In the near future, Gen II, III, and IV nuclear reactors will be in operation. UO2 is a common fuel for reactors in each of these generations and molten salts are used as coolant/fuel in Gen IV molten salt reactors. This thesis investigates potential ways to measure thermal conductivity for these materials: Raman thermometry for UO2 and a needle probe for molten salts. Four Raman thermometry techniques are investigated in this thesis: The Two Laser Raman (TLR), Time Differential Domain Raman (TDDR), Frequency Resolved Raman (FRR), and Frequency Domain Raman (FDR). The TLR is a steady state method used with a thin film. The TDDR and FRR are both time domain methods used with thin cantilever samples. The FDR is a frequency domain method used with a thermally thick sample. Monte Carlo like simulations are performed for each technique. In the simulations, the affect introduced uncertainty has on the measurement of thermal conductivity and thermal diffusivity is measured. From the results, it is recommended that the TLR should be used for measuring thermal conductivity and the FRR used for measuring thermal diffusivity. The TDDR and FDR were heavily affected by the uncertainty which resulted in inconsistent measured thermal properties. For measuring the thermal conductivity of molten salt, a needle probe was designed and manufactured to withstand the corrosive environment found in using molten salts. The probe uses modulated joule heating and measures the temperature rise in a thermocouple. The phase delay and temperature amplitude of the thermocouple are used in determining the thermal conductivity. A new thermal quadrupole based analytical solution, which takes into consideration convection and radiation, to the temperature rise of the probe is presented. The analytical solution is verified using a numerical solution found using COMSOL. Preliminary data was obtained with the probe in water.

