Experimental Study on the Separation of Selected Metal Elements (Sm, Co, Fe, and Cu) from Nitric Acid Leachate Using Specific Precipitants

Abstract

As more countries emphasize the importance of the circular economy, recycling resources from waste has become increasingly crucial. This study proposes a novel separation process for SmCo magnets, which can separate and recover metals by precipitation, thus reducing the amount of solvent used. The precipitation process involved the use of Na₂SO₄, NH₄OH, and H₂C₂O₄ to separate Sm, Fe, Cu, and Co, resulting in high precipitation efficiencies of 96.11%, 99.97%, 93.81%, and 98.15%, respectively. Moreover, the recovered metals can be directly used to create magnets after calcination, making this process a step towards achieving a circular economy.

1. Introduction

Rare earth elements include light rare earth elements (LREEs) such as Lanthanum (La), Cerium (Ce), Praseodymium (Pr), Neodymium (Nd), Promethium (Pm), Samarium (Sm), and Europium (Eu), and heavy rare earth elements (HREEs) such as Gadolinium (Gd), Terbium (Tb), Dysprosium (Dy), Holmium (Ho), Erbium (Er), Thulium (Tm), Ytterbium (Yb), and Lutetium (Lu). Yttrium (Y) and Scandium (Sc) are two other elements excluded from the LREEs and HREEs classifications [1,2]. The Sm in SmCo magnets belongs to the light rare earth elements.

In 2022, the global rare earth reserves amounted to approximately 130 million metric tons. Among them, China, Vietnam, Brazil, and Russia hold 44 million, 22 million, 21 million, and 21 million metric tons of rare earth reserves, respectively [3]. These four countries together possess more than 80% of the world’s rare earth reserves. Rare earth reserves are excessively concentrated in specific countries. In addition, the production of rare earth is also concentrated in specific countries. China’s rare earth production in 2022 was 210,000 metric tons, which represented about 70% of the total global rare earth production [3]. Furthermore, in 2010, China issued the “The Opinion on Promoting Mergers and Acquisitions of Enterprises” [4], which listed rare earth elements as a key industry for mergers, acquisitions, and reorganization, in an attempt to reduce the export of rare earths. This led to a historical high in the price of rare earths [5]. The most effective method currently to reduce the risk of rare earth supply is to recycle metals from waste [6,7].

Rare earth elements have become a significant topic of discussion among researchers, entrepreneurs, and politicians in recent years due to their crucial role in current technological advancements. Rare earths are utilized in a broad range of applications [8,9,10], including wind turbines, electric vehicles [11], mobile phones, hard disk drives, fluorescent and LED lamps, defense applications, catalysts, pharmaceuticals, and medicine [12]. Of these, over a quarter of rare earths are utilized in magnet manufacturing [13], which is the field that consumes the most rare earths. Consequently, if a large number of magnets can be recycled, rare earth elements can be effectively recovered.

Among the rare earth elements, Nd and Sm are commonly used in magnet production. Nd is primarily used to create NdFeB magnets, which are widely applied across industries due to their high energy density. These industries include acoustic transducers, wind turbines, electric vehicles, industrial motors, missiles, tanks, warplanes, submarines, microwaves, computers, printers, and hard disk drives [14,15,16]. In comparison, SmCo magnets exhibit unique properties, including higher coercivity and superior temperature resistance, making them ideal for specialized applications like aero engines [17,18,19]. While NdFeB magnets currently account for over 90% of annual rare earth magnet production, many experts predict an increase in demand for SmCo magnets in the coming years [20]. This shift is largely due to the high cost of dysprosium (Dy), a key component in NdFeB magnets. Consequently, innovative direct and indirect recycling methods for SmCo magnets must be developed.

