We proposed new separation and recovery process for RE metals from Nd magnet scraps using molten salt electrolysis and an alloy diaphragm (1). This new process was first applied to chloride melts, and the separation of Dy and Nd were investigated using Ni and Cu cathodic electrodes in molten LiCl-KCl-DyCl₃-NdCl₃ systems (2, 3). The highest mass ratio of Dy/Nd in Dy-Nd-Ni alloy sample was found to be 72 by ICP-AES. Moreover, in the present study we focused on the electrochemical formation of RE-Zn (RE=Dy, Nd) alloys with solid and liquid Zn electrodes in a molten LiCl-KCl system at 653-723 K (4). The alloy samples were prepared by potentiostatic electrolysis for 1 h using solid Zn plates in a molten LiCl-KCl added DyCl₃(0.50 mol%) and NdCl₃(0.50 mol%) at 653 K. The mass ratio of Nd/Dy in the alloy sample was the highest value 3.3 at 1.00 V(vs. Li⁺/ Li). In this study, the electrochemical formation of RE-Zn (RE=Dy, Nd) alloys was investigated with liquid Zn electrodes in a molten LiCl-KCl system at 723 K.

Experimental

All experiments were performed in LiCl-KCl eutectic melts under dry argon atmosphere at 723 K. DyCl₃ or NdCl₃ was added directly to these melts. The working electrodes were liquid Zn electrodes for the investigation of electrochemical behavior. The reference electrode was a Ag⁺/Ag electrode. All the potentials given hereafter were referred to Li⁺/ Li electrode potential on a Mo wire. The counter electrode was a glassy carbon rod (50 mm × f 5 mm). The alloy samples were prepared by potentiostatic electrolysis. After the electrolysis, the samples were analyzed by XRD, SEM and ICP-AES.

Results and discussion

Before the investigation of RE-Zn (RE=Dy, Nd) alloys formation, the electrochemical behavior of Li (I) was investigated using a liquid Zn electrode as a working electrode in a molten LiCl-KCl system at 723 K. The scanning rate was set at 0.05 Vs^-1. During the scan in the negative direction from 1.60 V, a small cathodic current was observed. Since Zn can form alloy with Li, the cathodic current corresponds to the formation Li-Zn alloy. On the other hand, taking into account the possibilities of the formation of various Dy,Nd-Zn alloys, the electrochemical behaviors of Dy (III) and Nd(III) were investigated using liquid Zn electrodes as working electrodes in molten LiCl-KCl systems at 723 K. A cathodic current was observed from 1.60 V using a liquid Zn electrode after the addition of NdCl₃(0.50 mol%). This cathodic current was larger than that cathodic current obtained before the addition of NdCl₃. Therefore, this cathodic current might corresponds to not only the formation Li-Zn alloy but also the reduction from Nd(III) to Nd(II). Moreover, a cathodic current peak was observed at 1.15 V. This cathodic current peak correspond to the formation of Nd-Zn alloys. When the potential was shifted to 0.80 V, a large cathodic current was observed. This cathodic current suggested the formation of Li-Zn and Nd-Zn alloys. When the potential scan direction was reversed at 0.50 V, two anodic peaks were observed. These anodic peaks suggested the Li and Nd dissolution from the different Li-Zn and Nd-Zn alloys. During the scan in the negative direction using a liquid Zn electrode after the addition of DyCl₃(0.50 mol%), a cathodic current was observed from 1.15 V. Since Zn can form alloy with Dy, the cathodic current corresponds to the formation of Dy-Zn alloys. When the potential was shifted to 0.80 V, a large cathodic current was observed. This cathodic current also suggested the formation of Li-Zn and Dy-Zn alloys. When the potential scan direction was reversed at 0.50 V, two anodic peaks were also observed. These anodic peaks also suggested the Li and Dy dissolution from the different Li-Zn and Dy-Zn alloys. Based on the above results, an alloy sample was prepared by potentiostatic electrolysis at 0.60 V, 0.80 V and 1.0 V for 1 h in a molten LiCl-KCl-DyCl₃-NdCl₃ system. From the EDX analysis, the alloy phases were identified as DyZn₁₂ or NdZn₁₁.

References

1. H. Konishi, H.Ono, T. Nohira and T. Oishi, MOLTEN SALTS, 54, 21 (2011).

2. H. Konishi, H.Ono, T. Nohira and T. Oishi, ECS Transactions, 50, 463 (2012).

3. H. Konishi, H.Ono, T. Nohira and T. Oishi, ECS Transactions, 53, 37 (2013).

4. H. Konishi, H.Ono, T. Nohira and T. Oishi, ECS Transactions, 61, 19 (2014).