Used nuclear fuels (UNFs) generated from nuclear power plants have characterized to have very high radiotoxicity but contain energy-producing materials such as uranium and transuranic elements. A recovery of energy producing materials in the UNFs with a pyroprocess is one of the options to manage the UNFs in the nations with nuclear power plants. The pyroprocess is an electrometallurgical process to recover uranium and transuranic elements from the UNFs. In Korea, Korea Atomic Energy Research Institute (KAERI) has developed the pyroprocess based on the molten LiCl-KCl eutectic to reduce radiotoxicity of the UNFs and to minimize a volume of the final radioactive wastes . The major parts of KAERI pyropreocess are composed of electrolytic reduction in a molten LiCl to reduce oxide into metallic form, electrorefining and electrowinning in a molten LiCl-KCl eutectic to recover uranium and transurnic elements, salt purification and recycle of molten LiCl and LiCl-KCl eutectic, and waste treatment process to generate waste package for disposal.
A salt waste generated from the pyroprocess after recovery of uranium and transuranic elements is classified to a high-level waste because it contains a residual actinide. The residual actinide recovery process such as reductive extraction can be employed to reduce radiotoxicity of the salt wastes. As a candidate process, a sequential process of electrolytic reductive extraction and selective oxidative extraction in molten salt-liquid metal system was proposed to recover the residual actinide from the salt waste . The performance prediction of the residual actinide recovery process requires the thermodynamic and kinetic data of lanthanide and actinide dissolved in the salt waste. Among the dissolved species in the salt waste, samarium may have either oxidation state of divalent ion or trivalent ion according to electrochemical conditions. The oxidation state of samarium ion may have an effect on the electrolytic reduction into liquid metal by disproportionate reaction.
Some work has been performed to determine redox potentials, electron transfer stoichiometry, and diffusion coefficient of samarium in the molten LiCl-KCl eutectic using conventional electroanalytical techniques. However, conventional electroanalytical techniques can rarely unequivocally identify electroactive species. In addition, the electroanalytical techniques provide only limited and indirect information on structural changes accompanying redox events. On the other hand, spectroscopic techniques could characterize the dissolved species in the molten salt. A spectroelectrochemical technique that is a combination of electroanalytical techniques with spectroscopic measure ment technique could provide a useful information to gain a knowledge on mechanism of the electron transfer reaction and the ionic state at the interface between electrode and solution. This technique has employed to investigate the electrochemical behaviors of the europium in the molten salt [3, 4].
In the study, the spectroelectrochemical technique under high temperature environments has been developed based on a concept of an optically transparent electrode  as a working electrode and employed to investigate an electrochemical behavior of samarium ion in the molten LiCl- KCl eutectic. Special electrochemical cell with optically transparent working electrode was designed to conduct the electrochemical experiment. Cyclic voltammetry was used to characterize the electrochemical conditions of samarium ions in the molten LiCl-KCl eutectic and then chronoabsormetry at fixed potential and potential step chronoabsorptometry was used to characterize the redox reaction between trivalent and divalent samarium ions. Thermochemical data for samarium ion dissolved in the molten LiCl-KCl eutectic was estimated on the basis of the experimental results.
2.Materials and Methods
2.1.Chemicals and Materials
Anhydrous SmCl3 and LiCl-KCl eutectic mixture (molar ratio of lithium to potassium = 59/41) were products of Sigma Aldrich and used without further purification. The appropriate amount of LiCl-KCl eutectic mixture poured into the quartz cell and then melted by raising the temperature of 773 K. Then, SmCl3 was added and dissolved in the molten salt. Prior to the any measurement, the cell was thoroughly shaken to eliminate any bubbles present in the molten salt.
The spectroelectrochemical measurement system under high temperature environments was designed and was fabricated to investigate electrochemical characteristics of the dissolved species in the molten LiCl-KCl eutectic. A measurement cell was made of 1.2 cm rectangular quartz cell with a long-necked cylindrical quartz tube as shown in Fig. 1. It was placed at the center of the electric furnace with optically accessible windows that was attached at the bottom of glove box filled with high purity argon gas (99.99%) continuously purified to remove oxygen and humidity. The content of O2 and H2O in the inside atmosphere of the glove box was maintained less than 1 ppm, respectively. To conduct electroanalytical and spectroscopic measurements at the same time, the three quartz windows in a 90° angle were installed to the lower part of the apparatus. The quartz windows were cooled by circulating water in order to prevent the thermal damage.
