Investigation of electrorefining of metallic alloy fuel onto solid Al cathodes

This work concerned the electrorefining of UZr and UPuZr alloys on a solid aluminium cathode, in the LiCl-KCl eutectic melt containing U³⁺, Pu³⁺, Np³⁺, Zr²⁺ or Zr⁴⁺, Am³⁺, Nd³⁺, Y³⁺, Ce³⁺ and Gd³⁺ chlorides. During constant current electrolyses, the use of a cathodic cut-off potential (−1.25 V versus Ag/AgCl) allowed to selectively deposit actinides (mainly U), while lanthanides remained in the salt. The aim was to determine the maximal load achievable on a single aluminium electrode. The total exchange charge was 4300 C, which represents the deposition of 3.72 g of actinides in 4.17 g Al, yielding a composition of 44.6 wt% An in Al. It was shown that the melting of the cathode contributed to increase the total amount of actinides deposited on the aluminium.     

Spent fuel radionuclide source term model for assessing spent fuel performance in geological disposal. Part II: Matrix alteration model and global performance

In the framework of the research conducted on the long term evolution of spent nuclear fuel under geological disposal conditions, a source term model has been developed to evaluate the instantaneous release of radionuclides (RN) (instant release fraction, IRF) and the delayed release of the RN which are embedded within the matrix. This model takes into account most of the scientific results currently available except the effect of hydrogen and the current knowledge of the uncertainties. IRF was assessed by considering the evolution with time of the RN inventories located within the fuel microstructure to which no confinement properties can be allocated over the long term (gap, rim, grain boundaries). This allows for bounding values for the IRF as a function of time of canister breach and burnup. The matrix radiolytic dissolution was modeled by a simple kinetic model neglecting the recombination of radiolytic species and the influence of aqueous ligands. The oxidation of the UO₂ matrix was assumed not to be kinetically controlled. Spent fuel performance was therefore demonstrated to mainly depend on the reactive surface area.     

Simulation study of carbon impurity dynamics on tungsten surfaces exposed to hydrogen ions

The depth profile of C impurity deposited on a W target exposed to H⁺ and C⁺ impurities at a concentration of C: 0.8% has been calculated in terms of segregation, diffusion and chemical erosion. For the segregation, the Gibbsian model has been used. For the diffusion, a concentration dependent diffusion model (C in WC and/or C) has been utilized. For the chemical erosion, the chemical erosion yield much lower than that for the H-C system has been applied. The calculated depth profiles at 653 K and 913 K are in good agreement with the XPS data. The agreement indicates that there is a significant contribution of segregation, which shifts the maximum C concentration to the top surface in the depth profiles. On the other hand, there are little contributions from diffusion and chemical erosion, which are related closely to formation of WC in the target.     

Enhanced diffusion under alpha self-irradiation in spent nuclear fuel: Theoretical approaches

Various theoretical approaches have been developed in order to estimate the enhanced diffusion coefficient of fission products under alpha self-irradiation in spent nuclear fuel. These simplified models calculate the effects of alpha particles and recoil atoms on mobility of uranium atoms in UO₂. They lead to a diffusion coefficient which is proportional to the volume alpha activity with a proportionality factor of about 10⁻⁴⁴ (m⁵). However, the same models applied for fission lead to a radiation-enhanced diffusion coefficient which is approximately two orders of magnitude lower than values reported in literature for U and Pu. Other models are based on an extrapolation of radiation-enhanced diffusion measured either in reactors or under heavy ion bombardment. These models lead to a proportionality factor between the alpha self-irradiation enhanced diffusion coefficient and the volume alpha activity of 2 × 10⁻⁴¹ (m⁵).     

Vaporization behavior and Gibbs energy of formation of Rb2ThO3

The thermodynamic stability of rubidium thorate, Rb₂ThO₃(s), was determined from vaporization studies using the Knudsen effusion forward collection technique. Rb₂ThO₃(s) vaporized incongruently and predominantly as Rb₂ThO₃(s)=ThO₂(s) + 2Rb(g) + 1/2 O₂(g). The equilibrium constant K=p_Rb²·p_O2^1/2 was evaluated from the measurement of the effusive flux due to Rb vapor species under the oxygen potential governed by the stoichiometric loss of the chemical component Rb₂O from the thorate phase. The Gibbs energy of formation of Rb₂ThO₃ derived from the measurement and other auxiliary data could be given by the equation, Δ_fG°(Rb₂ThO₃,s)=−1794.7+0.42T ± .     

