Comparison of the Preparation Process of Rare Earth Oxides from the Water Leaching Solution of Waste Nd-Fe-B Magnets’ Sulfate Roasting Products

Abstract

The new process developed here consisting of sulfurization roasting transformation and water immersion can effectively realize the separation of rare earth elements (REEs) and impurities from spent Nd-Fe-B magnets. For the industrial application of the new process, it is critical to determine how to economically and efficiently prepare rare earth oxide (RE_xO_y) products with higher purity from the obtained water leaching solution. Therefore, according to rare earth sulfate (RE₂(SO₄)₃) solution characteristics, the oxalic acid precipitation–calcination method, sodium carbonate precipitation–calcination method, and double sulfates precipitation–alkali conversion–calcination method were optimized and compared. The results show that the recovery efficiency of REE recovery via the oxalic acid precipitation–calcination method is 99.44%, and the purity of RE_xO_y is 99.83% under optimal technological conditions. However, the cost of oxalic acid precipitation is higher. The process consisting of the double sulfates precipitation–alkali conversion–calcination method is relatively complicated, the recovery efficiency of REEs is 97.95%, and the purity of the RE_xO_y is 98.04%. The recovery efficiency of the REEs and the purity of the RE_xO_y obtained from the sodium carbonate precipitation–calcination method are 99.12% and 98.33%, respectively. Moreover, the recycling cost of sodium carbonate precipitation is the lowest among the three processes for preparing RE_xO_y, so it has industrial application potential. The obtained results for REE recovery from spent Nd-Fe-B magnets in this research can provide theoretical guidance for the innovation of the recycling process for REEs as secondary resources.

1. Introduction

REEs are known as “industrial vitamins” and are widely used in aviation, aerospace, medicine, electronic information, and other high-tech fields. More than 40% of REEs are used to make Nd-Fe-B permanent magnets in China, and the average annual growth rate of the industry in over 22.5%. Statistics show that approximately 110 thousand tons of Nd-Fe-B waste are produced in China each year. However, >30% of the raw materials are abandoned in the production and processing of Nd-Fe-B permanent magnets [1,2,3]. This not only causes a waste of resources, but also creates potential harm to the environment. Therefore, the recycling of Nd-Fe-B waste has both economic and environmental benefits.

In recent years, much relevant research has been conducted on the recycling and utilization of Nd-Fe-B waste. Pyrometallurgical methods such as alloy methods [4,5], chloride methods [6,7], selective oxidation methods [8,9], and melting separation methods [10] have been developed, but these processes have the disadvantages of high energy consumption and low REEs recovery efficiencies. Thus, these pyrometallurgical processes are difficult to realize in industrial applications. Typical hydrometallurgical technologies have been widely promoted, including the hydrochloric acid selective leaching method [11,12], hydrochloric acid total leaching method [13,14], and sulfuric acid leaching method [15,16,17]. However, the above processes require high corrosion-resistant equipment, and the REEs are difficult to selectively separate. In order to achieve the efficient separation of REEs, two-stage ammonium sulfate roasting followed by a water leaching process [18] has been used. The leaching of REEs can reach 96% under optimal conditions, whereas the leaching rates of impurities such as Fe, Al, Cu, and Co are substantially lower, which results in the high-efficiency separation of REEs and metal impurities.

The ultimate objective of waste Nd-Fe-B magnet recovery is to obtain high purity RE_xO_y products from the REE-enriched leachate. Currently, the oxalate precipitation method [19,20,21] and double sulfates precipitation method [22,23] are often used in industry to further separate REEs from aqueous solutions. The oxalic acid precipitation method has the advantages of good crystallization performance, easy filtration, high REEs precipitation efficiencies, and high purity of obtained products. Nevertheless, the large amount of wastewater produced by the oxalic acid precipitation method needs to be treated to make it harmless, which inevitably increases the complexity and treatment cost for enterprises. The double sulfates precipitation method mainly relies on the fact that RE₂(SO₄)₃ and sulfates (Na₂SO₄) can undergo insoluble double sulfates precipitation. According to the difference in the solubility of double sulfates precipitation, REEs can be divided into the following three groups: (1) the cerium group (La, Ce, Pr, Nd, and Sm), which exhibits insoluble double sulfates precipitation; (2) the terbium group (Eu, Gd, Tb, and Dy), which exhibits slightly soluble double sulfates precipitation; (3) the yttrium group (Ho, Er, Tm, Yb, Lu, and Y), which exhibits easily soluble double sulfates precipitation. Therefore, the double sulfates precipitation method cannot ensure the efficient recovery of REEs, especially for terbium and yttrium group REEs.

