High-temperature molten salts not only demonstrate exceptional thermal and chemical stability but also offer significant advantages in catalyzing chemical reactions. Consequently, they have garnered attention as a promising medium for next-generation nuclear reactors and a wide range of electrochemical processes. Nevertheless, the challenging experimental conditions in molten salts make applying conventional analytical methods to understand reaction mechanisms a formidable task. This underscores the imperative need for more intuitive approaches to investigate molten salt chemistry. One of the simplest yet potent methods involves real-time visual monitoring of the reaction system as chemical reactions progress. In light of this, we have developed an experimental system enabling real-time visual monitoring of the internal dynamics of molten salt media. This system can capture high-resolution videos and images within molten salts, surpassing existing methodologies. We have applied this system in various electrochemical experiments using the molten LiCl-KCl eutectic salt medium. Among them, this study primarily focuses on two challenging experimental scenarios that became comprehensible through our proposed system’s application: (1) the transpassivation of Zr metal and the agglomeration of potassium hexachlorozirconate (K2ZrCl6) solid salt, and (2) the solvation of electrons during the oxidation of Li metal within the molten LiCl-KCl eutectic salt.

The separation efficiency of nuclides in molten salt systems was investigated, with a focus on the influence of apparatus configuration and experimental conditions. A prior study revealed that achieving effective Sr separation from simulated oxide fuel required up to 96 hours, reaching a separation efficiency of approximately 90% using a static dissolution reaction in a porous alumina basket. In this study, we explored the impact of agitation on improving Sr separation efficiency and dissolution rates. The simulated oxide fuel composition consisted of 2wt% Sr, 3wt% Ba, 2wt% Ce, 3wt% Nd, 3wt% Zr, 2wt% Mo, and 89wt% U. To quantify the Sr concentration in the salt, we utilized ICP analysis after salt sampling via a dip-stick technique. Furthermore, we conducted ICPOES analysis over a 55-hour duration to assess the separated nuclides. Complementing these analyses, SEM and XRD investigations were performed to validate the crystal structure and morphology of the oxide products.

Zirconium(Zr) alloys are commonly used in the nuclear industry for applications such as fuel cladding and pressure tubes. To minimize the levels and volumes of radioactive waste, molten salts have been employed for decontaminating Zr alloys. Recently, a two-step Zr metal recovery process, combining electrolysis and thermal decomposition, has been proposed. In the electrolysis process, potentiostatic electrorefining is utilized to control the chemical form of electrodeposits(ZrCl). Although Zr metals are expected to dissolve into molten salts, reductive alloy elements can also be co-dissolved and deposited on the cathode. Therefore, a better understanding of the anodic side’s response during potentiostatic electrorefining is necessary to ensure the purity of recovered Zr and long-term process operation. As the first step, potentiodynamic polarization curves were obtained using Zr, Nb, and Zr-Nb alloy to investigate the anodic dissolution behavior in the molten salts. Nb, which has a redox potential close to Zr, and Zr exhibit active or passivation dissolution mechanisms depending on the potential range. It was confirmed that Zr-Nb alloy also has a passivation region between -0.223 to -0.092 V influenced by the major elements Zr and Nb. Secondly, active dissolution of Zr-Nb was performed in the range of -0.9 to -0.6 V. The dissolution mechanism can be explained by percolation theory, which is consistent with the observed microstructure of the alloy. Thirdly, passivation dissolution of Zr, Nb, and Zr-Nb alloy was investigated to identify the pure passivation products and additional products in the Zr-Nb alloy case. K2ZrCl6 and K3NbCl6 were identified as the pure passivation products of the major elements. In the Zr-Nb alloy case, additional products, such as Nb and NbZr, produced by the redox reaction of nanoparticles in the high viscous salt layer near the anode, were also confirmed. The anodic dissolution mechanism of Zr-Nb alloy can be summarized as follows. During active dissolution, only Zr metal dissolves into molten salts by percolation. Above the solubility near the anode, passivation products begin to form. The anode potential increases due to the disturbance of passivation products on ion flow, leading to co-dissolution of Nb. When the concentration of Nb ion exceeds the solubility, a passivation product of Nb also forms. In this scenario, a high viscous salt layer is formed, which traps nanoparticles of Zr metal, resulting in redox behavior between Zr metal and Nb ion. Some nanoparticles of Zr and Nb metal are also present in the form of NbZr.

