Search

Uranium dioxide (UO2) is a widely studied material due to its use as fuel in pressurised water reactors (PWR), and Raman spectroscopy is a technique of choice to characterise the evolution of its microstructure. UO2 crystallises in a fluorite CaF2 (space group Fm‐3m) structure that gives rise to a unique Raman signature, the T2g band. However, several other bands are often detected whose attribution remains unclear. The present study gives new insights on the Raman spectrum of UO2 thanks to the combination of isotopic labelling with 18O and Raman imaging. In addition to the expected T2g, U2 (LO), 2LO and U3 bands, we have detected a doublet at 885 and 925 cm−1, a U* band at 555 cm−1 in some specific areas and two bands located at 367 and 1196 cm−1. All Raman bands shift under the effect of the replacement of 16O by 18O, excepting for the U* band that could not be detected anymore. The isotopic shift ratio could be determined for 20% and 30% 18O labelling. No discrepancy in band position is observed between grains and grain boundaries, except for the U2(LO) band. We also evidence a difference between the U defect bands and the 885 and 925 cm−1 doublet bands evolution under labelling, although the latter also seems to be connected to the presence of defects in the material. [ABSTRACT FROM AUTHOR]

The current distribution structure of the thermal conductivity (TC) enhancements in UO 2 matrix can be divided into three categories: dispersed structure, continuous structure and oriented structure. This paper develops a new semi-continuous structure to improve the TC of UO 2. Ti 3 SiC 2 is selected as the enhancement in this paper due to its excellent irradiation resistance and high TC at elevated temperature. The semi-continuous structure of Ti 3 SiC 2 is formed through a simple centrifugal spheroidization of UO 2 and a followed powder mixing process. Both the theoretical prediction and experimental results indicate the semi-continuous structure has a close TC enhancing efficiency to the continuous structure. Meanwhile, the semi-continuous structure has the advantages of simplifying the fabrication process and can be applied to more kinds of enhancements compared to the continuous structure. The TC of Ti 3 SiC 2 /UO 2 is remarkably improved at elevated temperature compared with pure UO 2 , which is beneficial to the fuel safety. [ABSTRACT FROM AUTHOR]

We analysed with different methods the densification of UO 2 nanopowders in SPS under constant heating rate (CHR) and isothermal sintering conditions. The apparent activation energy of densification in SPS (75 kJ/mol with CHR method) is significantly smaller than in conventional sintering. It is shown that this is likely not an effect of the applied current. We also observed a threshold stress at 64 MPa for the transition from pressure-insensitive sintering (stress exponent n ≈0) to pressure-assisted sintering, suggesting that the contribution of the capillary stresses in such nanopowders is comparable with the typical stress applied in SPS. [ABSTRACT FROM AUTHOR]

In order to investigate the impact of fission products on the oxidation of spent fuel in the cladding separation process, the oxidation behavior and kinetic mechanism of UO 2 and impact of M (M = Ce, Sn, Sr, Se) oxides on the oxidation of UO 2 to U 3 O 8 were investigated by TG/DSC, XRD, FT-IR and Raman. It was found that the oxidation of UO 2 to U 3 O 8 went through two stages: UO 2 →U 3 O 7 ; U 3 O 7 →U 3 O 8. Doping M oxide in UO 2 could not change the two-step mechanism, and significantly inhibited the formation of U 3 O 8. The final products of CeO 2 and SrO-doped UO 2 oxidation contained 20% U 3 O 7 and 47% U 3 O 7 , respectively. The doping of M oxide increased the onset oxidation temperature of the two oxidation stages of UO 2 by 5–65 °C. In addition, the kinetic mechanism of UO 2 oxidation was investigated and found that the first and second oxidation stages were controlled by 1st order reaction or nucleation and growth, and 2nd order reaction, respectively. The doping of M oxide significantly affected the kinetic mechanism of the first stage. [ABSTRACT FROM AUTHOR]

We report a simple and general method for the production of UO 2 nanoparticles by heat-treating an aqueous mixture of uranyl acetate dihydrate and ethylenediamine. The products were characterized, and the effects of heat treatment temperature and time were investigated. When heat treatment time was fixed at 24 h, UO 2 could not be obtained at temperature lower than 160 °C, both UO 2 and ianthinite were obtained at 200 and 240 °C, while pure UO 2 nanoparticles were collected at 280 °C and higher temperatures. The average size of these nanoparticles varied in the range of 250–450 nm, which could be adjusted by tuning the reaction temperature and incubation time. Longer incubation time led to slight size increase due to Ostwald ripening. The mixture precursor took shorter time to transform to UO 2 at higher temperatures. A possible synthesis mechanism was proposed, including three stages: (I) initial precursor formation and transformation, (II) mid-term formation of ianthinite and UO 2 , and (III) the final formation of UO 2 nanoparticles. This research provides a facile method for the synthesis of UO 2 nanoparticles with excellent size stability in a relatively wide range of experimental conditions. Moreover, this method provides a new way for synthesis of ianthinite, a rare and hard-to-synthesize hydrous uranyl oxide. [ABSTRACT FROM AUTHOR]

Uranium dioxide (UO2) is the primary fuel material that is used in current nuclear reactors. As one of the most fundamental material parameters, grain boundary (GB) energy strongly influences many fuel properties, and the influences depend on the characters and properties of individual GBs. Using molecular dynamics simulations, a high throughput survey of GB energy in UO2 was carried out for the purpose of elucidating the roles of GB geometry such as misorientation and inclination, as well as the bonding nature of UO2, in affecting GB energy. GB energies in CeO2 were calculated as well for comparison with UO2 to investigate the generality of GB energy anisotropy in fluorite phase oxides. The results show significant GB energy anisotropy in both UO2 and CeO2 that is associated with the cubic symmetry of the fluorite structure. More interestingly, the GB anisotropy is found to be dependent not only on the crystal structure but also the ionic bonding. As such, the GB energy anisotropy in fluorite oxides has significant differences compared with that in fcc metals. The data obtained and the increased knowledge on GB anisotropy will facilitate GB engineering for nuclear fuels with improved properties. [ABSTRACT FROM AUTHOR]

