Afin d’évaluer la sûreté d’ASTRID (Réacteur à Neutrons Rapide refroidi au sodium), les transitoires accidentels sont actuellement étudiés avec des codes de calculs déterministes coûteux en temps de calcul, comme SIMMER-III ou SAS-SFR. En complément de ces études, le CEA a entrepris le développement d’outils de calculs analytiques simulant les différents phénomènes physiques régissant ces transitoires. Ces outils permettent, compte-tenu de leur robustesse et des faibles temps de calculs, de prendre en compte par une approche probabiliste les incertitudes et d’analyser de manière statistique les résultats. Ce traitement s’avère en effet indispensable afin de tenir compte des incertitudes physiques et de la variabilité des scénarios de déroulement du transitoire accidentel. In fine, les études réalisées avec ce type d’outils, couplant une modélisation analytique de la physique à un traitement statistique des résultats, fourniront des informations quantitatives sur les marges de sûreté, vis-à-vis de critères donnés.Le développement et la validation de l’outil dédié aux transitoires de perte de débit primaire non protégé (ULOF - Unprotected Loss Of Flow), résultant du déclenchement des pompes primaires sans reprise de secours ni chute des barres de contrôle, fait l’objet de cette thèse. Cet outil a été nommé MACARENa (Modélisation de l'ACcident d'Arrêt des pompes d'un Réacteur refroidi au sodium). Au cours de cette thèse, seule la phase primaire de l’accident a été traitée.Le début de cette phase, enclenchée par la réduction du débit primaire, est gouverné par le couplage entre la thermohydraulique et les contre-réactions neutroniques. Le code MACARENa simule, selon les conditions initiales, l’établissement de la convection naturelle monophasique ou la stabilisation de l’ébullition dans la partie haute du cœur. Si l’écoulement est instable, l’excursion de débit conduisant à l’assèchement des aiguilles puis à leur dégradation est aussi modélisée. A la suite d’un tel transitoire, l’outil calcule la fusion et la relocalisation des gaines et du combustible ainsi que l’entraînement d’acier liquide par les vapeurs de sodium et le possible bouchage de l’assemblage par des matériaux resolidifiés, avant de suivre l’évolution de bains fondus qui conduisent à la rupture des tubes hexagonaux. Les mouvements de matériaux induisent aussi des effets neutroniques complexes qui sont traités dans la modélisation.Les modélisations effectuées pour construire l’outil MACARENa ont été validées sur des expériences à effets séparés (thermohydraulique ou de dégradation) et des résultats issus de code déterministes.Finalement, des études de propagation d’incertitudes et d’analyses de sensibilité sont réalisées avec cet outil afin d’illustrer son intérêt vis-à-vis de la démonstration de sûreté. / Within the framework of the Generation IV Sodium-cooled Fast Reactor (SFR) R&D program of CEA (French Commissariat à l’Energie Atomique et aux Energies Alternatives), safety in case of severe accidents is assessed.Such transients are usually simulated with mechanistic codes (such as SAS-SFR and SIMMER III). As a complement to these codes, which give reference accidental transient calculations, a new physico-statistical approach is currently followed by the CEA; its final objective being to derive the variability of the main results of interest for safety. This approach involves a fast-running description of extended accident sequences coupling physical models for the main phenomena to advanced statistical analysis techniques. It enables to perform a large number of simulations in a reasonable computational time and to describe all the possible bifurcations of the accident transient.In this context, this PhD work presents the physical tool (models and results assessment) dedicated to the initiation and primary phases of an Unprotected Loss Of Flow accident (i.e. until the end of sub-assemblies degradation and before large molten pools formation). The accident phenomenology during these phases is described and illustrated by numerous experimental evidences.It is underlined that the features of the new heterogeneous core concept (called CFV of the French ASTRID prototype) leads to different kinds of ULOF transients than those occurring in the previous past homogeneous cores (SuperPhenix, Phenix...). Indeed, its negative void effect drops the nuclear power when sodium heats-up and possibly boils. This enables three types of ULOF transients characterized by various core final states; the first two types leading to final coolable core states in natural circulation flow (the first one in single phase, the second one in stabilized two-phase flow) whereas the core undergoes a flow excursion followed by sub-assemblies degradation in the last type. In this study, a particular attention is paid to stabilize boiling occurrence which leads to minimize severe accident consequences.The phenomena occurring during the various ULOF transients are modelled in accordance to the level of details required to catch all the possible bifurcations of the transient. The tool coupled different (2D, 1D and 0D) models of thermics, thermo-hydraulics, core degradation (material melting and motions) and neutronics. The assumptions associated to these models are highlighted, discussed and validated. The physical tool capability of simulating the various realistic ULOF transients (without boiling, with stabilized boiling or flow excursion after boiling) is demonstrated by comparisons to experimental results (GR19, SCARABEE experiments) and to mechanistic simulations (CATHARE2 and SIMMER III).Parametric studies are then carried out on two variables: the fuel burn-up and the model of neutronic feedbacks. They underline the important influence of these parameters on the transient and the final core state. Finally, a preliminary sensitivity analysis (2000 simulations) is performed on 26 uncertain parameters (linked to initial core configuration, accident features, model uncertainties and radial nodalization). The variability of the final core state is underlined and quantified; only around 25% of cases lead to core degradation. The main influent parameters on transient phenomena are also identified, enabling to prioritize core design and safety studies.In the future, this tool will be used for safety-informed design and stability analyses of fast reactor systems, allowing to emphasize the main dominant phenomena and trends of significance for safety assessment.