Thermodynamic considerations and prediction of the primary coolant activity of Tc

A physical model has been developed to describe the coolant activity behaviour of ⁹⁹Tc, during constant and reactor shutdown operations. This analysis accounts for the fission production of technetium and molybdenum, in which their chemical form and volatility is determined by a thermodynamic treatment using Gibbs-energy minimization. The release kinetics are calculated according to the rate-controlling step of diffusional transport in the fuel matrix and vaporization from the fuel-grain surface. Based on several in-reactor tests with defective fuel elements, and as supported by the thermodynamic analysis, the model accounts for the washout of molybdenum from the defective fuel on reactor shutdown. The model also considers the recoil release of both ⁹⁹Mo and ⁹⁹Tc from uranium contamination, as well as a corrosion source due to activation of ⁹⁸Mo. The model has provided an estimate of the activity ratio ⁹⁹Tc/¹³⁷Cs in the ion-exchange columns of the Darlington Nuclear Generating Station, i.e., 6 × 10⁻⁶ (following ∼200 days of steady reactor operation) and 4 × 10⁻⁶ (with reactor shutdown). These results are consistent with that measured by the Battelle Pacific Northwest Laboratories with a mixed-bed resin-sampling device installed in a number of Pressurized Water Reactor and Boiling Water Reactor plants.     

Laser annealing in combination with mass spectroscopy, a technique to study deuterium on tokamak carbon samples, a tool for detritiation

In this study, a method is presented based on mass spectroscopy to measure the areal density of deuterium on a graphite surface exposed to tokamak discharges. The studied sample was cut from a bumper limiter exposed in the TEXTOR tokamak and annealed by a 1 J Excimer laser (KrF). The energy used was 400 mJ cm⁻², which is below the threshold for ablation, 1 J cm⁻². The release of HD and D₂ was measured by a mass spectroscopy set-up and no other species released from the sample were detected in this experiment. The amount of D released from the sample after 20 laser pulses was measured to 7 × 10¹⁶ D atoms per cm⁻² (for this particular sample) and most of the hydrogen at the surface was released in the first pulse, as checked by nuclear reaction analysis (NRA) techniques, which gave changes of the amount of deuterium before and after laser annealing. The sensitivity in this experiment was 5 × 10¹⁴ atoms per cm⁻² for HD and 5 × 10¹³ atoms per cm⁻² for D₂.     

Radionuclide release from high burnup spent fuel during corrosion in salt brine in the presence of hydrogen overpressure

In the case of a contact between groundwater and Fe-based spent fuel disposal containers in a repository large amounts of hydrogen will be produced by the corrosion of iron, which may result in significant hydrogen pressures. To quantify to what extent the hydrogen overpressure may counteract radiolysis enhanced matrix dissolution, related experimental work has been performed. High burnup spent fuel was corroded in 5.6 mol (kg H₂O)⁻¹ NaCl solution applying H₂ overpressures (experimental set 1) <0.17 bar by radiolysis, (experimental set 2) 2.8 bar by Fe corrosion, (experimental set 3) 3.2 bar by external H₂ gas. In the absence of Fe (experimental set 3) the UO₂ matrix dissolution rate decreased by a factor of about 10. In this test the concentrations of U, Np, Tc in solution were found to be decreasing by at least two orders of magnitude, and ranging within the same level as in the presence of Fe powder (experimental set 2). However, Pu and Am concentrations (experimental set 3) were less affected, due to the high sorption capacity for these radioelements onto Fe corrosion products.     

In situ crack growth observation and fracture toughness measurement of hydrogen charged Zircaloy-4

The effect of hydrogen on the fracture behaviour of a Zircaloy-4 alloy was analysed performing simultaneous fracture mechanics tests of small SE(B) specimens and in situ observation of crack initiation and propagation inside the chamber of a scanning electron microscope. Load and displacement were continuously measured and J_IC, J-R curves and CTOD determinations were obtained. Detailed images of the zone close to the crack tip were taken and the resistance to crack growth was correlated with hydrogen content and hydride morphology. The size and orientation of hydride precipitates showed an important influence on the fracture process. A good agreement with results obtained using standard CT specimens was met.     

Comparison of binding energy to the sorption enthalpy of radionuclides with trap materials used in LMFBRs

In liquid metal fast breeder reactors (LMFBR), traps are provided in the primary coolant circuit to reduce the contamination due to the deposition of long lived γ-emitting nuclides. The binding energies of the radionuclides with iron and nickel were estimated using Pauling’s electronegativity. The results are comparable to the sorption enthalpies derived from the experimental isotherms.     