Molten salts have gained significant attention as a potential medium for heat transfer or energy storage and as liquid nuclear fuel, owing to their superior thermal properties. Various fluoride- and chloride-based salts are being explored as potential liquid fuels for several types of molten salt reactors (MSRs). Among these, chloride-based salts have recently received attention in MSR development due to their high solubility in actinides, which has the potential to increase fuel burnup and reduce nuclear water production. Accurate knowledge of the thermal physical properties of molten salts, such as density, viscosity, thermal conductivity, and heat capacity, is critical for the design, licensing, and operation of MSRs. Various experimental techniques have been used to determine the thermal properties of molten salts, and more recently, computational methods such as molecular dynamics simulations have also been utilized to predict these properties. However, information on the thermal physical properties of salts containing actinides is still limited and unreliable. In this study, we analyzed the available thermal physical property database of chloride salts to develop accurate models and simulations that can predict the behavior of molten salts under various operating conditions. Furthermore, we conducted experiments to improve our understanding of the behavior of molten salts. The results of this study are expected to contribute to the development of safer and more efficient MSRs.

Radioactive wastes, including used nuclear fuel and decommissioning wastes, have been treated using molten salts. Electrochemical sensors are one of the options for in-situ process monitoring using molten salts. However, in order to use electrochemical sensors in molten salt, the surface area must be known. This is because the surface area affects the current of the electrode. Previous studies have used a variety of methods to determine the electrode surface area in molten salts. One method of calculating the electrode surface area is to use the reduction current peak difference between electrodes with known length differences. The method is based on the reduction peak and has the benefit of providing long-term in-situ monitoring of surfaces immersed in molten salt. A number of assumptions have been made regarding this method, including that there is no mass transport by migration or convection; the reaction is reversible and limited by diffusion; the chemical activity of the deposit should be unity; and species should follow linear diffusion. For the purpose of overcoming these limitations, a variety of machine learning algorithms were applied to different voltammogram datasets in order to calculate the surface area. Voltammogram datasets were collected from multiarray electrodes, comprising a multiarray holder, two tungsten rods (1 mm diameter) working electrodes, a quasi-reference electrode, and a counter electrode. The multiarray electrode holder was connected to the auto vertical translator, which uses a servo motor, for changing the height of the rod in the molten salts. To make big and diverse data for training machine learning models, various concentrations of corrosion products (Cr, Fe) and fission products (Eu, Sm) in NaCl-MgCl2 eutectic salts were used as electrolyte; electrolyte temperatures were 500, 525, 550, 575, and 600°C. This study will demonstrate the potential of utilizing machine learning based electrochemical in situ monitoring in molten salt processing.

Measurement of oxide ion (O2-) concentration is a basic technology required in molten salt fields, from energy storage systems to electrolytic reduction of rare earth elements or spent nuclear fuels. In a molten salt reactor, O2- ions react with actinide elements to form their oxides or oxy-chlorides to induce actinide precipitation, and promote metal corrosion to cause a failure of structural material. For these reasons, removal of O2- ions and monitoring of the O2- concentration in molten salt reactors are essential. In this study, methods using chemical and electrochemical methods were investigated for measuring the concentration of O2- ions in a molten salts. The acid-base neutralization reaction was used as a chemical analysis method. And electrochemical methods using the O2- diffusion limit current and YSZ (yttria stabilized zirconia) indicator electrode were used for measuring the O2- concentration. Finally, a modified method using porous membrane electrode was applied to monitor the O2- concentration. The O2- concentration was measured up to about 2wt% of Li2O by the method using the O2- diffusion current, up to about 4wt% by the YSZ indicator electrode, and about 6wt% by the porous membrane electrode in LiCl molten salts.

Separation of high heat generating-radioactive isotopes from spent nuclear fuel is an important issue because it can reduce the final disposal area. As one of the technologies that can selectively separate only high heat generating-radioactive isotopes without dissolving spent fuel, the methods using molten salt have recently attracted attention. Although studies on chemical changes of Sr oxides in molten salts have been reported, they have limitation in that alternative oxide reagents rather than oxide fuel were used. In this study, the separation behaviors of Sr from simulated oxide fuel using various molten salts were investigated. A powder type containing 95.7wt% of U and 0.123wt% of Sr was used as the simulated oxide fuel. LiCl, LiCl-CaCl2, MgCl2, LiCl-KCl-MgCl2 and NaCl-MgCl2 were used as molten chloride salts. The separation of Sr from the simulated oxide fuel was conducted by loading it in porous alumina basket and immersing it in a salt. The concentration of Sr in the salt was measured by ICP analysis after sampling the salt outside the basket using dip-stick technique. The separation efficiencies of Sr from simulated oxide fuel using the salts were compared. Furthermore, the causes of their separation efficiency were systematically investigated.