Raman spectroscopy is one of the most useful techniques for studying the structure of UO2 and changes due to specific defects caused by doping, changes in stoichiometry, irradiation, or heating under oxidizing conditions. In this paper, we illustrate several aspects of the application of Raman techniques to the study of UO2, including the use of wavelength‐dependent excitation (455, 532, and 785 nm) to assess the effects of doping (Nd, Th, and Zr), ion irradiation, and in situ heating and oxidation (UO2 to U3O8). Additionally, we show examples of how correlative microscopy is possible using electron backscatter diffraction combined with Raman maps of specific vibration bands or of laser‐induced luminescence generated by rare‐earth dopants in the matrix. For each of these applications, we suggest optimal excitation wavelengths that vary depending on the desired data. Blue (455 nm) excitation tends to promote oxidation even at low powers, but because Raman spectra change little with doping, irradiation‐induced changes are easier to observe. Green (532 nm) excitation is optimal for observing electron–phonon resonance effects in UO2 and offers a good compromise for high‐temperature oxidation experiments, delivering high‐quality spectra for both UO2 and U3O8. Infrared (785 nm) excitation is best for observing "defect" bands associated with doping in UO2, as changes with irradiation are small. Raman spectroscopy is particularly suited for studying the stability of UO2 towards oxidation in the presence of dopants simulating fission products, where electron–phonon resonant effects, dopant ion luminescence, and mapping can be used together to investigate structural rearrangement as a function of temperature. These techniques can offer insight into microstructural changes in UO2 fuels at higher burnups envisioned in future reactors. [ABSTRACT FROM AUTHOR]

The dissolution rate of spent nuclear fuel under environmental conditions has been studied heavily to understand the impacts of a failed waste package on potential radionuclide release from a geologic repository. Multiple countries have evaluated these scenarios in oxidizing and reducing environments respective to their relevant repository conditions. The repository environment (e.g., water chemistry, temperature, oxygen content) and the fuel itself (i.e., chemical composition of the fuel) both heavily impact the dissolution rate of the spent fuel. This work examined the impact that rare earth element dopants (Ce, Nd, Yb) have on the fuel dissolution under repository relevant conditions with decreasing oxidizing conditions using a single pass flowthrough system. UO 2 samples were doped with Ce, Nd, or Yb with concentrations between 1 and 5 at%. The addition of dopants to the samples reduced the dissolution rates on most samples relative to pure UO 2 samples. Preliminary experiments that occurred in a less oxidizing environment showed a reduction in dissolution rate compared to fully oxidizing conditions. The results within highlight the importance of dopant behavior in used fuel dissolution modeling. [ABSTRACT FROM AUTHOR]

[Display omitted] • UO 2 were dominated by U(V) after exposure to γ-ray in either H 2 O or 1 M Cl- / Br-. • XPS shows (meta)-studtite formed on UO 2 after exposure to γ-ray in H 2 O. • UPS shows U(VI) on UO 2 after exposure to γ-ray in either H 2 O or 1 M Cl- / Br-. • Br 3 - was observed in 1 M Br- solution exposed to γ-radiation. In this work, we have experimentally studied UO 2 dissolution in pure water and in 1 M aqueous solutions of either Cl- or Br- exposed to γ-radiation. It has previously been found that high ionic strength can facilitate adsorption of dissolved UO 2 2+ on UO 2 surfaces. The adsorption is also affected by the solution pH relative to the point of zero charge of UO 2. In our experiments, Br 3 - was observed in 1 M Br- solution exposed to γ-radiation. Experiments confirmed that Br 3 - can quantitively oxidize UO 2. XPS and UPS were used to characterize potential surface modifications after exposure. The XPS results show that the UO 2 surfaces after exposure to γ-radiation in pure water and in 1 M aqueous solutions of either Cl- or Br- were significantly oxidized with U(V) as the dominating state. U 4f 7/2 and O 1 s spectra of the UO 2 surface after exposure to γ-radiation in pure water demonstrates the formation of uranyl peroxide secondary phases. UPS results indicate that there is a large percentage of U(VI) on the ultra-thin outer layer of UO 2 after exposure to γ-radiation in 1 M aqueous solutions of Br- and Cl-, and 100 % of U(VI) in the pure water case. [ABSTRACT FROM AUTHOR]

• Thermal diffusion kinetics of cesium in ceramic microcell UO2 fuels were examind using diffusion couples. • Diffusion couple experiments provided the diffusion length with the temperature and time, thus Arrhenius equations. • A cell wall composition of Al-Si-O exhinited lower cesium diffusivity than that of Si-O. • Based on the kinetic findings, a diffusion model that describes the macroscopic diffusion behavior of cesium is suggested. This work investigates the diffusion kinetics of cesium in ceramic microcell UO 2 fuels at temperatures ranging from 650 to 1200 °C to predict its release behavior under actual fuel circumstances. Diffusion couples containing ceramic microcell UO 2 pellets with silica-based cell wall compositions (with and without Al) were prepared and subjected to a heat treatment with abundant gaseous cesium source. From the experiment, kinetic data were obtained in the form of an Arrhenius equation: the addition of Al lowers the degree of cesium diffusivity. The favorable contribution of Al may stem from the formation of the cesium compound as a solid form of Cs-Al-Si-O. The findings here provide a diffusion model describing diffusion mechanism of cesium in the cell wall. Also, this study suggest an evaluation method that can be used in UO 2 fuels with trapping agents from a kinetic standpoint. [ABSTRACT FROM AUTHOR]

This paper presents a comprehensive nuclear forensics signatures analysis of UO 2 fuel pellets, allowing their provenance to specific manufacturers. The study is based on a synthetic database designed for the 4th international Galaxy Serpent exercise circulation, which mimics real fuel pellet characteristics. The scenario of the exercise is completely fictional, including the data used and manufacturers' names, locations, etc. The signatures investigated include chemical form, pellet dimensions, equivalent boron content, Gd additive, impurities distribution, and isotopic composition. The consistency levels of the results, which determine the confidence in the provenance process, were divided into four levels due to the varying resolving power of the signatures. The paper emphasizes the importance of a graphical representation of the results to simplify interpreting multi-degree-of-freedom data, typical to nuclear forensics investigations, including the amalgamation of quantitative and qualitative information. The analysis connects 11 orphan pellets and a single UO 2 powder to five manufacturers. The presented methodology offers a straightforward approach to data interpretation in nuclear forensic investigations and highlights the need for continued research and development in this field. [ABSTRACT FROM AUTHOR]

