LiCl-Li2O Molten Research Articles

Development of a Li2O Sensor Based on a Yttria Stabilized Zirconia Membrane for Oxide Reduction in a Molten LiCl-Li2O Electrolyte at 650°C

LiCl-Li2O salt is a widely used electrolyte for the electrochemical reduction of spent oxide (mainly uranium oxide) fuels, and the Li2O concentration is usually controlled at about 1.41 mol % for optimum operation and corrosion mitigation of the platinum anode material. Due to the small difference in reduction potential between UO2 and Li2O, some Li2O will be reduced during the oxide reduction process, leading to a Li2O deviation from the desired 1.41 mol %. Monitoring of the Li2O concentration is desired for proper control of the electrochemical oxide reduction process. In this paper, a Li2O sensor based on a yttria stabilized zirconia (YSZ) membrane was developed and tested to evaluate the feasibility of using the Li2O sensor to monitor the Li2O concentrations in the range of 0.57 to 1.69 mol % Li2O in LiCl-Li2O salt systems—one without dissolved Li metal and another with 0.24 mol % dissolved Li metal. The experimental results show that the open circuit potential of the Li2O sensor logarithmically responded to the concentration of Li2O in LiCl-Li2O salts with or without the presence of Li metal, suggesting that YSZ appears promising for real-time monitoring of the Li2O concentration in LiCl-Li2O salt for an oxide reduction process. The Li2O sensor developed herein is based on a potentiometry measurement that requires a stable, reliable reference electrode (RE), particularly for long-term Li2O concentration monitoring. To this end, a novel Ag/AgCl RE that is contained in a high-density MgO tube with a closed end was developed and exhibited stable electrode potential, chemical compatibility with LiCl-Li2O, and good mechanical strength. The performance of the newly developed Ag/AgCl RE was demonstrated in Li2O monitoring by comparing it with a traditional Ni/NiO RE.

Comparative study of monolithic platinum and iridium as oxygen-evolving anodes during the electrolytic reduction of uranium oxide in a molten LiCl–Li2O electrolyte

A series of four electrolytic reduction runs was performed in molten salt at bench scale to compare the performance characteristics of monolithic platinum and iridium as oxygen-evolving anodes, while simultaneously reducing uranium oxide to metal. In each run, 25 g of uranium oxide particulate was loaded into a permeable steel basket, which, in turn, was immersed in a pool of LiCl—1 wt% Li2O at 650 °C. Both anodes, each 3 mm in diameter, were suspended vertically in the salt pool, adjacent to the steel cathode basket. The anodes were connected in parallel to separate direct current power supplies with the uranium oxide-loaded basket as the common cathode. A cell voltage (3.1 V) was intermittently applied to the system with both power supplies operating concurrently, effecting the reduction of uranium oxide to uranium metal at the cathode basket and the simultaneous oxidation of oxygen anions in the salt to oxygen gas at each anode. Anode and cathode potentials and currents were recorded to compare the performance of platinum vis-a-vis iridium. After completing the series of runs, both anodes were removed and subjected to dimensional, chemical, and microscopic analyses. Even though the accumulated charges on each anode over the series of four runs were similar, the platinum anode exhibited up to 29% reduction in cross-sectional area compared to < 3% for the iridium anode.

Hot corrosion behaviour of nickel-cobalt-based alloys in a lithium molten salt

In this study, hot corrosion behaviours of Haynes 282 (H282), Nimonic 90 (N90), and Haynes HR160 (HR160) in LiCl-Li2O molten salt were investigated at 650 °C under argon atmosphere. The corrosion rate was in the order of H282 < N90 < HR160, depending on the morphology and structure of the scale. H282 showed a localised corrosion behaviour, while N90 showed a uniform corrosion behaviour. For HR-160, intergranular corrosion was observed. The high Cr content caused intergranular corrosion in the corrosion layer. Thus, excess Cr has a deleterious effect on the corrosion kinetics during hot corrosion in LiCl-Li2O molten salt.