Currently, methods for recovering rare earth from NdFeB magnets include hydrometallurgical processes [21,22,23,24] and pyrometallurgical processes [25,26,27,28]. However, there have been few studies on the recovery of rare earths from SmCo magnets. In the literature [29,30,31] on the pre-treatment of SmCo magnets, waste magnets were first demagnetized in a high-temperature furnace. Sinha et al. (2017) [29] proposed demagnetizing them at a temperature of 850 °C for 6 h. Then, SmCo magnets should be crushed and ground using a hydraulic press and grinder, which is beneficial for follow-up research. Sinha et al. (2017) [29] also suggested mechanically crushing, grinding, and screening the SmCo magnet ring with a 150 μm sieve. After completing the pretreatment process, most studies [29,31,32,33] use hydrometallurgical methods to purify and separate valuable metals. First, the powder is leached with either an inorganic acid (such as sulfuric acid, hydrochloric acid, or nitric acid) [29,31,32] or an organic acid (such as citric acid) [34]. However, Zhou et al. [33] (2017) proposed using sulfuric acid to leach waste SmCo magnets. Unfortunately, Sm easily reacts with sulfuric acid to form samarium sulfate precipitation, which results in poor leaching efficiency. Therefore, most studies [29,32] have used hydrochloric acid and nitric acid for leaching. After the leaching experiment was completed, the study used solvent extraction [17,29,32,35], ion exchange [36], and precipitation [31,33,37,38] methods for metal separation. Finally, the recovered powder was calcined to meet the raw materials for making magnets. Sahoo et al. proposed that the separated Sm and Co solutions be precipitated with oxalic acid, and then the samarium oxalate and cobalt oxalate were calcined at 800 °C and 450 °C, respectively, to form metal oxides.

One of the primary strategies for circulating magnet materials throughout the supply chain involves proposing the most effective approach to refine valuable materials from discarded SmCo magnets. However, the above studies [17,29,31,32,33,35,37,38] only focused on the recovery of Sm and Co, and did not purify and separate all metals (Sm, Co, Fe, Cu, and Zr) in the waste. Therefore, this study hopes to propose a complete recovery method to deal with real waste. In addition, in the separation method, after the metal is separated by the solvent extraction method, metal precipitation is required, and the process is relatively complicated. The ion exchange method, on the other hand, requires diluting the aqueous solution, resulting in increased wastewater volumes. In this study, a simpler process was developed, using the precipitation method to directly separate the metal by precipitation, so the metal could be reprocessed into a magnet.

2. Results and Discussion

2.1. Selective Precipitation of Sm by Na₂SO₄

To facilitate subsequent research, precipitation experiments were carried out using a L₉(3⁴) orthogonal table to determine the precipitation efficiency of Sm. The results of these experiments, showing the precipitation efficiencies of Sm under different parameters, are presented in

Table 1. The orthogonal array experiment results of Sm precipitation.

No.	Temp.	Time	Na2SO4	Sm	Co	Fe	Cu
Unit	°C	min	(w/v)%	%	%	%	%
1	70	20	6	17.69	0.09	0.64	0
2	70	40	9	90.88	2.05	3.51	1.13
3	70	60	12	99.19	2.31	6.06	0.88
4	80	20	9	90.22	1.78	5.74	0.84
5	80	40	12	99.31	2.85	4.31	0.84
6	80	60	6	44.63	1.25	2.71	0
7	90	20	12	99.40	3.03	4.78	1.34
8	90	40	6	48.59	0	0	0
9	90	60	9	95.60	0	3.19	0

. The precipitation efficiency was observed to be affected by the temperature, addition of Na₂SO₄, and time [33].

In this study, the priority of each factor on Sm precipitation efficiency was analyzed through factor effect analysis using results from orthogonal table experiments. The precipitation efficiency at the addition of Na₂SO₄ of K3 (as shown in

Table 2. Factor response table for Sm precipitation.

Sm	Effect Factor	Temperature	Time	Na2SO4
	K1	69.25%	69.10%	36.97%
	K2	78.05%	79.60%	92.23%
	K3	81.20%	79.81%	99.30%
	Extreme Deviation	11.94%	10.70%	62.33%
	Priority Order	Na2SO4 > Temperature > Time

) represents the average precipitation efficiency with an addition of Na₂SO₄ of 12 (w/v)% (as shown in Table 1). The extreme deviation is the difference between the best and worst precipitation rates of Sm under K1 to K3 conditions. The order of priority for factors that affect precipitation efficiency was determined using extreme deviation [39,40].

Table 2 shows the factor response table for the precipitation rate, which revealed that the order of influence on precipitation efficiency was Na₂SO₄ > temperature > time. The preliminary optimal parameters determined were 12 (w/v)% Na₂SO₄ addition, 80 °C temperature, and 40 min time, which resulted in the highest precipitation efficiency. As the precipitation efficiency at 40 and 60 min was found to be equivalent, 40 min was chosen as the initial optimum parameter. Similarly, as the precipitation efficiency at 80 and 90 °C was also found to be equivalent, 80 °C was selected as the initial optimum parameter time.

