Low-Acid Leaching for Preferential Lithium Extraction and Preparation of Lithium Carbonate from Rare Earth Molten Salt Electrolytic Slag

Abstract

In this work, lithium was preferentially recovered through the low-acid leaching from rare earth molten salt electrolytic slag (REMSES) with a leaching temperature of 60 °C. The influence on lithium extraction was investigated in detail in different leaching conditions. The optimal conditions were as follows: liquid-to-solid ratio (10 mL/g), sulfuric acid concentration (0.8 mol/L), leaching time (60 min) and leaching temperature (60 °C). This yielded a lithium extraction rate of 98.52% and a lithium carbonate purity of 99.5%. It was fitted using an empirical model; the kinetics showed that internal diffusion control conformed to the low-acid leaching reaction, which had an apparent activation energy of 10.81 kJ/mol for lithium. The total profit from the whole process was USD 0.2576 when dealing with 1.0 kg of REMSES. Moreover, in the sulfuric acid system, the leaching reaction mechanism was carefully investigated between 30 and 90 °C. An innovative process of recovering lithium from REMSES was achieved with environmental friendliness and good economic returns. Compared to traditional leaching using concentrated sulfuric acid, this cleaner recycling method conforms to the concept of green, low-carbon sustainable development, with high lithium selectivity, low impurity content in the filtrate and low acid consumption.

1. Introduction

Rare earth metals are produced from the rare earth smelting industry through a molten salt electrolysis method, with high current efficiency, stable electrolyte composition, difficult hydrolyzation, a good rare earth recovery effect, and green environmental protection [1,2]. In China, about 5000 tons of rare earth oxides entered (in oxide terms) REMSES in this process every year, which not only included rare earth fluoride but also a lot of graphite powder, lithium fluoride, iron oxide and many other metal and non-metallic impurities [3,4]. If not disposed of reasonably, precious resources will be wasted and environmental pollution will increase. Lithium, as the main raw material in lithium batteries, is widely applied in new energy industry; its demand will also rapidly increase in future years [5,6]. Due to the immaturity of lithium extraction technology in domestic salt lakes and the high cost of lithium extraction from lepidolite, the yield of domestic lithium carbonate makes it difficult to meet the growing demand for lithium resources in China, and reasonable development and utilization of domestic lithium resources have become a research hotspot [7,8,9]. Therefore, a considerable amount of lithium extraction from REMSES not only improves the resource utilization rate and economic benefits of the rare earth molten salt electrolysis industry but also provides a new way to recover the lithium from REMSES, meanwhile alleviating the current imbalance between the demand and supply of domestic lithium resources to a certain extent.

In recent years, the processes of recycling REMSES have attracted a lot of attention due to their high rare earth content, which was mainly categorized into acidic and alkali recycling methods [10,11,12,13,14]. Generally, the content of lithium was low, relative to rare earth, in REMSES and was often abandoned, but the lithium content was comparable to the lithium content in lepiolite [8]. Ammonium sulfate was used for the transformation roasting of REMSES and extraction of lithium. The lithium extraction rate was above 90%, while the process has the disadvantage of high energy consumption, lengthy decontamination steps, a large of amount consumption of oxalic acid, high organic content in the lithium solution and difficulty removing impurities [15]. A new process of recovering lithium from REMSES was proposed, using aluminum sulfate and calcium oxide for synergistic roasting activation followed by acid immersion. Though >95% of lithium was achieved using this process, the recycling processes have the drawbacks of high energy consumption, incomplete fluorinated rare earth conversion and a long process flow [13]. Mineral phase reconstruction for REMSES was acquired by leaching in sulfuric acid solution, with the transformation rates of lithium reaching 95.88%. The process provides a way for the efficient utilization of fluoride ions. However, there were drawbacks such as low rare earth recovery and a long process [10]. An effective method for recycling lithium from molten salt electrolytic slag via selective nitration was proposed, with the effective leaching of lithium (99.3%), and impurities were selectively separated. Though the phase-selective transformation was acquired at a temperature of 250 °C, there are drawbacks, including more operation steps, a complicated process and a large amount of acid-containing wastewater for disposal [11]. The recovery of lithium as lithium fluoride with a purity of 99.8 wt% by vacuum distillation from REMSES via mineral phase reconstruction was carried out. This method has the advantage of being environmentally friendly for the high-value recycling of lithium from REMSES. But, a long roasting time, high roasting temperature, high energy consumption and high cost were generated in this process [16]. In addition, lithium mainly existed as lithium fluoride in REMSES, and lithium fluoride was easily decomposed by mineral acid and organic acid, while rare earth fluoride and rare earth oxyfluoride mostly do not react with low acid at low temperature (<100 °C) [17,18,19].