The cell is equipped with three electrodes that are composed of an optically transparent working electrode, a reference electrode, and a counter electrode. The optical transparence of working electrode was achieved with use of a tungsten gauze of 100 mesh (99.9%, Alfa Aesar) which is commercially available. The upper end of the tungsten gauze was pressed with silver wire for electrical connection. The silver wire was guided with Quartz tube for avoiding contact with the other electrode. The fabricated working electrode has 81% transmittance to light. The counter electrode was a tungsten rod (99.95% with 2mm in diameter, Sigma Aldrich). The reference electrode was Ag (99.99% Ag wire with 1.0 mm in diameter)/AgCl (1.0 wt% in LiCl-KCl) system. A solution of 1.0 wt% of AgCl in the molten LiCl-KCl eutectic was encapsulated in a thin-end Pyrex glass tube. A silver wire was inserted into the Pyrex tube to be immersed in the solution. All reported potentials were referenced versus the Ag/AgCl (1.0 wt%) reference electrode.
The spectroscopy measurement system was composed of light source (DH-2000-BAL, Ocean Optics), optical fiber (GH4001, Mitsubishi Rayon), collimating lens (84-UV- 25, Ocean Optics) and UV-Vis spectrometer (multichannel spectrometer, Avantes) with Avasoft software (Avantes). The light beam travels out from a light source through an optical fiber that were connected to a collimating lens that makes a 3 mm φ straight beam. The light beam that travels through the cell is caught by another collimating lens at the opposite side of the furnace and is lead to the spectrometer by another optical fiber. Electroanalytical measurements were carried out with Solatron Potentiostat controlled with CorrWare software (Scribner Assoc.)
3.Results and Discussion
Cyclic voltammetry was conducted to investigate electrochemical characteristics of samarium ion in the molten LiCl-KCl eutectic. Fig. 2 shows cyclic voltammograms for scan rates ranging from 0.05 to 2.00 V/s with the potential window from -1.4 V to 0.2 V at the temperature of 723 K on the optically transparent electrode of tungsten gauze. The voltammograms have single well-defined peaks for cathodic reaction at about -0.92 V and for anodic reaction at about -0.67 V, respectively and exhibits a soluble-soluble system presumably related to Sm3+/Sm2+ redox system. The reversibility test was performed with a dependency of peak current versus square root of the potential scan rate and a slope of the log of the peak current versus log of the potential scan rate [6, 8]. The plots of the peak currents versus the square root of the potential scan rate, as shown in Fig. 3, show that the peaks of cathodic current and anodic current are linearly proportional to the square root of the scan rates, respectively. The slope of the log of the peak current versus the log of the potential scan rate is equal to approximately 0.5 by the regression analysis.
The reversibility of the redox reaction of Sm3+/Sm2+ allows us to determine the formal potential with the voltammograms. The formal potential, E0, of the redox reaction of Sm3+/Sm2+ was estimated from the data of the anodic and cathodic peak potentials obtained from the cyclic voltammograms using the following equations :(1)
where Epc is a peak potential of cathodic reaction, and Epa is a peak potential of anodic reaction. The peak potentials measured against the Ag/AgCl reference electrode were transformed into the Cl2 /Cl- reference electrode scale using correlation of EAg+/Ag = - 1.0659 - 0.0001805 × T (V vs. Cl2 / Cl-). Analysis of voltammograms exhibited that the formal potential of Sm3+/Sm2+ redox couple was -1.99 V vs. Cl2 / Cl-.
The diffusion coefficient of Sm3+ ions in molten LiCl- KCl eutectic was determined using the cathodic peak current, according to the Randles-Sevcik equation for reversible soluble-soluble system:(2)
where D is the diffusion coefficient[cm2∙s-1], S is the electrode surface area[cm2], C0 is the solute concentration [mol·cm-3], F is the Faraday constant [C·mol-1], R is the ideal gas constant [J·K mol-1], n is the number of exchanged electrons, v is the potential scan rate [V∙s-1] and T is the absolute temperature [K]. Analysis of voltammograms exhibited that the diffusion coefficient of Sm3+ was 2.53 × 10-6cm2∙s-1.