Figure 1 illustrates the results of the Sm precipitation confirmation experiment. As shown in Figure 1a, Sm precipitation efficiency was less than 50% when 6 (w/v)% of sodium sulfate was added. However, when the amount of sodium sulfate was increased to 10 (w/v)%, the precipitation efficiency of Sm significantly improved, reaching 96.25%. Therefore, the amount of sodium sulfate added was identified as the most crucial parameter, as it had a significant impact on precipitation effectiveness.

Compared with the amount of sodium sulfate added, the temperature had a negligible effect on Sm precipitation efficiency, as depicted in Figure 1b. When the Sm temperature was 60 °C, the precipitation efficiency reached 93.44%, and increasing the temperature did not result in a significant improvement in precipitation efficiency. However, at a temperature of 60 °C, the co-precipitation efficiency of Fe was 18.56%. Thus, 60 °C was not the optimal precipitation parameter. The primary reason for Fe co-precipitation was that Fe readily forms ferric sulfate precipitation at lower temperatures [41,42,43], resulting in increased co-precipitation rates of Fe. Therefore, the optimal temperature was 70 °C.

Figure 1c shows that the Sm precipitation reaction was relatively slow, and it took 50 min to achieve a Sm precipitation rate higher than 95%.

Combining the above results, the optimal conditions for precipitation were determined to be the addition of 10 (w/v)% of sodium sulfate, a temperature of 70 °C, and a precipitation time of 50 min. Under these conditions, the precipitation rates of Sm, Co, Fe, and Cu were 96.11%, 0.05%, 0%, and 0%, respectively. This successful separation of Sm resulted in a co-precipitation rate of other metals that were less than 0.1%.

The findings of this study were compared to those of Zhou et al. [33] (

Table 3. Comparison of results for Sm precipitation.

No.	This Study	Zhou et al. [33]
Temp.	70 °C	80 °C
Time	50 min	60 min
Molar ratio	6.69:1	4:1
Leachate	HNO3	H2SO4
Ions in the leachate	Sm, Co, Fe, and Cu	Sm and Co
The precipitation rate of Sm	96.11%	93.4%

). Zhou et al. [33] utilized sulfuric acid for leaching, resulting in a solution with a high concentration of sulfate ions, which promoted the precipitation of sodium samarium bisulfate. Consequently, less sodium sulfate was added in their study. The precipitation of sodium samarium bisulfate is demonstrated in Formula (1). Despite the higher molar ratio used in this study, a greater amount of Sm was successfully precipitated at lower temperature and time conditions. Moreover, this study effectively optimized the parameters for Sm precipitation.

$${\mathrm{S}\mathrm{m}}_2{\left({\mathrm{S}\mathrm{O}}_4\right)}_3+{\mathrm{N}\mathrm{a}}_2{\mathrm{S}\mathrm{O}}_4+2{\mathrm{H}}_2\mathrm{O}\leftrightarrow 2{{\mathrm{N}\mathrm{a}\mathrm{S}\mathrm{m}(\mathrm{S}\mathrm{O}}_4)}_2·{\mathrm{H}}_2\mathrm{O}$$

(1)

2.2. Selective Precipitation of Fe by NaOH

This section analyzes the impact of pH and time on Fe precipitation. Figure 2a shows the effect of pH on Fe precipitation, revealing that the efficiency of Fe precipitation was less than 10% when the pH was below 2.5, and the co-precipitation rate of Cu exceeded 50% when the pH was above 4. Thus, pH was a crucial factor in Fe precipitation. The optimal Fe precipitation rate of 98.28% was achieved at a pH of 3.5, with less than 1% co-precipitation of other metals. Therefore, a pH of 3.5 was selected as the optimal parameter.

The results regarding the effect of time on Fe precipitation are shown in Figure 2b. The efficiency of Fe precipitation was faster compared with that of Sm precipitation. At a time of 10 min, the precipitation rate of Fe reached 99.97%.

Combining the aforementioned results, it can be concluded that the precipitation rates of Fe, Cu, and Co were 99.97%, 0%, and 0%, respectively, when the pH was 3.5 and the time was 10 min. Fe was successfully separated, and the co-precipitation rate of other metals was less than 0.1%.