Because the solubility of rare earth carbonates (RE₂(CO₃)₃) is smaller than that of rare earth oxalates (RE₂(C₂O₄)₃) and rare earth double sulfates (NaRE(SO₄)₂), the yield of REEs is higher when using the carbonate precipitation method. The carbonate precipitation reagents mainly include ammonium bicarbonate (NH₄HCO₃) [24,25,26], sodium bicarbonate (NaHCO₃) [27,28], and sodium carbonate (Na₂CO₃) [29,30]. Although the NH₄HCO₃ precipitation method is relatively mature, a large amount of ammonia nitrogen wastewater is produced, which has adverse impacts on aquatic ecosystems. The precipitation method using NaHCO₃ and Na₂CO₃ as precipitants can solve the problems of ammonia nitrogen pollution. However, the amount of NaHCO₃ needed as a precipitant is large, and the production cost is high. In contrast, Na₂CO₃ as a precipitant not only reduces the amount of precipitant needed but also reduces the introduction of sodium ions, resulting in a lower cost and better development advantages than NaHCO₃. When considering the recovery cost, the Na₂CO₃ precipitation method has great industrial application potential.

Considering that the previous development of an ammonium sulfate roasting–water leaching process showed significant advantages, in order to ensure the continuity of the process, this research further optimizes the sodium carbonate precipitation process and at the same time compares it with the oxalic acid precipitation and double sulfates precipitation processes so as to obtain a more economical, environmentally friendly, and efficient REE recovery and separation process. This research also provides a valuable reference for the separation of REEs from sulfate leachate in REE recycling smelters.

2. Experimental Materials and Methods

2.1. Experimental Materials

The raw material used in this study was the leachate obtained from Nd-Fe-B waste treatment after using a two-stage ammonium sulfate roasting transformation followed by a water leaching method [18], and its main components and contents are shown in

Table 1. The main composition and content of leachate obtained after sulfated roasting followed by a water leaching process.

Composition	Ce	Pr	Nd	Gd	SO42−	Al	Fe	Co	Cu	Si
Content (mg/L)	1720	1700	5654	1510	17,290	3.12	1.10	2.45	2.08	2.05

.

In Table 1, it can be seen that the concentrations of impurities (Fe, Al, Co, Cu, and Si) in the REEs leaching solution are all lower than 5 mg/L, while the concentrations of Ce, Pr, and Gd are above 1500 mg/L and that of Nd is up to 5500 mg/L. Therefore, RE_xO_y is directly prepared by the oxalic acid precipitation–calcination method, sodium carbonate precipitation–calcination method, and double sulfates precipitation–alkali conversion–calcination method.

2.2. Experimental Method

Oxalic acid precipitation–calcination method: According to existing literature reports and industrial production process parameters, the most suitable temperature for REE precipitation is 80 °C. Due to the difference in the concentration of rare earth ions (RE³⁺), this paper investigates the influence of oxalic acid amount and reaction time on the precipitation efficiency of REEs. RE₂(C₂O₄)₃ precipitate under optimum conditions was calcined for 1 h to obtain RE_xO_y at 850 °C, and the impurity content and RE_xO_y purity were characterized.

Double sulfates precipitation–alkali conversion–calcination method: As reported previously, a reaction time of 1 h was used to investigate the influence of the Na₂SO₄ amount and the reaction temperature on the precipitation efficiency of REEs. The obtained NaRE(SO₄)₂ was added to boiling sodium hydroxide (NaOH) solution for alkali conversion to produce rare earth hydroxides (RE(OH)₃) under a reaction temperature of 90 °C, a reaction time of 1 h, m(RE₂O₃):m(30%NaOH) = 1.0:1.0. and a stirring speed of 350 r/min. The alkali conversion products (RE(OH)₃) were roasted for 1 h to obtain RE_xO_y at 700 °C, and the impurity content and RE_xO_y purity were tested.

