Este trabalho tem como objetivo estudar as modificações introduzidas, ao longo de sucessivas versões, nos modelos empíricos do programa computacional FRAPCON utilizado para a simulação do comportamento sob irradiação de varetas combustíveis de Reatores a Água Leve Pressurizada (Pressurized Water Reactor - PWR) em regime de estado estacionário e sob condições de alta queima. No estudo, foram analisados os modelos empíricos utilizados pelo FRAPCON e que são apresentados em sua documentação oficial. Um estudo bibliográfico foi conduzido sobre os efeitos da alta queima em combustíveis nucleares visando melhorar o entendimento dos modelos utilizados pelo FRAPCON nestas condições. Foram feitas simulações do comportamento sob irradiação de uma vareta combustível típica de um reator PWR utilizando as versões 3.3, 3.4 e 3.5 do FRAPCON. Os resultados apresentados pelas diferentes versões do programa foram comparados entre si de forma a verificar as consequências das mudanças de modelo nos parâmetros de saída do programa. Foi possível observar que as modificações introduzidas trouxeram diferenças significativas nos resultados de parâmetros térmicos e mecânicos da vareta combustível, principalmente quando se evoluiu da versão FRAPCON-3.3 para a versão FRAPCON-3.5. Nessa ultima versão, obteve-se menores temperaturas na vareta combustível, menores tensões e deformações no revestimento, menor espessura da camada de oxido formada no revestimento a altas queimas na vareta combustível. / This work studies the modifications implemented over successive versions in the empirical models of the computer program FRAPCON used to simulate the steady state irradiation performance of Pressurized Water Reactor (PWR) fuel rods under high burnup condition. In the study, the empirical models present in FRAPCON official documentation were analyzed. A literature study was conducted on the effects of high burnup in nuclear fuels and to improve the understanding of the models used by FRAPCON program in these conditions. A steady state fuel performance analysis was conducted for a typical PWR fuel rod using FRAPCON program versions 3.3, 3.4, and 3.5. The results presented by the different versions of the program were compared in order to verify the impact of model changes in the output parameters of the program. It was observed that the changes brought significant differences in the results of the fuel rod thermal and mechanical parameters, especially when they evolved from FRAPCON-3.3 version to FRAPCON-3.5 version. Lower temperatures, lower cladding stress and strain, lower cladding oxide layer thickness were obtained in the fuel rod analyzed with the FRAPCON-3.5 version.

If international agreements regarding the need to significantly reduce greenhouse gas emissions are to be met then there is a high probability that the shipping industry will have to reduce its greenhouse gas emissions. For emission reductions from ships greater than around 40% then alternatives to fossil fuels - such as nuclear energy - will very likely be required. Whilst nuclear powered ships have successfully operated at sea for a number of decades, these have been primarily naval systems (or derivatives of naval systems such as icebreakers) and a few demonstration projects using reactors with low power outputs. The operational requirement for large civilian vessels (for example high capacity factors and limited personnel) mean the naval and past demonstration reactor systems are ill-suited for use in the current fleet of commercial container ships. There have been relatively few studies performed addressing the likely requirements upon core design a marine reactor would have to meet. This study addresses those issues and also implements a Pressurised Water Reactor core design capable of achieving these requirements. Furthermore, in order to simplify reactor operation for a limited number of personnel on board, the chemical reactivity control system has been eliminated during power operation. This has resulted in a novel low power density core that does not require refuelling for 15 years. The neutronic and fuel performance behaviour of this system has been studied with conventional UO2 fuel and thorium-uranium oxide ((Th,U)O2) fuel. With respect to (Th,U)O2 fuel there has been limited analysis comparing the performance of key fuel characteristics, such as fission gas release and thermal conductivity, as a function of uranium content in (Th,U)O2 fuel and their impact on fuel behaviour. Furthermore, the performance of neutronic codes for modelling Th-232 and U-233 from a variety of experiments using modern nuclear data libraries (post 1990) is lacking. Both of these issues are addressed in this study. Whilst it is frequently stated that thorium-based oxide fuel is superior to UO2 fuel it was found that due to the sensitivity of thermal conductivity on temperature and uranium content this was not true for the core designed in this study. The (Th,U)O2 core showed no net economic benefits with respect to the UO2 core and it was found that the fuel performance of (Th,U)O2 fuel was worse than the UO2 fuel in the reactor designed here. The UO2 core design, however, was able to satisfactorily meet the majority of requirements placed upon the system.

