The authors have been investigating a new recycling process for Nd and Dy from wasted Nd–Fe–B magnets utilizing selective permeation phenomenon through alloy diaphragms in molten salts. In this process, alloy diaphragms such as Ni–rare earth (RE) alloy film, functions as a bipolar electrode through which RE ions permeate from the anolyte to the catholyte in three steps: (a) reduction of RE ions to form alloys on the anolyte side of the diaphragm, (b) diffusion of RE atoms inside the diaphragm, and (c) oxidation of RE atoms to dissolve into the catholyte as RE ions on the catholyte side of the diaphragm [1]. Because the alloying and dealloying potentials for each RE element are different, a specific RE element can selectively permeate through the alloy diaphragm under an appropriate condition. Actually, the selective permeation of Dy or Nd has already been demonstrated [2, 3]. However, in actual recycling from end-of-life products, the wasted Nd–Fe–B magnets may contain a variety of impurities, such as base metals and other RE elements. Although most of impurities can be easily separated from RE elements due to differences in electrochemical properties, RE impurities such as La, Ce, Pr, Gd, and Tb are more difficult to separate from Nd and Dy. Thus, in this study, the permeation behavior of La, Ce, Pr, Gd, and Tb was investigated in the presence of Nd and Dy in LiCl–KCl eutectic melts.A schematic drawing of the experimental setup is described in Figure 1. The electrolytic cell was made of alumina and separated to the anole and cathode compartments by a Ni film of thickness 0.1 mm which is used as the starting material for the alloy diaphragm. The electrolyte was a molten LiCl–KCl eutectic dissolved at 723 K under Ar flow, and RECl3 (RE = La, Ce, Pr, Nd, Gd, Tb, and Dy) anhydrous were added to the anolyte. In the catholyte, DyCl3 was added in order to prevent the dendritic deposition of RE metal on the cathode. The potentials of each side of the diaphragm were controlled by two potentio-galvanostats (PGS-A and B). The permeation experiment was performed under two experimental conditions that were expected to enable the selective permeation of Nd and Dy, respectively, which were determined based on our previous studies [2, 3]. In the first experiment, the potential of the anolyte side of the diaphragm (WE-A) was controlled at 0.65 V vs. Li+/Li by PGS-A. At this potential, a part of added RE ions can be reduced and form RE–Ni alloys (mostly RENi2). The catholyte side potential of the diaphragm (WE-B) was not controlled initially. As the alloy formation proceeded, the potential of WE-B negatively shifted. When the potential of WE-B reached 0.80 V vs. Li+/Li, this value was fixed. At this potential, all the added RE elements can be dissolved from the alloy phase if they exist on the catholyte side surface. In the second experiment, the anolyte and catholyte side of the diaphragm were controlled at 0.50 V and 0.70 V vs. Li+/Li, respectively. At this condition, most of the added RE elements can form RENi2, and only a part of RE elements can be dissolved from the diaphragm on the catholyte side.The permeated amount of RE elements were evaluated by ICP-AES analysis. As a result, Gd and Tb showed a similar permeation tendency to Dy, whereas Ce and Pr behaved like Nd in both experimental conditions. La hardly permeated probably because it does not form RENi2 alloy phase. These tendencies are discussed based on the applied potential and the equilibrium potential for RENi2/RENi3 (RE = Ce, Pr, Nd, Gd, Tb, and Dy) and La7Ni16/LaNi3. Keywords Rare earth, Recycle, Alloy diaphragm, Separation, Molten salt electrolysis References [1] T. Oishi et al., Kagaku Kogaku Ronbunshu, 36, 299 (2010) [in Japanese].[2] T. Oishi et al., J. Electrochem. Soc., 167, 163505 (2020).[3] T. Oishi et al., J. Electrochem. Soc., 168, 103504 (2021). Acknowledgments This work was partly supported by the New Energy and Industrial Technology Development Organization (NEDO) of Japan. Figure 1
Read full abstract