Methods for the determination of the actinide elements in water, biological, soil and sediment samples have been developed using on-line solid phase extraction and high performance liquid chromatography (HPLC) coupled with inductively coupled plasma mass spectrometry (ICP-MS). Initial applications utilised a commercially available resin, namely TRU-Spec resin, for efficient removal of the matrix prior to elution of uranium and thorium analytes. Comparative analyses of reference materials and natural water samples from Plymouth and Dartmoor demonstrated significant improvement in precision and speed of analysis by using TRU-Spec coupled to ICP-MS compared with alpha spectrometry. Further applications of the TRU-Spec resin for the determination of the transuranic actinide elements neptunium, plutonium and americium, resulted in the successful determination of 239Pu and 237Np in biological reference materials. Detection limits were 700, 850, and 600 attograms (ag) for 237Np, 233Pu, and 241Am, respectively, for a 0.5 ml sample injection, and better than 200 ag/g with 50 ml pre-concentration when sector field (SF) ICP-MS was used. A method for the selective sequential elution of uranium and plutonium was also developed to facilitate the determination of 239Pu without interference due to the 238U1H+ polyatomic ion, caused by high concentrations of 238U in sediment samples. Investigations were performed into the use of a polymeric substrate, which was dynamically coated with chelating dyes such as xylenol orange and 4-(2-pyridylazo) resorcinol, and a silica substrate coated with permanently bonded iminodiacetic acid. The latter was used for the successful determination of uranium and thorium in certified reference material waters. However, the column was found to have a high affinity for iron, making it unsuitable for the determination of the actinides in soil and sediment samples. Subsequently, a polystyrene substrate which was dynamically coated with dipicolinic acid was used for HPLC coupled with SF-ICP-MS. Using this column it was possible to separate the various actinides from each other and from the matrix. In particular, it was possible to separate plutonium and uranium to facilitate interference-free determination of the former. The column also exhibited some selectivity for different oxidation states of Np, Pu and U. Two oxidation states each for plutonium and neptunium were found, tentatively identified as Np(V) and Pu(III) eluting at the solvent front, and Np(IV) and Pu(IV) eluting much later. Detection limits were 12, 8, and 4 fg for 237Np, 239Pu, and 241Am, respectively, for a 0.5 ml injection, and the system was successfully used for the determination of 239Pu in water, biological and soil reference materials.

Les réactions d'oxydoréduction de l'antimoine (V) et (III) avec la mackinawite (FeS) et du plutonium (III) et (V) avec plusieurs minéraux à fer(II) et des oxydes de fer(III) ont été étudiées dans des conditions d'anoxie. La spectroscopie d'absorption des rayons X fut utilisée pour l'analyse de l'état d'oxydation et de la structure locale du Sb et Pu associés à la phase solide. Après réaction avec la mackinawite, la chukanovite et la magnétite, PuO2, Pu(III) ou des mixtures des deux états d'oxydation ont été observé. A la surface de la magnétite un complexe tridenté du Pu(III) a pu être mis en évidence à l'aide des spectres EXAFS couplé à une simulation de Monte-Carlo utilisant le code de calcul Feff. La quantité relative de Pu(III) est fonction de l'espèce minérale, du ratio solide/liquide, des valeurs pe et pH du système et, potentiellement, de la taille de particule et de la cristallinité de la phase solide de PuO2 en présence de laquelle