Search

Direct electrochemical reduction of solid ZrO2 to Zr metal assumes importance in the context of pyrochemical reprocessing of spent oxide nuclear fuel. In this context, electrochemical reduction experiments were carried out with powder compacted and sintered ZrO2 pellets as cathode in LiCl-x wt.% Li2O (x = 0, 0.1, 0.25 and 0.5) melts against graphite anode and in LiCl-(0.5 and 1.0) wt.% Li2O melts against platinum anode at 650 °C and the samples were analysed using XRD and SEM combined with EDS techniques. The pellets were found reduced to Zr in pure LiCl melt and to Zr and smaller amounts of Li2ZrO3 in LiCl-Li2O melts under specific experimental conditions. Similar cathodic polarization experiments carried out with Li2ZrO3 pellets converted those to a mixture of Zr and Li2ZrO3. Contrary to the results reported in the literature by earlier workers, the results of this study make it quite clear that both solid ZrO2 and Li2ZrO3 can be electro-reduced to Zr in LiCl-Li2O melts. However, the study carried out with two different cathode configurations reveal that the efficiency of electro-reduction is strongly influenced by the design of the oxide electrode and to some extent the Li2O content of the melt.

The CaCl 2 deficient ternary eutectic melt LiCl–KCl–CaCl 2 (50.5: 44.2: 5.3 mol %) was electrochemically characterised by cyclic voltammetry and polarization techniques in the context of its probable use as the electrolyte in the electrochemical reduction of solid UO 2 to uranium metal. Tungsten (cathodic polarization) and graphite (anodic polarization) working electrodes were used in these studies carried out in the temperature range 623 K–923 K. The cathodic limit of the melt was observed to be set by the deposition of Ca 2+ ions followed by Li + ions on the tungsten electrode and the anodic limit by oxidation of chloride ions on the graphite electrode (chlorine evolution). The difference between the onset potential of deposition of Ca 2+ and Li + was found to be 0.241 V at a scan rate of 20 mV/s at 623 K and the difference decreased with increase in temperature and vanished at 923 K. Polarization measurements with stainless steel (SS) cathode and graphite anode at 673 K showed the possibility of low–energy reactions occurring on the UO 2 electrode in the melt. UO 2 pellets were cathodically polarized at 3.9 V for 25 h to test the feasibility of electro-reduction to uranium in the melt. The surface of the pellets was found reduced to U metal.

Graphite anode used in the direct electrochemical de-oxidation of solid oxides in molten calcium chloride medium (FFC Cambridge process) degrades by formation of carbon oxides and erosion. The problem was studied by using graphite as anode in the electrochemical de-oxidation of powder compacted and sintered Nb 2 O 5 pellet electrodes in molten CaCl 2 at 1173 K and at a constant voltage of 3.1 V. The degradation of the graphite anodes, recovered from the electro-de-oxidation cells at predetermined intervals of time (2, 9, 22, 26, 35, 44 and 56 h) during the long-duration electrolysis, was determined by mass loss measurements and the morphological changes by SEM imaging. The graphite rods were found corroded differently along the lengths, with the maximum corrosion being at the melt-immersed surface of the electrode close to the melt–gas interface. The prominent carbon consuming reaction C + 2O 2− → CO 2 + 4e − apart, the impregnation of salt into the defects of the graphite anode and convection currents existing in the melt appeared to enhance the corrosion and hence the mass loss of graphite towards the later stages of electrolysis.

Electrochemical reduction behaviour of TiO 2 powder samples, sintered at 1423 K and 1673 K and weighing ∼ 1.5 g and ∼ 10 g, was studied by constant voltage electrolysis at 3.1 V at 1173 K. Both the samples were found to undergo reduction, but the low-temperature sintered and low-mass samples showed better overall reduction than the high-temperature sintered and high-mass samples. The reduction of the bulk parts of the samples became difficult as the sample size was increased from 1.5 g to 10 g. The size and location of the TiO 2 particles in the powder electrode played an important role in the reduction of the oxide to the metal.