In this study, the recovery of lithium by low-acid leaching from REMSES was investigated and elucidated in depth. The process parameters, including the reaction temperature, reaction time, liquid-to-solid ratio and sulfuric acid concentration, on the leaching rate of lithium from REMSES were investigated in detail. Impurities, such as calcium, aluminum, iron, silicon and fluorine, were separated from the lithium-containing leaching solution. Small amounts of leached RE elements could be recycled in subsequent processes. The surface morphology, phase evolution and element content were analyzed throughout this study using scanning electron microscopy (SEM) images, X-ray diffraction patterns (XRD) and inductively coupled plasma–optical emission spectrometry (ICP–OES), respectively. Finally, the solid waste of REMSES was regenerated and utilized, and lithium carbonate was prepared via precipitation. A novel process for the selective recovery of lithium from REMSES was proposed based on low temperature using dilute sulfuric acid. In addition, the economic analysis showed that the proposed process was economical and environmentally friendly.

2. Materials and Methods

2.1. Materials

In this work, the REMSES was supplied from a company (Ganzhou, China). REMSES was crushed via a sample pulverizer (XY-100, Xinyun corporation, Hebi, China), and the sample was sieved via a vibrating sieve with a particle size of 100 mesh (150 μm). All chemicals were of analytical grade, such as sodium hydroxide, sodium carbonate, calcium hydroxide and sulfuric acid, and ultrapure water (electrical conductivity < 0.055 μs/cm) was used to prepare the solutions. The mass fraction of raw materials is presented in

Table 1. The mass fraction of the REMSES (wt.%).

Element	C	Li	Na	Al	Fe	Ca	Si	F	TREO *
%(w/w)	9.62	1.23	16.49	1.47	6.31	2.20	5.20	16.44	26.69

* TREO represents total rare earth oxides.

.

2.2. Experimental Process

A three-neck flask with a 250 mL polytetrafluoroethylene vessel was used for acid leaching experiments, and the experiments were conducted in a water bath equipped with stirring, temperature controlled at a predetermined liquid-to-solid ratio (4–12 mL/g) for a given time (20–100 min) at certain temperature (30–90 °C). The sulfuric acid solutions (0.4–1.2 mol/L) were prepared as the leaching medium via diluting concentrated sulfuric acid. The sample (10 g) obtained from above process was tested, vacuum filtration was used to filter the slurry after the acid leaching, and the volume of filtrate was collected and recorded. Finally, the extraction rates of lithium and rare earth (RE) were characterized.

The extraction of lithium and RE ($\eta$) was calculated from above filter liquor using Equation (1):

$$\eta =\left(c_i\times v\right)/\left(m_o\times {\omega }_o\right)\times 100\%$$

(1)

where $c_i$ and $v$ represented the concentration of element ($i$) and the volume of the leaching liquor; ${\omega }_o$ and $m_o$ (g) are the compositions of element ($i$) and the mass of the REMSES, respectively.

The filtrate was first purified by adding calcium hydroxide to remove fluoride ion and then by adding sodium hydroxide, adjusting the pH to 3.0–4.0, reacting under agitation for 30 min. After vacuum filtration, lithium solution and iron hydroxide slime were obtained. Sodium hydroxide was further added into the lithium-containing solution to adjust the pH to 12.0 with stirring at 60 °C for 40 min; lithium-containing solution was acquired after filtration and alkaline impurity removal residue, which was evaporated and concentrated to about 24 g/L, and saturated sodium carbonate solution was added dropwise into the concentrated solution at 95 °C for 60 min to precipitate out lithium carbonate. The obtained solids were washed three times using deionized water at 95 °C, which was put in a drying oven for 4 h at 105 °C. The purity and impurity contents of lithium carbonate were detected according to methods proposed by the Standardization Administration of PRC [20]. A flow chart for lithium recovery via acid leaching process from REMSES is shown in Figure 1.