Este trabalho tem como objetivo estudar as modificações introduzidas, ao longo de sucessivas versões, nos modelos empíricos do programa computacional FRAPCON utilizado para a simulação do comportamento sob irradiação de varetas combustíveis de Reatores a Água Leve Pressurizada (Pressurized Water Reactor - PWR) em regime de estado estacionário e sob condições de alta queima. No estudo, foram analisados os modelos empíricos utilizados pelo FRAPCON e que são apresentados em sua documentação oficial. Um estudo bibliográfico foi conduzido sobre os efeitos da alta queima em combustíveis nucleares visando melhorar o entendimento dos modelos utilizados pelo FRAPCON nestas condições. Foram feitas simulações do comportamento sob irradiação de uma vareta combustível típica de um reator PWR utilizando as versões 3.3, 3.4 e 3.5 do FRAPCON. Os resultados apresentados pelas diferentes versões do programa foram comparados entre si de forma a verificar as consequências das mudanças de modelo nos parâmetros de saída do programa. Foi possível observar que as modificações introduzidas trouxeram diferenças significativas nos resultados de parâmetros térmicos e mecânicos da vareta combustível, principalmente quando se evoluiu da versão FRAPCON-3.3 para a versão FRAPCON-3.5. Nessa ultima versão, obteve-se menores temperaturas na vareta combustível, menores tensões e deformações no revestimento, menor espessura da camada de oxido formada no revestimento a altas queimas na vareta combustível. / This work studies the modifications implemented over successive versions in the empirical models of the computer program FRAPCON used to simulate the steady state irradiation performance of Pressurized Water Reactor (PWR) fuel rods under high burnup condition. In the study, the empirical models present in FRAPCON official documentation were analyzed. A literature study was conducted on the effects of high burnup in nuclear fuels and to improve the understanding of the models used by FRAPCON program in these conditions. A steady state fuel performance analysis was conducted for a typical PWR fuel rod using FRAPCON program versions 3.3, 3.4, and 3.5. The results presented by the different versions of the program were compared in order to verify the impact of model changes in the output parameters of the program. It was observed that the changes brought significant differences in the results of the fuel rod thermal and mechanical parameters, especially when they evolved from FRAPCON-3.3 version to FRAPCON-3.5 version. Lower temperatures, lower cladding stress and strain, lower cladding oxide layer thickness were obtained in the fuel rod analyzed with the FRAPCON-3.5 version.

L'énergie nucléaire constitue aujourd'hui une partie importante de la production d'électricité dans le monde, notamment en France. Dans les réacteurs nucléaires, le combustible le plus utilisé est le dioxyde d'uranium. Dans cette thèse, nous nous intéressons plus particulièrement aux modifications de la structure cristalline du combustible irradié liées à l'augmentation de son taux de combustion. Nous avons confirmé que, pour des conditions proches du combustible irradié, une lacune de miscibilité existe dans le ternaire U-Nd-O. Comme (U,Nd)O2 est un matériau modèle du combustible, nous avons recherché l'existence d'une lacune de miscibilité pour le combustible irradié, qui serait alors considéré comme un point dans un pseudo diagramme de phases. Des mesures par diffraction des rayons X sur combustible irradié ont donné des résultats cohérents avec cette hypothèse. Fort de ce résultat nous proposons une nouvelle interprétation de l'évolution de la microstructure du combustible irradié en fonction du taux de combustion qui s'appuie sur l'existence de cette lacune de miscibilité. / The nuclear energy represents today an important fraction of electricity production in the world and especially in France. The most used nuclear fuel today is the uranium dioxide UO2. In this thesis, we have studied the crystallographic structure evolution of this material related to the increase of its burn-up.We have confirmed that, under conditions similar of those of irradiated nuclear fuel, a miscibility gap exists in the (U1-yNdy)O2 system. As (U1-yNdy)O2 system can be considered as a model of the fuel, we have search for the existence of a miscibility gap in the irradiated fuel, which would be considered as a ternary pseudo diagram de phases. XRD measurements of theses system give us results consistent with this hypothesis.Based on this evidence, we propose a new interpretation of the microstructure evolution of the irradiated fuel as a function of the burn-up.

