Peter A. Zink

Invited) Solid Oxide Membrane Process for the Reduction of Uranium Oxide Surrogate in Spent Nuclear Fuel

Electrometallurgical processes are currently employed to treat spent metallic nuclear fuel. In an effort to extend this technology to spent oxide fuels, it would be useful to electrolytically reduce the oxide fuel to metal, thereby facilitating its subsequent electrorefining using traditional pyroprocessing methods. Traditional electrolytic processes for oxide reduction require a halide feed preparation and higher dissociation potentials compared to the solid oxide membrane (SOM) electrolysis process [1]. By employing an yttria-stabilized zirconia (YSZ) SOM tube [2, 3], metal oxides can be fed into a molten flux for electrolysis with a relatively low dissociation potential. The byproducts from the SOM process are also relatively environmentally friendly. As uranium (U) is a radioactive element in the actinide series, the oxide of one of the lanthanides, ytterbium (Yb), was chosen as the surrogate for uranium oxide. This is based on similarities in multivalent states, atomic structure, atomic size and other physical properties between U and Yb and their respective oxides. Varying amounts of Yb2O3 were added to lithium fluoride-ytterbium trifluoride (LiF-YbF3) flux (21 w% LiF and 79 w% YbF3), and differential thermal analysis – thermo-gravimetric analysis (DTA-TGA) was performed to determine the solubility of Yb2O3 in the flux below the melting point of Yb (820°C). Melting point (liquidus) and weight loss data of the various flux compositions was collected and analyzed. An example of the DTA result is shown in figure 1. Electron microprobe analysis of the quenched flux samples revealed two distinct phases, one Yb2O3 rich, and the other Yb2O3 poor. The distributions of Yb and oxygen in each of the phases were obtained using wavelength dispersive spectroscopy (WDS). Optical analysis of the flux samples indicate that the Yb2O3 rich phase increases with increasing Yb2O3 additions to the flux. Based on analysis of the data from TGA, WDS and the optical micrographs, it was determined that the maximum solubility of Yb2O3 in the flux, below the melting point of Yb, is 11 w%. In order to confirm that YSZ tube was compatible and stable in the LiF-YbF3-Yb2O3 flux during the SOM process, a stability test was completed in which the YSZ tube was submerged in the flux and heated to the SOM operating temperature for 6 hours. Optical micrographs of the sectioned sample show that YSZ is stable in the flux. As Yb metal is expected to form as a solid deposit during the SOM process, interactions between Yb metal, flux and stainless steel crucible were evaluated with a second stability test, performed in a very low pO2 environment. X-ray diffraction (XRD) analysis of the second stability test showed that ytterbium trifluoride (YbF3) was mostly converted into ytterbium difluoride (YbF2) by the Yb metal. A SOM run was performed at 800°C, below the melting temperature of the Yb metal. A potential was applied between the SOM-YSZ anode membrane and a stainless-steel cathode tube immersed in the flux and bubbling Argon. Yb metal was reduced on the cathode from the Yb2O3 in the flux and the associated oxygen was removed though the YSZ (SOM) tube. It is to be noted that some excess Yb metal was added to the flux to convert the YbF3 to YbF2. Before potentiostatic electrolysis was performed to produce Yb metal from the LiF-YbF2-Yb2O3 flux, a potentiodynamic scan was performed to determine the dissociation potential of Yb2O3. The result shows that dissociation potential (figure 2) obtained is consistent with the theoretical value of 1.6V. Based on this result, electrolysis was performed for 4 hours with 2.5V applied across the electrodes. Yb metal deposit was seen on the stainless steel cathode (figure 3). Analysis of the deposit shows evidence of Yb metal. Additional analysis will be performed to determine the condition of the YSZ tube and composition of flux after the SOM electrolysis. Future work will involve measurement of flux conductivity, analysis of long-term membrane stability, measurement of the efficiency of electrolysis, and optimization of experimental parameters for the purpose of designing larger scale reactors.

Analysis of the Electronic and Ionic Conductivity of Calcium-Doped Lanthanum Ferrite

Lanthanum ferrites with high mixed ionic and electronic conductivities are promising materials for solid oxide fuel cell (SOFC) cathodes. A-site deficient calcium-doped lanthanum ferrite (LCF) powders were synthesized and tested to characterize their electrochemical properties for use as cathode materials in single-step cofired SOFCs. Four-probe conductivity and oxygen permeability tests indicate superior mixed ionic and electronic conductivities compared to conventional SOFC cathode materials. Thermogravimetric data measured as a function of temperature and oxygen partial pressure (PO 2 ) along with the conductivity and permeability measurements were consistent with the defect model of LCF.