le Pu(III) existe. Avec la mackinawite à pH 6,2 et à une occupation de surface de 67 nmol/m2 et avec la magnétite jusqu'à pH 8.4 et 56 nmol/m2 de Pu, uniquement du Pu(III) fut trouvé associé à la phase solide. Avec la maghémite contenant du fer(II) résiduel à pH6 Pu(III) et Pu(IV) était, probablement, présents dans des complexes de surfaces similaire à celui formé par le Pu(III) sur la magnétite. Dans les conditions expérimentales (couverture de surface ≤ 77 nmol/m2), aucune formation de PuO2 ne fut observé. Après réaction avec l'hématite et la goethite Pu(IV) était l'état d'oxydation prédominant associé à la phase solide. La sorption et la réduction du Sb(V) avec la mackinawite était fortement fonction du pH. A pH acide la sorption était rapide et Sb(V) fut complètement réduit en Sb(III), formant un complexe Sb(III)-S3 probablement associé à la surface de la mackinawite. La réduction du Sb(V) était couplée à l'oxydation de la mackinawite et la greigite (Fe3S4) fut détectée par XRD. A pH basique, la sorption du Sb(V) est lente et la réduction vers le Sb(III) n'était complète qu'à des ratios de Sb/FeS très bas. Pour des valeurs plus élevé de Sb/FeS la sorption de Sb se faisait en partie par la réduction envers le complexe de Sb(III)-S3 et en partie par une co-précipitation avec le Fe(III). Il a pu être démontré que les minéraux à fer(II) peuvent effectivement contribuer à la réduction et à l'immobilisation de l'antimoine et du plutonium qui sont des contaminants environnementaux d'importance croissante. / Redox reactions of Sb(V) and Sb(III) with mackinawite (FeS) and of aqueous Pu(III) and Pu(V) with various Fe(II)-bearing minerals and Fe(III)-oxides have been investigated under anoxic conditions. X-ray absorption spectroscopy was used to analyze oxidation state and local coordination environment of Sb and Pu associated with the solid phase. After reaction with mackinawite, chukanovite and magnetite, PuO2, Pu(III) or mixtures of the two oxidation states were observed. On magnetite, a tridentate Pu(III) surface complex could be identified from EXAFS combined with Feff-Monte-Carlo simulation. The relative amount of Pu(III) depends on the type of mineral, the solid/solution ratio, the system pe and pH, and, potentially, the particle size and crystallinity of the formed PuO2 solid phase. With mackinawite at pH 6.2 and a surface loading of 67 nmol/m2 and with magnetite up to pH 8.4 and a surface loading of 56 nmol/m2, only Pu(III) was identified associated with the solid phase. With maghemite containing residual Fe(II) at pH6, Pu(III) and Pu(IV) were present in, probably, inner-sphere surface complexes similar to the one formed by Pu(III) on magnetite. Under the given conditions (surface loadings ≤ 77 nmol/m2) formation of PuO2 was not observed. After reaction with hematite and goethite, Pu(IV) was the predominant oxidation state associated with the solid phase. Sorption and reduction of Sb(V) on mackinawite were strongly pH dependent. At acidic pH, sorption was fast and Sb(V) was completely reduced to an Sb(III)-sulfide complex associated with the solid phase. Reduction of Sb(V) was coupled to oxidation of mackinawite and formation of a greigite (Fe3S4) phase could be observed by XRD. At basic pH, Sb(V) was slowly removed from solution and reduction to Sb(III) was complete only at very small Sb/FeS ratios. At higher Sb/FeS, Sb(V) removal occurred partly through reduction to solid phase associated Sb(III)-S3 and partly through co-precipitation with Fe(III). In conclusion, it could be shown that Fe(II) bearing minerals can effectively contribute to the reduction and immobilization of antimony and plutonium, two contaminants of growing environmental importance.