Energy needs are projected to continue to increase in the coming decades, and with the drive to use more clean energy to combat climate change, nuclear energy is poised to become an important player in the energy portfolio of the world. Due to the unique nature of nuclear energy, it is always vital to have safe and efficient generation of that energy. In current light water reactors, the most common fuel is uranium dioxide (UO2), an oxide ceramic. There is also ongoing research examining uranium-based based metallic fuels, such as uranium-molybdenum (U-Mo) fuels with low uranium (U) enrichment for research reactors as part of a broader effort to combat nuclear proliferation, and uranium-zirconium-based fuels for Generation IV fast reactors. Each nuclear fuel has weaknesses that need to be addressed for safer and more efficient use. Two major challenges of using UO¬2 are the fission gas (e.g. xenon) release and the decreasing thermal conductivity with increasing burnup. In UMo alloys, the major weakness is the breakaway swelling that occurs at high fission densities. The challenges presented by both fuel types are heavily impacted by microstructure, and several studies have identified that the initial microstructure of the fuel in particular (e.g. initial grain size and grain aspect ratio) plays a large role in determining when and how quickly these processes occur. Thus, knowledge of how such initial microstructures evolve is paramount in having stable and predictable fission gas release and thermal conductivity decrease (in UO2) and fuel swelling (in UMo alloys). Mobility is a critical grain boundary (GB) property that impacts microstructural evolution. Existing literature examines GB mobility for a few specific boundaries but does not (in general) identify the anisotropy relationships that this property has. This work first examined the anisotropy in GB mobility, specifically identifying the anisotropy trend for the low-index rotation axes for tilt GBs in BCC γ U, and fluorite UO2 via molecular dynamics simulation. GB mobility is calculated using the shrinking cylindrical grain method, which uses the capillary effect induced by the GB curvature to drive grain growth. The mobilities are calculated for different rotation axes, misorientation angles, and temperatures in these systems. The results indicated that the density of the atomic plane perpendicular to the (tilt) GB plane (which is also perpendicular to the rotation axis) significantly impacts which GB rotation axis has the fastest boundaries. Specifically, the atomic plane that has a higher density tends to have a faster mobility, because it is more efficient for atoms moving across the GB along such planes. For example, for body-centered cubic materials, the <110> tilt GBs are determined to have the fastest mobilities, while face-centered cubic (FCC) and FCC-like structures such as fluorite have <111> tilt GBs as the fastest. Knowledge of GB mobility and its anisotropy in pure materials is helpful as a baseline, but real materials have solutes or impurities (both intentionally and unintentionally) which are known to affect GB mobility by processes such as solute drag and Zener pinning. Additionally, in reactors, nuclear fission can produce many fission products, each of which acts as an additional impurity that will interact with the GB in some way. Because the initial microstructure and its subsequent evolution are vital for addressing the challenges of using nuclear fuel as described above, knowledge of the impacts of these impurities on GB mobility is required. Therefore, this work examined the impact of solutes and impurities on GB mobility and its anisotropy. In particular, the solute effect was examined using the UMo alloy system, while the impurity effect was examined using Xe (a very common fission product) in the γ U, UMo, and UO2 systems. It is found that both Mo and Xe can cause a solute drag effect on GB mobility in the γ U system, with the effect of Xe being stronger than Mo at the same solute/impurity concentration. Xe also causes a solute drag effect in UO2, though the magnitude of the effect is interatomic-potential-dependent. The mobility anisotropy trend was found to disappear at high solute and impurity concentrations in the metallic U and UMo systems but was largely unaffected in the UO2 system. These results not only increase our fundamental understanding of GB mobility, its anisotropy, and solute/impurity drag effects, but also can be used as inputs for mesoscale simulations to examine polycrystalline grain growth with anisotropic GB mobility and in turn examine how the fuel performance parameters change with these properties. / Doctor of Philosophy / Worldwide, energy needs continue to increase each year. Concerns related to climate change have led to an increased emphasis on renewable energies such as solar and wind, but the limitations of these resources prevent them from being the only energy sources. Nuclear energy is uniquely positioned to address several energy concerns: it is clean (no carbon emissions and air pollution), reliable (for example, 24/7 energy production, independent of weather), and energy-dense (one kilogram of fissile uranium provides roughly the same amount of energy as 3000 metric tons of coal). Currently, nuclear energy provides roughly 20% of the energy of the United States, but future predictions show a decrease in the total share of energy generation due to aging systems and a limited number of new reactors being built. The safety and efficacy of existing and future reactors are among the primary concerns for being able to allow nuclear energy to increase its energy share. To determine the safety and efficacy of new reactor designs, a computer simulation tool called fuel performance modeling has been used over the last few decades. This tool requires several material properties as input, one of which is how the nuclear reactor fuel microstructure changes based on a variety of conditions. A significant process contributing to microstructural change is grain growth. Grains (crystallites that make up the whole material) meet at interfaces called grain boundaries (GBs), and these GBs have two properties that largely determine how grain growth occurs: energy and mobility. Significant effort is being put into understanding these properties and their anisotropy, or how they change based on the GB character which is the relative mismatch between the two grains. This work contributes additional understanding of GB mobility anisotropy in two nuclear fuels: uranium dioxide (UO2, the primary fuel in current reactors) and a uranium-molybdenum (UMo) alloy (the primary fuel for newer research reactors). In particular, computer simulation is used to determine GB mobility for several unique GB systems. It is found that for pure nuclear fuels, GB mobility anisotropy is largely determined by which atomic plane has the highest density perpendicular to the GB. When the fuel is no longer pure (through the addition of alloying elements or other impurities) the anisotropy changes significantly in UMo fuels, such that at high concentrations of solute or impurities there is little to no anisotropy, while very little change is observed in the anisotropy in UO2.

Nuclear energy is a proven large-scale, emission-free, around-the-clock energy source. As part of improving the nuclear energy efficiency and safety, a significant amount of effort is being expended to understand how the microstructural evolution of nuclear fuels affects the overall fuel performance. Grain growth is an important aspect of microstructural evolution in nuclear fuels because grain size can affect many fuel performance properties. In this work, the anisotropy of grain boundary energy and mobility, which are two important properties for grain growth, is examined for the light water reactor fuel uranium dioxide (UO₂) by molecular dynamics simulations. The dependence of these properties on both misorientation angle and rotation axis is studied. The anisotropy in grain boundary energy is found to be insignificant in UO₂. However, grain boundary mobility shows significant anisotropy. For both 20º and 45º misorientation angles, the anisotropy in grain boundary mobility follows a trend of M₁₁₁>M₁₀₀>M₁₁₀, consistent with previous experimental results of face-centered-cubic metals. Evidences of grain rotation during grain growth are presented. The rotation behavior is found to be very complex: counterclockwise, clockwise, and no rotation are all observed. / M.S. / Energy needs in the world increase year after year. As part of the effort to address these increasing needs, an increasing effort is needed to study each aspect of energy generation. For energy generated via nuclear fission, i.e., nuclear energy, many things need to be understood to gain maximum efficiency with maximum safety. At the core of a nuclear reactor, transport of energy generated by nuclear fission is heavily dependent on the microscopic structure (microstructure) of the materials being used as fuel. Thus, this work examines the microstructure of the most common nuclear fuel, uranium dioxide (UO₂). The microstructure changes based on at least two properties: grain boundary energy, and grain boundary mobility. This work examines how these properties change based on the orientation of individual crystallites within the polycrystalline material. An additional aspect of microstructural evolution, namely grain rotation, is briefly discussed.