The study of SmCo magnets has not yielded any other studies on Fe precipitation separation so far. Therefore, no comparison of experimental results was made, highlighting the novelty of this study.

2.3. Selective Precipitation of Cu by NaOH

This section discusses the impact of pH and time on the efficiency of Cu precipitation. The effect of pH on the Cu precipitation efficiency is demonstrated in Figure 3a. The pH level played a crucial role in Cu precipitation as it significantly influenced precipitation efficiency. The optimal precipitation pH value was found to be 6.5, resulting in a Cu precipitation efficiency of 94.93%.

Figure 3b shows the effect of time on Cu precipitation efficiency. The Cu precipitation rate was slow, taking more than 70 min to achieve over 90% efficiency.

By combining the above findings, it was determined that a pH of 6.5 and a precipitation time of 70 min resulted in Cu and Co precipitation rates of 93.81% and 0.95%, respectively, indicating successful Cu separation with less than 1% Co co-precipitation.

The study of SmCo magnets has not yielded any other studies on Cu precipitation separation so far. Therefore, no comparison of experimental results was made, highlighting the novelty of this study.

2.4. Selective Precipitation of Co by Oxalic Acid

This part discusses the influence of molar ratio and time on the Co precipitation effect, as shown in Figure 4. As the molar ratio increased from 0.5 to 1.5, the Co precipitation efficiency increased from 76.76% to 96.84%. However, increasing the precipitation time did not significantly increase the precipitation efficiency. Therefore, the molar ratio had a greater influence on Co precipitation.

In the study of the influence of time on Co precipitation, it was found that the Co precipitation rate was very fast. The Co precipitation efficiency reached 98.15% in 1 min.

Based on the above results, the precipitation rate of Co reached 98.15% under the condition of a molar ratio of 1.5 and a time of 1 min. These results were compared with those of Wang et al. [36] (

Table 4. Comparison of results for Co precipitation.

No.	This Study	Wang et al. [36]
Temp.	25 °C	60 °C
Time	1 min	60 min
Molar ratio	1.5:1	1:1
Leachate	HNO3	H2SO4
The precipitation rate of Co	98.15%	96.20%

). Despite the higher molar ratio used in this study, more Co was successfully precipitated under lower temperature and time conditions. Moreover, this study effectively optimized the parameters for Co precipitation.

2.5. Characterization of NaSm(SO₄)₂

As NaSm(SO₄)₂ releases sulfur oxides during TGA analysis, which can damage the instrument, the oxidation temperature was not determined. However, Denisenko et al. [44] proposed that heating NaSm(SO₄)₂ to 1187 °C would decompose it into liquid sodium sulfate and solid samarium oxide. In this study, NaSm(SO₄)₂ can be used to obtain samarium oxide, which can be reused to make magnets.

The Sm precipitate obtained was analyzed by XRD, and the results are displayed in Figure 5. The major peaks of NaSm(SO₄)₂ H₂O at 25.699°, 29.736°, 42.325°, and 49.278° were observed, confirming the presence of crystalline NaSm(SO₄)₂ H₂O as the dominant phase.

2.6. Preparation of Iron Oxide and Characterization

The TGA diagram and oxidation formula of Fe(OH)₃ are shown in Figure 6a and Formula (2), respectively. It can be observed that there is no significant weight loss when the temperature reached 500 °C, and the weight remained at about 59.40% of the original weight. Based on the theory of weight loss after oxidation, it can be concluded that the final oxidation temperature of Fe(OH)₃ was 500 °C.

Fe(OH)₃ was calcined at 500 °C for 4 h, and the XRD analysis was performed (Figure 6b). The results showed that after calcination, the sample exhibited the main peak of Fe₂O₃ at 33.152°, confirming that Fe(OH)₃ had transformed into Fe₂O₃.

$$2\mathrm{F}\mathrm{e}{\left(\mathrm{O}\mathrm{H}\right)}_3\to {\mathrm{F}\mathrm{e}}_2{\mathrm{O}}_3+3{\mathrm{H}}_2\mathrm{O}$$

(2)

2.7. Preparation of Copper Oxide and Characterization

Figure 7a and Formula (3) depict the TGA diagram and oxidation formula of Cu(OH)₂, respectively. Upon analysis, it was observed that there was no significant weight loss when the temperature reached 500 °C. At this temperature, the weight was approximately 80.36% of the original weight, which was consistent with the theoretical value of weight loss after oxidation. Thus, 500 °C was determined to be the final oxidation temperature of Cu(OH)₂.

