Ce travail s’inscrit dans le programme de recherche sur les réacteurs nucléaires de « génération IV » à caloporteur sodium. Un des verrous technologiques à lever concerne le contrôle de la corrosion des matériaux (matériaux de structure, gaines des combustibles, etc.) par le sodium liquide. Un des paramètres clé est la teneur en oxygène dissous dans le caloporteur. Cette thèse porte sur l’élaboration et la caractérisation de matériaux céramiques de type ThO2 dopé avec Y2O3 pour la réalisation de sondes potentiométriques à oxygène utilisables dans le sodium liquide. Des travaux ont été effectués et des sondes ont été testées par le passé, cependant les sondes présentaient généralement, à l’époque, une absence de fiabilité dans le temps. L’objectif de cette thèse est d’élaborer et de caractériser des électrolytes à base d’oxyde de thorium dopé à l’oxyde d’ytrium à l’aide de techniques de synthèse permettant de contrôler la pureté, la taille des grains, la compacité, etc. Pour mettre en place les protocoles expérimentaux une céramique modèle a été choisie, la cérine dopée à l’oxyde d’yttrium. Les processus de transport ont été étudiés à l’aide de la spectrocopie d’impédance. Une interprétation des phénomènes de blocage de la conduction ionique dans les deux céramiques étudiées en fonction de la pression partielle d’oxygène a pu être proposée. / This PhD thesis was prepared within the framework of the research program on "Generation IV" nuclear reactors with sodium as coolant. One of the main technological problem concerns the control of the corrosion processes of the materials (structural materials, fuel claddings, etc.) by liquid sodium. A key parameter is the dissolved oxygen content in the coolant. This thesis focuses on the development and characterization of ceramic materials based on ThO2 doped with Y2O3 for making potentiometric oxygen sensor used in liquid sodium. Work has been carried out and probes were tested in the past, however, the probes had at the time, a lack of reliability. The objective of this thesis is to develop and characterize electrolytes based on thorium oxide doped with yttrium oxide using specific synthesis techniques to control purity, grain size, compactness, etc. To develop experimental protocols a ceramic model has been chosen, i.e., yttria-doped ceria. Transport processes were studied using the impedance spectroscopy technique. An interpretation of the blocking phenomena of the ionic conduction in both ceramics as a function of the oxygen partial pressure has been given.

Minimization of HLW inventories and U consumption are key elements guaranteeing nuclear energy expansion. The integration of complex nuclear systems into a viable cycle yet constitutes a challenging multi-parametric optimization problem. The reactors and fuel cycle performance parameters may be strongly dependent on minor variations in the system's input data. Proven discrepancies in nuclear data evaluations could affect the validity of the system optimization metrics. This study first analyzes various advanced AP1000-VHTR fuel cycle scenarios by assessing their TRU destruction and their U consumption minimization capabilities, and by computing reactor performance parameters such as the time evolution of the effective multiplication factor keff, the reactors' energy spectrum or the isotopic composition/activity at EOL. The performance metrics dependence to prompt neutron fission spectrum discrepancies is then quantified to assess the viability of one strategy. Fission spectrum evaluations are indeed intensively used in reactors' calculations. Discrepancies higher than 10% have been computed among nuclear data libraries for energies above 8MeV for 235U. TRU arising from a 3wt% 235U-enriched UO2-fueled AP1000 were incinerated in a VHTR. Fuels consisting of 20%, 40% and 100% of TRU completed by UO2 were examined. MCNPX results indicate that up to 88.9% of the TRU initially present in a VHTR fueled with 20% of TRU and 80% of ThO2 were transmuted. Additionally, the use of WgPu instead of RgPu should reduce the daily consumption of 235U by 1.3 and augment core lifetime. To estimate the system metrics dependence to fission spectrum discrepancies and validate optimization studies outputs, the VTHR 235U fission spectrum distribution was altered successively in three manners. keff is at worst lowered by 1.7% of the reference value and the energy spectrum by 5% between 50meV and 2MeV when a significantly distorted fission spectrum tail is used. 233U, 236Pu and 237Pu inventories and activities are multiplied by 263, 523 and 34 but are still negligible compared to 239Pu mass or the total activity. The AP1000-VHTR system is in conclusion not dependent on the selected fission spectrum variations. TRU elimination optimization studies in AP1000-VHTR systems will be facilitated by freeing performance metrics dependency from 1 input parameter.