Les codes de simulation du cycle du combustible nucléaire sont des outils permettant d’évaluer les stratégies futures du cycle du combustible nucléaire et de comprendre la physique de ce cycle. Dans le contexte d’incertitude entourant l’évolution future du parc nucléaire français, notamment concernant le déploiement de Réacteurs à Neutrons Rapides au sodium (RNR-Na), la problématique de cette thèse est d’étudier des solutions alternatives de gestion du plutonium et des autres noyaux lourds, basées sur les Réacteurs à Eau Pressurisés (REP). Les stratégies étudiées s’appuient sur deux hypothèses. La première suppose un retard important dans le déploiement des RNR-Na, impliquant une stratégie d’attente visant à stabiliser l’inventaire en plutonium. La deuxième hypothèse suppose un abandon de la stratégie de déploiement des RNR. Dans ce cadre, une stratégie d’incinération du plutonium a été étudiée pour quantifier la capacité de réduction de l’inventaire par les REP. Le code de simulation CLASS, développé par le CNRS/IN2P3 et l’IRSN, est utilisé. Le multi-recyclage du plutonium en REP requiert un combustible dédié. Des développements ont été réalisés pour modéliser le combustible étudié, composé de MOX sur un support d'uranium enrichi. Une méthodologie innovante d’évaluation de scénarios nucléaires basée sur l’analyse globale de sensibilité a été appliquée. Cette méthode a permis d’identifier des scénarios de référence pour la stabilisation et la réduction de l’inventaire en plutonium et actinides mineurs. Des simulations du cycle détaillées ont été réalisées afin d'analyser la capacité des REP à gérer le plutonium à l’échelle du cycle. / Nuclear fuel cycle simulation codes are used to evaluate fuel cycle future strategies and understand the nuclear fuel cycle physics. In the context of uncertainty related to the future of French nuclear fleet, especially on theSodium Fast Reactor (SFR) deployment, the present work aims to study alternative solutions for plutonium and heavy isotopes management, based on Pressurized Water Reactor (PWR). Two hypothesis have been formulated to identify strategies. First, a delay has been expected in SFR deployment which induces a stabilization of plutonium inventory before SFR integration. The second hypothesis is based on the assumption that SFR won’t be deployed in France. For this specific case, a plutonium incineration strategy has been studied to quantify the PWR plutonium inventory reduction capacity. Fuel cycle simulations are performed using the fuel cycle simulator CLASS developed by the CNRS/IN2P3 in collaboration with IRSN. Plutonium multi-reprocessing in thermal reactor requires an innovative fuel. Developments have been made to simulate a fuel composed of MOX on enriched uranium support. An innovative methodology for fuel cycle simulation evaluation, based on Global Sensitivity Analysis, has been applied. This methodology leads to reference scenarios identification for plutonium and minor actinides inventories stabilization and reduction. Fuel cycle detailed simulations have been performed to produce fuel cycle data, to analyze PWR plutonium management at the cycle scale.

L'utilisation du plutonium (Pu(IV) pour des applications militaires et civiles peut engendrer des contaminations internes chez les personnes exposées. Plusieurs voies de contamination sont possibles : par ingestion, par inhalation ou par blessure. En cas d'inhalation, le plutonium se présente le plus souvent sous forme de particules d'oxyde de plutonium qui vont se localiser au sein des alvéoles pulmonaires. Par un mécanisme de phagocytose, les particules sont internalisées par les macrophages de l'épithélium pulmonaire. Actuellement, le seul agent de décorporation administré en cas de contamination au plutonium est le DTPA (l'acide diethylenetriaminepentaacetique). Il est administré en France sous forme de CaNa3-DTPA par injection intraveineuse et est efficace pour les contaminations par ingestion et par blessure. Les nanoparticules fonctionnalisées à base de polymères naturels sont un concept innovant de décorporation du Pu(IV) solubilisé dans les macrophages pulmonaires et ouvrent la voie au développement de nouvelles familles de décorporants. C'est dans ce contexte que deux stratégies ont été développées : des nanoparticules à base de N-trimethyl chitosan fonctionnalisées par le ligand DTPMP (l'acide diéthylènetriaminepentamethylene phosphonique) qui est l'analogue phosphonique du DTPA et des nanoparticules chélatantes à base de -cyclodextrines amphiphiles anioniques. Ce travail a consisté en la synthèse et la caractérisation des nano-objets puis de l'étude de leur complexation avec les actinides (Th/Pu) en utilisant la spectroscopie EXAFS. Et enfin, des tests préliminaires biologiques in vitro ont été réalisés. Les résultats obtenus, avec les nanoparticules à base de chitosan et de DTPMP montrent que