O objetivo deste trabalho é verificar a validade e a acurácia dos resultados fornecidos pelos programas computacionais FRAPCON-3.4a e FRAPTRAN-1.4, utilizados no processo de simulação do comportamento de varetas combustíveis de reatores a água leve pressurizada PWR (Pressurized Water Reactor), sob situações operacionais de regimes permanente e transiente, em condições de alta queima (high burnup). Para realizar a verificação, foi utilizada a base de dados FUMEX-III, que fornece dados relativos a experimentos realizados com diversos tipos de combustíveis nucleares, submetidos a diversas condições operacionais. Através dos resultados obtidos nas simulações computacionais com os programas FRAPCON-3.4a e FRAPTRAN-1.4 e da sua comparação com os dados experimentais da base FUMEX-III, foi possível constatar que os programas empregados possuem um boa capacidade de predizer o comportamento operacional de varetas combustíveis de PWR em regime permanente a altas queimas e sob condição de transiente inicializado por reatividade (Reactivity Initiated Accident RIA). / The objective of this work is to verify the validity and accuracy of the results provided by the computer programs FRAPCON-3.4a and FRAPTRAN-1.4, used in the simulation process of the irradiation behavior of Pressurized Water Reactors (PWR) fuel rods in steady-state and transient operational conditions at high burnup. To perform the verification, the database FUMEX-III was used to provide data on experiments with different nuclear fuel types, under various operating conditions. Through the comparison of the computational simulation results of the programs FRAPCON-3.4a e FRAPTRAN-1.4 with the experimental data of the database FUMEX III, it was found that the computer programs used have good ability to predict the operational behavior of PWR fuel rods in high burnup steady-state conditions and under Reactivity Initiated Accident (RIA).

The swelling mechanisms of U3Si2 under neutron irradiation in reactor conditions are not unequivocally known. The limited experimental evidence that is available suggests that the main driver of the swelling in this material would be fission gases accumulation at crystalline grain boundaries. The steps that lead to the accumulation of fission gases at these locations are multiple and complex. However, gradually, the gaseous fission products migrate by diffusion. Upon reaching a grain boundary, which acts as a trap, the gaseous fission products start to accumulate, thus leading to formation of bubbles and hence to swelling. Therefore, a quantitative model of swelling requires the incorporation of phenomena that increase the presence of grain boundaries and decrease grain sizes, thus creating sites for bubble formation and growth. It is assumed that grain boundary formation results from the conversion of stored energy from accumulated dislocations into energy for the formation of new grain boundaries.This thesis attempts to develop a quantitative model for grain subdivision in U3Si2 based on the above mentioned phenomena to verify the presence of this mechanism and to use in conjunction with swelling codes to evaluate the total swelling of the pellet in the reactor during its lifetime.