Despite the growing interest in the spark plasma sintering (SPS) of uranium dioxide, its sintering mechanisms have yet to be studied in great detail. Herein we propose a direct method to calculate the apparent activation energy for densification, Q act , and the stress exponent, n , for SPS of nearly stoichiometric UO 2. A set of experiments performed at different heating rates (CHR) and different pressures levels allowed us to calculate Q act and n , respectively, though we were limited to a theoretical density between 50% to 75 %. The master sintering curve was employed as a complementary method to compare Q act. The average values were Q act =96 kJ/mol (CHR), Q act = 100 kJ/mol (MSC) and n = 1.4. We have therefore proposed grain boundary diffusion coupled with grain boundary sliding as the densification mechanism. The activation energy in SPS tends to be lower compared with that in other processes like conventional sintering (250−450 kJ/mol), creep (350−550 kJ/mol) and hot pressing (222 kJ/mol and 480 kJ/mol). This decrease could be due to the effect of the electric field combined with the higher heating rates, typical of SPS. [ABSTRACT FROM AUTHOR]

Predicting the nuclear fuel microstructure at each moment of its irradiation cycle (nominal, power transient, accidental conditions) is a significant nuclear safety issue. For that, it is necessary to understand the impact of irradiation parameters on the microstructure. This study provides insight about the temperature effect on dislocations. In situ thermal annealing up to 1400°C on pre-irradiated polycrystalline UO 2 thin foils was performed inside a TEM for the first time. The aim of the current study is to establish the kinetic and the mechanisms of thermal recovery of extended defects induced by irradiation. Whatever the initial irradiation conditions, extended defect recovery was observed around 1000-1100°C, in good agreement with literature data. Dislocation line disappear mainly by climb and dislocation loops move by pencil glide along the <110> Burger vector directions. Defect growth by coalescence of dislocation loops is also observed. Image, graphical abstract [ABSTRACT FROM AUTHOR]

• Void change during ion implantation in UO 2 was studied using in situ TEM. • The measured void size strongly depends on the TEM capabilities and settings. • When damage increases, small voids nucleate and saturate after 0.2–0.3 dpa. • Void size remains stable during the ion irradiations. • Temperature and xenon atom incorporation do not impact void change. Transmission electron microscopy experiments coupled with 4 MeV Au and 390 keV Xe implantations were conducted on polycrystalline UO 2 thin foils in order to study the void change with damage, temperature (ranging from −180 to 1100 °C) and exogenous xenon atoms incorporation. Void size is weakly dependent on the ion dose in the range of [0.01–10] × 1014 i/cm2, whereas void density increases strongly at the early beginning of irradiation and then saturates after 3–4 × 1013 i/cm2. Change in void size and density are shown as weak functions of temperature and xenon atoms incorporation in these experimental conditions. These results indicate a heterogeneous void nucleation and a change monitored by ballistic effects. Studying voids is very tricky. This work also highlights the strong dependence of the measured void size on the TEM settings and capabilities. Thus, it is important to discuss only data obtained in the same conditions to derive influence parameters. [ABSTRACT FROM AUTHOR]

Microcrystalline CeO 2 , ThO 2 , and UO 2 were irradiated with 198 MeV 132Xe ions to the same fluence at temperatures ranging from 25 °C to 700 °C then characterized by synchrotron X-ray diffraction and X-ray absorption spectroscopy. All samples retain crystallinity and their nominal fluorite-type phase at a fluence of 1.5 × 1013 ions/cm2. Both CeO 2 and ThO 2 display defect-induced unit cell expansion after irradiation at room temperature (∼0.15% and ∼0.10%, respectively), yet as irradiation temperature increases, the maximum swelling produced decreases to ∼0.02%. Alternatively, UO 2 shows an initial contraction in unit cell parameter (approximately −0.05%) for room temperature irradiation, most likely related to irradiation-enhanced annealing or irradiation-induced oxidation. At higher temperatures (above 200 °C) UO 2 begins to swell, surpassing its unit cell parameter prior to irradiation (∼0.05%), an effect which could be attributed to minor reduction in uranium oxidation state in vacuum. However, while CeO 2 irradiated at room temperature undergoes partial reduction, both UO 2 and ThO 2 exhibit no measurable change in cation oxidation state as evidenced by X-ray absorption spectroscopy. All samples display a decrease in irradiation-induced heterogeneous microstrain as a function of increasing irradiation temperature. [ABSTRACT FROM AUTHOR]

The escape of radionuclides from underground spent nuclear fuel disposal facilities will likely result from anoxic dissolution of spent nuclear fuel by intruding groundwater. Anoxic dissolution of various forms of uranium dioxide (UO 2), namely bulk pellet, powder and thin film, has been investigated. Long-duration static batch dissolution experiments were designed to investigate the release of uranium ions in deionized water and any surface chemistry that may occur on the UO 2 surface. The dissolved uranium concentration for anoxic dissolution of nearly stoichiometric UO 2 was found to be of the order of 10−9 mol/l for the three different sample types. Further, clusters (∼500 nm) of homogenous uranium-containing precipitates of ∼20–100 nm grains were observed in thin film dissolution experiments. Such a low solubility of UO 2 across sample types and the observation of secondary phases in deionized water suggest that anoxic UO 2 dissolution does not only occur through a U(IV) (solid) to U(VI) (aqueous) process. Thus, we propose that dissolution of uranium under anoxic repository conditions may also proceed via U(IV) (solid) to U(IV) (aqueous) , with subsequent U(IV) (precipitates) in a less defective form. Quantitative analysis of surface-sensitive EBSD diffractograms was conducted to elucidate lattice-mismatch induced cracks observed in UO 2 thin film studies. Variable temperature anoxic dissolution was conducted, and no increased uranium concentration was observed in elevated temperatures. • Extended dissolution of different forms of UO 2 was conducted under anoxic conditions. • Dissolved uranium of the same magnitude (10−9 mol/l) was obtained for all three different physical solid forms of UO 2. • Different secondary phases of uranium were observed from dissolution in di-water. • These U(IV) secondary phases limit dissolved concentrations of uranium. [ABSTRACT FROM AUTHOR]

Thermally conductive microcell UO 2 pellets in which metallic networks enclose UO 2 granules have been highlighted due to their enhanced thermal performance for a potential candidate of accident-tolerant fuels. However, studies on thermal expansion effects of microcell UO 2 pellets are still lacking. Here, stress and strain distributions of the fabricated UO 2 -5 vol% Mo microcell pellet under thermal expansion were investigated. After we confirmed that the numerical results agreed well with experimental measurement in terms of linear thermal expansion, stress and strain distributions of the microcell pellets were calculated under maximum stress and maximum strain conditions. As a result, formation of Mo network decreases the maximum hoop stress and the maximum radial strain of UO 2 pellets by 50.9% and by 29.4%, respectively under a linear heat generation rate of 200 W/cm. Finally, the UO 2 -5 vol% Mo microcell pellet can decrease the maximum cladding stresses by 36.7%. • Stress and strain distributions of microcell UO 2 -5 vol% Mo pellets under thermal expansion were investigated. • Formation of Mo network in UO 2 pellets could reduce the maximum hoop stress and radial strain of UO 2 pellets. • Microcell UO 2 -5 vol% Mo pellets could considerably reduce the maximum cladding stresses. [ABSTRACT FROM AUTHOR]