les nanoparticules présentent des tailles et une stabilité compatible avec l’application visée. D’autre part, leur affinité pour les actinides (IV) (Th,Pu) est comparable à celle du chélatant de référence, le DTPA. Enfin, les tests, effectués sur deux lignées de macrophages montrent que les nanoparticules sont internalisées très rapidement et que la matrice polysaccharidique semble se dégrader, permettant le relargage du chélateur DTPMP au niveau des sites de rétention du Pu(IV). Cette thèse constitue un travail préliminaire au développement d'une nouvelle famille d’agents décorporants plus ciblés pour une contamination au plutonium par inhalation. / The use of plutonium (Pu(IV) for military and civil applications can lead to internal contamination. There are several possible routes of contamination: ingestion, inhalation or injury. In case of plutonium inhalation, the plutonium forms oxide particles that reach the pulmonary alveoli. Through a phagocytosis mechanism, the particles are internalized by the macrophages of the pulmonary epithelium and continue to exert their toxicity. Currently, the only decorporating agent administered in the event of contamination with plutonium is DTPA (diethylenetriaminepentaacetic acid). in France, it is administered as CaNa3-DTPA by intravenous injection. This standard is effective for contamination by ingestion and injury. However, it is not effective in case of contamination by inhalation. Functionalized nanoparticles based on natural polymers constitute an innovative concept for decorporating Pu(IV) solubilized in pulmonary macrophages and open the way for the development of new families of decorporants. We investigated two strategies: chitosan-based nanoparticles functionalized by the DTPMP (diethylenetriaminepentamethylene phosphonic acid) which is the phosphonic analog of DTPA and self-organized chelating β-cyclodextrin-based nanoparticles. This work was first focused on the synthesis and characterization of the nano-objects and then on the study of their complexation abilities with actinides (Th/Pu) using EXAFS spectroscopy. Finally, preliminary in vitro biological tests were carried out. Our obtained results with DTPMP and chitosan based nanoparticles showed that these aggregates exhibit size and stability compatible with the application. Furthermore, we demonstrate their affinity for the actinides(IV) (Th, Pu) is comparable to the reference DTPA. Finally, in vitro tests realized onto macrophages show that our nanoparticles are rapidly internalized through phagocytosis and that the polysaccharide matrix seems to undergo degradation which allows the DTPMP to be released and targeted right into the sequestration sites of Pu(IV). This work constitutes a first step in the development of new family of decorporating agents with a higher efficiency in case of plutonium contamination through inhalation.

For the first time microbial tDNA could be isolated from 50 g unperturbed Mont Terri Opalinus Clay. Based on the analysis of the tDNA the bacterial diversity of the unperturbed clay is dominated by representatives of Firmicutes, Betaproteobacteria, and Bacteriodetes. Firmicutes also dominate after treatment of the clay with R2A medium. Bacteria isolated from Mont Terri Opalinus Clay on R2A medium were related to Sporomusa spp., Paenibacillus spp., and Clostridium spp.. All further investigations are concentrated on the unique isolates Sporomusa sp. MT-2 and Paenibacillus sp. MT-2. Cells of the type Sporomusa sp. MT-2 and Paenibacillus sp. MT-2 were comprehensively analyzed in terms of growing, morphology, functional groups of the cell envelope, and cell membrane structure. Strong actinide(An)/lanthanide(Ln)-interactions with the Opalinus Clay isolates and the Äspö-strain Pseudomonas fluorescens (CCUG 32456) could be determined within a broad pH range (2-8). The metals bind as a function of pH on protonated phosphoryl, carboxyl and deprotonated phosphoryl sites of the respective cell membrane. The thermodynamic surface complexation constants of bacterial An/Ln-species were determined and can be used in modeling programs. Depending on the used An different interaction mechanisms were found (U(VI): biosorption, partly biomineralisation; Cm(III): biosorption, indications for embedded Cm(III); Pu: biosorption, bioreduction and indications for embedded Pu). Different strategies of coping with U(VI) were observed comparing P. fluorescens planktonic cells and biofilms under the chosen experimental conditions. An enhanced capability of the biofilm to form meta-autunite in comparison to the planktonic cells was proven. Conclusively, the P. fluorescens biofilm is more efficient in U(VI) detoxification. In conclusion, Mont Terri Opalinus Clay contains bacterial communities, that may influence the speciation and hence the migration behavior of selected An/Ln under environmental conditions.