Identification, measurement, and mitigation of key impurities in LiCl-Li2O used for direct electrolytic reduction of UO2

Direct electrolytic reduction (DER) of UO2 utilizes a molten LiCl-Li2O electrolyte operating at 650 °C.Productive processes at the cathode are UO2 and Li2O reduction (followed by metallothermic reduction of UO2 by Li metal). Non-productive reduction current has been observed that could be attributed to impurities produced from interaction of the salt with water. Cyclic voltammetry (CV) scans of molten mixtures of LiCl containing Li2O, LiOH, and/or Li2O2 indicate that LiOH is the primary cause of reduction current preceding UO2 and Li2O reduction. LiOH reduction appears as a broad soluble-soluble transition and forms a pre-reduction plateau characteristic of micro-electrode reactions. Electrochemical response to LiOH in the salt is reduced with addition of Li0, providing a practical approach to eliminating LiOH from LiCl-Li2O salt before the DER process. Generation of H2 gas was detected while heating Li0 with LiCl-Li2O-LiOH, consistent with a reaction mechanism where LiOH reduction causes formation of Li2O and H2.

On the Corrosion Performance of Monel 400 in Molten LiCl-Li2O-Li at 923 K

The corrosion resistance of a Ni-Cu alloy, Monel 400, in molten LiCl-Li2O-Li at 923 K (650 °C) was investigated. Exposure testing of Monel 400 samples submerged in molten LiCl-2 wt pct Li2O solutions with Li concentrations between zero and 1 wt pct was performed at 923 K (650°C) for 20 hours. Post exposure surface analysis was performed using X-ray diffraction, scanning electron microscopy, energy dispersive X-ray spectroscopy, optical microscopy, micro-Vickers hardness testing, and X-ray photoelectron spectroscopy, while inductively coupled plasma optical emission spectroscopy was used to quantify the rate of material leaching. The extent of material degradation was observed to be strongly correlated to the concentration of metallic Li in the molten LiCl-Li2O system.

High-temperature corrosion characteristics of yttria-stabilized zirconia material in molten salts of LiCl-Li2O and LiCl-Li2O-Li

ABSTRACT The isothermal and cyclic corrosion behavior of yttria (Y2O3)-stabilized zirconia (ZrO2) in a LiCl-Li2O molten salt were investigated at 650 °C in an argon atmosphere. During isothermal and cyclic corrosion tests in the molten salt of LiCl-Li2O for 168 h and 7 thermal cycles, the corrosion rate was very low, whereas under the molten salt of LiCl-Li2O-Li for 168 h, the corrosion rate was almost 10 times higher than that in the molten salt of LiCl-Li2O. No corrosion product was detected until 168 h for the isothermal corrosion test, however, after 7 thermal cycles, a very-low-intensity Li2ZrO3 peak was detected at the beginning stage of the chemical reaction between ZrO2 and Li2O. Additionally, in the molten salt of LiCl-Li2O-Li for 168 h, a large amount of Li2ZrO3 was formed, with evidence of marked cracks, pores, and spallations on the corroded surface. The introduction of Y2O3-stabilized ZrO2 was beneficial in increasing the hot corrosion resistance of the structural materials used to handle molten salts containing Li2O at elevated temperature without forming a lithium at the cathode during the electrolytic reduction process.