The thermoelastic effects on the behaviour of subsecond laser heating experiments on UO 2 to temperatures approaching 3000 K are explored using a thermoelastic model coupling heat transport and large deformation hyperelasticity, with consideration of thermoelastic effects. A series of steady state experiments at temperatures up to ∼2600 K are conducted and used to calibrate heat loss models. A series of subsecond transient laser flash melting experiments are then conducted and simulated. Simulation results reproduce the observations well without tuning of the transient model. The importance of considering thermoelastic effects in modelling the behaviour of solid materials with temperature and stress gradients typical of laser flash heating experimentation is discussed. [ABSTRACT FROM AUTHOR]

• Introduction of a kinetic evolution of the irradiation defects density in the formulation of the thermal conductivity of UO 2. • Calibration of this model on dedicated separate effects tests. • Quantification of the modelling uncertainty based on inverse method technique. Thermal conductivity of the fuel is a key parameter to assess the temperatures in nuclear fuel. Several empirical or mechanistic models exist in the literature. The objective of this work is to improve one of the empirical formulation by introducing a kinetic evolution of the irradiation defects density in the irradiated fuel. This evolution is calibrated amongst some dedicated experimental data. An inverse probabilistic method is also applied to quantify the modelling uncertainty of the obtained formulation, considering the scattering and the uncertainty of the experimental database available for both un-irradiated and irradiated materials. [ABSTRACT FROM AUTHOR]

UO 2 fuel pellets are often doped with chromium oxide to obtain favourable properties such as higher density, improved thermal stability, large grain sizes, improved pellet-clad interaction margins, and increased fission gas retention during transients. Chromium has a low solubility limit in UO 2 , with past experimental work reporting solubility limits ranging between 0.004 to 0.06 wt.% Cr. Due to its low solubility, segregation of Cr ions to the grain boundary may occur. Further, the complexity of these boundaries may be high as observed in other ceramics resulting in disordered or amorphous regions along the boundary, affecting a range of material and operational properties of the fuel pellet. To assess these disordered regions, in this work we study amorphous undoped and Cr doped UO 2 systems (containing 10–50 at.% Cr3+) that have been modelled using classical molecular dynamics methods incorporating Cr3+ into the well-used CRG potential library. Diffusion coefficients, pre-exponential factors, and activation energies for diffusion were computed for oxygen ions, assessing the impact of structure and extrinsic species on migration. Oxygen diffusion was observed to be much faster in the undoped amorphous system compared to its crystalline counterpart. Oxygen diffusion in doped systems decreased with increasing Cr concentration, highlighting the importance of additives to retain fission products and other migratory species. [ABSTRACT FROM AUTHOR]

To understand the thermodynamic behavior of xenon (Xe) bubbles in uranium dioxide (UO 2), the pressure of Xe nanobubbles as a function of temperature was systematically investigated using molecular dynamics (MD) simulations at temperatures ranging from 300 to 2100 K. The results show that the maximum pressure decreases with increasing temperature and bubble size. Interestingly, for a critical Xe/Schottky Defect (SD) ratio of 1 to 1.25, within a temperature range from 300 to 1500 K, the increase of pressure attributed to the change of kinetic energy offsets the decrease of pressure due to the change of potential energy with respect to volume, resulting in a roughly constant bubble pressure. The critical Xe/SD ratio slightly decreases with increasing bubble size. As Xe/SD ratios are lower or higher than the critical ratio, the bubble pressure increases or decreases with increasing temperature, respectively. Based on MD simulations of the pressure of unconfined (free) Xe at temperatures of 300, 900, 1500 and 2100 K, a virial equation of state (EOS), which was expanded to the fifth term, is reported for gas, fluid and solid, unconfined Xe at temperatures higher than 300 K. The EOS for free Xe can also be used to assess the pressure of large bubbles in a matrix, since the confinement effect can be negligible. The confinement effect of the UO 2 matrix causes an overestimation of Xe density in highly pressurized nanobubbles. Considering the interaction between Xe and the UO 2 lattice, the free Xe EOS is modified to calculate the pressure of small intragranular Xe nanobubbles in UO 2. [ABSTRACT FROM AUTHOR]

• Dissolution of monocrystalline (111) uraninite surface in H 2 O 2 solutions. • In situ characterization identifies secondary precipitates. • Schoepite forms in 0.005 M H 2 O 2 ; only studtite forms in 0.5 M H 2 O 2. • Orientation of precipitates is dictated by uraninite (111) surface crystallography. • Surface emersion at high [H 2 O 2 ] results in the formation of schoepite. The (111) surface of uraninite obtained by monolith cleavage was reacted in H 2 O 2 solutions of increasing concentrations ([H 2 O 2 ]) from 0.005 to 0.5 M. The reacted interface was characterized in situ by µRaman spectroscopy, or, for 0.5 M H 2 O 2 , Grazing-Incidence X-Ray Diffraction (GI-XRD). At low [H 2 O 2 ], schoepite and studtite coexisted on the reacted surface. With increasing [H 2 O 2 ], studtite predominated, and a pattern of black strips cross-cutting at 60° suggested control of the solid neoformation by the (111) surface structure. In situ GI-XRD for 0.5 M H 2 O 2 revealed the presence of studtite only, with an initial anisotropy in particle orientation (c* axes parallel to the uraninite surface), and a subsequent reorientation of growing particles. Upon draining out the solution, the formation of schoepite was observed after a few hours of emersion. This result reveals that apparently benign post-experimental manipulations can lead to significant changes in the nature of observed secondary precipitates. [ABSTRACT FROM AUTHOR]