Plutonium Uranium Reduction EXtraction (PUREX) technology is a solvent extraction process used to recover plutonium and uranium from spent nuclear fuel. The solvent system is composed of an aqueous nitric acid phase in contact with an organic phase made up of tributyl phosphate in an organic diluent. During the separation process, the PUREX solvent system is subject to an intense multi-component radiation field (gamma rays, alpha particles, beta particles, neutrons, and fission fragments) rendering it susceptible to radiolytic degradation, which reduces its performance. Despite the PUREX process being used for over sixty years, a complete quantitative mechanistic understanding of the radiolytic degradation processes is not available. Nitrous acid is the most significant radiolytic degradation product of nitric acid, especially as its chemical and physical properties alter the formulation of the PUREX solvent system. Furthermore, nitrous acid exhibits complex redox relationships with a number of actinides, with plutonium being of greatest concern to the performance of the PUREX process. A combination of experimental and computational (stochastic and deterministic) techniques have been used to investigate the radiolysis of the PUREX solvent system's aqueous phase, specifically the radiolytic formation of nitrous acid, and its conjugate base nitrite, as a function of solvent system formulation, absorbed dose (up to 1.7 kGy), and radiation quality (cobalt-60 gamma rays and alpha particles from plutonium and americium alpha decay). The research presented in this thesis focuses on: (i) the experimental radiation chemistry of solutions of nitric acid and sodium nitrate over the range of concentrations 1 × 10-3 to 6 mol dm-3, and (ii) the development of a multi-scale modelling approach for evaluating the radiolysis of aqueous systems in terms of reaction mechanisms. The experimental and modelling studies provide insight into the radiation chemistry of the PUREX solvent system's aqueous phase, mechanistically demonstrating how the radiation chemical yield of nitrous acid and nitrite is dependent upon the interplay between non-homogeneous radiation track chemistry and secondary bulk homogeneous chemistry. This interplay is influenced by low pH, the presence of chemical scavengers and redox active metal ions, and radiation quality. These findings will act as a benchmark for the development of advanced reprocessing schemes, which must seriously consider how modifications in solvent system formulation and fuel composition may affect this dynamic interplay, and ultimately the generation of secondary highly active liquid waste.

Several methods have been developed previously for estimating cumulative energy production and plutonium production from graphite-moderated reactors. The Graphite Isotope Ratio Method (GIRM) is one well-known technique. This method is based on the measurement of trace isotopes in the reactor’s graphite matrix to determine the change in their isotopic ratios due to burnup. These measurements are then coupled with reactor calculations to determine the total plutonium and energy production of the reactor. To facilitate sensitivity analysis of these methods, a one-group cross section and fission product yield library for the fuel and graphite activation products has been developed for MAGNOX-style reactors. This library is intended for use in the ORIGEN computer code, which calculates the buildup, decay, and processing of radioactive materials. The library was developed using a fuel cell model in Monteburns. This model consisted of a single fuel rod including natural uranium metal fuel, magnesium cladding, carbon dioxide coolant, and Grade A United Kingdom (UK) graphite. Using this library a complete sensitivity analysis can be performed for GIRM and other techniques. The sensitivity analysis conducted in this study assessed various input parameters including 235U and 238U cross section values, aluminum alloy concentration in the fuel, and initial concentrations of trace elements in the graphite moderator. The results of the analysis yield insight into the GIRM method and the isotopic ratios the method uses as well as the level of uncertainty that may be found in the system results.