Advanced Characterization of Molten Salts

Standard methods of analysis are frequently incapable of characterizing molten salts due to the high temperatures and chemically reactive nature of these systems. As a result, fundamental properties of molten salt solutions, and material interactions with them, remain unknown despite their industrial significance. The molten LiCl, LiCl-Li2O, and LiCl-Li2O-Li systems are of specific interest to the electrolytic reduction process used to treat oxide nuclear fuels prior to pyrochemical reprocessing. In this study, we employ two advanced techniques, impedance sensitive electrochemistry and synchrotron based high energy X-ray diffraction (HEXRD), to characterize molten LiCl and to demonstrate the benefits of these methods of analysis. Impedance dependent electroanalytical chemistry was used to study the electrochemical behavior of molten LiCl-Li2O, and to investigate the processes associated with Li+ reduction from the melt. During this study, a unique behavior of the AC signal was observed to coincide with Li+ reduction. This effect has been investigated in terms of the electrochemical reversibility of the Li|Li+ redox couple, and has yielded information that could not be obtained using standard DC voltammetry techniques. The advantages of employing impendence sensitive methods of electroanalysis for the study of molten salts over typical DC voltammetry methods will be discussed. The nature of elemental Li suspended in molten LiCl has been the subject of research for more than 50 years without yielding conclusive results. The study of molten solutions containing Li is exceptionally challenging due to the reactivity of molten Li, and as a result requires advanced methods of analysis. In attempt to circumvent these challenges, a system has been developed for containing molten salts and liquid metals during synchrotron characterization at the Advanced Photon Source (APS). By employing the APS, the system is capable of in situ characterization of molten LiCl while electrochemically forming Li in the melt. Standard methods of synchrotron analysis are incapable of characterizing molten LiCl due to the low energy state of the electrons of interest (55 eV in the case of Li). HEXRD offers a distinct advantage in this regard, as it is capable of yielding the structure factor and the corresponding pair distribution function of low energy electron states using high energy X-rays. In the current experiment, 100 KeV X-rays were propagated through an atmospheric chamber, the melt containing crucible, and a molten LiCl electrochemical cell, in order to characterize the molten solution during electrochemical analysis. This apparatus enables the study of the physical chemistry of the melt, as well as in situ characterization of material interactions with the high temperature fluid. The ability of this technique to characterize molten LiCl-Li2O-Li at 650°C will be presented along with an analysis of the chemistry of the melt itself.

Nanocatalytic conversion of CO2 into nanodiamonds

Carbon dioxide was converted into Li2CO3 nano-single crystals with particle sizes below 30 nm using a reactive LiCl-Li2O molten salt method. The Li2CO3 nanocrystals were then in-situ encapsulated into carbon layers. Upon heating of the core-shell nanostructured carbon material, the nucleation of diamond crystallites within the nanostructure of carbon encapsulated Li2CO3 was demonstrated by high resolution transmission and scanning electron microscopy, X-ray diffraction and Raman spectrometry. The diamond crystallites could grow to micrometer-sized octahedral crystals by a further heating. These findings are of significance since they give insight into low-pressure nanocatalytic diamond nucleation and may also provide economic motivations towards the reduction of CO2 emissions.

Hot corrosion behavior of magnesia-stabilized ceramic material in a lithium molten salt

The isothermal and cyclic corrosion behaviors of magnesia-stabilized zirconia in a LiCl-Li2O molten salt were investigated at 650 °C in an argon atmosphere. The weights of as-received and corroded specimens were measured and the microstructures, morphologies, and chemical compositions were analyzed by scanning electron microscopy, X-ray energy dispersive spectroscopy, and X-ray diffraction. For processes where Li is formed at the cathode during electrolysis, the corrosion rate was about five times higher than those of isothermal and thermal cycling processes. During isothermal tests, the corrosion product Li2ZrO3 was formed after 216 h. During thermal cycling, Li2ZrO3 was not detected until after the completion of 14 cycles. There was no evidence of cracks, pores, or spallation on the corroded surfaces, except when Li was formed. We demonstrate that magnesia-stabilized zirconia is beneficial for increasing the hot corrosion resistance of structural materials subjected to high temperature molten salts containing Li2O.

Review—Metallic Lithium and the Reduction of Actinide Oxides

Extensive research and process development has been conducted on the electrolytic reduction of actinide oxides in molten LiCl-Li2O. It is now accepted that the reduction of these metal oxides occurs via two separate reduction mechanisms: direct electro-chemical reduction and mediated chemical reduction by metallic lithium. The deposition of metallic lithium at the cathode (mediated chemical reduction mechanism) during the process is known to be essential in order to achieve high process throughputs and reduction yields, and yet a knowledge gap exists regarding the nature of metallic lithium in this system. This review summarizes the formation of lithium during the process and its dispersion into the molten salt electrolyte. Previously reported aspects of the physical chemistry of the LiCl-Li2O-Li system are presented with a specific focus on the dispersion of Li in the solution. Finally, issues regarding the effect of the presence of lithium on the electrolytic reduction process are discussed. Evidence shows that electrochemically generated metallic lithium is likely a significant source of experimental uncertainty, low current efficiency and Li2O consumption in the oxide reduction process.