Thermochemical conversion of UO 2 to UCl 3 using AlCl 3 and Al metal in LiCl–KCl melt at 773 K is reported here. The reaction product was assessed qualitatively by UV–Vis spectroscopy, cyclic voltammetry (CV) and square wave voltammetry (SWV). The extent of conversion efficiency was quantified through UV–Vis spectroscopy by measuring the absorbance of U(VI) at 420 nm after dissolving the reaction product in 1 M HNO 3 + 1 M H 3 PO 4 medium. The conversion process was optimized by varying concentrations of AlCl 3 (stoichiometric × 1 to 4) and the duration of reactions (45 min–5 h). About 93% conversion of UO 2 was achieved by doubling the stoichiometric requirement of AlCl 3 in 50 g of LiCl–KCl salt in 2.5 h. • The conversion of UO 2 to UCl 3 is achieved using AlCl 3 and Al metal in LiCl–KCl. • Cyclic and square wave voltammetry ascertain the conversion through redox peaks. • UV–Vis spectrophotometry is used to quantify the conversion of UO 2 to UCl 3. • More than 92% conversion is achieved at optimized experimental conditions. [ABSTRACT FROM AUTHOR]

Models for calculating dose rates of spherical particles as well as in fuel cracks are important for radionuclide source term estimations. Dose rates from UO 2 based fuels were calculated for planar, spherical, and crack geometries. The escape probability for α-particles in spherical UO 2 particles was derived as a continuous equation. The dose rate increased with increasing spherical radius due to the decreasing relative volume of the surrounding water layer. The model produced escape probabilities that were closely predicted by the theoretical derivation. It was shown that the dose rate in water filled fuel cracks with width smaller than the range of an α-particle led to dissolution rates that were lower per unit surface area. • Alpha dose rates of spherical and crack geometries of UO 2 based fuel were studied. • An expression for escape probability as function of emission depth was derived. • Dissolution rates as a function of crack width and particle radius were modelled. • Comparison with empirical treatments of the stopping power spectrum was performed. [ABSTRACT FROM AUTHOR]

To improve the fundamental understanding of the processes controlling spent fuel alteration under deep geological conditions, the influence of dopants (Cr and Cr/Al) on UO 2 matrix stability is evaluated for "modern" types of light water reactor (LWR) fuels with additives. The uranium dissolution behavior of as-prepared 0.06 wt%Cr and 0.05 wt%Cr/0.02 wt%Al doped UO 2 pellets is studied in simplified groundwater containing 19 mM HCO 3 − in autoclaves under hydrogen atmosphere. Sintered disks were exposed to simulated highly carbonated conditions, representative of a repository scenario of water intrusion after a hypothetical canister failure. The uranium concentration released was ∼10−7 M for the 0.05 wt%Cr/0.02 wt%Al doped UO 2 pellet and ∼10−6 M for 0.06 wt%Cr doped UO 2 pellets, after 170 days. The results indicate that the amount of dissolved uranium is slightly lower compared to previous studies in absence of a reductant gas phase, but clearly above the solubility of UO 2 (am, hyd). The initial measured pH was 8.9 ± 0.1, which gradually approached a constant value of ∼9.2 ± 0.1. Solid characterization at the end of the dissolution experiment, by SEM, Raman spectroscopy and XRD shows that the surface of all pellets remains almost unaltered. The experimental results indicate a potential oxidative dissolution of UO 2 , which could be attributed to the presence/intrusion of dissolved oxygen in the prepared synthetic groundwater. In order to identify the mechanism of uranium release, the datasets from the batch experiments are simulated with a PHREEQC model previously calibrated with results of existing spent fuel UOx leaching experiments. The model includes the geochemical processes that are relevant for the studied experimental conditions: (i) non-oxidative dissolution of UO 2 , (ii) UO 2 oxidation with O 2 , (iii) dissolution of U(VI) by carbonate water, (iv) reduction of oxidized U(VI) on the surface pellet by activated H 2. The ability to activate the dissolved H 2 is studied by implementing a kinetically controlled process of H 2 activation on the Cr surface in the model. [ABSTRACT FROM AUTHOR]

UO 2 corrosion is a key issue in the study of the safe storage of spent nuclear fuel (SNF). In this work, corrosion experiments were performed on UO 2 samples exposed to 18O labelled water during 290 days at room conditions, thereafter a complete analysis of the solid and leachant (labelled water) allow us obtaining new data and exploring new methods. Thus, a new application of the Nuclear Reaction Analysis (NRA), as a simple and low-cost approach, has been developed to estimate the oxidized layer thickness and the diffusion coefficient of oxygen through the oxidized layer. These results were verified by comparison with the ones obtained by traditional Secondary Ion Mass Spectrometry (SIMS). The diffusion coefficient obtained is two orders of magnitude larger than the scarce published ones at room temperature, highlighting the great effect of the composition of the oxidized layer due to the UO 2 corrosion on this parameter. In addition, the water used in the corrosion experiments (leachant) has been analyzed by Raman spectroscopy, being the main purpose for the in situ analysis. The data obtained in this study bring out the great ability (higher than expected) of the oxygen to break through the oxidized layer which could endanger the SNF pellet integrity. Besides, this work shows new methods for analyzing corrosion experiments which can be used to obtain new data for completing the database. • A complete analysis of the solid and leachant (labelled water) of the UO 2 corrosion during 290 days. • A new application of the Nuclear Reaction Analysis (NRA), as a simple and low-cost approach. • New data of the oxidized layer thickness and the diffusion coefficient of oxygen through the oxidized layer. • New methods for analyzing corrosion experiments. [ABSTRACT FROM AUTHOR]

The exploitation of the data acquired by Raman spectroscopy on UO 2 related materials requires the understanding of the so-called defect peaks. To do so, we measured the Raman spectrum of a He-implanted UO 2 sample as a function of temperature. Two annealing steps are evidenced that correspond to known annealing temperatures reported in literature. These results are discussed in order to identify what defect could be at the origin of the Raman defect peaks. [ABSTRACT FROM AUTHOR]

A model investigating the role of geometry on the alpha dose rate of spent nuclear fuel has been developed. This novel approach utilises a new piecewise function to describe the probability of alpha escape as a function of particulate radius, decay range within the material, and position from the surface. The alpha dose rates were produced for particulates of radii 1 μ m to 10 mm, showing considerable changes in the 1 μ m–50 μ m range. Results indicate that for decreasing particulate sizes, approaching radii equal to or less than the range of the α -particle within the fuel, there is a significant increase in the rate of energy emitted per unit mass of fuel material. The influence of geometry is more significant for smaller radii, showing clear differences in dose rate curves below 50 μ m. These considerations are essential for any future accurate prediction of the dissolution rates and hydrogen gas release, driven by the radiolytic yields of particulate spent nuclear fuel. • An approach to simulate α-dosimetry from spherical particulates is presented. • Probability of α-particle escape for curved surfaces has been derived. • The role of geometry has a considerable influence for particulate radii <50 μm. • Results comparable with previous MCNPX model to within 0.7% of total dose rate. [ABSTRACT FROM AUTHOR]