The sensitivity of two operational output parameters, criticality and isotopic composition during burnup, to specific design and operational reactor parameters in a Sodium Cooled Fast Reactor, is investigated. The computational simulation tool Serpent is used. The parameters varied include Uranium enrichment, Plutonium content, rod thickness, fuel temperature, and sodium density. In burnup, the development of the fraction of fissile isotopes, isotopes used for measurements, the isotopic composition of Plutonium, and isotopes that complicate fuel reprocessing is displayed. A surrogate model, optimized for use in determining how criticality develops between data points, is used. The results are displayed as plots created in Matlab. The results are discussed, with a focus on how large an effect varying different parameters have on different outputs related to the reactor's operation. It is concluded that the Plutonium content has the largest effect on the isotopic composition and that, based on the performed simulations, MOX fuel is potentially safer than Zirconium alloy fuel in a practical setting.

In this thesis, the feasibility to use a medium sized sodium cooled fast reactor fully loaded with MOX fuel for efficient transmutation of americium is investigated by simulating the safety performance of a BN600-type fast reactor loaded with different fractions of americium in the fuel, using the safety parameters obtained with the SERPENT Monte Carlo code. The focus is on americium mainly due to its long-term contribution to the radiotoxicity of spent nuclear fuel and its deterioration on core's safety parameters. Applying the SAS4A/SASSYS transient analysis code, it is demonstrated that the power rating needs to be reduced by 6% for each percent additional americium introduction into the reference MOX fuel, maintaining 100 K margin to fuel melting, which is the most limiting failure mechanism.Safety analysis of a new Accelerator Driven System design with a smaller pin pitch-to-diameter ratio comparing to the reference EFIT-400 design, aiming at improving neutron source efficiency, was also performed by simulating performance for unprotected loss of flow, unprotected transient overpower, and protected loss-of-heat-sink transients, using neutronic parameters obatined from MCNP calculations. Thanks to the introduction of the austenitic 15/15Ti stainless steel with enhanced creep rupture resistance and acceptable irradiation swelling rate, the suggested ADS design loaded with nitride fuel and cooled by lead-bismuth eutectic could survive the full set of transients, preserving a margin of 130 K to cladding rupture during the most limiting transient. The thesis concludes that efficient transmutation of americium in a medium sized sodium cooled fast reactor loaded with MOX fuel is possible but leads to a severe power penalty. Instead, preserving transmutation rates of minor actinides up to 42 kg/TWhth, the suggested ADS design with enhanced proton source efficiency appears like a better option for americium transmutation. / QC 20110318

No description available.

No description available.

Safety and reliability are one of the key objectives for future Generation IV nuclear energy systems. The neutron flux monitoring system forms an integral part of the safety design of a nuclear reactor and must be able to detect any irregularities during all states of reactor operation. The work in this thesis mainly concerns the detection of in-core perturbations arising from unwanted movements of control rods with in-vessel neutron detectors in a sodium-cooled fast reactor. Feasibility study of self-powered neutron detectors (SPNDs) with platinum emitters as in-core power profile monitors for SFRs at full power is performed. The study shows that an SPND with a platinum emitter generates a prompt current signal induced by neutrons and gammas of the order of 600 nA/m, which is large enough to be measurable. Therefore, it is possible for the SPND to follow local power fluctuations at full power operation. Ex-core and in-core detector locations are investigated with two types of detectors, fission chambers and self-powered neutron detectors (SPNDs) respectively, to study the possibility of detection of the spatial changes in the power profile during two different transient conditions, i.e. inadvertent withdrawal of control rods (IRW) and one stuck rod during reactor shutdown (OSR). It is shown that it is possible to detect the two simulated transients with this set of ex-core and in-core detectors before any melting of the fuel takes place. The detector signal can tolerate a noise level up to 5% during an IRW and up to 1% during an OSR.