Effects of alloying elements of nickel-based alloys on the hot-corrosion behavior in an electrolytic reduction process

In this study, the isothermal and cyclic corrosion behaviors of custom-made nickel-based alloys (N105, N107, and N108) with various elemental compositions in a LiCl-Li2O molten salt were investigated at 650 °C in argon atmosphere. The isothermal and cyclic corrosion behaviors were studied by measuring the mass changes, surface morphologies, cross-sectional morphologies with elemental distribution, and the compositional changes of the subscale and substrate as well as the spalled oxide scale. Under the isothermal and cyclic corrosion test, the corrosion rate was in the order of N108 < N105 < N107, and N107 < N105 < N108, respectively, which depended on the morphology and structure of the crystals formed on the surface of the outer scale in contact with the molten salt, and was also related to the region of Al and Cr depletion. After the isothermal corrosion test, the hot corrosion products formed in N105 were Cr2O3, NiCr2O4, and Li2Ni8O10 and those in N107 were Cr2O3 and NiCr2O4, while Cr2O3, NiCr2O4, and LiTiO2 were identified as the hot corrosion products of N108. However, the hot corrosion products of N105, N107, and N108 were Cr2O3, NiCr2O4, and LiFeO2 after the cyclic corrosion test.

Electrochemical properties of noble metal anodes for electrolytic reduction of uranium oxide

Noble metals (Rh, Pd, Ir, and Au) were investigated as O2-evolving anodes for electrolytic reduction of UO2 in LiCl–Li2O molten salt to replace Pt anodes, which are gradually consumed owing to Li2PtO3 layer formation. Anodic behaviors of these metals were examined by cyclic voltammetry. Au only showed O2 evolution in a moderate potential range without side reactions, suggesting better electrochemical stability relative to Pt anodes. With Au anodes, UO2 was electrochemically reduced to metallic U. No oxide layer was observed on the surface after the reduction. However, local dissolution remains a potential issue from a stability viewpoint.

Development of Carbon Anodes for Use in Electrolytic Reduction of Uranium(IV) Oxide

Recycling of used nuclear fuel using pyrochemical methods involves a series of molten salt processes performed at elevated temperatures and results in the separation of actinides from fission products. The first step in the pyrochemical treatment of light water reactor (LWR) oxide fuel is electrolytic reduction. This electrochemical operation converts used fuel oxides to metal before electrochemical separations can be performed by electrorefining. Conversion of oxides to metal is commonly performed in molten LiCl-Li2O at 650℃. Electrolytic reduction of UO2, the primary component of used nuclear fuel, has been successfully demonstrated at the kilogram-scale using platinum anodes1-2, but progression toward engineering scale presents different anode material requirements. Carbon, specifically graphite, is commonly used for ore refining in commercial molten salt processes because it is cost effective, readily available, and easily fabricated to meet scale-up requirements. Use of carbon anodes in LiCl-Li2O results in different salt chemistries, with respect to using platinum or other oxygen evolving electrode materials. In particular, the evolved gaseous byproduct of carbon anodes is CO and CO2, which can react with Li2O to form Li2CO3. If the carbonate concentration is allowed to increase, the desired reaction, namely the transport of oxide from the cathode to the anode via Li2O, is replaced by the undesirable transport of carbon from the anode to the cathode via Li2CO3. This report will discuss use of carbon as an anode material for the electrolytic reduction of UO2 to uranium metal in molten LiCl-Li2O at 650℃. A discussion of the salt chemistry, minimizing Li2CO3 formation, method developments, cell operation, and reduction efficiency will be presented.