2.3. Characterization

The morphology and element distribution of raw materials, leaching residue, and the product were detected using field-emission scanning electron microscopy–energy-dispersive spectrometry (SEM–EDS, MIRA 3 LMH, TESCAN Brno, S.r.o, Brno, Czech Republic). The average mass fraction of the REMSES was measured by inductively coupled plasma–optical emission spectrometry (ICP–OES, Perkin Elmer Optima 7100 DV, Waltham, MA, USA), and the content of carbon was measured using a carbon–sulfur analyzer (EMIA-820 V, Horiba, HORIBA Corporation, Kyoto, Japan). X-ray fluorescence (XRF) analysis with a Panalytical (AXIOS model) sequential spectrometer was used to identify the major elements. The content of fluoride ions in the solution was measured using an ion meter (PXSJ-216F, Leici, Shanghai, China). The main mineral phases of the REMSES X-ray diffraction patterns (XRD, PANalytical X’Pert Pro Powder, Almelo, The Netherlands) were used to characterize the phase of leaching residue and the product, which has a 40 kV acceleration potential and current of 40 mA, with a CuKα radiation source. MDI Jade 6.0 software was used for analyzing XRD diffractograms. The products of lithium carbonate was determined via a fourier transform infrared spectrometer (FTIR, ALPHA, Bruker Corporation, Bremen, Germany).

3. Results and Discussion

3.1. Low-Acid Leaching Process

3.1.1. The Effect of Acid Concentration

The effects on the extraction rate of lithium and RE metals were investigated in a low-acid leaching experiment. This investigation was carried out using 10 g of raw materials treated using the above-mentioned method, the concentration of sulfuric acid (0.4–1.2 mol/L), the liquid-to-solid ratio (4–12 mL/g), leaching time (20–100 min) and leaching temperature (30–90 °C).

Sulfuric acid is widely used for lithium-containing compounds [21]; the following dissolution reaction was proposed as the most possible, as shown in Equation (2):

2LiF + H₂SO₄ → Li₂SO₄ + 2HF

(2)

Figure 2 proposed diagram of reaction mechanism. Generally, lithium fluoride and rare earth fluoride were decomposed using lithium, fluoride and RE ions in the concentrated acid leaching process, and lithium and RE were recovered in steps as precipitate. However, in this low-acid leaching process, lithium in REMSES was first entered into solution, and RE remained in the acid leaching residue. Moreover, compared to the concentrated acid leaching process, the impurity of iron, calcium, and aluminum was only partly dissolved in the solution in the low-acid leaching process, which reduced the burden of impurities’ removal in subsequent processes. Then, fluorine, calcium, aluminum and iron were removed as the precipitate of calcium fluoride, calcium sulfate, aluminum hydroxide and iron hydroxide. Lithium was precipitated out as lithium carbonate. Therefore, lithium can be decomposed from REMSES and transformed into water-soluble lithium ions in the sulfuric acid leaching process.

Figure 3a shows the effect on the leaching of REMSES with a sulfuric acid concentration at 60 °C for 60 min, with a 10 mL/g liquid-to-solid ratio. The results demonstrating the extraction rate of lithium increased markedly from 88.9% to 98.52% when the concentration of sulfuric acid increased from 0.4 mol/L to 0.8 mol/L, after which a plateau was reached. Minor changes in the lithium extraction were observed when the sulfuric acid concentration was further increased from 0.8 to 1.2 mol/L. Compared to the lithium leaching rate, the leaching behavior of RE remains at a low level (<2%) when the sulfuric acid concentration is below 0.8 mol/L because rare earth fluoride does not react with dilute sulfuric acid under this condition. To achieve a higher lithium extraction rate and simultaneously lower the leaching rate of rare earth, 0.8 mol/L sulfuric acid was determined as the optimal leaching concentration for follow-up research. Figure 3b demonstrates that the XRD results of the leaching residue obtained at the optimal sulfuric acid concentration were mainly composed of neodymium fluoride and carbon, and the lithium fluoride phase disappeared.