O objetivo deste trabalho é verificar a validade e a acurácia dos resultados fornecidos pelos programas computacionais FRAPCON-3.4a e FRAPTRAN-1.4, utilizados no processo de simulação do comportamento de varetas combustíveis de reatores a água leve pressurizada PWR (Pressurized Water Reactor), sob situações operacionais de regimes permanente e transiente, em condições de alta queima (high burnup). Para realizar a verificação, foi utilizada a base de dados FUMEX-III, que fornece dados relativos a experimentos realizados com diversos tipos de combustíveis nucleares, submetidos a diversas condições operacionais. Através dos resultados obtidos nas simulações computacionais com os programas FRAPCON-3.4a e FRAPTRAN-1.4 e da sua comparação com os dados experimentais da base FUMEX-III, foi possível constatar que os programas empregados possuem um boa capacidade de predizer o comportamento operacional de varetas combustíveis de PWR em regime permanente a altas queimas e sob condição de transiente inicializado por reatividade (Reactivity Initiated Accident RIA). / The objective of this work is to verify the validity and accuracy of the results provided by the computer programs FRAPCON-3.4a and FRAPTRAN-1.4, used in the simulation process of the irradiation behavior of Pressurized Water Reactors (PWR) fuel rods in steady-state and transient operational conditions at high burnup. To perform the verification, the database FUMEX-III was used to provide data on experiments with different nuclear fuel types, under various operating conditions. Through the comparison of the computational simulation results of the programs FRAPCON-3.4a e FRAPTRAN-1.4 with the experimental data of the database FUMEX III, it was found that the computer programs used have good ability to predict the operational behavior of PWR fuel rods in high burnup steady-state conditions and under Reactivity Initiated Accident (RIA).

L’objectif de cette thèse est d’étudier et de reproduire les caractéristiques spécifiques de la microstructure du combustible nucléaire irradié à fort taux de combustion, appelée structure HBS (High Burnup Structure). Il s’agit d’étudier les différents paramètres pertinents impliqués dans la formation d’une telle structure, en évaluant leur importance, et en clarifiant leurs éventuelles synergies. Cet objectif a été réalisé en utilisant un système de modèle ultra simplifié, à savoir des monocristaux de dioxyde d’uranium (UO₂) irradiés par des ions de basse énergie (quelques centaines de keV) de Lanthane (La) ou de xénon (Xe) à une température de 773 K, correspondant à celle de la périphérie des véritables pastilles de combustible en réacteur. Les énergies et les masses des ions ont été choisies pour étudier la déstabilisation du solide en fonction de deux paramètres-clefs: (i) les collisions nucléaires élastiques et (ii) la contribution chimique de l'incorporation d'impuretés à forte concentration. Les deux espèces ont été choisies délibérément pour leurs solubilités très différentes dans le dioxyde d’uranium: les ions La sont solubles dans l'UO₂ jusqu’à de très fortes concentrations, tandis que les ions Xe sont insolubles. Les techniques de la Microscopie Électronique en Transmission (TEM) et de Spectrométrie de Rétrodiffusion Rutherford en canalisation RBS/C ont été conduites in situ couplée avec l’irradiation. Ces deux techniques utilisées pour visualiser, quantifier et fournir des informations concernant la fraction des défauts induits par l’irradiation et la formation des bulles, de cavités ou de précipités dans le solide. Les données de canalisation ont été analysées par simulation Monte-Carlo en supposant l’existence de deux catégories de défauts : (i) des atomes aléatoirement déplacés (RDA) et (ii) des distorsions des rangés atomiques (BC). L’évolution de la fraction de défaut de type RDA montre une forte augmentation entre 0.4 à 4.0 dpa (correspondant à une très faible concentration des ions implantés), indépendamment de la nature des ions. Elle est suivie par une saturation de la fraction de RDA pour les deux ions sur une large gamme d’irradiation quoi s’étend jusque 100 dpa. Une forte élévation de la fraction de RDA est observée en particulier pour les cristaux implantés avec des ions Xe pour une concentration élevée dépassant les 4%. En ce qui concerne l’évolution de BC, elle augmente fortement jusqu’à 4 dpa et sature ensuite deux ions La et Xe. Les résultats de microscopie électronique in situ montrent que des défauts identiques pour les deux ions induits par l’irradiation apparaissent, et présentent la même évolution en fonction de la fluence. Les différents