Les barres de commande du futur démonstrateur de réacteur à neutrons rapides refroidi au sodium (RNR – Na) nommé ASTRID sont constituées de pastilles de B4C enfermées dans une gaine en acier inoxydable AIM1 (15Cr-15Ni-0,4Ti). En service, les pastilles de B4C sont plongées dans le sodium liquide à une température allant de 500 à 600 °C. Les retours d’expérience des RNR - Na ont mis en évidence que la durée de vie des barres de commande était limitée par leur cinétique de carburation. Cependant, un phénomène de boruration des gaines a été observé lors d’essais réalisés « hors réacteur / hors irradiation ». Afin de maîtriser la durabilité des barres de commandes, il est donc nécessaire d’évaluer précisément la nature de l’interaction entre les gaines en acier et le B4C dans le sodium liquide. Ainsi, deux campagnes d’essai ont été menées : 1. Trois aciers inoxydables (AIM1, 316L et EM10) ont été exposés dans du sodium liquide fortement carburant (ac > 1) à 500, 600 et 650 °C. 2. Les mêmes nuances d’aciers ont été exposées dans du sodium liquide contenant de la poudre de B4C en excès à 500 et 600 °C. La première campagne a été réalisée pour avoir une meilleure compréhension des mécanismes et des cinétiques de carburation des barres de commande. Tout d’abord, l’état de carburation a été caractérisé finement au moyen de différentes techniques d’analyse (microsonde de Castaing, diffraction des rayons X du rayonnement synchrotron, microscopie électronique en transmission). Ensuite, la cinétique de carburation a été simulée à l’aide d’un modèle analytique simplifié de la carburation puis grâce à un outil commercial plus complet de simulation numérique de la diffusion à l’équilibre thermodynamique (DICTRA). Des écarts ont été observés entre les simulations des états de carburation réalisées avec DICTRA et les mesures expérimentales (profil de concentration en carbone et population de carbures). Afin de prédire au mieux l’état de carburation des aciers rencontré à 500 et 600 °C, il a notamment été démontré qu’il est probablement nécessaire de prendre en compte la diffusion du carbone dans les joints de grains et un écart à l’équilibre thermodynamique entre le carbone piégé dans les carbures et le carbone dissout dans la matrice. La deuxième campagne expérimentale a concerné l’étude du système : acier – B4C – Na. Des caractérisations couplées à des études thermodynamique et cinétique ont permis de proposer un mécanisme de carburation et de boruration des aciers. Après la dissolution du B4C dans le sodium, deux phénomènes ont été observés. Le bore réagit avec les aciers pour former une couche duplexe de borures à la surface (MB, M2B) et des borures dans les joints de grains du substrat. La cinétique de formation de la couche de borures dans les aciers suit une loi parabolique. Le carbone entraine une légère carburation des aciers plus en profondeur et le degré de carburation des aciers s’est avéré constant entre 250 et 3000 h d’exposition, ce qui suggère que le phénomène de carburation s’opère probablement avant la formation d’une couche continue de borures. / Pellets of boron carbide, B4C, enclosed in AIM1 (15Cr-15Ni-0.4Ti) stainless steel tubes are constitutive materials of the control rods in the future French Advanced Sodium Technological Reactor for Industrial Demonstration (ASTRID). During reactor operation, the B4C pellets are immersed in liquid sodium in the temperature range 773-873 K. Based on the feedback from operation of former Sodium Fast Reactors (SFR), the lifetime of the control rods has been shown to be limited by their carburization kinetics. Although, boriding of the steels was observed in out-ofpile studies. In order to increase the lifetime prediction of the aforementioned components in service, detailed information on the chemical interaction between the steel and B4C in liquid sodium is required. As a result, two sets of out-of-pile experiments were conducted: 1. Three stainless steels (AIM1, 316L, EM10) were exposed to highly carburizing sodium (ac > 1) at 773, 873 and 923 K. 2. The same grades were exposed to high purity B4C powder in liquid sodium at 773 and 873 K. The first campaign was performed in order to have a better understanding of the carburization phenomenology and kinetics of the control rods. The extent of carburization was evaluated. A good description of the carburization kinetics was obtained by means of two models and a simulation tool (DICTRA). The limits of the simulation tools were exposed. It was shown that the grain boundary diffusion of carbon had to be taken into account. The second set of experiments was carried out in order to study the system: steel – B4C – Na. A thorough examination of the nature of the chemical interaction was performed. The characterizations were combined with a thermodynamic and kinetic study to propose a carburization and boriding mechanism. The B4C powder dissolved in liquid sodium and reacted with the steels to form a boride layer (MB and M2B) at the surface, borides in the grain boundaries and a carburized zone underneath. The growth kinetics of the boron affected zone was shown to be parabolic. The carburization depth did not evolve between 250 and 3000 h and suggested that this phenomenon occurred during a transient stage.