Le retraitement du combustible nucléaire irradié passe par une étape de séparation de l’uranium, du plutonium et des produits de fission qui le constituent, notamment par une étape de dissolution en milieu nitrique. Dans une démarche d’amélioration continue et pour optimiser le procédé quel que soit le combustible, il est nécessaire de comprendre les phénomènes physico-chimiques, cinétique et hydrodynamiques mis en jeu lors de la dissolution, pour permettre une modélisation de ce procédé à des fins de prévision. L’état de l’art ne permet de donner que des indications limitées car il repose sur des études macroscopiques dans des réacteurs de plusieurs centaines de millilitres. Les conclusions qui peuvent en être tirées sont donc soumises à la superposition de phénomènes microscopiques liés à la complexité du milieu nitrique, à des solides à dissoudre dont la composition et plus généralement la nature sont mal définies. Il est donc nécessaire de passer par une autre démarche qui consiste à décomposer et analyser les différents processus mis en jeu. Un modèle mettant en œuvre un couplage entre hydrodynamique et cinétique de dissolution d’un solide en présence d’espèces autocatalytiques est alors proposé. Ce modèle a permis de mettre en évidence la nécessité de réaliser des observations des concentrations des espèces au niveau de l’interface réactionnelle. Un réacteur miniaturisé a alors été conçu, et des expériences ont été menées sur des billes de cuivre, simulant le combustible, et ont permis d’obtenir de premières observations de bulles de gaz formées en cours de dissolution. Une méthode originale de suivi du pH in-situ au niveau de l’interface a été mise au point : un marqueur fluorescent a permis de visualiser les acidités in-situ et une cartographie du pH a pu être dressée en cours de dissolution, de même qu’une visualisation directe des processus de transfert avec mesure des couches-limites de diffusion. Cette méthode a enfin pu être transposée en zone nucléarisée sur du dioxyde d’uranium et a conduit à la compréhension et la modélisation du procédé de dissolution en milieu nitrique. / The reprocessing of irradiated nuclear fuel passes through a stage of separation of uranium, plutonium and fission products by dissolution in nitric acid. To be able to optimize the process regardless of the fuel used, it is necessary to understand physical and chemical phenomena, kinetics and hydrodynamic parameters involved in the process, to allow its modelling and to be able to forecast behaviours during the operation. The state of the art can only provide limited guidance because it is based on macroscopic studies in reactors of hundreds of millilitres. The conclusions that can be drawn are therefore subject to the overlay of microscopic phenomena related to the complexity of the nitric mid, and to the composition and nature of the solid to dissolve that are generally poorly defined. It is therefore necessary to use another approach which is to separate and analyze the various processes involved. A model implementing a coupling between hydrodynamics and kinetics involved in the dissolution of a solid in the presence of autocatalytic species is then proposed. This model was used to highlight the need for observations of the concentrations of the species at the level of the reactive interface. A miniaturized reactor was designed, and experiments were conducted on copper beads, simulating the fuel, and provided initial observations of gas bubbles formed during dissolution. A novel method for monitoring pH in-situ at the level of the interface has been developed: a fluorescent marker enabled to visualize in-situ acidity and pH mapping during dissolution, and a direct visualization of the transfer process with diffusion layers. This method could be transposed in the nuclear area on uranium dioxide and has led to the understanding and modelling of the process of dissolution in nitric environment.

