In safety assessments for the deep geological disposal of high-level nuclear waste, the unlikely ingress of water and corrosion of spent nuclear fuel (SNF) need to be considered. For many ceramics, grain boundary dissolution plays an important role, which can continuously increase the reactive surface area and sometimes even leads to a disintegration of the microstructure. Model systems were developed to enable a detailed study of single effects occurring during SNF corrosion. The most important phase in this aspect is the UO₂ matrix. Here we present a detailed investigation of UO₂ based model systems which were doped with Nd₂O₃ or Gd₂O₃.
Polycrystalline UO₂ ceramic pellets were synthesized using a co-precipitation method, as well as Nd₂O₃ and Gd₂O₃-doped samples. Doping levels ranged from 0.5 wt.-% to 4 wt.-% of the respective dopant, representing the fission products of SNF. These were characterized by the determination of physical, chemical, structural, and microstructural properties via density measurements, XRD, SEM and EDX. A focus was put on the comprehensive analysis of grain size, grain shape and pore size.
Pure UO₂ ceramics are characterized by a monomodal grain size distribution. The doping changes the average grain size for Nd₂O₃ from 11 µm (pure UO₂) to 9 µm at 4 wt.-% Nd₂O₃ and from 11 µm to 10 µm at 4 wt.-% Gd₂O₃. At lower dopant concentrations, clusters of small grains with low dopant content tend to accumulate in addition to larger grains with a higher dopant composition. This effect decreases with increasing dopant concentration. As a result, the grain size distribution of the samples becomes bimodal between 1 wt.-% and 2 wt.-%, while it is nearly homogeneous at 0.5 wt.-% and 4 wt.-%. In conclusion, there is no simple linear relationship between doping level and microstructural changes in the studied systems.

Recent research has shown that fission products and actinides present in UO₂-based spent nuclear fuel significantly improve its resistance to oxidation in air and to oxidative dissolution in aqueous media compared to pure UO₂. However, the mechanisms behind these retardation effects are not yet fully understood. In this study, we have developed thermodynamic models for the oxidation of pure UO₂ as well as Ln- and Pu-doped UO₂ solid solutions in air, with reference to measured data on the oxygen partial pressure at equilibrium.

The doped systems are primarily distinguished from the pure UO₂-UO₃ system by an additional degree of freedom that allows for a decrease in the total free energy by redistributing dopants between the MO₂, M₄O₉ and M₃O₈ phases. Our modelling shows that the cubic phases tend to be significantly enriched in Ln or Pu, while the M₃O₈ phase tends to have the smallest fraction of the impurity component considered. The achievement of a large O/M ratio in a doped system at thermodynamic equilibrium requires this fractionation to be developed. Consequently, the equilibrium oxidation of doped UO₂ is necessarily coupled with the transport of Ln or Pu between the constituent oxides. Thus, the slow diffusion of Ln or Pu within any of the relevant phases is proposed to be the cause of the enhanced resistance to oxidation.

The UO₂-ZrO₂ system has been considered as a fuel additive for accident tolerant fuels due to increased resistance to oxidation and corrosion in high temperature and humid environments. In literature, at higher doping levels (>20 at.% Zr) one or more zirconia phases have been found in addition to (U_1-yZr_y)O_2-x phases which may help to stabilize against oxidation. However, it has also been shown that the addition of several percent of Zr, maintaining a single (U_1-yZr_y)O_2-x phase, can accelerate oxidation in air at temperatures ~200° C. Under normal reactor conditions a small fraction of Zr from Zircalloy cladding can migrate into the fuel pellet and into the UO₂ matrix, potentially forming phases within the outer edge of fuel pellets which are more susceptible to oxidizing. Presented in this study are two series of (U_1-yZr_y)O_2-x materials synthesized via a nitrate coprecipitation reaction. A low-doped series containing 0.1-1 at.% Zr in 0.1% increments, and a high-doped series containing 2-20 at.% Zr in 2% increments. The doped ammonium diuranate materials were calcined either in air to form U₃O₈ or in H₂/Ar to reduce to UO₂ prior to pressing pellets and sintering. A portion of the sintered pellets were then powdered and reoxidized to U₃O₈ allowing the comparison of two sets of doped U₃O₈ processed at low (800° C) and at high (1700° C) temperature. The evolution of phases present across the doping range is shown by Rietveld refinements of X-ray diffraction patterns and compared with thermal analysis. Defect signatures are shown by Raman and infrared spectroscopy. Select samples are analyzed using electron microscopy and in-situ Raman mapping during oxidation, previously shown to correlate well with thermal analysis results.