• Dislocation loop change under dual-beam irradiation was studied using in situ TEM. • With ion fluence increase, small dislocation loops nucleate and grow. • Similar evolution is studied regardless of the energy losses regime. • Electronic excitations lead to an acceleration of the extended defect formation kinetics. During in reactor operations, the slowing down of fission fragments generates most of the radiation damage in uranium dioxide. Fission fragments deposit their energy to both atomic and electronic subsystems through nuclear and electronic interactions. To study the possible synergistic effects of nuclear and electronic energy deposition on the microstructure, UO 2 thin foils have been irradiated with 0.39 MeV Xe and/or 6 MeV Si ions at 298 K using single or dual beam ion irradiations. The evolution of extended defects was characterized by in situ Transmission Electron Microscopy. Results show a similar evolution with fluence increase, irrespective of the ion beam: a nucleation of small dislocation loop that increase in density up to a saturation value, followed by an increase of the average loop size. However, the kinetics differs according to the irradiation conditions. In the case of the dual beam irradiation, all the phenomena occur at a lower value of damage level. The local increase of the temperature along the 6 MeV Si ion path seems to lead to an enhanced defect mobility, as in the case of irradiations performed at higher temperatures. [ABSTRACT FROM AUTHOR]

• Modeling and burnup analysis nuclear fuel assemblies comprising both solid and annular fuel was carried out. • Lattice pitch of the models was optimized in terms of infinite multiplication factor. • The burnup dependent fission rate and infinite multiplication factor were determined. • Reactivity and temperature sensitivity of reactivity were investigated. • Production and depletion of isotopes were also studied. Modeling and burnup analysis of Uranium dioxide ( U O 2) and Mixed Oxide (MOX) fuel in square assemblies comprising both solid and annular rods was carried out. The lattice pitch of the models was optimized first in terms of infinite multiplication factor (k ∞) using OpenMC. The burnup analysis was performed at the optimized pitch by lattice physics code Dragon for 1620 EFPD. The burnup dependent fission rate density was studied, and the thermal fission rate was found higher for the models as compared to reference designs. The values of k ∞ and reactivity were determined and higher values were found for the models at low fuel burnup period. However, models showed a higher criticality period than the reference design consisting of only the annular rod. The fuel and moderator temperature sensitivity of reactivity was investigated for different fuel burnup periods. The burnup-dependent atomic concentrations of fissionable materials, minor actinides, and neutron poisons were also studied. The production of neutron poison, namely Xe-135 and Sm-149 and minor actinides viz. Np-237, Am-241, Cm-242 and Cm-244 were found lower in models as compared to the reference designs. [ABSTRACT FROM AUTHOR]

A model to determine the dose rate of a planar alpha-emitting surface has been developed. The approach presented is a computationally efficient mathematical model using stopping range data from the Stopping Ranges of Ions in Matter (SRIM) software. The alpha dose rates as a function of distance from irradiated UO 2 spent fuel surfaces were produced for benchmarking with previous modelling attempts. This method is able to replicate a Monte Carlo (MCNPX) study of an irradiated UO 2 fuel surface within 0.6% of the resulting total dose rate and displays a similar dose profile. • Bragg peak shift due to attenuation through fuel can be approximated as linear. • Bragg curve maintains functional form with peak shift. • Single function can be manipulated to describe all particle paths. • Clarification over error made in literature regarding the probability of alpha escape. • Results comparable to Monte Carlo attempts while improving computational efficiency. [ABSTRACT FROM AUTHOR]

UO 2 pellets exposed to 1.85 MBq and 3.30 MBq Am-241 sources with a 30 μm separation in solution were studied under Ar and H 2 atmospheres. The 1.85 MBq Am-241 source was shown to have a too low radiolytic production to cause any measurable change in UO 2 surface oxidation state. However, a significant oxidation was observed after exposure to the 3.30 MBq source under Ar atmosphere through the production of U(V), as seen from the U4f 7/2 -peak deconvolution and valence band peak analysis. The H 2 atmosphere was shown to suppress the oxidation of the UO 2 pellet surface as well as the dissolution in both Milli-Q and NaHCO 3 solutions. The dissolved uranium concentration was reduced by a factor 10 under H 2 atmosphere as compared to the Ar atmosphere experiments. [ABSTRACT FROM AUTHOR]

• New fits were created for the average grain boundary mobility and energy in UO 2 • Our mechanistic grain growth model includes the curvature driving force and resistance due to mobile pores and bubbles. • Our model has been implemented in BISON and considers the temperature history. The purpose of this work is to develop a mechanistic model of grain growth in UO 2 fuel during reactor operation. The model development builds on published experimental data on UO 2 grain growth, as well as atomistic and mesoscale simulation results. We begin by developing new fits with temperature T (in K) for the average grain boundary (GB) energy and mobility in UO 2 using literature data, where the GB energy γ ¯ = (1.56 − 5.87 × 10 − 4 T) ± 0.3 J/m2 and the GB mobility M ¯ = (2.14 ± 0.15 × 10 − 7) exp (− (290 ± 22 kJ / mol) / R T) m4/(Js). We then discuss the pinning of grain boundaries by porosity, including porosity left over after sintering and fission gas bubbles that form during operation. We present our mechanistic model and validate it using existing grain growth data. Finally, we implement the model in the BISON fuel performance code and quantify its impact for constant and transient power cases. Our model produces similar results to an empirical model for the constant power case, but it predicts more grain growth than the empirical model for the transient power case. We attribute this discrepancy to the new mechanistic model's ability to account for the impact of the temperature and power history. [ABSTRACT FROM AUTHOR]

• SIMS and TEM techniques were combined to compare the thermal behaviour of Cs and Xe in UO 2. • Both elements form bubbles with different growth kinetics. • At 1600 °C, caesium is found to be highly mobile in the UO 2 matrix while Xe distribution does not evolve. Xenon and caesium are among the most impacting fission products when studying the nuclear fuel: xenon for its role on the fuel rod thermomechanical behaviour during reactor operation and caesium in the case of atmospheric radioactive release during an accident in a nuclear power plant. This paper focuses on the comparison of caesium and xenon thermal behaviour in polycrystalline uranium dioxide (UO 2) pellets. Caesium-133 or xenon-136 stable isotopes were introduced in depleted UO 2 samples by ion implantation at a maximum concentration of 0.08 at% at a depth of around 140 nm below the sample surface. Annealing under reducing atmosphere (Ar/H 2 5 %) was performed at 1000 °C or 1600 °C, which corresponds respectively to a representative temperature during nuclear reactor operation (at the centre of the fuel pellets) and during an accident. The caesium migration in UO 2 was investigated by Secondary Ion Mass Spectrometry and compared to the thermal behaviour of xenon in UO 2 at 1600 °C. Transmission Electron Microscopy was performed in order to characterise UO 2 microstructure before and after annealing. The results indicate that caesium has a different behaviour than xenon with which it is often compared for its release from the nuclear fuel. In particular, we highlight a difference between the growth kinetics of caesium and of xenon bubbles at 1600 °C which can be correlated to the availability of thermal vacancies in UO 2 and to the different ability of Xe and Cs atoms for thermal resolution. [ABSTRACT FROM AUTHOR]

