Studies on nuclear technology have been ongoing since nuclear power became uniquely important to meet climate change goals while phasing out fossil fuels. Research on the fluoride salt cooled high temperature reactor (FHR), which is funded by the United States Department of Energy (DOE), has developed smoothly with the ultimate goal of a 2030 deployment. One challenge presented by FHR is that the primary coolant salt can acquire contamination from fuel failure and moisture leaking into the system. If contamination happens, it will result in a low concentration of fission products, fuel, transuranic materials and oxide impurities in the coolant. These impurities will then affect the properties of the molten salt in the long term and need to be removed without introducing new impurities. Most of the research conducted recently has focused on impurity separation in chloride molten salts. More research urgently needs to be conducted to study the impurity separation method for the fluoride molten salts. In this study, the La₂O₃-LiF-NaF-KF (La₂O₃-FLiNaK) system is used to demonstrate impurity separation in molten fluoride salt. Since lanthanum oxide needs to be dissolved in the fluoride molten salt and studies in this field are still not complete, the solubility of lanthanum oxide in FLiNaK have been measured at different temperatures to obtain the temperature-dependent solubility and understand the corresponding dissolution mechanisms first. In the solubility related experiments, Inductively Coupled Plasma Mass Spectrometry (ICP-MS) is utilized to analyze the concentration of lanthanum ions in the molten FLiNaK salt, while X-ray powder diffraction (XRD) was applied to determine the phase patterns of molten salt. Second, electrochemical experiments with tungsten and graphite as working electrodes were conducted individually to demonstrate the separation of the dissolved oxide from the salt. When the tungsten working electrode was applied, the lanthanum ions were reduced to lanthanum metal at the tungsten cathode, while the fluorine ions reacted with the tungsten anode to form tungsten fluoride. In the experiments, the production of tungsten fluoride could lead to increasing current in the cell, even overload. Moreover, theoretically, tungsten fluoride WF4 is soluble in the fluoride salt thus introducing new impurities. All these issues make tungsten not the best choice when applied to the separation of oxygen ions. Therefore, another common working electrode graphite is used. It not only has all the advantages of tungsten, but also has good performance on separation of oxygen ions. When the graphite electrode was applied, the lanthanum ions were separated in the form of lanthanum carbide (LaC₂), while the oxygen ions can be removed in the form of carbon dioxide (CO₂) or carbon monoxide (CO). In addition, only graphite was consumed during the whole separation process, which is why the graphite anode electrode is called the “sacrificial electrode”. Third, First Principle Molecular Dynamics (FPMD) simulations with Vienne Ab initio Simulation Package (VASP) was conducted to study the properties of the fluoride molten salt. In this study, the structure information and enthalpy of formation were obtained. Generally, the simulation process can be divided into four steps: (1) the simulation systems are prepared by packing ions randomly via Packmol package in the simulation cell; (2) an equilibrium calculation is performed to pre-equilibrate the systems; (3) FPMD simulations in an NVT ensemble are implemented in VASP; (4) based on the FPMD simulations results, the first peak radius and the first-shell coordination number were evaluated with partial radial distribution function (PRDF) analysis to determine the statistics of molten salt structure information, while the transport properties, e.g., the self-diffusion coefficient was calculated according to the function of mean square displacement (MSD) of time generated by the Einstein-Smoluchowshi equation. The viscosity and ionic conductivity were obtained by combining the self-distribution coefficient with the Einstein-Stokes formula and Nernst-Einstein equation. / Master of Science / With the fast development of modern society and economy, more and more energy is urgently needed to meet the growth of industry. Since the traditional energy, such as nature gas, coal, has limited storage and not sustainable, nuclear energy has attracted much attention in the past few decades. Although lots of study has been conducted by thousands of researchers which has attributed to application of nuclear power, there are still some concerns in this field, among which, impurities removal is the most difficult part. Fluoride salt cooled high temperature reactor (FHR) is one of the most promising Gen IV reactor types. As the name indicates, molten salt is the coolant to serve as the heat exchanger intermedium. In addition, it’s inevitable that fission products, i.e. lanthanum, moisture, would leak into the coolant pipe, thus affect the molten salt properties, even degrade reactor performance, therefore, those impurities must be removed without introducing new impurities. In this study, the La₂O₃-LiF-NaF-KF (La₂O₃-FLiNaK) system is used to demonstrate impurity separation into molten fluoride salt. First, solubility of lanthanum oxide in FLiNaK has been measured at different temperatures to understand its dissolution mechanisms. Then, electrochemical experiments with tungsten and graphite as working electrodes were conducted individually to demonstrate the separation of the dissolved oxide from the salt. It has been concluded that tungsten performed well to separate La3+, while failed in the separation of O2-. However, graphite working electrode has succeeded in the removal of La³⁺ and O²⁻. Finally, molecular dynamic simulation with first principle was also conducted to further understand the local structure and heat of formation in the molten FLiNaK and La₂O₃-FLiNaK salt.