After calcining Cu(OH)₂ at 500 °C for 4 h, the XRD analysis was performed, and the results are shown in Figure 7b. The XRD pattern of the calcined Cu(OH)₂ matches with the main peaks of CuO at 35.495° and 38.730°, confirming the transformation of Cu(OH)₂ to CuO upon calcination.

$$\mathrm{C}\mathrm{u}{\left(\mathrm{O}\mathrm{H}\right)}_2\to \mathrm{C}\mathrm{u}\mathrm{O}+{\mathrm{H}}_2\mathrm{O}$$

(3)

2.8. Preparation of Cobalt Oxide and Characterization

The TGA diagram and oxidation formula of CoC₂O₄ are shown in Figure 8a and Formula (4), respectively. Upon analysis, it was observed that there was no significant weight loss when the temperature reached 470 °C. At this temperature, the weight was approximately 38.98% of the original weight, which was consistent with the theoretical value of weight loss after oxidation. Therefore, 470 °C was determined to be the oxidation temperature of CoC₂O₄.

After calcination of CoC₂O₄ at 500 °C for 4 h, the XRD analysis was performed, and the results are shown in Figure 8b. The XRD pattern of the calcined CoC₂O₄ corresponds to the main peaks of Co₃O₄ at 36.935°, 59.508°, and 65.406°, confirming the transformation of CoC₂O₄ to Co₃O₄ upon calcination.

$$3\mathrm{C}\mathrm{o}{\mathrm{C}}_2{\mathrm{O}}_4+2{\mathrm{O}}_2\to {\mathrm{C}\mathrm{o}}_3{\mathrm{O}}_4+6\mathrm{C}{\mathrm{O}}_2$$

(4)

3. Materials and Methods

3.1. Materials

Tai Chuan Metal Co., Ltd. (Taipei, Taiwan) provided the SmCo magnet scraps used in this study, which were abundant in Sm, Co, Fe, Cu, and Zr.

Table 5. The content of the SmCo magnet and the concentration in the leachate.

Element	Sm	Co	Fe	Cu	Zr
The weight percent of element (wt.%)	22.70	51.08	14.51	5.12	4.31
Metal concentration in leachate (ppm)	15,756	31,088	8762	3174	-

displays the composition of the SmCo magnet scrap after digestive analysis and the concentrations of Sm, Co, Fe, and Cu in the leachate, which only contains these metals since Zr, which cannot be leached [45], was initially separated during the leaching process. The leachate was produced at a temperature of 25 °C, a nitric acid concentration of 2.5 mol L⁻¹, a solid-to-liquid ratio of 65 g L⁻¹, and a time of 1 min, resulting in a leaching efficiency of 99.52%, 92.45%, 92.84%, and 94.30% for Sm, Co, Fe, and Cu, respectively.

Nitric acid (HNO₃, 70%) and hydrochloric acid (HCl, 37%) for aqua regia digestion, nitric acid (HNO₃, 70%) for leaching, sodium sulfate (Na₂SO₄, 99.5%) for Sm precipitation, ammonia (NH₄OH, 29%) for precipitation of Fe and Cu, and oxalic acid (H₂C₂O₄, 99%) for Co precipitation were purchased from Echo Chemical Co., Ltd. (Miaoli, Taiwan). All aqueous solutions were prepared in deionized water.

