Improved room-temperature-selectivity between Nd and Fe in Nd recovery from Nd-Fe-B magnet

The sustainable society requires the recycling of rare metals. Rare earth Nd is one of rare metals, accompanying huge consumption especially in Nd-Fe-B magnets. Although the wet process using acid is in practical use in the in-plant recycle of sludge, higher selectivity between Nd and Fe at room temperature is desired. We have proposed a pretreatment of corrosion before the dissolution into HCl and the oxalic acid precipitation. The corrosion produces $\gamma$-FeOOH and a Nd hydroxide, which have high selectivity for HCl solution at room temperature. Nd can be recovered as Mn$_{2}$O$_{3}$-type Nd$_{2}$O$_{3}$. The estimated recovery-ratio of Nd reaches to 97\%.


I. INTRODUCTION
Rare-earth is widely consumed in glass industry, catalysts, Nd magnets and so on 1 . Especially the demand of Nd-Fe-B magnets is rapidly growing, viewing from sustainable and/or low-carbon society, in motors of electric vehicle and wind turbine et al. Major productcountry of rare earths is now China, often controlling the export quota. Therefore, the other countries urgently work through the recycle of Nd from used magnets 2,3 , as one of provisions for stock.
The recycle method can be divided into two main classes: wet and dry processes. Associated with the development of ore dressing technologies, the wet process has already been put to practical use in the recycle of sludge of in-plant scrap. The scrap is dissolved in acid such as HCl, HNO 3 and H 2 SO 4 . After the filtration of insoluble materials mainly containing Fe, acid solution is reacted with oxalic or carbonic acid to form the precipitate containing Nd element 4 . The calcined precipitate becomes Nd oxide, which can be returned to the initial manufacture process of Nd-magnet. The oxidation of Nd-Fe-B magnet as the pretreatment improves the selectivity between Nd and Fe, however the recovery ratio of Nd is rather low 5,6 . Nearly 100% recovery is achieved when the acid solution is heated at 180 As for the dry process, Takeda  is a promising dry process for high purity and high extraction ratio of rare-earth oxide 11 . Although many attempts based on the dry process have been proposed so far, they are still in the fundamental stage.
Returning to the wet process, it is desired to increase the selectivity between Nd and Fe at room temperature. We have found that the corrosion process leads to the high selectivity at room temperature and nearly 100% recovery of Nd. Our method is compatible with the present recovery method in the in-plant, because the oxidation process is merely replaced with the corrosion one. In this report, we have investigated the Nd-recovery ratio in the recovery method using corrosion as the pretreatment. The ratio is compared with that obtained by the conventional method in the in-plant.  hydroxide is probably responsible for the rest of the diffraction peaks in corroded sample.

II. EXPERIMENTAL METHOD
It should be noted that the NaCl concentration is not optimized. We merely suppose the sea water, which is abundant resource. The preliminary result using more concentrate NaCl solution (10 %) also leads to the corroded state. The further study of another NaCl concentration is needed in the future.
We checked the XRD pattern of residue left after HCl dissolution of the corroded sample as shown in Fig. 3(a). The diffraction peaks are in good agreement with the XRD pattern of γ-FeOOH. Oxidatively-roasted γ-FeOOH exhibits the XRD pattern of α-Fe 2 O 3 (see Fig.   3(b)). Figure 3(a) indicates that the Nd hydroxide would be selectively dissolved into HCl solution, in which Nd ions are formed. The oxalic acid precipitation was carried out to recover Nd. The precipitate has been oxidatively roasted and evaluated by XRD pattern, which is displayed in Fig. 3 For each HCl concentration, R increases with prolonged immersion-time, and approximately saturates, however exceeds 100% at some conditions. The excess above 100% would be due to the existence of impurity phases in c-Nd 2 O 3 . The compositions of c-Nd 2 O 3 and α-Fe 2 O 3 analyzed by the ICP spectroscopy are summarized in Table I Fig. 5 is 97%. The 100% recovery ratio would be achieved with increasing HCl concentration and/or immersion time, however the amount of impurity phases would be increased. We have to introduce a sophisticated step such as pH adjustment in the oxalic acid precipitation to reduce the incorporation of impurity phases.
We revisited the conventional Nd-recovery-method using oxidation process. The coarselyground commercial Nd-Fe-B magnet after the oxidization at 600 • C for 5 h was dissolved into 0.2 mol/L HCl solution, at room temperature, with varying immersion-time from 30 min to 2 h. The oxidized Nd-Fe-B decomposes into Nd 2 O 3 and α-Fe 2 O 3 . In Fig. 6(a), the XRD pattern of recovered sample after the oxalic acid precipitation for each condition is displayed. Although c-Nd 2 O 3 is recovered, several impurity peaks, which are different from those in Fig. 4(b), appear (see arrows in the inset of Fig. 6(a)). Figure 6 Even though the preparation of α-Fe 2 O 3 by the oxidation process prevents the incorporation of Fe in the recovered material 5,6 , Nd-recovery ratio is rather low. According to the patent 7 , the high-temperature HCl-dissolution improves the recovery ratio near to 100%.

6
This means that the solubility of Nd 2 O 3 is low at room temperature. On the other hand, our proposed method, where the oxidation process is replaced with the corrosion one, have no need of temperature-rise to obtain enough solubility of Nd into HCl solution. After the corrosion, Nd-Fe-B magnet decomposes into γ-FeOOH and Nd hydroxide. The preparation of these compounds is the key factor to realize the high selectivity between Nd and Fe even at room temperature.
There is not yet a study, carrying out the proposed Nd recovery process for used magnets.
The elemental component of Nd-magnet depends on the kind of products, and we speculate that the process condition should be finely tuned for each composition. Therefore the sort of the kind of products might be necessary. In addition, the study for Nd-magnet with elemental component other than that in this study is further needed. We note here that the corrosion process allows the wide applicable range of Nd-magnet condition; for example, the proposed process is highly compatible with a magnet corroded due to peeled coating in using for many years. Furthermore our process may also allow a direct corrosion of a roughly disassembled product, and the recycle cost can be reduced.

IV. SUMMARY
We have demonstrated the Nd-recovery from Nd-Fe-B magnets by improving the in-plant recycle method, based on the wet process using HCl solution and oxalic acid precipitation.
The pretreatment of corrosion plays the important role for the high selectivity between Nd and Fe even at room temperature. After the oxalic acid precipitation, c-Nd 2 O 3 is obtained.
The recovery ratio of Nd reaches to 97%, which is much higher than that obtained by the conventional recovery method using oxidation. We emphasize that our proposed method is compatible with the present in-plant recycling of sludge.