Sodium carbonate precipitation–calcination method: Na₂CO₃ was added to the solution of REEs, and the effects of feeding method, Na₂CO₃ amount, reaction temperature, Na₂CO₃ concentration, feed liquid flow rate, aging time, seed crystal addition amount, and product size were systematically investigated. The RE₂(CO₃)₃ precipitate was roasted for 1 h to obtain RE_xO_y at 850 °C, and the impurity content and RE_xO_y were tested.

During the preparation process, the resulting purity of the RE_xO_y and the cost of the abovementioned three RE_xO_y preparation processes were comprehensively compared to find the optimal production process. The main flowsheet for this research is shown in Figure 1.

2.3. Experimental Analysis Method

In this study, a Thermo ICAP-AES 7400 inductively coupled plasma emission spectrometer produced by HORIMA was used to analyze the REEs’ concentration (Ce, Pr, Nd, and Gd) in the leachate. The main phases of the samples were identified by X-ray diffraction (XRD; PANalytical X’Pert Pro Powder, Almelo, The Netherlands) using a Cu Kα radiation source with a 40 kV acceleration potential and current of 40 mA. XRD diffractograms were analyzed using Jade 6.5 software. The morphology and element distribution behaviors in the leach residues and roasted products were determined by scanning electron microscopy with energy dispersive spectroscopy (SEM-EDS; MIRA 3 LMH, TESCAN Brno, S.r.o, Brno, Czech Republic). The particle size distribution of the REE precipitation products was analyzed using a Mastersizer 3000 laser particle size analyzer from Malvern (UK). In order to determine the thermal decomposition behavior of RE₂(C₂O₄)₃, RE(OH)₃, and RE₂(CO₃)₃, thermogravimetric analysis (TG) and differential thermal analysis (DTA) were undertaken (STA449F5, Netzsch, Germany) with a 10 K/min heating rate under a 100 mL/min dry air flow in order to analyze the reaction steps.

3. Results and Discussion

3.1. Oxalic Acid Precipitation Method

The effects of oxalic acid amount (actual amount (mol)/theoretical amount (mol)) and reaction time on the transformation of the precipitation efficiency of REEs were investigated at 80 °C, and the results are displayed in Figure 2.

The results shown in Figure 2a indicate that the precipitation efficiencies of Ce, Pr, Nd, and Gd increased from 99.48%, 99.61%, 99.80%, and 99.79%, respectively, to 99.94%, 99.95%, 99.97% and 99.97%, respectively as the molar ratio of the oxalic acid increased from 1.0:1.0 to 1.2:1.0. A further increase in the H₂C₂O₄ amount from 1.2:1.0 to 1.4:1.0 resulted in only minor changes in the precipitation efficiencies of the REEs. Therefore, an oxalic acid amount of 1.2:1.0 was selected for the separation of REEs from the sulfate leachate. In Figure 2b, the precipitation of REEs from the sulfate leachate is relatively rapid, as 98% is achieved within 10 min. Further increases in the amount of oxalic acid within 30 min resulted in a gradual increase to over 99.94% precipitation efficiency. In order to achieve the maximum precipitation efficiencies of the REEs, the optimum reaction time was 30 min.

Based on the results from the experiments, the optimum conditions for the oxalic acid precipitation method are as follows: the optimum oxalic acid amount is 1.2:1.0, the optimum reaction temperature is 80 °C, and the optimum reaction time is 30 min. Under these conditions, the total precipitation efficiencies of the REEs reach 99.94%. The obtained RE₂(C₂O₄)₃ precipitation was washed twice with hot water at a liquid-to-solid ratio of 2:1 (mL/g). During the washing stage, the REE loss was 0.52%, and the total recovery efficiency of REEs was 99.44%. The XRD analysis of RE₂(C₂O₄)₃ conducted under optimal conditions (Figure 3) shows that the main phases present in the RE₂(C₂O₄)₃ were Pr₂(C₂O₄)₃·10H₂O, Ce₂(C₂O₄)₃·10H₂O, Nd₂(C₂O₄)₃·10H₂O, and Gd₂(C₂O₄)₃·10H₂O. It had an excellent crystallinity and a particle size within the range of 1–5 μm (Figure 4).