Among all the fission products formed in UO₂-based spent nuclear fuels, molybdenum is one of the most abundant due to its high fission yield. Although its radiotoxicity is low, it has been studied during the last years because of its relevance on the fuel oxidation and other fission products migration. In fact, the oxygen potential of Mo/MoO₂ is very similar to that of the fuel, hence, the excess oxygen created during fission could be neutralized by the oxidation of metallic Mo to Mo(IV), buffering the oxidation of UO₂. Therefore, the distribution of Mo between metallic particles and dissolved in the UO₂ matrix as MoO₂ is of great importance. In this work, electrochemical experiments were performed to study the influence of molybdenum on the oxidation of UO₂.

UO₂ and Mo powders were mixed and compacted at 700 MPa into pellets, which were then sintered at 1740ºC for 4 hours in a reducing atmosphere (5%H₂/95%Ar). Microstructure characterization of the pellets by SEM evidenced the formation of Mo channels throughout the whole UO₂ pellet, whereas no Mo was found inside the UO₂ matrix. The Mo-doped UO₂ pellet was used in electrochemical experiments as a working electrode. Ag/AgCl (3M KCl) and a Pt wire were used as a reference and counter electrodes, respectively. Test solutions were prepared at pH 10 with NaCl 0.1 mol·dm^-3 in the presence of NaHCO₃, Na₂SiO₃ and/or CaCl₂. The corrosion process was studied by performing cyclic voltammetry, potentiostatic experiments and corrosion potential experiments.

Preliminary results indicate that the presence of Mo significantly decrease the reactivity of UO₂, when compared to that of non doped UO₂. XPS analysis will be performed on the electrode after potentiostatic experiments, to determine the surface oxidation state of both U and Mo by the deconvolution of the U4f band and the Mo 3d band.

Particle analysis is one of the key-techniques used in the field of nuclear safeguards. Beyond traditional uranium isotopic ratio measurement, other methodologies are implemented to better characterize nuclear materials. Among them, age dating at the particle scale enables to determine the time elapsed since the last chemical step of separation/purification or enrichment, for example through the ²³⁰Th-²³⁴U radiochronometer. During this work, uranium-thorium mixed oxide microspheres were synthesized as potential reference materials for nuclear safeguards using a wet chemistry route. The hydrothermal conversion of aspartate precursors at T = 433 K led to mixed dioxide micro-particles with controlled spherical morphology and size, up to 5 mol.% in thorium. In order to remove impurities, densify the micro-particles, and control the chemical form of the final compounds, heat treatments were performed under various atmospheres. Nearly stoichiometric (U,Th)O₂ dioxides were obtained under reducing conditions (Ar-4%H₂) while U₃O₈-based samples were formed under air, with thorium incorporated in the structure up to 2 mol.%. Last, the homogeneity of the cation distributions in the samples was evaluated by various methods, including PERALS α-scintillation counting, as well as X-EDS and LG-SIMS analyses of individual particles, leading to consistent results. Particularly, the relative external reproducibility (2σ) of the ²³²Th⁺/²³⁸U⁺ ion ratios measured at the particle scale remained below 10%, paving the way to use these mixed oxide particles in the field of nuclear safeguards.

The safeguards laboratories at Forschungszentrum Jülich provide the International Atomic Energy Agency (IAEA) with well-defined microparticulate uranium oxide reference materials for mass spectrometric verification measurements to support a sustainable and reliable quality control system for particle analysis in nuclear safeguards. For specific applications, such as chronometrical measurements further development of analytical methods including the quality control process for the evaluation of analytical data itself as well as of novel mixed uranium oxide microparticulate reference materials is required. But due to the extremely low production quantity of the microparticles (few µg) the characterization of doped uranium oxide microparticles is very challenging. Therefore, a co-precipitation method was adjusted to produce doped bulk-scale materials as “internal reference materials” which can be investigated by state-of-the-art analytical techniques to unravel the structural incorporation mechanism of relevant dopants, such as lanthanides, Th, and Pu, into uranium oxide. The determination whether a solid solution or segregated phases are formed in dependence on the chemical properties, ionic radii as well as the amount of tri- and tetravalent dopants will provide essential information about the applicability of these mixed compounds as reference materials in nuclear safeguards. Regarding the transferability to the particle production process in Jülich, the phase transformation from UO₃ to U₃O₈ is of particular interest. Therefore, the pristine materials (doped ammonium diuranate) were investigated with TG-DSC to identify the temperature of the phase transformation of UO₃ to U₃O₈ for the doped materials. Subsequently, the materials were calcined at the identified temperatures and structurally characterized with XRD, Raman, and IR spectroscopy. This presentation will provide an insight regarding the incorporation of tri- and tetravalent dopants, such as lanthanides, Th and Pu, into the uranium oxide structures applying ex situ and in situ techniques.