• Confocal laser scanning profilometry/focussed ion beam characterization of laser-UO 2 interaction. • Quantitative evaluation of laser ablation (LA) craters generated by ns laser irradiation. • Removal rate and surface texture highly reproducible. • Depth resolution limited by bumpy shape and surface roughness of LA craters as well as by their conical silhouette. • Microscopic characterization revealed porosity formation and melting during laser-sample interaction. A total number of 960 laser ablation (LA) craters produced by a ns-LA system were characterized systematically for their depth profiling capability of UO 2 single crystals. For gaining improved understanding of laser-UO 2 interaction, a variable number of circular laser pulses, having a nominal diameter of 25 µm, were shot repeatedly (N = 5) on the surface of a UO 2 single crystal. To this end, performance at both diverse fluence settings and different number of individual laser pulses was evaluated at laser repetition rates of 1 Hz and 2 Hz. As such, the impact of cumulative fluence of 50, 150, 250, 350, 450, 550, and 650 J cm−2 on the integrity of a UO 2 single crystal was investigated. Confocal laser scanning profilometry and a dual beam focused ion beam microscope were employed to assess essential parameters such as LA rate, surface roughness, silhouette and depth of LA craters. While the LA rate decreased with increasing single shot fluence, surface roughness continuously intensified with expanded cumulative fluence. The average depth of LA craters was higher, the higher was the cumulative fluence. The conical geometry and the variability of the bumpy bottom of LA craters limited the experimentally achievable depth resolution to ~ 1 µm. Identical LA experiments carried out weeks apart, produced very similar data, proving high reproducibility of the laser-UO 2 interaction. [ABSTRACT FROM AUTHOR]

High thermal conductivity additives are being explored to make next generation accident tolerant fuels for light water reactors (LWRs). The goal is to create composite fuels that will more effectively dissipate heat, thus lowering the high centerline temperatures that develop in current LWR fuel. These fuels can employ various additives and the fuel microstructures vary depending on the manufacturing process. In this work, a thermal resistor model is developed to quickly estimate the effective thermal conductivity of a composite with a high thermal conductivity additive. This model is unique in that it employs a parameter, the continuousness of the secondary constituent, to consider the composite microstructure in its predictions. A genetic algorithm was developed to measure a composite's continuousness. The model and algorithm are validated using data from literature from four experimental composites composed of UO 2 with beryllium oxide and silicon carbide additives. The model predictions have an absolute mean error across a wide temperature range of less than 0.50 W / m K in most cases. These tools are then used to create a set of guidelines for microstructure design. [ABSTRACT FROM AUTHOR]

We report here on the generation of simulated defects on sintered uranium dioxide fuel pellets by taking advantage of the underwater laser ablation procedure and their subsequent characterization. This work, we believe, can play a role towards the validation of the performance of fuel pellet inspection machines. A repetitive fiber laser, capable of delivering pulses of nanosecond duration, in conjunction with a galvo-scanner served as the machining tool in the experiment. The study of the dependence of mass ablation rate on laser fluence, water column height, repetition rate, and beam scanning speed formed the bulk of this work. A water column of height ∼3 mm above the pellet surface in combination with a laser fluence of lying within 6–7 J/cm2 yielded the maximum ablation rate. The generated defects were analyzed using optical and electron microscopy. Clean defects, e.g. longitudinal cracks, circumferential cracks, pits, and end caps of different sizes and depths were created on the pellet surface by varying the laser parameters, beam travel trajectory, and beam scanning parameters. The mechanical pressure arising out of shock wave generated as a result of the confinement of the laser-produced plasma and the collapse of cavitation bubbles together with the microfluid jet formed due to the implosion of the bubbles pushed the molten ablated products from the interaction zone thereby facilitating the generation of cleaned machined surfaces. An insight into the physical processes' operative in the underwater laser machining process has been offered, albeit qualitatively. • Simulated defects were generated on sintered UO 2 pellets using a pulsed fiber laser. • The defects, e.g. pits, cracks, missing surface area were formed by laser ablation. • The dimension of the fabricated defects can be controlled. • Underwater ablation resulted in cleaner defects without any bulging near the edge. • Defective pellets can be utilized for authentication of pellet inspection machines. [ABSTRACT FROM AUTHOR]

In UO 2 , due to the existence of U 5 f electrons, the strong-correlation effects are significant, and correction of on-site Coulomb interaction is necessary. For strongly-correlated electronic system like UO 2 , however, it is a challenge to consider such correction during simulation of their dynamic processes under radiation. In this study, a new method, i.e., AIMD + U, in which ab initio molecular dynamics method and correction of on-site Coulomb interaction between U 5 f electrons are combined, is employed to investigate the response of UO 2 to low energy radiation. The threshold displacement energies for O and U atoms have been determined, and defect creation and distribution have been provided. New mechanism for defect generation, and novel defect types, including uranium (oxygen) dumbbell interstitials and uranium antisite defects, are revealed from this study. • A new AIMD + U method is employed to investigate the radiation response of UO 2. • Novel dumbbell interstitials and antisite defects are revealed. • Oxygen interstitial prefers to occupy octahedral sites. [ABSTRACT FROM AUTHOR]