3.2. Double Sulfates Precipitation–Alkali Conversion Method

3.2.1. Double Sulfates Precipitation of REEs

The effect of the reaction temperature and Na₂SO₄ amount (actual amount (mol)/theoretical amount (mol)) on the precipitation of REEs over 1 h and the results are displayed in Figure 5.

The results shown in Figure 5a indicate that the precipitation efficiency of Gd increased from 84.90% to almost 94.45% as the reaction temperature increased from 70 °C to 80 °C. In contrast to Gd, the precipitation of Ce, Pr, and Nd changed slightly within the temperature range investigated under the same conditions. This is mainly because the double sulfates precipitation of cerium group elements such as Ce, Pr, and Nd are insoluble sulfates. However, Gd double sulfates is a slightly soluble terbium group element, and its solubility decreases with increasing temperature, so the temperature has a considerable influence on Gd precipitation. From the perspective of energy saving and REE recovery, a temperature of 80 °C was selected to be the most suitable process condition for the precipitation of REEs. Figure 5b shows that the precipitation efficiencies of Ce, Pr, Nd, and Gd increased from 98.02%, 98.18%, 97.98%, and 94.44% to 98.71%, 98.82%, 98.80%, and 98.67%, respectively, as the Na₂SO₄ amount increased from 1.0:1.0 to 1.3:1.0. A further increase in the sodium sulfate amount from 1.3:1.0 to 1.4:1.0 resulted in only minor changes in the precipitation efficiencies of the REEs.

Based on the results of the above experiments, the optimum conditions for the double sulfate precipitation method are as follows: a sodium sulfate amount of 1.3:1.0, a reaction temperature of 80 °C, and a reaction time of 1 h. Under these conditions, the precipitation efficiencies of Ce, Pr, Nd, and Gd are 98.71%, 98.82%, 98.80%, and 98.67%, respectively, and the total precipitation efficiency of the REEs is 98.75%. The XRD analysis (Figure 6) of double sulfates precipitation of REEs under optimal conditions shows that the main phases present in NaRE(SO₄)₂ were NaNd(SO₄)₂·H₂O, NaPr(SO₄)₂·H₂O, NaCe(SO₄)₂·H₂O, and NaGd(SO₄)₂·H₂O.

3.2.2. Alkali Transformation of Double Sulfates Precipitation

In order to prepare RE_xO_y, it is necessary to transform the refractory NaRE(SO₄)₂·H₂O into RE(OH)₃, which not only avoids the production of SOx but also increases the product purity of RE_xO_y. The main reaction is as follows:

NaRE(SO₄)₂·H₂O + 3NaOH = RE(OH)₃↓+ 2Na₂SO₄ + H₂O

(1)

In the stage of alkali transformation, the NaRE(SO₄)₂·H₂O was added to a NaOH solution, and the slurry was filtered under the following reaction conditions: the m(RE_xO_y):m(30%NaOH) was 1.0:1.0, the reaction temperature was 90 °C, the reaction time was 1 h, and products with a gray color were obtained. The obtained products were further dried, and the gray precipitate immediately transformed into a yellow precipitate (Figure 7). This is mainly because Ce(OH)₃ is easily oxidized to yellow Ce(OH)₄ when exposed to air. Under the above optimal conditions, an alkali conversion rate of REEs of 99.55% was calculated by measuring the concentration of REEs in the filtrate. The prepared RE(OH)₃ was washed twice with hot water (liquid-to-solid (L/S, 100 mL distilled water/50 g hydroxide): 2:1) to ensure the purity of the subsequent RE_xO_y. During the washing stage, the loss of REEs was 0.36%. Thus, the total recovery efficiency of REEs after transformation and washing was 97.95%. The obtained RE(OH)₃ crystals were irregularly granular and had a relatively dispersed particle size distribution (Figure 8).

3.3. Sodium Carbonate Precipitation Method

In order to obtain the optimal conditions for RE₂(CO₃)₃ precipitation, the effects of the following factors on the precipitation efficiencies of REEs were investigated: feeding method of Na₂CO₃, Na₂CO₃ amount (n(RE³⁺)/n(Na₂CO₃)), reaction temperature, Na₂CO₃ concentration, feed liquid flow rate, aging time, and seed crystal addition amount. The results are shown in Figure 9.