A series of accident simulations were performed using INL's thermal hydraulics code RELAP5-3D to analyze steady-state and transient behavior of a sodium cooled fast reactor. The reactor chosen for this study was General Electric's S-PRISM, which is a 1,000 MWt pool-type sodium-cooled fast reactor, designed for either an Oxide or Metal fueled core. Once key core characteristics including power profiles, reactivity feedback coefficients and delayed neutron parameters were calculated, S-PRISM was redesigned for a Nitride fueled core to take advantage of the Nitride fuel's high thermal conductivity and melting temperature. Loss of flow, loss of heat sink, loss of power and inadvertent control rod withdrawal accidents were simulated for each core at beginning, middle and end of cycle to determine if one fuel type provides significant safety advantages over the others.

This work considers the impact of fast reactor fuel cycles on the UK's nuclear waste inventory, focusing on the disposition of the UK's plutonium stockpile and spent fuel from new build nuclear reactors. Reprocessing spent fuel from nuclear reactors has led to a large stockpile of civil plutonium in the UK. At the end of reprocessing the stockpile was estimated to be 112 tonnes. This large stockpile of separated plutonium poses a proliferation concern and there is no strategy at present for UK plutonium disposition. The NDA's position paper in 2014 stated the re-use of plutonium in a reactor as a preferred option. These options included Mixed OXide (MOX) fuelled Pressurised Water Reactors (PWR) and the use of plutonium in a Sodium-cooled Fast Reactor (SFR), PRISM, operated as a once-through plutonium burning fast reactor. As yet a preferred option has not been selected by the government. Nuclear power is the UK's largest source of low-carbon electricity. Current plans aim to build 16 GWe of new reactors by 2050 to replace the UK's current fleet. This work considered PWR MOX and once-through SFRs for UK plutonium disposition, comparing their relative merits to the direct disposal of the plutonium stockpile in a geological repository. The waste performance of disposition options were compared using assessment criteria based on: Technology Readiness Level (TRL), final stockpile mass, repository size and radiotoxicity. To maximise the reduction of the UK's plutonium stockpile, closed SFR fuel cycles were also considered with scenarios aimed at improving waste performance. Once-through and closed SFR fuel cycles were also considered for the disposition of spent fuel from new build reactors. Research presented in this thesis shows that UK waste disposition options are highly dependent on fuel cycle operating parameters. In once-through plutonium disposition options all scenarios increased repository size compared to direct disposal. Once-though SFRs increased repository size the least, where as PWR MOX reduced the stockpile mass most significantly. The most significant improvement in waste performance, using a closed fuel cycle up to 2150, required short reprocessing times and americium reprocessing. There were no additional improvements of significance with curium reprocessing and the choice of metallic or MOX fuelled SFRs had little impact on waste performance. Preferred fuel cycle scenarios are dependent on the priority given to different assessment criteria. To compare fuel cycle scenarios on an even basis, decision analysis methods were presented using assessment criteria results from the fuel cycles modelled in this work. Decision analysis methods were designed so that the reader can apply their own priorities, through the use of weightings, to the assessment criteria to determine preferable fuel cycle scenarios.

The thesis deals with a project of a steam generator heated with liquid sodium. The first section describes some types of steam generators at nuclear power plants which have been projected or put into use. The next part presents a draft concept of the steam generator, solved by this work. The implementation of the steam generator module has been selected from several options, based on the thermal, hydraulic and stress calculations and on the chosen criteria. The conclusion of this thesis deals with the evaluation of the final solution in terms of nuclear safety and technical solutions, and compares this solution to similar projects following various criteria.