Development and evaluation of a new Eu/SmPd3 source for use with Gd Mössbauer spectroscopy

A ¹⁵⁵Eu/¹⁵⁴SmPd₃ (about 231 MBq) source for use with ¹⁵⁵Gd Mössbauer spectroscopy was developed by a novel method. In the novel method, the isotopically enriched ¹⁵⁴SmPd₃ compound was prepared by the conventional solid state reaction of ¹⁵⁴Sm(HCOO)₃ and PdH_x in a hydrogen atmosphere at 1273 K for 18 h, which is simpler than the previously reported method. In order to increase the reaction areas, palladium fine particles used to synthesize the PdH_x hydride were prepared by a chemical solution process. Performance of the newly developed source was evaluated by observing the ¹⁵⁵Gd Mössbauer spectra of known compounds, GdPd₃ and cubic Gd₂O₃ at 12 K. The obtained results indicated that the developed source is fine enough to investigate the structural characteristic of various materials containing gadolinium.     

A creep rupture criterion for Zircaloy-4 fuel cladding under internal pressure

The usual criterion which limits the cladding strain to 0.01 to prevent the creep rupture under internal pressure seems too conservative for application to transport and interim storage. So we have analysed CEA’s data on this subject for CWSR Zircaloy-4 in order to find a less conservative criterion. Temperatures between 350 and 470 °C were studied for stresses between 100 and 550 MPa, according to the irradiation level from 0 to 9.5 × 10²⁵ n m⁻². Except for high stressed irradiated material (because of low ductility), the plastic instability appears as the major mechanism of rupture. For the unirradiated material, it is essentially due to the stress increase with strain. This instability is accelerated by annealing for the irradiated one at moderate or low stress. From these considerations, we propose a new rupture criterion for CWSR Zircaloy-4 cladding submitted to internal pressure, for both unirradiated and irradiated materials.     

Determining the shear properties of the PyC/SiC interface for a model TRISO fuel

The fracture behavior of TRISO-coated fuel particles is dependent on the shear strength of the interface between the inner pyrolytic carbon (PyC) and silicon carbide coatings. This study evaluates the interfacial shear properties and the crack extension mechanism for TRISO-coated model tubes using a push-out technique. The interfacial debond shear strength was found to increase with increasing sample thickness and finally approached a constant value. The intrinsic interfacial debond shear strength of ∼280 MPa was estimated. After the layer is debonded, the applied load is primarily transferred by interfacial friction. A non-linear shear-lag model predicts that the residual clamping stress at the interface is ∼350 MPa, and the coefficient of friction is ∼0.23, yielding a frictional stress of ∼80 MPa. These relatively high values are attributed to the interfacial roughness. Of importance in these findings is that this unusually high interfacial strength could allow significant loads to be transferred between the inner PyC and SiC in application, potentially leading to failure of the SiC layer.     

Characterization of hydrogen concentration in Zircaloy claddings using a low-frequency acoustic microscope with a PVDF/LFB transducer

Hydrogen embrittlement is one of the major mechanisms responsible for the degradation of ductility of Zircaloy cladding materials. Currently the characterization of hydrogen concentration (HC) very often relies on destructive methods that are time-consuming and costly. In this research, an ultrasound-based nondestructive evaluation (NDE) technique is reported for the determination of HC in Zircaloy claddings. This ultrasound-based NDE technique employs a low frequency acoustic microscope (AM) with a PVDF/LFB transducer and a Fourier-based signal processing technique. With this AM technique, a relation between the ultrasound wavespeed and the HC of Zircaloy is established. A resolution of HC measurements with the current technique is demonstrated to be better than 200 ppm. This NDE technique has been developed with an aim to have a better resolution and also to be potentially applied to poolside inspection.     

Heats of formation of (U,Mo)Al3 and U(Al,Si)3

The heats of formation of (U,Mo)Al₃ intermetallic compounds were obtained by measuring the reaction heats of U-Mo/Al dispersion samples by differential scanning calorimetry. Based on literature data for the reaction heats of U₃Si/Al and U₃Si₂/Al dispersion samples, the heats of formation of U(Al,Si)₃ as a function of the Si content were calculated. The heat of formation of (U,Mo)Al₃ becomes less negative as the Mo content increases. Conversely, the heat of formation of U(Al,Si)₃ becomes more negative with increasing Si content.     