3.1.2. The Effect of Liquid-to-Solid Ratio

The effects on the extraction rate of lithium and rare earth were investigated at different liquid-to-solid ratios (4–12 mL/g) under unchanged experimental conditions. Figure 4a shows that the lithium extraction rate increased gradually as the liquid-to-solid ratio rose in the range of 4 to 8 mL/g. The extraction of lithium slowly increased and then reached a peak with a liquid-to-solid ratio increment, while the RE extraction rate was always stabilized at a particular low level as the liquid-to-solid ratio was below 10 mg/L. Considering the amount of sulfuric acid and the evaporation and concentration process in the follow-up experiment, the volume of leached liquid could be as low as possible; therefore, a liquid-to-solid ratio (10 mL/g) was determined for subsequent research. Figure 4b illustrates that the XRD diffraction peaks of lithium fluoride were largely undetected at different liquid-to-solid ratios. It could be that lithium fluoride was mostly decomposed by sulfuric acid.

3.1.3. The Effect of Acid Leaching Temperature

The effect on the extraction rate was tested at different temperatures (30–90 °C), with the other experimental conditions unchanged. Figure 5a indicates that as the reaction temperature increased from 30 °C to 60 °C, the lithium leaching rate increased from 83.4% to 98.52%, reaching a level point as the reaction temperature was further increased. The extraction rate of RE increased with the temperature increment because a higher temperature results in more violent collisions between molecules because of higher kinetic energy. The reaction temperature was further increased but did not markedly improve the extraction rate, since the reaction was mostly complete [22,23]. Therefore, the reaction temperature of 60 °C was determined to be the optimal condition. The XRD diffraction peak of lithium fluoride in the raw material cannot be found, as presented in Figure 5b, indicating that it was basically dissolved by sulfuric acid, which was also consistent with the results of ICP detection.

3.1.4. The Effect of Acid Leaching Time

The effect on the extraction rate of lithium was carried out at different acid leaching times, with the other experimental conditions unchanged. Figure 6a shows that the lithium extraction rates increased from 86.7 to 98.52% and then smoothed out as the leaching time increased from 20 to 60 min, indicating that increasing time was beneficial for the leaching process; i.e., the extent of dissolution of the raw material was increased with a prolonged reaction time [24]. To achieve the maximum extraction rate of lithium, 60 min was found to be the optimal reaction time. Figure 6b demonstrates that only the diffraction peaks of neodymium fluoride and carbon in the raw material matched the results of ICP detection.

Therefore, the liquid-to-solid ratio and reaction temperature were the most influential factors in the lithium leaching process, and the optimal experimental conditions were as follows: a sulfuric acid concentration of 0.8 mol/L, a 10 mL/g of liquid-to-solid ratio, 60 min of leaching time and 60 °C of leaching temperature. Under this condition, 98.52% of lithium was extracted, while RE mostly retained the leaching residue in optimal conditions, and selective lithium extraction was achieved to a large extent.

3.2. Analysis of Raw Materials and Leaching Residue

The SEM-EDS spectra of the REMSES before and after acid leaching are shown in Figure 7a–f. Figure 7a shows that the raw materials through grinding as a whole were irregular and independent of each other. Most granular matter existed as monomers, but some was agglomerated together, and the map distribution was basically uniform, as shown in Figure 7c. The EDS spectrum of the REMSES (Figure 7b) illustrates that the surface of elements mainly includes oxygen, fluorine, sodium, aluminum, silicon, calcium, iron, neodymium and praseodymium. The results were basically consistent with those of XRD in Figure 7g. The mineral particles were gradually dissolved after leaching using dilute sulfuric acid, and the leaching residue was observed to exhibit uniformly shaped particles and agglomerates (Figure 7d). Figure 7f shows that the map distribution of oxygen, fluorine, sodium, aluminum, silicon, calcium, iron, neodymium and praseodymium was also uniform, while the proportion of fluorine and sodium in the leaching residue decreased significantly in the EDS energy spectrum (Figure 7e). The results illustrated that the sodium fluoride in REMSES was dissolved under the action of sulfuric acid. Figure 7g shows that only the diffraction peak of neodymium fluoride, praseodymium oxyfluoride and carbon was discovered through XRD, while the diffraction peak of lithium fluoride, sodium fluoride and praseodymium oxide completely disappeared, as discussed in the above results, which indicated neodymium fluoride, praseodymium oxyfluoride and carbon in the acid leaching residue. Therefore, lithium was successfully extracted from REMSES.

