcement of the elastic matrix that can be fully described by micromechanical models. However, below a certain size, the predictions don’t work anymore and air bubbles can restructure the emulsion and further increase its elastic modulus. In the second part, I study the gravity driven destabilization of the same system through systematic experiments of emulsion creaming and drainage of the foam. I set up an experiment to quantify the evolution of the three fluid fractions (oil, water and air). I show the existence of three drainage regimes with foamed emulsions. Thirdly, I study bubble coarsening in viscous fluids using a model experiment where the permeability of the films and the diffusion of the gas can be neglected. Combining theoretical predictions and experiments, I estimate the characteristic relaxation times that are due to the rheology of the continuous phase. These results could give insights for manufacturing new materials such as very stable foams or light and rigid foamed emulsions.

In most finite element analysis, a uniform mesh is not the optimum way to model the problem. Mesh adaptation is the ability to modify a finite element model to include regions of the mesh with higher and lower node density. Mesh adaptation has received extensive study in both computational mechanics and computer graphics to increase the resolution or accuracy of the solution in specific areas. The algorithm developed in this thesis, the Automated Quadrilateral Coarsening by Ring Collapse (AQCRC) algorithm, provides a unique solution to allow conformal coarsening of both structured and unstructured quadrilateral finite element meshes. The algorithm is based on dual chord operations and dual chord removal. The AQCRC algorithm follows six steps: 1) input of a coarsening region and factor, 2) selection of coarsening rings, 3) improvement of mesh quality, 4) removal of coarsening rings, 5) mesh clean-up and 6) coarsening iterations. Examples are presented that show the application of the algorithm.

This thesis presents a new approach to conformal all-quadrilateral mesh adaptation. In finite element modeling applications, it is often desirable to modify the node density of the mesh; increasing the density in some parts of the mesh to provide more accurate results, while decreasing the density in other parts to reduce computation time. The desired node density is typically determined by a sizing function based on either the geometry of the model or the results of a finite element solution. Although there are numerous mesh adaptation methods currently in use, including initial adaptive mesh generation, node redistribution, and adaptive mesh refinement, there are relatively few methods that modify the mesh density by adding and removing mesh elements, and none of these guarantee a conformal, all-quadrilateral mesh while allowing general coarsening. This work introduces a new method that incorporates both conformal refinement and coarsening strategies on an existing mesh of any density or configuration. Given a sizing function, this method modifies the mesh by combining existing template based quadrilateral refinement methods with recent developments in localized quadrilateral coarsening and quality improvement into an automated mesh adaptation routine.

No description available.

No description available.

Selective dissolution also known as dealloying is a corrosion process in which one component of a binary alloy system is selectively removed through an electrochemically controlled process which leads to the formation of a porous metal "sponge" with a porosity that is completely interconnected and random in direction. Nanoporous metals are desirable since they have larger surface areas than an equal volume of non-porous material. Because of their enormous surface area per volume, these highly porous metal electrodes are superior materials for high surface area applications such as in biomedical devices, microfilters and catalysts. Understanding the kinetic processes governing the development of porosity during dealloying and having ability to change the electrochemical conditions will allow us to better control over the average ligament size and distribution in porosity. The basic kinetic processes involved in the formation of these structures are related to such issues as environmental effects and electrochemical conditions on diffusion, microscopic coarsening phenomenon at room temperature and elevated temperatures, alloy passivation, and Gibbs-Thomson effects. The average pore size and distribution was found to depend on the electrolyte composition, dealloying rate, applied potential and time. The porosity was found to significantly coarsen at room temperature during the dealloying process and this coarsening was highly dependent on the applied potential. It is showed that the commonly accepted measurement of the critical potential for alloy dissolution calculated based on extrapolation of anodic polarization data results in an overestimation of this quantity. A series of constant applied potential experiments prove to be a more accurate method for critical potential determination. / Ph. D.

An optical fiber has been developed that can be used as a chemical gas sensor. Fabrication of the optical fiber produces a fiber that has a solid core with a porous cladding. The porous cladding region is made from a spinodally phase separable glass where the secondary phase is removed through dilute acid leaching. A non-phase separable glass composition is used for the core region. The properties of the phase separable glass are dependent on the processing conditions and the thermal history of the glass after the porosity has been achieved. Investigation of how processing conditions affected the pore structure was conducted to determine what pore characteristics are achievable for the glass composition used. Phase separation temperature, removal of silica gel deposited in the pores, and the post fabrication heat treating were used as experimental processing conditions. A maximum useable average pore size of approximately 29 nm was achieved. Maximum pore volume in the experimental groups was 0.4399 cc/g. Most heat treatments of the porous glass caused consolidation of the pore structure, with some conditions producing pore coarsening. / Master of Science