Electrochemical behavior of graphite in LiCl and LiCl melts containing up to 1 wt% Li2O was studied by using cyclic voltammetry and electrolysis, in the context of its probable use as anode in the electrochemical reduction of solid UO2 in the molten salt at 650◦C. The lithium deposition and chlorine evolution potentials of LiCl melt were determined by CV of tantalum (cathodic polarization) and graphite (anodic polarization) working electrodes as −2.19 V and +1.21 V vs Ni/NiO reference electrode respectively. In LiCl-Li2O melts, carbon dioxide was formed on the graphite WE from > +0.75 V onwards and its current and potential range of formation increased with increase in the Li2O content of the melt. Cathodic polarization of a UO2 disc WE in LiCl-1 wt% Li2O melt showed that the reduction of UO2 to U occurs at −2.18 V and the lithium metal deposition on it at −2.36 V. Electro-reduction experiments carried out with a dense UO2 pellet cathode and graphite anode in both LiCl and LiCl-Li2O melts proved that the surface of the pellet was reduced to U metal at ≥ −2.2 V. The reduction became increasingly difficult with increase in the Li2O content of the melt. © The Author(s) 2015. Published by ECS. This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 License (CC BY, http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse of the work in any medium, provided the original work is properly cited. [DOI: 10.1149/2.0521506jes] All rights reserved.

Electrochemical behavior of platinum in LiCl and LiCl-(1-3) wt% Li 2 O melts at 650 °C was studied by cyclic voltammetry in the context of its use as anode in the direct electrochemical reduction of solid uranium oxide to uranium metal. By CV measurements with graphite and platinum (anodic polarization) and tungsten (cathodic polarization) working electrodes, the decomposition potential of LiCl and electrochemical potential of the reaction Pt + 2LiCl → PtCl 2 + 2Li were determined as 3.46 V and 3.14 V respectively. Three anodic reactions viz. (i) formation of Li 2 PtO 3 , (ii) oxygen evolution and (iii) platinum dissolution were found to occur on the platinum electrode in Li 2 O containing LiCl melts. The Li 2 PtO 3 formation was found to be a fast reaction with diffusion coefficient of O 2− ions in the melt as 4.53 × 10 −7 cm 2 /s. Oxygen evolution showed very high current densities when compared to that of the surface area limiting Li 2 PtO 3 formation. The platinum electrode surface was corroded by formation of Li 2 PtO 3 when the electrode was polarized for a longer period of time in the melt. The results of the study indicated that LiCl-2 wt.% Li 2 O melt could be the optimum electrolyte composition in the electro-reduction of uranium oxide with platinum anode.

Electrochemical behaviour of powder compacted and sintered Nb 2 O 5 pellet electrodes (~4 mm thick, open porosity ~10%) in molten CaCl 2 medium at 1173 K was studied using cyclic voltammetry and polarisation measurements. Similar dense Nb 2 O 5 pellets were subsequently electrolysed by using them as cathodes against graphite counter electrode in the molten CaCl 2 electrolyte medium with an aim to gain insight into the electro-reduction mechanism. The chemical changes that are taking place in the cathode and also in the electrolyte melt at different stages of electrolysis were analysed. The electro-reduction experiments were carried out at an applied voltage of 3.1 V and terminated after 2, 9, 22, 26 and 35 h of electrolysis and the visibly distinct layers present in the partially reduced samples were analysed. The analysis showed presence of Ca x Nb y O z compounds, sub-oxides of Nb like NbO 2 , NbO, etc. and metallic Nb in the samples. A pellet, which was electrolysed for 44 h was found to be reduced completely to Nb metal. The CaO, Ca and CaCO 3 produced and released to the melt during the cathodic polarisation was found to be associated with the electrochemical changes that are taking place on the electrode and this was ascertained by determination of the ‘total solubilised basicity (TSB)’ and ‘total insolubilised basicity (TIB)’ of the melt as a function of time during electrolysis. Based on the results, a three-stage mechanism comprising (i) cathodic insertion of Ca 2+ into the oxide pellet, (ii) reduction of the oxide electrode via formation and decomposition of ternary intermediates and (iii) chemical reduction of the cathode with electro-generated calcium metal has been proposed for electro-reduction of dense Nb 2 O 5 pellets.