3.2. Equipment

The pH of the solution was monitored by a pH meter (PL-700PVS, Dogger Science, Taipei, Taiwan). In order to control the precipitation and leaching temperature, the precipitation and leaching process was carried out in a thermostatic bath with magnetic stirring (Shin-Kwang Precision Industry Ltd., New Taipei, Taiwan). Subsequently, the solid–liquid separation was carried out through a filter device (Chemker 300, Rone Scientific Co., Ltd., New Taipei, Taiwan). An atomic absorption spectrometer (AAS, PinAAcle 900F AA Spectrometer, PerkinElmer Inc., Waltham, MA, USA) was used to analyze the concentration of metals in the aqueous solution. The oxidation temperature of Fe(OH)₃, Cu(OH)₂, and CoC₂O₄ in the precipitation were analyzed by thermogravimetry analysis (TGA, TGA-51, Shimadzu, Japan). The crystal structure of the product (NaSm(SO₄)₂, Fe₂O₃, CuO, and Co₃O₄) were analyzed via X-ray diffractometry (XRD, DX-2600, Dandong, China), with Cu Kα radiation ranging from 2θ = 10–70°, to identify the phases present. The precipitate was calcined using a high-temperature furnace (Thermo Scientific Lindberg/Blue M BF51866C Muffle Furnace, Hogentogler & Co., Inc., Howard County, Columbia, MD, USA).

3.3. Experimental Method

3.3.1. The Composition of the Magnet

To confirm the powder composition, aqua regia digestion was utilized. In detail, 1 g of SmCo magnet powder was added to a 250 mL conical flask containing 10 mL of nitric acid and 30 mL of hydrochloric acid. The flask was then placed in a thermostatic bath at 90 °C for one day to guarantee complete metal dissolution. The metal concentration in the resulting aqueous solution was analyzed using AAS. The contents of Sm, Co, Fe, Cu, and Zr in the magnet are displayed in Table 5.

3.3.2. Leaching of Magnets

The hydrometallurgical method was employed in this study to recover the metals from the magnet. For leaching, SmCo magnet powder and 100 mL of an acid solution (a mixture of deionized water and the required amount of acid) were added to a 250 mL conical flask, and the experimental temperature was regulated using a thermostatic bath. The resulting solution was filtered, and the metal concentration was analyzed using AAS. The leaching efficiency (L%) was determined using Formula (5) [33].

$$\mathrm{L}\left(\%\right)={\displaystyle \frac{{\mathrm{V}}_1\left(\mathrm{L}\right)\times {\mathrm{C}}_1(\mathrm{g} {\mathrm{L}}^{-1})}{{\mathrm{M}}_1\left(\mathrm{g}\right)\times {\mathrm{W}}_1}}\times 100\%$$

(5)

where M₁ is the mass of the alloy sample, W₁ is the metal content of the alloy sample, V₁ is the volume of the leach solution, and C₁ is the mass concentration of the metal in the leach solution.

3.3.3. Precipitation Method

For the precipitation experiment, this study added 100 mL of leachate and precipitant to a 250 mL Erlenmeyer flask. The experimental temperature and stirring rate were controlled using a thermostatic bath. The resulting solution was filtered, and the metal concentration was analyzed using AAS. The precipitation efficiency (P%) was determined using Formula (6) [33].

$$\mathrm{P}\%={\displaystyle \frac{{\mathrm{V}}_2\times {\mathrm{C}}_2-{\mathrm{V}}_3\times {\mathrm{C}}_3}{{\mathrm{V}}_2\times {\mathrm{C}}_3}}\times 100\%$$

(6)

where P is the precipitation percentage of metal, V₂ and V₃ are the volume of the leachate and the filtrate, respectively, and C₂ and C₃ are the mass concentrations of metal in the leachate and in the filtrate, respectively.

3.3.4. Sm Precipitation

The previous study [33] revealed that sodium sulfate exhibited selective separation characteristics for Sm precipitation under low pH conditions. Therefore, this study first employed sodium samarium sulfate to investigate Sm separation. In contrast to the study by Zhou et al. [33], which focused solely on the separation of Sm and Co, this study aims to investigate the separation of Sm, Co, Fe, and Cu with more complex components. Furthermore, this study employed the Taguchi method to determine the priority order of each parameter on precipitation efficiency and determine the optimal precipitation parameters. In comparison to the one-factor-at-a-time experimental method and the full-factor experiment method, the Taguchi method requires fewer experiments to obtain useful statistical information. Therefore, the Taguchi method was adopted for this study.

The efficiency of Sm precipitation is influenced by various factors, such as temperature, the addition of Na₂SO₄, and time [33]. To determine the priority order of these three factors on precipitation efficiency, this study used the Taguchi method. First, the orthogonal array experiment was implemented. This study used a L₉(3⁴) orthogonal table with selected control factors and levels of temperature (70–90 °C), an addition of Na₂SO₄ (6–12 g 100 mL⁻¹), and time (20–60 min), as shown in

Table 6. A L₉(3⁴) orthogonal table of Sm precipitation.