Several methods have been developed previously for estimating cumulative energy production and plutonium production from graphite-moderated reactors. The Graphite Isotope Ratio Method (GIRM) is one well-known technique. This method is based on the measurement of trace isotopes in the reactor’s graphite matrix to determine the change in their isotopic ratios due to burnup. These measurements are then coupled with reactor calculations to determine the total plutonium and energy production of the reactor. To facilitate sensitivity analysis of these methods, a one-group cross section and fission product yield library for the fuel and graphite activation products has been developed for MAGNOX-style reactors. This library is intended for use in the ORIGEN computer code, which calculates the buildup, decay, and processing of radioactive materials. The library was developed using a fuel cell model in Monteburns. This model consisted of a single fuel rod including natural uranium metal fuel, magnesium cladding, carbon dioxide coolant, and Grade A United Kingdom (UK) graphite. Using this library a complete sensitivity analysis can be performed for GIRM and other techniques. The sensitivity analysis conducted in this study assessed various input parameters including 235U and 238U cross section values, aluminum alloy concentration in the fuel, and initial concentrations of trace elements in the graphite moderator. The results of the analysis yield insight into the GIRM method and the isotopic ratios the method uses as well as the level of uncertainty that may be found in the system results.

Dual phase magnesia-zirconia ceramics were developed, characterized, and evaluated as a potential matrix material for use in light water reactor inert matrix fuel intended for the disposition of plutonium and minor actinides. Ceramics were fabricated from the oxide mixture using conventional pressing and sintering techniques. Characterization of the final product was performed using optical microscopy, scanning electron microscopy, x-ray diffraction analysis, and energy-dispersive x-ray analysis. The final product was found to consist of two phases: cubic zirconia-based solid solution and cubic magnesia. Evaluation of key feasibility issues was limited to investigation of long-term stability in hydrothermal conditions and assessment of the thermal conductivity. With respect to hydrothermal stability, it was determined that limited degradation of these ceramics at 300^oC occurred due to the hydration of the magnesia phase. Normalized mass loss rate, used as a quantitative indicator of degradation, was found to decrease exponentially with the zirconia content in the ceramics. The normalized mass loss rates measured in static 300^oC de-ionized water for the magnesia-zirconia ceramics containing 40, 50, 60, and 70 weight percent of zirconia are 0.00688, 0.00256, 0.000595, 0.000131 g/cm2/hr respectively. Presence of boron in the water had a dramatic positive effect on the hydration resistance. At 300^oC the normalized mass loss rates for the composition containing 50 weight percent of zirconia was 0.00005667 g/cm2/hr in the 13000 ppm aqueous solution of the boric acid. With respect to thermal conductivity, the final product exhibits values of 5.5-9.5 W/(m deg) at 500^oC, and 4-6 W/(m deg) at 1200^oC depending on the composition. This claim is based on the assessment of thermal conductivity derived from thermal diffusivity measured by laser flash method in the temperature range from 200 to 1200^oC, measured density, and heat capacity calculated using rule of mixtures. Analytical estimates of the anticipated maximum temperature during normal reactor operation in a hypothetical inert matrix fuel rod based on the magnesia-zirconia ceramics yielded the values well below the melting temperature and well below current maximum temperatures authorized in light water reactors.

La production de métal à partir de son oxyde est basée sur une réduction en plusieurs étapes. Un procédé à une seule étape a été récemment proposé pour la formation de Ti métal par électrolyse en chlorures fondus. Dans le procédé FFC (Fray, Farthing and Chen), le titane métallique est produit à la cathode pendant que les ions oxyde sont oxydés à une anode en graphite en dioxyde de carbone. Dans le domaine du nucléaire, ce procédé est intéressant car le sel peut être ré-utilisé sans retraitement particulier. Ce travail présente les études électrochimiques réalisées dans CaCl2-KCl et CaCl2 permettant de clarifier le mécanisme de réduction de CeO2, utilisé comme simulant des actinides. Des électrolyses sur de faibles quantités de CeO2 sont réalisées afin d'obtenir les conditions opératoires optimales. La réduction de plusieurs centaines de grammes de CeO2 est aussi discutée.