Yes / Uranyl complexes of a large, compartmental N8-macrocycle adopt a rigid, “Pacman” geometry that stabilizes the UV oxidation state and promotes chemistry at a single uranyl oxo-group. We present here new and straightforward routes to singly reduced and oxo-silylated uranyl Pacman complexes and propose mechanisms that account for the product formation, and the byproduct distributions that are formed using alternative reagents. Uranyl(VI) Pacman complexes in which one oxo-group is functionalized by a single metal cation are activated toward single-electron reduction. As such, the addition of a second equivalent of a Lewis acidic metal complex such as MgN″2 (N″ = N(SiMe3)2) forms a uranyl(V) complex in which both oxo-groups are Mg functionalized as a result of Mg−N bond homolysis. In contrast, reactions with the less Lewis acidic complex [Zn(N″)Cl] favor the formation of weaker U−O−Zn dative interactions, leading to reductive silylation of the uranyl oxo-group in preference to metalation. Spectroscopic, crystallographic, and computational analysis of these reactions and of oxo-metalated products isolated by other routes have allowed us to propose mechanisms that account for pathways to metalation or silylation of the exo-oxogroup.

The photo-oxygenation of 2-methylpropane and cyclohexane using visible light in aqueous acidic uranyl ion solutions at ambient temperature and pressure has been undertaken. For 2-methylpropane in the absence of oxygen, the main product (≈90%) is 2-methyl-2-propanol with a quantum yield of 0.021 ± 0.001. In the presence of molecular oxygen, both 2-methyl-2-propanol and 2-propanone (acetone) are found. Based on this, and results of gamma radiolysis of aqueous 2-methylpropane to give 'tert'-butyl radical by electronically excited uranyl ion is proposed. In the absence of oxygen, the quantum yield of 2-methyl-2-propanol shows a sigmoidal dependence on the concentration of perchloric acid. A two-species kinetic model involving an acid-base dissociation of the uranyl ion in the excited state accounts for the observed features. The addition of sodium perchlorate to the 2-methylpropane system has an inhibitory effect on the quantum yield. Excited-state ion pairing between the uranyl ion and perchlorate anion is proposed. Consequently, the two-species acid-base model is expanded upon to yield a three-species acid-base-perchlorate model that seems to account for the results from 0.01-0.4 M perchloric acid concentration. Potassium peroxydisulfate is proven effective to increase the quantum yield of 2-methyl-2-propanol from 0.021 ± 0.001 to greater than unity (1.5 ± 0.1) indicating the existence and importance of thermal chain reactions involving sulfate radical anion. The quantum yield of 2-methyl-2-propanol is found to be dependent on the concentrations of 2-methylpropane, perchloric acid and potassium peroxydisulfate, and inversely dependent on the light intensity. The net consumption of uranyl ion is zero in the presence of potassium peroxydisulfate meaning that the uranyl ion is regenerated, making this a catalytic cycle in terms of uranyl ion. The oxygenation of cyclohexane using photo-excited aqueous uranyl ion gives cyclohexanol and cyclohexanone as the two main isolated products (54%). The overall mechanism is expected to be similar to that for the 2-methylpropane system. Refluxing of this substrate with a reducing agent (CaH2 or LiAlH4) is required prior to photolysis to achieve consistent quantum yields of both products due to thermal autoxidation reactions. The presence of molecular oxygen is found to be important in determining the ratio of alcohol to ketone in the product distribution. Potassium peroxydisulfate significantly enhances the quantum yield of cyclohexanone, leaving the quantum yield of cyclohexanol relatively unchanged (as compared to without added potassium peroxydisulfate), while uranyl ion is not consumed. Possible reactions involving cyclohexyl and cyclohexyl peroxyl radicals are given to account for the experimental results.