The effects of the feeding method, including positive precipitation (the sodium carbonate solution is added to the leaching solution of RE₂(SO₄)₃), reverse precipitation (the leaching solution of RE₂(SO₄)₃ is added to the sodium carbonate solution), and co-precipitation (the sodium carbonate solution and the leaching solution of RE₂(SO₄)₃ were added simultaneously) on the recovery efficiencies of the REEs were investigated. The results shown in Figure 9a indicate that the precipitation efficiencies of the REEs are >99% when using counter-precipitation, which is much higher than the other two feeding methods. Moreover, the XRD results of the three precipitation products of the REEs (Figure 10) showed that the precipitation of the REEs obtained by reverse precipitation had good crystallinity and mainly consisted of Ce₂(CO₃)₃·8H₂O, Pr₂(CO₃)₃·8H₂O, Nd₂(CO₃)₃·8H₂O, and Nd₂(CO₃)₃·8H₂O. RE₂(CO₃)₃ shows better filtration performance when using counter-precipitation when compared with positive precipitation and co-precipitation. The XRD results of RE₂(CO₃)₃ obtained from the positive precipitation and co-precipitation had no obvious diffraction peaks, and the Na₂CO₃ precipitate was amorphous and had worse filtration performance.

As is shown in Figure 9b, when the molar ratio of n(RE³⁺)/n(Na₂CO₃) was increased from 1.0:1.5 to 1.0:1.6, the precipitation efficiencies of the REEs increased substantially, and the precipitation efficiencies of all the REEs were over 99.90%. When the molar ratio was further increased to 1.0:1.8, there was no significant change in the precipitation efficiencies of the REEs, so the suitable molar ratio was determined to be 1.0:1.6. The effect of Na₂CO₃ concentration on the precipitation efficiencies of the REEs was investigated and is presented in Figure 9c. The results showed that the precipitation efficiencies of the REEs did not significantly change when the Na₂CO₃ concentration was increased from 0.2 mol/L to 0.3 mol/L, and when the Na₂CO₃ concentration was increased to 0.4 mol/L, the precipitation efficiencies of the REEs decreased significantly. Therefore, the appropriate Na₂CO₃ concentration was determined to be 0.3 mol/L.

Figure 9d shows that the precipitation of REEs did not change significantly when the flow rate of RE₂(SO₄)₃ leaching solution was increased from 2 mL/min to 4 mL/min. However, the precipitation efficiencies of the REEs decreased considerably when the flow rate of the RE₂(SO₄)₃ leaching solution was increased further. The REEs’ precipitation efficiency decreases because the Na₂CO₃ is wrapped by the RE₂(CO₃)₃ precipitate produced due to the rapid local precipitation reaction with the fast flow rate of RE₂(SO₄)₃ leaching solution, which leads to unreacted REEs. However, due to the practical considerations of production, the flow rate of the feed solution should not be too slow, so the suitable flow rate of the RE₂(SO₄)₃ leaching solution was determined to be 4 mL/min.

The effect of reaction temperature on the precipitation efficiencies of the REEs is presented in Figure 9e. The results showed that an increase in the temperature from 30 °C to 70 °C resulted in only minor changes in the precipitation efficiencies of the REEs, but a higher temperature means that the crystalline shape of the RE₂(CO₃)₃ will deteriorate (Figure 11), so the suitable reaction temperature for RE₂(CO₃)₃ precipitation was determined to be 30 °C.

From Figure 9f, it can be seen that the introduction of seed crystals had no obvious effect on the precipitation of the REEs, and the precipitation efficiencies of Ce, Pr, Nd, and Gd were all above 99% throughout the investigated range. Thus, seed crystals were not added during the RE₂(CO₃)₃ precipitation. In addition, although the aging time has little effect on the precipitation efficiencies of the REEs (Figure 9g), the particle size of RE₂(CO₃)₃ increased considerably with increased aging time (Figure 9h). In terms of industrial production, the aging time is determined to be 4 h.