pellets(52%TD).Therateofdeoxidationincreasedwithincreaseintheelectrolysistemperaturefrom923Kto1023K,butdecreasedonfurtherincreaseto1173K,duetosinteringofUparticlesonthesurfaceofthepellet.ThethicknessofUmetallayerincreased with increase in the applied voltage from 3.1 to 3.3 V whereas no reduction of the oxide occurred at 2.8 V. The behaviorof the graphite anode, in relation to the electro-reduction of the UO

Electro-reduction of solid UO 2 to U has been studied with molten CaCl 2 or LiCl as the electrolyte medium. Electro-reduction of thick (>3 mm), powder compacted and sintered pellets of UO 2 showed incomplete reduction resulting in a mixture of uranium metal and UO 2 . The extent of reduction of UO 2 to U was determined by employing a novel method called ‘metal estimation by hydrogen sensor (MEHS)’, in which the hydrogen evolved during the reaction of U metal in the reduced product with con. HBr was measured using an in-house developed polymer electrolyte based amperometric hydrogen sensor. The results of our investigations on incompletely reduced UO 2 pellets in both CaCl 2 and LiCl melts showed that the extent of reduction of different regions of the oxide pellet was different. It varied from 88.3% on the surface of the pellet as against 3.7% towards the centre bulk during electro-reduction in CaCl 2 (at 1173 K). The metallisation was found restricted to the surface of the pellets reduced in LiCl melt (at 923 K). Electro-reduction of small chunks of UO 2 pellet in CaCl 2 melt resulted in products with lower extent of reduction. Based on the measurements, a probable mechanism on the propagation of reduction through the solid UO 2 matrix during the electrochemical reduction process has been proposed.

Molten salt electrorefining based pyrochemical reprocessing is ideally suited for treating the spent metallic fuels. Development of all aspects of the pyroprocess flow sheet is in progress at IGCAR to enable the commercialisation of the process. Following the initial laboratory scale studies, engineering scale studies have been taken up on the molten salt electrorefining process and the consolidation process for recovering the actinides from cathode deposit. Studies are also being carried out on the direct electrochemical reduction of actinide oxides, methods for the fabrication of waste forms and materials and coatings.

The reduction of zirconium dioxide pellets by electro-deoxidation in molten calcium chloride-calcium oxide (900 °C) has been studied. In this technique, the solid oxide is cathodically polarized against a graphite counter electrode under a constant applied potential. Unlike other metal oxides that have been reduced by this technique, only a small area around the cathodic current-collector wire was reduced to zirconium metal with zirconia pellets sintered at ~1100 °C; the rest of the sample was largely calcium zirconate. Pellets sintered above 1200 °C showed better reduction near the cathode wire and the reduction extended to the entire surface of the pellet with the passage of time. However, reduction of the inner core was found to be increasingly difficult, because the surface metal layer thickened on continuous electro-deoxidation. An analysis of the experimental results showed that the poor electrical conductivity of the intermediate compound, CaZrO3 and its blocky morphology inhibited the electro-deoxidation process. The increase in the sintering temperature of the pellet made it better conducting. However, the pores formed in the thick zirconium metal layer in such samples were too small for an ideal contact between the inner core and the molten electrolyte and hence the reduction of the inner core remained incomplete. Within the scope of this study, it is concluded that preforms with good grain growth and porosity are necessary for the electro-deoxidation of solid zirconium oxide.

Aluminium chloride intercalation in graphite was studied by anodic oxidation of compacted graphite (rod) and graphite powder electrodes in sodium chloroaluminate melt saturated with sodium chloride at 175 °C. The studies carried out by employing both galvanostatic and cyclic voltammetric techniques had shown that the intercalation reactions take place only beyond the chlorine evolution potential of +2.2 V vs. Al on both the electrodes. The extent of intercalation reaction was directly related to the anodic potential and probably to the amount of chlorine available on the graphite anodes. In the case of graphite powder electrode, a distinctly different redox process was observed at sub-chlorine evolution potentials and this was attributed to the adsorption of chlorine on its high surface area. This finding contradicts a report in the literature that the intercalation reactions occur at potentials below chlorine evolution in the chloroaluminate melt.