EBSD investigation of SiC for HTR fuel particles

Electron back-scattering diffraction (EBSD) can be successfully performed on SiC coatings for HTR fuel particles. EBSD grain maps obtained from thick and thin unirradiated samples are presented, along with pole figures showing textures and a chart showing the distribution of grain aspect ratios. This information is of great interest, and contributes to improving the process parameters and ensuring the reproducibility of coatings.     

Fuel oxidation and thermal conductivity model for operating defective fuel rods

A model has been developed to describe the fuel oxidation behaviour, and its influence on the fuel thermal conductivity, in operating defective nuclear fuel rods. The fuel-oxidation model is derived from adsorption theory and considers the influence of the high-pressure environment that results from coolant entry into the fuel-to-clad gap. This model is in agreement with the fuel-oxidation kinetics observed in high-temperature annealing experiments conducted at 1473-1623 K in steam over a range of pressure from 0.001 to 0.1 MPa. Using a Freundlich adsorption isotherm, the current model is also consistent with recent experiments conducted at a higher pressure of 7 MPa. The model also considers radiolytic effects as a consequence of fission fragment bombardment in the fuel-to-clad gap. This treatment suggests that radiolysis-assisted oxidation is insignificant in operating defective rods (as compared to thermal effects), as supported by limited in-reactor data. The effects of diffusion of the interstitial oxygen ions in the solid in the operating rod is further discussed.     

Spent fuel radionuclide source-term model for assessing spent fuel performance in geological disposal. Part I: Assessment of the instant release fraction

A source-term model for the short-term release of radionuclides from spent nuclear fuel (SNF) has been developed. It provides quantitative estimates of the fraction of various radionuclides that are expected to be released rapidly (the instant release fraction, or IRF) when water contacts the UO₂ or MOX fuel after container breaching in a geological repository. The estimates are based on correlation of leaching data for radionuclides with fuel burnup and fission gas release. Extrapolation of the data to higher fuel burnup values is based on examination of data on fuel restructuring, such as rim development, and on fission gas release data, which permits bounding IRF values to be estimated assuming that radionuclide releases will be less than fission gas release. The consideration of long-term solid-state changes influencing the IRF prior to canister breaching is addressed by evaluating alpha self-irradiation enhanced diffusion, which may gradually increase the accumulation of fission products at grain boundaries.     

Behavior of molybdenum in pyrochemical reprocessing: A spectroscopic study of the chlorination of molybdenum and its oxides in chloride melts

The high temperature reactions of molybdenum and its oxides with chlorine and hydrogen chloride in molten alkali metal chlorides were investigated between 400 and 700 °C. The melts studied were LiCl-KCl, NaCl-CsCl and NaCl-KCl and the reactions were followed by in situ electronic absorption spectroscopy measurements. In these melts Mo reacts with Cl₂ and initially produces MoCl₆²⁻ and then a mixture of Mo(III) and Mo(V) chlorocomplexes, the final proportion depending on the reaction conditions. The Mo(V) content can be removed as MoCl₅ from the melt under vacuum or be reduced to Mo(III) by Mo metal. The reaction of Mo when HCl gas is bubbled into alkali chloride melts yields only MoCl₆³⁻. MoO₂ reacts in these melts with chlorine to form soluble MoOCl₅²⁻ and volatile MoO₂Cl₂. MoO₃ is soluble in chloride melts and then decomposes into the oxychloride MoO₂Cl₂, which sublimes or can be sparged from the melt, and molybdate. Pyrochemical reprocessing can thus be employed for molybdenum since, after various intermediates, the end-products are chloride melts containing chloro and oxychloro anions of molybdenum plus molybdate, and volatile chlorides and oxychlorides that can be readily separated off. The reactions were fastest in the NaCl-KCl melt. The X-ray diffraction pattern of MoO₂Cl₂ is reported for the first time.     

Analysis for hydrogen particle balance of plasma-wall system in the large helical device

The hydrogen particle balance of the plasma-wall system in the large helical device (LHD) is analyzed, using a zero dimensional model for plasma particles, neutrals in vessel and hydrogen inventory in wall. Based on the measurement of neutral gas pressure, plasma density and the pumping speed of the cryo-pumping system, it is found that the hydrogen retained in the wall desorbes with short and long time constant. The short term desorption is of order of 10²¹ atoms with a time constant of a few minutes, which is much smaller than the wall pumping for one shot, 10²² atoms. In a long time scale of about one experimental day, the wall absorbs significantly large amounts of hydrogen, up to 10²⁴ atoms. One of the possible reasons for the large wall pumping is a carbon deposition layer on the first wall surface. The effect of hydrogen retention on density control is also discussed.