No.	Temp.	Time	Na2SO4
Unit	°C	min	(w/v)%
1	70	20	6
2	80	40	9
3	90	60	12

. After completing the orthogonal array experiment, this study performed a factor effect analysis to understand the order of priority of each factor affecting the Sm precipitation efficiency and found the preliminary optimal experimental parameters. Finally, this study used the priority ranking of precipitation by each factor to conduct confirmation experiments and obtained the best precipitation parameters.

3.3.5. Fe and Cu Precipitation

This study examines the precipitation efficiency of Fe, Cu, and Co at different pH values. The results [38] show that Fe precipitation occurs at a pH range of 2–2.7, while Cu precipitation occurs at pH 4.5–6.5, and Co precipitation at pH 7.5–9. Based on these findings, it is believed that adjusting the pH value is an effective method for separating these metals. Furthermore, the research findings [38] indicate that precipitation of each metal can be achieved at room temperature. As a result, this study opted to investigate the precipitation of Fe and Cu at 25 °C.

Given that the parameters that need to be adjusted for Fe and Cu precipitation are pH and time, and based on previous studies [38], it is clear that pH affects metal precipitation more than time. Therefore, the optimal precipitation parameters for Fe and Cu were determined without evaluating the relative importance of each parameter. For Fe precipitation parameter adjustments, the pH range was from 2 to 5 and the time range was from 1 to 30 min. For Cu precipitation parameter adjustments, the pH range was from 5 to 7 and the time range was from 10 to 90 min.

3.3.6. Co Precipitation

To produce a reusable product, this study employed oxalic acid for Co precipitation. For Co precipitation parameter adjustments, the molar ratio ranged from 0.5 to 4, and the time ranged from 1 to 9 min.

3.3.7. Product Preparation and Characterization

To obtain raw materials suitable for magnet production, this study analyzed the oxidation temperature of the precipitate using TGA to confirm the oxidation temperature of each metal. However, sodium samarium bisulfate was not analyzed using TGA due to the generation of sulfur oxides during the analysis process, which could cause damage to the instrument.

After confirming the oxidation temperature of each metal, this study proceeded with the oxidation of the precipitate. Specifically, 5 g of the precipitate was placed into a ceramic vessel, which was put into a high-temperature furnace. The temperature was raised to the required level in order to obtain the oxides of each metal. Finally, the XRD [46] was used to confirm the composition of the product (NaSm(SO₄)₂, Fe₂O₃, CuO, and Co₃O₄).

3.3.8. Separation Process

Figure 9 shows the research process. First, this study added sodium sulfate to precipitate Sm from the leachate. The Sm formed a precipitate of NaSm(SO₄)₂ which was separated from the leachate. Secondly, this study added ammonia to precipitate Fe from the leachate by adjusting the pH, and the Fe formed a Fe(OH)₃ precipitate. Thirdly, this study added ammonia to precipitate Cu by adjusting the pH and obtained a Cu(OH)₂ precipitate. Fourth, the Co was precipitated as CoC₂O₄ by the addition of oxalic acid. Fifth, this study carried out a TGA analysis and calcination experiments to obtain the metal oxides of the precipitates. Finally, this study carried out XRD analysis to confirm the composition of the precipitates.

4. Conclusions

In this study, Na₂SO₄, NH₄OH, and H₂C₂O₄ were used to precipitate and separate Sm, Fe, Cu, and Co. The precipitation efficiencies for Sm, Fe, Cu, and Co were 96.11%, 99.97%, 93.81%, and 98.15%, respectively, with co-precipitation efficiency below 1%. The XRD analysis indicated that the precipitates were NaSm(SO₄)₂, Fe₂O₃, CuO, and Co₃O₄, respectively. This study proposes a new method for recycling SmCo magnets, enabling the recovered metals to be converted back into magnets to support a circular economy.

Compared with solvent extraction and ion exchange techniques, the precipitation method proposed in this study significantly reduces wastewater output and enables a one-step recovery process, thereby streamlining the recovery procedure. Nevertheless, the approach to wastewater treatment in this study warrants further investigation. Given the advanced state of current industrial wastewater treatment technologies, it is anticipated that effective solutions can be implemented to address this aspect with clarity.