Corrosion of Stainless Steel 316L in Molten LiCl-Li2o-Li

Material interactions with molten solutions of salt and metal are of interest to a wide verity of commercial electrochemical processes. Specifically, molten solutions of LiCl-Li2O containing dissolved elemental Li are applicable to the electrolytic reduction process used to incorporate used oxide nuclear fuels into pyroprocessing based fuel cycles [1, 2]. Additionally, electrolytic reduction of TiO2 [3, 4], SiO2 [5], Ta2O5 [6] and Nb2O5 [7] has been achieved in molten LiCl-Li2O electrolytes. While these processes have been shown to be highly effective, the co-reduction of Li at potentials more anodic than the Li+|Li standard potential is also well documented [8]. As a result of the formation of Li during electrochemical processing in molten LiCl-Li2O, and the known miscibility of Li in molten LiCl, a complex ternary molten solution of LiCl-Li2O-Li forms as these processes are conducted. Material interactions with molten LiCl-Li2O-Li are complicated by the simultaneous presence of dissolved oxidizing (Li2O) and reducing (Li) species in the melt. Coupons of stainless steel 316L (SS316L) were exposed to molten LiCl-Li2O-Li for periods of 20, 50 and 100 hours. Post exposure surface analysis was conducted using Raman and X-ray photoelectron (XPS) spectroscopies, as well as by X-ray diffraction and electron microscopy. Additionally, the leaching of alloying elements that occurred during the exposures was quantified using ICP-OES. Additional studies evaluated the effect of varying crucible materials and the presence of trace quantities of H2O on the corrosion behavior of SS316L in molten LiCl-Li2O-Li. Cr and Mo were observed to be preferentially leached from SS316L exposed to molten LiCl-Li2O containing high concentrations of Li. Alternatively; no material leaching was detected from SS316L exposed to molten LiCl-1wt%Li2O containing 0, 0.1 or 0.3wt%Li for 50 or 100 hours. The surface of SS316L was observed to be hardened as a result of the exposure; however the cause of the hardening is currently unknown. No alteration to the austenite crystal lattice perimeter, or trend in ferrite phase transformation was observed by XRD analysis of the samples. EDS analysis demonstrated a preferential enrichment of Fe and Ni, accompanied by a depletion in Cr and Mo, on the surface of SS316L following exposure. XPS analysis confirmed the presence of LiCrO2 on SS316L exposed to melts containing low concentrations of Li, however, the surfaces of samples exposed to melts containing excess quantities of Li were observed to be metallic. A mechanistic understanding of material interactions with molten LiCl-Li2O-Li will be discussed based on these and other observations obtained from this study. Acknowledgements: This work was performed under the auspices of the Department of Energy (DOE) under contracts DE-NE0008262, and DE-NE0008236 as well as the US Nuclear Regulatory Commission (USNRC) under contracts NRCHQ-11-G-38-0039 and NRC-38-10-949. A.M. acknowledges the Fellowship Award from the USNRC. Dr. Kenny Osborne serves as the program manager for the DOE award and Ms. Nancy Hebron-Isreal serves as the grants program officer for the NRC awards.

On the Formation of Clusters of Li8 in Molten Solutions of LiCl-Li

Numerous base metals can be obtained via electrolytic reduction of their respective oxides in molten LiCl-Li2O electrolytes. In these processes, often times, Li+ is co-reduced to metallic Li. The formation of Li0 is undesirable and leads to a loss of current efficiency and therefore is of significant interest. In spite of its importance, molten mixtures of LiCl and Li are still poorly understood and previous research has demonstrated a variety of phenomena that are seemingly unexplainable. We have custom built a unique setup to conduct in-situ Raman spectroscopy of such molten mixtures. In this work, we show that a Raman active phase forms in the fused salt when Li is electrochemically reduced from Li2O in LiCl. The spectrum was comprised of three fundamental features at 285.5, 302.8, and 318.2 cm-1, with overtones of decreasing intensity at approximately integer multiples of these Raman shifts as Shown in Figure 1. This spectrum was found to be consistent with previously-reported characteristic spectrum of the lithium cluster Li8. The presence of a colloidal suspension of lithium clusters in the molten salt is hypothesized to be the reason behind anomalous physical behavior of LiCl-Li solutions. Acknowledgements: This work was performed under the auspices of the Department of Energy (DOE) through grant DE-NE0008262, and the US Nuclear Regulatory Commission (NRC) under contracts NRCHQ-11-G-38-0039, NRC-38-10-949 and NRC-HQ-13-G-38-0027. Mr. Kenny Osborne serves as the program manager for the DOE awards and Ms. Nancy Hebron-Isreal serves as the grants program officer for the NRC awards. Figure 1

Electrochemical Measurement of Li2O in Molten LiCl Salt

Methods to prepare and condition a reference electrode for precise potential measurements in molten LiCl-Li2O are discussed in this paper. The Ni/NiO redox couple is commonly encased in MgO tubes containing porous, low-density MgO plugs that support ionic transport. Open circuit potential measurement between a source of fixed lithium activity and these reference electrodes is a promising approach to achieving real time Li2O concentration measurement in the salt. But the reference potentials must be extremely stable and reproducible in order for concentration correlations to be accurate for any given reference electrode. It was found in our study that pre-soaking of the reference electrodes is needed to eliminate transient effects. And equalizing the length of the porous MgO plugs is also critical to minimizing variability between different reference electrodes.