défauts évoluent en fonction de la fluence : la première étape correspond à la formation de ‘black dots’ ; la deuxième étape est caractérisée par la formation de boucles puis de lignes de dislocations, qui évoluent finalement jusqu’à commencer à devenir moins différenciables; le processus de restructuration se poursuit et forme un réseau de dislocations enchevêtrées. Une forte densité de bulles de gaz, de taille nanométrique et avec un diamètre moyen de 2 nm est observée pour le cristal Xe implanté à une dose seuil de 4 dpa. Le couplage des deux techniques conduites in situ montre que la différence entre les valeurs à saturation des fractions RDA des deux ions, d’une part, et l'augmentation drastique de RDA à très forte concentration d'ions Xe implantés d’autre part peuvent être attribuées à : (i) la solubilité des ions La vis-à-vis des ions Xe, conduisant à la formation des bulles de gaz de taille nanométrique et (ii) la taille des espèces implantées dans la matrice UO₂, pour laquelle les atomes Xe insolubles ont un rayon atomique beaucoup plus grand que le rayon cationique des atomes U⁴⁺(les atomes La³⁺ ont un rayon atomique similaire à celui des atomes U⁴⁺), responsable de plus de contraintes supplémentaires dans la structure cristalline. / The aim of this thesis is to investigate and reproduce the specific features of the microstructure of the high burnup structure of the irradiated nuclear fuel and to explore the various relevant parameters involved in the formation of such a structure, in evaluating their importance, and in clarifying the synergies between them. This have been performed by using a very simplified model system – namely uranium dioxide single crystals- irradiated with low energy La and Xe ions at 773 K, corresponding to the temperature at the periphery of the genuine fuel. The energies and masses of bombarding ions were chosen to investigate the destabilization of the solid due to: (i) the elastic nuclear collisions and by (ii) the chemical contribution of implanting impurities at high concentrations by implanting different ions in UO₂, namely Xe and La, having very different solubility: La species are soluble in UO₂ while Xe ions are insoluble. In situ Transmission electron Microscopy (TEM) and in situ Rutherford Backscattering Spectrometry in the channeling mode (RBS/C), both techniques coupled to ion irradiation, were performed to visualize, quantify and provide information with respect to the fraction of radiation-induced defects and the formation of bubbles, cavities, or precipitates. The channeling data were analyzed afterwards by Monte Carlo simulations assuming two class of defects comprising (i) randomly displaced atoms (RDA) and (ii) bent channels (BC) defects. Regarding the RDA evolution, a sharp increase step appears from 0.4 to 4.0 dpa (corresponding to a low concentration of implanted ions) regardless of nature of ions followed by a saturation of the fraction of RDA for both ions over a wide range of irradiation. A sharp increase of RDA fraction is observed specifically for crystals implanted with Xe ions at a high concentration exceeding 1.5 % (corresponding to the dose of more than 125 dpa). Regarding the BC evolution, for both ions, the evolution shows an increase in the fraction of BC up to 4.0 dpa then the fraction of BC almost saturates for Xe and La ions. In situ TEM results show that a similar radiation-induced defects appear for both ions and the same evolution of defects as a function of the fluence is observed. The various defects evolved as a function of the fluence: starting from the black dot defects formation that were observed as a first type of defects created, then dislocation lines and loops appeared and evolved until they started to be become less distinguishable, the restructuring process continued by forming a tangled dislocation network. A high density of nanometer-sized gas bubbles with a mean diameter 2 nm were observed at room temperature for the Xe-implanted crystal at a threshold dose of 4 dpa. The coupling between both techniques (in situ RBS/C and TEM) demonstrates that the difference between the two plateaus of saturation between the two ions and the dramatic increase of RDA at high concentration of implanted Xe ions can be ascribed to: (i) the solubility of La compared to Xe ions leading to the formation of nanometer-sized gas bubbles and (ii) the size of implanted species in UO₂ matrix where insoluble Xe atoms have an atomic radius much larger than the cationic radius of U⁴⁺ atoms, (La³⁺ atoms have a similar atomic radius as U⁴⁺ atoms) responsible for more stress in UO₂ crystal.