Ce travail, effectué dans le cadre d’une collaboration entre l’IPN d’Orsay et EDF, contribue aux études destinées à améliorer la compréhension du comportement des radioéléments en production (centrale en fonctionnement) et à l’aval du cycle électronucléaire (stockage géologique profond des déchets). Le comportement et l’évolution des radioéléments sont fortement dépendants des interactions aux interfaces eau / surface minérale, phénomènes complexes et souvent difficiles à caractériser in situ (en particulier, dans le cas du circuit primaire des centrales REP). La dynamique moléculaire basée sur la théorie de la fonctionnelle de la densité apporte des éléments de compréhension sur l’évolution des structures d’équilibre en prenant en compte explicitement la solvatation et les effets de la température sur les mécanismes d’interaction. Dans un premier temps, le comportement de l’ion uranyle en solution et à l’interface d’un système modèle eau / TiO2 à température ambiante a été simulé et validé par la confrontation avec des résultats expérimentaux et des calculs de DFT statiques. Dans un deuxième temps, cette approche a été employée sur ce même système, à des fins prédictives, pour étudier l’effet d’une élévation de la température. La rétention de l’ion augmente avec la température en accord avec les données expérimentales obtenues sur d’autres systèmes, et conduit également à une modification du complexe de surface. Dans un troisième temps, une étude similaire a été effectuée à l’interface eau / NiO, produit de corrosion présent dans le circuit primaire des centrales nucléaires, pour lequel peu de données expérimentales sont disponible actuellement. / This study, performed within the framework of an EDF and IPN of Orsay partnership, contributes to the studies intended to improve the understanding of the radioelement behaviour in service (nuclear power plant) and at the end of the uranium fuel cycle (deep geologic repository). The behaviour and the evolution of radioelement depend mainly on the interactions at the water / mineral interfaces, which are complex and often difficult to characterize in situ (in particular, in the PWR primary circuit). Molecular dynamic simulations based on the Density Functional Theory provide some insight to understand the evolution of the structures against the solvation and the effects of the temperature on the interaction mechanisms. At first, the behaviour of the uranyl ion at room temperature in solution and at the water / TiO2 interface, as a system model, has been studied and validated by the systematic comparisons with the experimental and static DFT calculations data. Secondly, this approach was used on the same system, in predictive purposes, to study the effect of a temperature rise. The retention of the ion increases with the temperature in agreement with the experimental data obtained on other systems, and led also to a modification of the surface complex. Finally, a similar study has been performed at the water / NiO interface, which corresponds to a corrosion product present in the primary circuit of nuclear power plants, but for which few experimental data are currently available.

The chemistry of actinides in solution is a very important aspect of the nuclear fuel cycle, especially as the energy needs of the world continue to increase. However, the radio-active nature of the actinides makes experimentation very difficult and dedicated expensive instruments are required. In addition, the disposal of radio-active waste materials requires a proper understanding of their chemistry at a molecular level. To tackle the problem, and to underpin the experimental studies, in this thesis we have studied the redox chemistry and disproportionation mechanism of actinyl complexes in solution using state-of-the art computational methods. Reduction potentials of actinyl complexes in solution have been estimated in solution using density functional theory (DFT) approaches. Solvation effects were included in the quantum chemistry calculations with the conductor like polarisable continuum model (CPCM) solvation method. First of all, we have validated our computational method by studying a variety of solute cavity definitions within the CPCM solvation model and assessed the performance of a range of DFT functionals to suitable to accurately describe the actinide chemistry in solution. Penta-valent uranyl(V) ions are unstable and readily disproportionate; in this study we have explored outer-sphere electron transfer and disproportionation mechanisms to determine the stability of these ions in solution. We have found that the process of outer-sphere disproportionation is unlikely to occur in non-aqueous solutions, such as DMSO, DMF, DCM, acetonitrile and pyridine, when the uranyl(V) ion is bound with a multi-dentate organic ligand. However, our computational results hypothesise that the presence of a trace of water in the experimental conditions can promote a disproportionation reaction by protonating the uranyl(V) ‘yl’ oxygen atoms and then the electron transfer process would proceed through either inner or outer sphere mechanism. In addition, the effect of alkali metal cations on the outer-sphere disproportionation mechanisms was also studied. Overall it has been shown that DFT can be used to accurately predict the redox properties of actinyl complexes in solution and thus contributing for an effective and efficient design of nuclear material separations, proper as well as safer radioactive waste disposal.