3.3. Exploration of Leaching Kinetics

The reaction mechanisms of low-acid leaching lithium were explored. The leaching lithium kinetics at different temperatures (20–80 °C) and times (2–30 min) was analyzed with the unreacted shrinking core model, whose mechanisms are generally represented by the mixed, diffusion and chemical reaction control models [25]. In view of the fact that the lithium fluoride in the raw material was quickly decomposed in the acid leaching process while the rare earth fluoride was hardly dissolved, in this experiment, the RE leaching kinetics was not considered. As the process was controlled by diffusion through the raw material layer, the model could be described using Equation (3):

$$1-3{(1-x)}^{{\displaystyle \frac{2}{3}}}+2(1-x)=k_{d}t$$

(3)

Assuming that the chemical reaction was the speed control step of the process, the shrinking core model can be expressed as Equation (4):

$$1-{(1-x)}^{{\displaystyle \frac{1}{3}}}=k_{c}t$$

(4)

where $k_d$ and $k_c$(min⁻¹) are the rate constant for the diffusion- and chemical-controlled process, $t$ is the chemical reaction time and $x$ represents the extraction rate, respectively.

The lithium extraction rate rises with the increase in temperature and time, as presented in Figure 8a and Table S1, which is consistent with previous studies [26,27]. To explore the kinetics of this research, data were matched with Equations (3) and (4). However, the correlation coefficients of these models were relatively low as well as the low-acid leaching in this work. Therefore, the “empirical model” (Equation (5)) was used as the fitting model [28].

$${(-\ln (1-x))}^2=k_{e}t$$

(5)

where $x$ represents the extraction rate of lithium, $k_e$ is the reaction rate constant and $t$ is the acid leaching time.

The correlation coefficients (R²) of Equation (3) and Equation (4) were lower than those of Equation (5). The linear fitting results and R² of Equation (5) were all more than 0.98, as presented in Figure 8b and Table S2. Therefore, Equation (5) was considered the most compatible model.

The Arrhenius Equation (6) described the connection between the values of the rate constants obtained in the kinetic plots at different temperatures and the apparent activation energy.

$$k=A{\exp }^{-{\displaystyle \frac{E_a}{RT}}}$$

(6)

where $A$ is the pre-exponential factor, $T$ means the thermodynamic temperature (K), $R$ represents the gas constant (8.314 J/(mol⋅K)), $E_a$ is the apparent reaction activation energy and $k$ is the rate constant.

By plotting $\ln k$ vs. 1000/$T$ in Figure 8c, the slope of the matching line was −1.3002 for lithium, and R² exceeded 0.98. Therefore, the $E_a$ of lithium was 10.81 kJ/mol by calculation, which was far below 30 kJ/mol. This result further verifies that the control step was internal diffusion control [28]. Raising the temperature and reducing the particle size of REMSES were effective methods to strengthen the acid leaching reaction process.

3.4. Purified and Precipitated Lithium Carbonate

The effects of the impurity removal ratio for the aluminum, iron, calcium, silicon, RE and fluorine and loss ratio of lithium were investigated under optimal conditions. The Lixivium-obtained acid leaching process was conducted by neutralizing at an initial pH of 0.83, which was neutralized to 8.0 of pH using 3 mol/L sodium hydroxide solution at room temperature; then, a mixture of calcium hydroxide and calcium chloride (molar ratio = 1:1) was added to solution to remove the impurity of fluoride with a stirring speed of 200 rpm for 30 min. After the reaction, the pH was adjusted for the second time to neutralize the solution by adding 3 mol/L sodium hydroxide to adjust the pH to 12.0. Then, the purified lithium-containing solution was evaporated and concentrated to a lithium ion concentration of about 24 g/L to improve the recovery rate of lithium. Following this, the lithium carbonate was precipitated out via introducing saturated sodium carbonate. The trace impurity elements in precipitated lithium carbonate were removed and washed three times with 95 °C deionized water. The contents of the main elements in the lixivium-purified and concentrated solution are shown in

Table 2. The concentration of main elements in the lixivium, purified and concentrated solution.