Measurement of Solubility of Metallic Lithium Dissolved in Molten LiCl–Li2O

The solubility of lithium metal in molten LiCl–Li2O mixtures has been measured at various concentrations of Li2O ranging from 0 wt.% to 2.7 wt.% at a temperature of approximately 670–680°C. After contacting molten lithium with molten LiCl–Li2O for several hours to achieve equilibrium saturation, samples were taken by freezing the salt onto a room-temperature steel rod and dissolving in water for analysis. Both volume of hydrogen gas generated and volume of titrated HCl were measured to investigate two different approaches to calculating the lithium concentration. There appeared to be no effect of Li2O concentration on the Li solubility in the salt. But the results vary between different methods of deducing the amount of dissolved Li. The H2 collection method is recommended, but care must be taken to ensure all of the H2 has been included.

Chemical Stability of Conductive Ceramic Anodes in LiCl–Li 2 O Molten Salt for Electrolytic Reduction in Pyroprocessing

Conductive ceramics are being developed to replace current Pt anodes in the electrolytic reduction of spent oxide fuels in pyroprocessing. While several conductive ceramics have shown promising electrochemical properties in small-scale experiments, their long-term stabilities have not yet been investigated. In this study, the chemical stability of conductive La 0.33 Sr 0.67 MnO₃ in LiCl-Li₂O molten salt at 650℃ was investigated to examine its feasibility as an anode material. Dissolution of Sr at the anode surface led to structural collapse, thereby indicating that the life time of the La 0.33 Sr 0.67 MnO₃ anode is limited. The dissolution rate of Sr is likely to be influenced by the local environment around Sr in the perovskite framework.

TiN Anode for Electrolytic Reduction of UO2 in Pyroprocessing

Developing novel anode materials to replace the Pt anode currently used in electrolytic reduction is an important issue on pyroprocessing. In this study, the electrochemical behavior of TiN was investigated as the conductive ceramic anode which evolves O2 gas during the reaction. The feasibility and stability of the TiN anode was examined during the electrolytic reduction of UO2. The TiN anode could electrochemically convert UO2 to metallic U in a LiCl–Li2O molten salt electrolyte. No oxidation of TiN was observed during the reaction; however, the formation of voids in the bulk section appeared to limit the lifetime of the TiN anode.

Graphite Anodes for Electroreduction of Uranium Oxide

Recycling of used nuclear oxide fuel is under investigation using pyrochemical methods, which involve a series of molten salt processes performed at elevated temperatures and result in the separation of actinides from fission products. During this process, used fuel oxides are electrochemically reduced to metals prior to electrochemical separation of the actinides from fission products in a subsequent process step. This electroreduction is typically performed at 650°C in LiCl-Li2O. Reduction of UO2, the primary component of used nuclear fuel, has been successfully demonstrated at the kilogram-scale using platinum anodes1-2, but progression toward larger scales presents different requirements for anode materials. Carbon (graphite) is commonly used in commercial molten salt processes within the ore refining industry. It is cost effective, readily available, and a logical choice for scale-up needs. However, different chemistries in molten LiCl-Li2O result from this change of anode material in both the salt and off-gas streams. Instead of forming oxygen at a platinum anode, the byproduct gas becomes CO and CO2. Although slightly soluble, these species can react with Li2O to form Li2CO3 as a result of residence time within the salt. The carbonate concentration can build up in the salt during the reduction process, which is problematic because of potential reaction between actinides and carbonate ions to form undesired byproducts in the reduced product. This report will discuss use of carbon as an anode material for the electroreduction of UO2 to uranium metal in molten LiCl-Li2O at 650°C. We will present design challenges and a discussion of salt chemistry with respect to carbonate formation in the system, along with results on reduction efficiency.