In summary, the optimal process conditions for the sodium carbonate precipitation method are as follows: the feeding method is reverse precipitation, the n(RE³⁺)/n(Na₂CO₃) ratio is 1.0:1.6, the reaction temperature is 30 °C, the Na₂CO₃ concentration is 0.3 mol/L, the flow rate of the solution of REEs is 4 mL/min, the aging time is 4 h, and seed crystals are not added. Under these conditions, the total precipitation efficiency of the REEs was 99.93%, and the obtained carbonate products were feathery with a more uniform particle size distribution and good filtration performance (Figure 12). The RE₂(CO₃)₃ was washed twice with 90 °C water (liquid-to-solid (L/S, 100 mL distilled water/50 g hydroxide): 2:1) to ensure the purity of the subsequent RE_xO_y. During the washing stage, the loss of REEs is 0.81%. Thus, the total recovery efficiency of REEs is 99.12% after transformation and washing.

3.4. Rare Earth Oxides Preparation

The prepared RE₂(C₂O₄)₃, RE₂(CO₃)₃, and RE(OH)₃ are calcined to obtain RE_xO_y. The main reactions are as follows:

RE₂(C₂O₄)₃·10H₂O + 3/2O₂ = RE₂O₃ + 6CO₂↑ +10H₂O↑

(2)

RE₂(CO₃)₃·8H₂O = RE₂O₃ + 3CO₂↑ + 8H₂O↑

(3)

2RE(OH)₃ = RE₂O₃ + 3H₂O↑

(4)

To confirm the optimum calcination temperature, TG and DTA were performed on RE₂(C₂O₄)₃·10H₂O, RE₂(CO₃)₃·8H₂O, and RE(OH)₃. The results shown in Figure 13 indicate that the RE₂(C₂O₄)₃·10H₂O was completely decomposed into RE_xO_y with a weight loss of 55.2% at 850 °C, which was consistent with the theoretical value of 54.1%. Similarly, the RE₂(CO₃)₃·8H₂O completely decomposed into RE_xO_y with a weight loss of 42.1% at 850 °C, which is relatively consistent with the theoretical weight loss value of 45.1%. In contrast to the RE₂(C₂O₄)₃·10H₂O and RE₂(CO₃)₃·8H₂O, the RE(OH)₃ started to lose weight at 600 °C, and the weight loss basically stopped at 700 °C. Thus, to ensure the purity of RE_xO_y, the most suitable calcination temperature for RE₂(CO₃)₃·8H₂O and RE₂(C₂O₄)₃·10H₂O was determined to be 850 °C, and the most suitable calcination temperature for RE(OH)₃ was 700 °C.

The RE_xO_y obtained from the calcination of RE₂(CO₃)₃·8H₂O and RE₂(C₂O₄)₃·10H₂O at 850 °C and the calcination of RE(OH)₃ at 700 °C were analyzed via XRD to confirm the main mineral phase composition. The results shown in Figure 14 indicate that the roasting products of the three REE precipitates are all composed of Pr₆O₁₁, Ce₆Nd₂O₁₅, and (Ce_0.7Gd_0.3)O_1.85. The SEM analysis of the roasted products (Figure 15a) showed that the REEs obtained from the calcination of RE₂(C₂O₄)₃·10H₂O were found to be irregularly cubic in shape, ranging in size from 2–5 μm. Moreover, the EDS results (Figure 15b) of RE_xO_y indicated that the main components of the products are REEs, and no other metal impurities were detected. The structure of the RE_xO_y obtained via the calcination of RE₂(CO₃)₃·8H₂O is a polyhedron with a random shape and a size range of 0.5–2 μm (Figure 15c). However, the calcination products of RE₂(CO₃)₃·8H₂O contain a small amount of S in addition to REEs (Figure 15d). The structure of the RE_xO_y obtained by the calcination of the RE(OH)₃ is mostly irregular cone in terms of shape and ranges in size from 1–4 μm (Figure 15e). The EDS results (Figure 15f) showed that the calcination products contain a small amount of Si. To compare the purity of these calcined RE_xO_y products, the main chemical composition of the above three products of RE_xO_y were further analyzed in detail, and the results are shown in

Table 2. Chemical composition of RE_xO_y (mass, %).