Les phénomènes d’adsorption interviennent dans les processus géochimiques gouvernant ainsi le transport des contaminants. Par ailleurs, les variations de température sont susceptibles d’influencer significativement leur comportement vis-à-vis de la surface des minéraux. Aussi, l'influence de la température sur la sorption doit être étudiée afin de mieux appréhender le devenir des éléments dans l'environnement. Dans cette optique, l'interaction entre un ion modèle, l'uranyle et deux minéraux a été étudiée. Dans un premier temps, un composé méthodologique, la monazite, a été choisi afin de déterminer la démarche à suivre pour étudier l'influence de la température sur la sorption de l'uranyle dans trois milieux plus ou moins complexants. Puis, des tests préliminaires ont été réalisés pour étudier la sorption de l'uranyle sur un composé d'intérêt industriel, la magnétite, en appliquant la démarche mise en place. Le solide a d'abord été caractérisé d'un point de vue massif puis les caractéristiques acido-basiques de sa mise en suspension dans les trois électrolytes (NaClO4, NaNO3 et Na2SO4) ont été étudiées en fonction de la température. Les constantes d'équilibre associées aux réactions de déprotonation des sites de surface ont été déterminées entre 25°C et 95°C par modélisation de courbes de titrages potentiométriques. Les simulations ont été effectuées en limitant au maximum le nombre de degrés de liberté du système. Le modèle 1-pK a donc été préféré au modèle 2-pK en raison du nombre de paramètres ajustables plus limité dans ce modèle. Des contraintes expérimentales telles que le pH de point de charge nulle ou les enthalpies déterminées par mesure directe des chaleurs associées par microcalorimétrie de mélange ont été imposées pour déterminer les constantes d'équilibre acido-basiques. La sorption de l'uranyle en fonction du pH sur le même intervalle de température a été étudiée en alliant l'acquisition de données macroscopiques telles que les sauts de sorption et la spéciation en solution à une étude structurale menée par analyse par spectrofluorimétrie laser permettant l'identification des espèces sorbées. La simulation des sauts de sorption permettant d'accéder aux constantes associées aux réactions de sorption a été réalisée en imposant les caractéristiques acido-basiques préalablement déterminées. Des mesures directes, par microcalorimétrie de mélange, des chaleurs mises en jeu lors de la sorption de l'uranyle ont permis de tester la validité de la loi de Van't Hoff sur ce phénomène. La même démarche a été suivie pour étudier l'influence de la température sur la sorption de l'uranyle sur la magnétite en milieu NaClO4 et NaNO3. Cependant, l'étude structurale par spectrofluorimétrie laser n'a pu être réalisée en raison de la couloration noire de la magnétite. / The migration of radiotoxic elements in the geosphere is mainly regulated by chemical parameters which control the partitioning of the elements between mineral phases and aqueous solutions. Variation in temperature may affect the retention properties of a mineral surface and requires a careful investigation in order to understand the radionuclides behavior in the geosphere. In this way, the interaction mechanisms between uranium(VI) and two minerals (LaPO4 and Fe3O4) have been studied. In a first step, the monazite (LaPO4) has been chosen as methodological solid in order to clearly define all the different stages needed to completely characterize the influence of temperature on the sorption phenomena. To reach that goal, three media, more or less complexants towards aqueous uranyl and the mineral surface, have been considered. Physico-chemical as well as surface acid-base properties of the solid surface have been studied by considering three electrolytes (NaClO4, NaNO3 and Na2SO4) and temperatures ranged from 25°C to 95°C. The point of zero charge has been found to be identical for perchlorate and nitrate media (pHPZC=2.1) but it was found to be one pK unit higher for the sulfate medium indicating a sorption of the background electrolyte ions. The reaction heats associated to the hydration of the solid have been measured by using microcalorimetry and the nature of the reactive surface sites has been determined by carrying out Time-Resolved Laser-Induced Fluorescence Spectroscopy (TRLFS). On the basis of these experimental constraints, the titration curves obtained for the monazite suspensions were fitted by using the Constant Capacitance Model and the 1-pK model was preferred to characterize the surface charge evolution, due to the limited number of adjustable parameters. The surface protonation constants being determined, the behavior of U(VI) towards the monazite surface in the three electrolytes has been investigated. On the basis of both U(VI) speciation in solution and the results of a structural study carrying out by using TRLFS together with calorimetric measurements, the sorption edges have been modeled and the corresponding sorption constants determined. Since these values take into account a wide number of experimental results (both structural and thermodynamical ones) they appear to be accurate and could be extrapolated more confidently to other physico-chemical conditions. The experimental approach being validated with the methodological solid, preliminary tests have been carried out to study uranyl sorption onto a second substrate, the magnetite, more relevant than monazite in the field of radionuclides migration in the geosphere.