Dissolution of the UO2 matrix is one of the potential routes for radionuclide release in a future deep geological repository for spent nuclear fuel. This doctoral thesis focuses on interfacial reactions of relevance in radiation-induced dissolution of UO2 and is divided in two parts: In the first part, we sought to explore the effects of solid phase composition: The impact of surface stoichiometry on the reactivity of UO2 towards aqueous radiolytic oxidants was studied. H2O2 reacts substantially faster with stoichiometric UO2 than with hyperstoichiometric UO2. In addition, the release of uranium from stoichiometric UO2 is lower than from hyperstoichiometric UO2. The behavior of stoichiometric powder changes with exposure to H2O2, approaching the behavior of hyperstoichiometric UO2 with the number of consecutive H2O2 additions. The impact of Gd-doping on the oxidative dissolution of UO2 in an aqueous system was investigated. A significant decrease in uranium dissolution and higher stability towards H2O2 for (U,Gd)O2 pellets compared to standard UO2 was found. In the second part, we sought to look at the effect of external factors: The surface reactivity of H2 and O2 was studied to understand the overall oxide surface reactivity of aqueous molecular radiolysis products. The results showed that hydrogen-abstracting radicals and H2O2 are formed in these systems. Identical experiments performed in aqueous systems containing UO2 powder showed that the simultaneous presence of H2 and O2 enhances the oxidative dissolution of UO2 compared to a system not containing H2. The effect of groundwater components such as bentonite and sulfide on the oxidative dissolution of UO2 was also explored. The presence of bentonite and sulfide in water could either delay or prevent in part the release of uranium to the environment. The Pd catalyzed H2 effect is more powerful than the sulfide effect. The poisoning of Pd catalyst is not observed under the conditions studied. / <p>QC 20170421</p>

Le dioxyde d'uranium (UO2) est le combustible nucléaire le plus largement répandu dans le monde pour alimenter les centrales nucléaires et plus particulièrement les réacteurs à eau pressurisée (REP). En réacteur, la fission des atomes d'uranium crée des produits de fission et des défauts ponctuels dans le matériau combustible. La compréhension de l'évolution de ces dégâts d'irradiation nécessite une approche de modélisation multi-échelle, de l'échelle de la pastille combustible à l'échelle atomique. Nous avons utilisé une méthode de calcul de structure électronique (DFT), pour modéliser les dégâts d'irradiation dans UO2 à l'échelle atomique. Un terme d'interaction Coulombienne de type Hubbard est ajouté au formalisme de la DFT standard pour prendre en compte les fortes corrélations des électrons 5f dans l'UO2. Cette méthode a été utilisée pour étudier les défauts ponctuels dans différents états de charge ainsi que l'incorporation et la diffusion du krypton dans le dioxyde d'uranium. Cette étude nous a permis d'obtenir des données clés pour les modèles aux échelles supérieures mais aussi pour interpréter des résultats expérimentaux. En parallèle de cette étude, trois pistes d'amélioration de l'état de l'art des calculs pour la description de l'UO2 ont été explorées : la prise en compte du couplage spin-orbite, l'application de fonctionnelles permettant la prise en compte des interactions non locales telles que les interactions de van der Waals importantes pour les gaz rares et l'utilisation de la théorie de champ dynamique moyen (Dynamical Mean Field Theory) combinée à la DFT afin de prendre en compte les corrélations dynamiques des électrons 5f. / Uranium dioxide (UO2) is worldwide the most widely used fuel in nuclear plants in the world and in particular in pressurized water reactors (PWR). In-pile the fission of uranium nuclei creates fission products and point defects in the fuel. The understanding of the evolution of these radiation damages requires a multi-scale modelling approach of the nuclear fuel, from the scale of the pellet to the atomic scale. We used an electronic structure calculation method based on the density functional theory (DFT) to model radiation damage in UO2 at the atomic scale. A Hubbard-type Coulomb interaction term is added to the standard DFT formalism to take into account the strong correlations of the 5f electrons in UO2. This method is used to study point defects with various charge states and the incorporation and diffusion of krypton in uranium dioxide. This study allowed us to obtain essential data for higher scale models but also to interpret experimental results. In parallel of this study, three ways to improve the state of the art of electronic structure calculations of UO2 have been explored: the consideration of the spin-orbit coupling neglected in current point defect calculations, the application of functionals allowing one to take into account the non-local interactions such as van der Waals interactions important for rare gases and the use of the Dynamical Mean Field Theory combined to the DFT method in order to take into account the dynamical effects in the 5f electron correlations.