REO	Ce	Pr	Nd	Gd	Al	Ca	Co	Cu	Fe	Si	SO42−
a	30.51	10.03	31.54	11.67	0.0070	0.020	0.0012	0.0047	0.002	0.0008	0.081
b	30.38	9.86	30.89	11.34	0.0054	0.024	0.0020	0.0021	0.006	0.011	0.12
c	30.16	9.68	31.61	10.78	0.01	0.034	0.00019	0.0021	0.006	0.052	0.090

(a) calcination products of RE₂(C₂O₄)₃·10H₂O; (b) calcination products of RE₂(CO₃)₃·8H₂O; (c) calcination products of RE(OH)₃.

.

As can be seen from Table 2, the purities of the RE_xO_y obtained via the calcination of RE₂(C₂O₄)₃·10H₂O, RE₂(CO₃)₃·8H₂O, and RE(OH)₃ are 99.83%, 98.33%, and 98.04%, respectively. Moreover, the content of SO₄^2- impurities is higher in the RE_xO_y obtained via the calcination of RE₂(CO₃)₃·8H₂O, while Al and Si impurities are more prevalent in the RE_xO_y obtained via the calcination of RE(OH)₃.

3.5. Cost Comparison of Three Preparation Methods

In this study, RE_xO_y was prepared via the oxalic acid precipitation–calcination method, sodium carbonate precipitation–calcination method, and double sulfates precipitation–alkali conversion–calcination method. The total recovery efficiency of REEs, the purity of the RE_xO_y, and the cost of producing 1 t of RE_xO_y were compared. The results are shown in

Table 3. Comparison of the three RE_xO_y oxide preparation methods.

Precipitation Method	REEs Recovery Rate/%	Product Purity/%	H2C2O4 Dosage/t	H2C2O4 Price (CNY/t)	Sum (CNY)
a	99.44	99.83	0.96	5500	5280
Precipitation method	REEs recovery rate/%	Product purity/%	Na2CO3 dosage/t	Na2CO3 price (CNY/t)	Sum (CNY)
b	99.12	98.33	1	2600	2600
Precipitation method	REEs recovery rate/%	Product purity/%	Na2SO4 dosage/t	Na2SO4 price (CNY/t)	NaOH dosage (t)	NaOH price (CNY/t)	Sum (CNY)
c	97.95	98.04	0.55	520	1	3300	3586

(a) oxalic acid precipitation–calcination method; (b) sodium carbonate precipitation–calcination method; (c) double sulfates precipitation–alkali conversion–calcination method; the amount of each reagent required is calculated based on the amount needed to produce 1 t of RE_xO_y.

.

The comparison results showed that although the RE_xO_y prepared via the the oxalic acid precipitation–calcination method has the highest recovery efficiencies of the REEs and the greatest purity of RE_xO_y, the production cost of the RE_xO_y products is also obviously higher than that of the other two processes. Compared with the oxalic acid precipitation–calcination method, the cost of preparing RE_xO_y via the sulfate method is relatively low, but it is still higher than the carbonate precipitation method, and the process flow is more complicated and has the lowest REE recovery efficiency and RE_xO_y purity. Therefore, REEs recovery from RE₂(SO₄)₃ leaching solution via the sodium carbonate precipitation–calcination method has the greatest potential for industrial application. This process has the lowest cost for preparing RE_xO_y, and the REEs’ recovery efficiencies and RE_xO_y purity are ideal.

4. Conclusions

(1)

Under optimal conditions, the REEs recovery efficiencies of the oxalic acid precipitation–calcination, sodium carbonate precipitation–calcination, and double sulfates precipitation–alkali conversion–calcination methods were 99.44%, 99.12%, and 97.95%; the purities of obtained the RE_xO_y were 99.83%, 98.33%, and 98.04%, respectively.

(2)

The recovery of REEs from RE₂(SO₄)₃ leaching solution via the sodium carbonate precipitation–calcination method has the greatest potential for industrial application. This process has the lowest cost for preparing RE_xO_y, and the recovery efficiencies of the REEs and the purity of the RE_xO_y are ideal.

(3)

By combining the obtained research results with previous studies, a new process for the selective recovery of REEs from NdFeB waste can be formed, which mainly involves ammonium sulfate roasting, REEs water leaching separation, sodium carbonate precipitation, and calcination. The results can also provide theoretical guidance for innovations in secondary resource recovery processes for REEs.