During the dissolution of the spent nuclear fuel (SNF) some radionuclides are released to the solution simultaneously from different sources in the fuel. This is of particular importance to some radionuclides that contribute to the Instant Release Fraction (IRF), which govern the initial radiation dose during the dissolution of the SNF. In this work a model that is able to discriminate between the different contributions responsible for the total concentration of a radionuclide in solution was developed. The model permits to establish that uranium and radionuclides that dissolved congruently with the UO 2 matrix came from two sources as a function of time: oxidized phases on the surface of the SNF including fines and the matrix itself. Other radionuclides such as Ru and Rh were released from metallic precipitates with dissolution rates lower than the matrix dissolution rate. In the case of radionuclides that were expected to contribute to the IRF, this work showed that Cs, Rb and Sr had initial release rates higher than uranium because a fraction of such radionuclides were segregated from the matrix during the irradiation. Actually it was calculated that the fraction of those radionuclides in the grain boundaries in a BWR SNF powder sample from the center part of a pellet (burnup 42 MWd/kgU) was 2.1%, 0.9%, and 0.6% for Cs, Rb, and Sr, respectively. In addition, the model permitted to calculate the duration of their contribution to the IRF, matrix dissolution governed the release of such radionuclides after 137 days, 75 days, and 164 days for Cs, Rb, and Sr, respectively (at these times, the contribution of the release from grain boundaries was lower than the 0.1%). • Model developed to establish the contribution of segregated phases to SF dissolution. • Mo, Tc Pu, Am, Ce and La dissolve congruently with the matrix. • Ru and Rh segregated from the matrix. • Cs, Rb, Sr partially segregated to grain boundaries (2.1, 0.9 and 0.6%, respectively). [ABSTRACT FROM AUTHOR]

The solubility of UO 2 in molten LiF–BeF 2 (2:1 mol) (FLiBe) eutectic salt at 600 °C was studied by chemical and electrochemical methods. The results of dissolution experiment showed that the saturated solubility of UO 2 in this melt was 2.37 × 10−3 mol/kg and its corresponding apparent solubility product ( K s p ' ) was approximately 1.67 × 10−5 mol3/kg3. When more Li 2 O were added to the FLiBe–UF 4 system, the cathodic peak current of U(IV) in the electrochemical cyclic voltammetry (CV) curve accordingly decreased because of precipitate formation. The precipitate corresponded to UO 2 as determined by stoichiometric ratio (concentration variation of U4+ and O2−) and X-ray diffraction (XRD) analysis. Compared to the chemical analysis method, the CV technique was confirmed to be more feasible for accurate determination of concentration of U(IV). Meanwhile, the K s p ' value was also obtained to be 1.33 × 10−5 mol3/kg3 during the whole oxide titration procedure, which was highly consistent with that from the dissolution experiment. With the value of K s p ' , the allowable amount of dissolved oxide ions (oxide tolerance) can be theoretically estimated in the FLiBe–UF 4 system. [ABSTRACT FROM AUTHOR]

A series of sintered UO 2 pellets doped with lanthanide (Ce, Nd, Yb) elements were investigated using powder X-ray diffraction, Raman spectroscopy, thermogravimetric analyses and differential scanning calorimetry. A combination of electron microprobe and thermogravimetric analyses, for oxygen content, enabled precise determination of the hypostoichiometry for lanthanide-doped samples at 1 and 5 atom percent. Two Raman laser wavelengths (785 and 455 nm) have afforded greater sensitivity to spectroscopic signatures of the phonon bands (1LO and 2LO) associated with oxidation of (U 1-y M y)O 2-x and the anion defects introduced by lanthanide substitution. Oxygen hypostoichiometry forces a reduction in the average coordination number surrounding (U,M) sites, which is compensated by a decrease in U–O bond length, and concomitantly the lattice parameter, consistent with the obtained Raman spectra. The evolution of O/M ratio up to (U 1-y M y)O 2 after oxidation was also examined using Raman spectroscopy, revealing that the 'defect band', including a component attributed to oxygen vacancies (∼540 cm−1) and the 1LO phonon (∼575 cm−1) increased in intensity with increasing dopant concentration and upon oxidation. The lanthanide dopants inhibited oxidation to U 3 O 8 , most prominently for Yb 5 at%, having been delayed by ∼180 °C. Thermogravimetric analyses reveal an early oxidation feature that may be related to influx of O to satisfy hypostoichiometry up to (U 1-y M y)O 2 , possibly stabilizing a U 4 O 9 or U 3 O 7 intermediate, delaying oxidation to U 3 O 8. • Oxidation characteristics for hypostoichiometric Ln-doped UO 2-x revealed. • Two Raman lasers provide spectrally selective information on UO 2-x. • This is the first oxidation study on hypostoichiometric Yb-doped UO 2-x. • Yb-doping strongly inhibits U 3 O 8 formation by 180 °C relative to pure UO 2. • New spectroscopic and oxidation signatures compared with prior studies of UO 2-x. [ABSTRACT FROM AUTHOR]

During reactor operation, the formation of bubbles at the grain boundaries can alter the fracture properties and subsequently the thermo-mechanical behavior of nuclear fuels. In this work, we discuss a phase-field fracture model, its implementation in MOOSE (Multiphysics Object-Oriented Simulation Environment), and its application to simulate porosity dependent brittle intergranular fracture. The presence of gas bubbles on grain boundaries has a non-negligible impact on crack initiation and propagation. Three-dimensional simulations are performed to investigate the influence of bubble geometry, porosity and loading conditions on the fracture strength. The correlation between fracture strength and porosity suggested by numerical simulations has a good agreement with experimental measurements. [ABSTRACT FROM AUTHOR]

To evaluate the stability, potential reactivity, and relaxation mechanisms on different uraninite surfaces, surface energy values were calculated and structural relaxation was determined for the (111), (110), and (100) crystallographic faces of uranium dioxide (UO2) using quantum mechanical (density functional theory) and empirical potential computational methods. Quantum mechanical results are compared with empirical potential results, which use surface slab models with two different geometries, as well as various different empirical force fields. The strengths and weaknesses of the different approaches are evaluated, and surface stabilizing mechanisms such as relaxation, charge redistribution, and electronic stabilization are investigated. Quantum mechanical (q.m.) surface energy results are in agreement with the relative surface energy trends resulting from calculations using three different empirical potential sets for uranium and oxygen (two from Catlow 1977; one from Meis and Gale 1998), and with empirical force-field values published in the literature (Abramowski et al. 1999, 2001). The (111) surface consistently has the lowest surface energy (0.461 J/m² from q.m. calculations) and the smallest amount of surface relaxation, followed by the (110) surface (0.846 J/m²; q.m.), and the (100) surface (1.194 J/m²; q.m.) (quantum mechanical surface energy values in parentheses are for surface slabs with a thickness of four UO2 units). Differences exist, however, in the absolute values of surface energies calculated as a function of potential set used. Quantum mechanical values are consistently lower than values calculated using empirical potential methods. A fourth potential set is presented that is derived from fitting electrostatic and short-range repulsive parameters to experimental bulk properties and surface energies and relaxations from quantum mechanical calculations. [ABSTRACT FROM AUTHOR]