Molten salt reactors and pyroprocessing are widely considered for various nuclear applications. The main challenges for monitoring these systems are high temperature and strong radiation. Two harsh environments make the monitoring system needs to measure nuclides at a long distance with sufficient resolution for discriminating many different elements simultaneously. Among available methodologies, laser-induced breakdown spectroscopy (LIBS) has been the most studied. The LIBS method can provide the required stand-off and desired multi-elemental measurable ability. However, the change of the level for molten salts induces uncertainty in measuring the concentration of the nuclides for LIBS analysis. The spectra could change by focusing points due to the different laser fluence and plasma shape. In this study, to prepare for such uncertainties, we evaluated a LIBS monitoring system with machine learning technology. While the machine learning technology cannot use academic knowledge of the atomic spectrum, this technique finds the new variable as a vector from any data including the noise, target spectrum, standard deviation, etc. Herein, the partial least squares (PLS) and artificial neural network (ANN) were studied because these methods represent linear and nonlinear machine learning methods respectively. The Sr (580–7200 ppm) and Mo (480–4700 ppm) as fission products were investigated for constructing the prediction model. For acquiring the data, the experiments were conducted at 550°C in LiCl-KCl using a glassy carbon crucible. The LIBS technique was used for accumulating spectra data. In these works, we successfully obtained a reasonable prediction model and compared each other. The high linearities of the prediction model were recorded. The R2 values are over 0.98. In addition, the root means square of the calibration and cross-validation were used for evaluating the prediction model quantitatively.

Molten salt immersion technique has been tested with several Sr oxides, SrZrO3, SrMoO4 and U2SrOy, and MgCl2 based molten salts for the Sr nuclide separation. Reaction time, temperature, and salt composition were varied to effectively separate Sr in chloride forms. ICP-OES, XRD, and SEM analysis were conducted for the conversion efficiency and structure and morphology analysis. It is confirmed that all experiments of SrZrO3 with MgCl2 at 800°C for reaction time 5, 10, 20 hours showed higher conversion efficiency than 99% and in LiCl-KCl-MgCl2 and NaCl-MgCl2 molten salts at 500°C or 600°C, conversion efficiency higher than 97% was obtained. SrMoO4 in MgCl2 immersion experiments for 10 hours showed higher conversion efficiency than 99% when the molar ratio of salt/oxide powder is 7. U2SrOy was also tested with MgCl2 molten salt at 800°C and higher efficiency than 99% and mainly MgUO4 were produced as a reaction product.

For diamond/metal composites it is better to use diamond particles coated with metal carbide because of improved wettability between the diamond particles and the matrix. In this study, the coating of diamond particles with a chromium carbide layer is investigated. On heating diamond and chromium powders at 800~900 oC in molten salts of LiCl, KCl, CaCl2, the diamond particles are coated with Cr7C3. The surfaces of the diamond powders are analyzed using X-ray diffraction and scanning electron microscopy. The average thickness of the Cr7C3 coating layers is calculated from the result of the particle size analysis. By using the molten salt method, the Cr7C3 coating layer is uniformly formed on the diamond particles at a relatively low temperature at which the graphitization of the diamond particles is avoided. Treatment temperatures are lower than those in the previously proposed methods. The coated layer is thickened with an increase in heating temperature up to 900 oC. The coating reaction of the diamond particles with chromium carbide is much more rapid in LiCl-KCl-CaCl2 molten salts than with the molten salts of KCl-CaCl2.

Ceramic powder, such as MgO, is added as a binder to prepare the green compacts of molten salts of an electrolyte for a thermal battery. Despite the addition of a binder, when the thickness of the electrolyte decreases to improve the battery performance, the problem with the unintentional short circuit between the anode and cathode still remains. To improve the current powder molding method, a new type of electrolyte separator with porous MgO preforms is prepared and characteristics of the thermal battery are evaluated. A Spherical PMMA polymer powder is added as a pore-forming agent in the MgO powder, and an organic binder is used to prepare slurry appropriate for tape casting. A porous MgO preform with 300 μm thickness is prepared through a binder burnout and sintering process. The particle size of the starting MgO powder has an effect, not on the porosity of the porous MgO preform, but on the battery characteristics. The porosity of the porous MgO preforms is controlled from 60 to 75% using a pore-forming agent. The batteries prepared using various porosities of preforms show a performance equal to or higher than that of the pellet-shaped battery prepared by the conventional powder molding method.