Le dioxyde d'uranium irradié (UO2), combustible nucléaire des réacteurs à eau pressurisée, contient deux populations de cavités saturées par des gaz de fission : i. des cavités intergranulaires plutôt lenticulaires, dont la taille varie de quelques dizaines à plusieurs centaines de nanomètres, ii. des cavités intragranulaires plutôt sphériques, dont la taille est de l'ordre du nanomètre. Des travaux récents ont montré qu'il existe un effet de surface à l'échelle des cavités nanométriques qui modifie le comportement élastique effectif du combustible. Ce travail vise à proposer un modèle micromécanique analytique capable de tenir compte de cette microstructure hétérogène ainsi que de l'effet de surface afin de décrire le comportement élastique macroscopique de l'UO2 irradié. La démarche mise en oeuvre est fondée sur une modélisation multi-échelles et s'appuie sur des techniques d'homogénéisation en mécanique des matériaux. L'UO2 irradié est décrit comme un matériau poreux contenant des nanocavités sphériques (cavités intragranulaires) et sphéroïdales (cavités intergranulaires), sous pression et orientées aléatoirement. L'effet de surface présent à l'échelle nanométrique est pris en compte via un modèle d'interface imparfaite cohérente entre la matrice et les cavités. Un modèle original fondé sur l'approche par motifs morphologiques représentatifs a été développé afin de décrire le comportement élastique effectif de ce milieu hétérogène. Le modèle analytique proposé repose sur des hypothèses simplificatrices dont la pertinence est évaluée à partir de simulations numériques par éléments finis qui s'appuient sur une formulation spécifique afin de tenir compte de la présence d'interfaces imparfaites cohérentes. / The irradiated uranium dioxide (UO2), which is the nuclear fuel of pressurized water reactors, contains two populations of cavities saturated by fission gaz: i. intergranular cavities almost lenticular in shape whose size ranges between few tens to several hundred nanometers, ii. intragranular cavities, almost spherical in shape whose size is of the order of the nanometer. Recent studies have shown the existence of a surface effect at the scale of nanometric cavities, which influences the effective elastic behavior of the nuclear fuel. In this work, an analytical micromechanical model, which is able to take into account this heterogeneous microstructure and the surface effect at the nanometric scale, is proposed to describe the macroscopic behavior of the irradiated UO2. The approach is based on a multiscale modeling and homogenization techniques in mechanics of materials. The irradiated UO2 is described as a porous media, which contains pressurized spherical nanocavities (intragranular cavities) and randomly oriented pressurized spheroidal cavities (intergranular cavities). The surface effect is taken into account with imperfect coherent interfaces between the matrix and the cavities. A novel model based on the morphologically representative pattern approach has been developed to describe the effective elastic behavior of this heterogeneous medium. The proposed model relies on assumptions whose relevance is evaluated with finite element simulations which require a specific formulation to take into account the imperfect coherent interfaces.

First-principles calculations such as density functional theory (DFT) employ numerical approaches to solve the Schrodinger equation of a system. Standard functionals employed to determine the cohesive system energy, specifically the local density and generalized gradient approximations (LDA and GGA), underestimate the correlation of 5f electrons to their ions in AO₂ systems (A=U/Pu/Np). The standard correction, the "Hubbard +U," causes the multidimensional energy surface to develop a large number of local minima which do not correspond to the ground state (global minimum). Because all useful energy values derived from DFT calculations depend on small differences between relatively large cohesive energies, comparing systems wherein one or more of the samples are not in the ground state has the potential to introduce large errors. This work presents an analysis of the fundamental issues of metastable states in both pure and binary AO₂ systems, investigates novel methods of handling them, and describes why current literature approaches which appear to work well for the pure compounds are not well-suited for systems containing multiple actinide species.

No description available.

Made available in DSpace on 2014-10-09T12:43:11Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T13:56:49Z (GMT). No. of bitstreams: 1 06476.pdf: 9927984 bytes, checksum: 071861dcaed4ce3370a5065fdd2ae525 (MD5) / Tese (Doutoramento) / IPEN/T / Instituto de Pesquisas Energeticas e Nucleares - IPEN/CNEN-SP