Magnesium Recycling of Partially Oxidized, Mixed Magnesium-Aluminum Scrap through Combined Refining and Solid Oxide Membrane Electrolysis Processes

Pure magnesium (Mg) is recycled from 19g of partially oxidized 50.5 wt.%Mg-Aluminum (Al) alloy. During the refining process, potentiodynamic scans (PDS) were performed to determine the electrorefining potential for magnesium. The PDS show that the electrorefining potential increases over time as the Mg content inside the Mg-Al scrap decreases. Up to 100% percent of magnesium is refined from the Mg-Al scrap by a novel refining process of dissolving magnesium and its oxide into a flux followed by vapor phase removal of dissolved magnesium and subsequently condensing the magnesium vapors in a separate condenser. The solid oxide membrane (SOM) electrolysis process is employed in the refining system to enable additional recycling of magnesium from magnesium oxide (MgO) in the partially oxidized Mg-Al scrap. The combination of the refining and SOM processes yields 7.4g of pure magnesium; could not collect and weigh all of the magnesium recovered.

Study of an Innovative Energy Storage and Recovery System based on the W/WO3 Oxidoreduction Reaction

Energy storage and recovery using the redox reaction is proposed. The system will store energy as tungsten metal (W) by reducing tungsten oxide with hydrogen. Thereafter steam will be used to re-oxidize the metal and recover hydrogen. The volumetric energy density of W for storing hydrogen is 21kWh/L of W based on the LHV of hydrogen. Theoretical treatment of isothermal kinetics has been extended in the present work to the reduction of tungsten oxide in powder beds. Experiments were carried out using a thermogravimetric technique under isothermal conditions at different temperatures. The reactions at 800°C were found to take place in the following sequence: WO3→WO2.9→WO2.72→WO2→W. Expressions for the last three reaction rate constants and activation energies have been calculated based on a model in which the intermediate reactions proceed as a front moving at a certain velocity while the first reaction occurs simultaneously in the entire bulk of the oxide.

Defect Chemistry and Electrical Properties of (La[sub 0.8]Ca[sub 0.2])[sub 0.95]FeO[sub 3−δ]

A–site deficient calcium doped lanthanum ferrite (LCF) powders are synthesized and tested to characterize their electrochemical properties for use as cathode materials in single–step co–fired solid oxide fuel cells (SOFCs). Four–probe conductivity and oxygen permeability tests show superior performance compared to conventional SOFC cathode materials. A point–defect model is applied to interpret the experimental data yielding concentrations of the predominant defects as a function of pO2 and temperature.

Defect Chemistry and Electrical Properties of ( La0.8Ca0.2 ) 0.95FeO3 − δ

A–site deficient calcium doped lanthanum ferrite (LCF) powders are synthesized and tested to characterize their electrochemical properties for use as cathode materials in single–step co–fired solid oxide fuel cells (SOFCs). Four–probe conductivity and oxygen permeability tests show superior performance compared to conventional SOFC cathode materials. A point–defect model is applied to interpret the experimental data yielding concentrations of the predominant defects as a function of pO2 and temperature.

Refractory Cathode Investigation for Single-Step Co-fired Solid Oxide Fuel Cells

A single-step co-firing process for fabricating planar solid oxide fuel cells (SOFCs) requires refractory electrodes to prevent excessive sintering of the electrode while facilitating full-density sintering of the electrolyte. Single cell current-potential curves and impedance measurements indicate that the majority of the performance losses occur in the cathode and are due to activation polarization. A-site deficient calcium and cerium doped lanthanum ferrite cathode powders were synthesized and investigated as possible refractory cathode materials with low activation polarization losses. Four-probe conductivity measurements indicated that all compositions were suitable as cathodes. However, reactivity with YSZ reduced the conductivity by as much as two orders of magnitude. Future experiments will investigate the applicability of a doped- ceria barrier layer to prevent reaction of lanthanum ferrite cathodes with YSZ electrolyte.

Defect chemistry and electrical properties of (La0.8 Ca0.2)0.95 Fe O3-δ

A–site deficient calcium doped lanthanum ferrite (LCF) powders are synthesized and tested to characterize their electrochemical properties for use as cathode materials in single–step co–fired solid oxide fuel cells (SOFCs). Four–probe conductivity and oxygen permeability tests show superior performance compared to conventional SOFC cathode materials. A point–defect model is applied to interpret the experimental data yielding concentrations of the predominant defects as a function of pO2 and temperature.