Samples	Index	Li	RE	Al	Ca	Fe	Na	Si	F
Lixivium	Extraction rate (%)	98.52	1.57	61.20	66.30	9.77	99.95	0.01	12.30
	Ionic concentration (g/L)	1.21	0.42	0.90	1.46	0.62	16.68	5.2 *	2.02
Purified solution	Removal or loss ratio (%)	10.59	99.98	99.01	99.64	99.59	1.50	100	98.91
	Ionic concentration (g/L)	1.08	0.08 *	8.91 *	5.25 *	2.51 *	16.43	ND	22.04 *
Concentrated solution	Removal or loss ratio (%)	10.80	/	/	/	/	69.40	/	/
	Ionic concentration (g/L)	24.15	ND	0.22	0.13	63*	127.0	ND	0.551

* (mg/L).

. The purity of depositing lithium carbonate was higher than that of the relevant Chinese standard (YS/T 582-2023, Li₂CO₃ > 99.5%). Figure 9a illustrates SEM images of the precipitated lithium carbonate, which existed as massive agglomerates in a flower-shaped distribution-like cluster. The XRD pattern of the lithium carbonate was consistent with lithium carbonate (JCPDS#87-0728 of standard PDF card), as shown in Figure 9b. Additionally, the relatively notable symmetric peaks of 1500.84 cm⁻¹ and 1432.65 cm⁻¹ appeared in the FTIR spectrum in Figure 9c, indicating that carbonate ion exists in the products.

3.5. Economic Analysis

The proposed chart of sulfuric acid leaching for preferential lithium recovery from REMSES is presented in Figure 1, and the proposed process of economic analysis was carried out. The costs of equipment, factory construction and human labor were not considered in this calculation. In addition, according to the local market situation in China, the price of raw materials was mainly determined by the value of rare earth. The small amount of rare earth generated during the acid leaching process in this study can be recycled, so the price of raw materials was not included in the calculation. In the experimental process, the products and chemicals involved (H₂SO₄, Ca(OH)₂, CaCl₂, NaOH, Na₂CO₃, Li₂CO₃) were all from China Guidechem Network (https://china.guidechem.com/). Taking 1.0 kg of rare earth REMSES as a calculation basis, the total cost was USD 0.8394 in this recovery process, and the costs of power, water and chemicals were all included. The gross revenue of the product reached USD 1.097. Thus, a profit of about USD 0.2576 can be obtained based on recycling 1.0 kg of REMSES. This method has the advantage of a reduced process, low acid, easy operation and large scale, which reduces economic costs from the source to a degree. In brief, the low-acid leaching method was environmentally friendly and has broad application prospects.

4. Conclusions

A low-acid leaching process was proposed in this study to enhance the selective recovery of lithium from REMSES. The results revealed that with low acid as a leaching agent, 98.52% of lithium could be converted to water-soluble lithium salts from REMSES in the following conditions: 0.8 mol/L of sulfuric acid concentration, 10 mL/g of liquid-to-solid ratio, 60 °C of the leaching temperature and 60 min leaching time. The RE element could be maintained in the form of neodymium fluoride and praseodymium oxyfluoride in the REMSES by low-acid leaching in this study, which has the advantages of reprocessing REMSES and recovery of the RE element. Simultaneously, the leaching kinetics shows that the internal diffusion-controlled process conformed to the low-acid leaching process at 20–80 °C, which has an apparent activation energy of 10.81 kJ/mol for lithium. The method is suitable for large-scale manufacturing, as a profit of USD 0.2576 was obtained, treating REMSES of 1.0 kg, and a feasible basis was provided for its industrial application. This method has the advantages of environmental friendliness, high profits, high-value-added products, less wastewater discharge and low-acid usage; meanwhile, in this study, the proposed process also conforms to the idea of green chemistry.