3.2.Chronoabsortometry at fixed potential
The absorption spectra of samarium ion in molten LiCl- KCl eutectic was measured under an applied potential of -1.5 V for cathodic reduction at 723 K during 120 seconds. Fig. 4 shows the absorption spectra of samarium in the motlen LiCl-KCl eutectic. Initially, absorption of LiCl-KCl- SmCl3 melt was appeared at 390 ~ 440 nm without the applied potential. After the potential was applied at -1.5 V, the reduction of Sm3+ to Sm2+ resulted in the color change from pale yellow to red-brown and a broad band at 520 ~ 560 nm appeared in the spectra. Peak findings of absorption spectra show that characteristic peaks are 408.08 nm and 545.62 nm that correspond to Sm3+ and Sm2+, respectively . Absorp tion intensities at 408.08 nm and 545.62 nm were analyzed after baseline correction of spectra. Fig. 5 shows the change of absorption intensities at 408.08 nm and 545.62 nm. When the cathodic reduction at -1.5 V is progressed, absorbance at 408.08 nm is gradually decreased and absorbance at 545.62 nm is gradually increased. After turn off the applied potential, broad band at 520~560 nm gradually diminished indicating that the product of reduction of Sm3+ is stable in the melt only under the cathodic potential. Absorption intensities at each wavelength are proportional to the square root of time.
3.3.Potential Step Chronoabsorptometry
The double potential step chronoabsorptometry was conducted in LiCl-KCl-SmCl3 at 723K. The potential step from -0.67 V to -0.92 V were selected from the voltammograms, which correspond to anodic and cathodic peak potentials, respectively. An initial potential was set to -0.67 V to ensure that the entire samarium concentration in the salt was in the fully oxidized. The potential was subsequently changed to -0.92 V during 50 seconds to reduce Sm3+ to Sm2+ at a diffusion controlled rate, which was followed by changing the potential back to -0.67 V to reoxidize the sample to Sm3+. Absorption spectra were recorded during both reduction and oxidation steps. Fig. 6 shows the increase in absorbance at 545.62 nm upon lowering the potential to -0.92 V, indicating a reduction of Sm3+ to Sm2+. The magnitude of absorbance reaches a maximum after 25 seconds, indicating complete electrolysis of Sm3+ within the optical path of the cell. After 50 seconds the applied potential was stepped-up from -0.92 V to -0.67 V. This resulted in the oxidation of the electro-generated Sm2+ back to parent Sm3+. It is confirmed by the decrease in the absorbance at 545.62 nm and subsequent leveling off after 25 seconds.
Absorbance is related to diffusion coefficient of the dissolved ion in the solution under the double potential step chronoabsorptometry. Diffusion-limited current in potential step method is expressed using Cottrel equation and Absorbance is related to concentration by Beer’s law. Combining two relations, the diffusion coefficient for Sm3+ was determined from the slope of the plot of absorbance vs t1/2. Initially the variation of absorbance is diffusion controlled (up to t = 25s), which can be used to calculate the diffusion coefficient in accordance with the following equation :(3)
where A is the absorbance, ϵr is the molar absorptivity of Sm2+, D is the diffusion coefficient of Sm3+, C is the concentration of Sm3+, and t is time. The molar absorptivity of Sm2+ in LiCl-KCl was calculated to be 5385 M-1·cm-1 using the maximum absorbance of a fully reduced Sm2+ solution and Beer’s law. Applying this equation to slope of the linear portion of the graph gives a diffusion coefficient of 2.15 × 10-6cm2∙s-1 for Sm3+, which is comparable to the value determined by cyclic voltammetry at the same temperature (2.53 × 10-6cm2∙s-1).
The spectroelectrochemical technique under high temperature has been developed with the optically transparent working electrode of tungsten gauze in order to investigate the electrochemical behavior in the electrolytic reduction process for actinide recovery process from the salt waste. It was first applied to samarium ion dissolved in the molten LiCl-KCl eutectic. In the study, the electrochemical characteristics of samarium ion dissolved in the molten LiCl-KCl eutectic were obtained with cyclic voltammetry, chronoabsorptometry at fixed potential, and potential step chronoabsorptometry. Based on the reversibility of the redox reaction of Sm3+/Sm2+ to be determined from the cyclic voltammograms, the formal potential and diffusion coefficient were calculated to be -1.99 V vs.Cl2/Cl- and 2.53 × 10-6cm2∙s-1, respectively. The characteristic peaks of absorption for samarium ions were determined to be 408.08 nm for Sm3+ and 545.62 nm for Sm2+ from chronoabsorptometry at the applied potential of -1.5 V vs. Ag/ AgCl (1wt%). Potential step chronoabsorptometry was conducted using the anodic and cathodic peak potentials form the voltammograms. Absorbance analysis at 545.63 nm exhibits that diffusion coefficient of Sm3+ is 2.15 × 10-6cm2∙s-1 , which is comparable to the value determined by cyclic voltammetry at the same temperature. Experimental results show that the spectroelectrochemical technique with the optically transparent working electrode is a useful tool to investigate the electrochemical behaviors of lanthanides dissolved in the molten LiCl-KCl eutectic.