Extracting Lithium from Brazilian α-Spodumene via Chlorination Roasting

Abstract

The lithium market has been expanding due to the high demand for lithium-ion batteries, which are essential for electric and hybrid vehicles as well as portable devices. This has driven the search for new lithium ore deposits and the development of more efficient extraction and processing technologies. The main methods used for lithium extraction from hard rock ores include the acid process, the alkaline process, and chlorination roasting. This study investigated a chlorination process applied to α-spodumene extracted in Brazil for lithium chloride (LiCl) production. The ore underwent thermal treatment in the presence of calcium chloride (CaCl₂) and magnesium chloride (MgCl₂), followed by water leaching at 90 °C. The thermodynamics of the α-Li₂O·Al₂O₃·SiO₂ system, combined with calcium and magnesium chlorides, was analyzed using HSC 5.1 software. The main objective of this study was to produce lithium chloride from alpha spodumene and avoid decrepitation of the ore to the beta phase before mixing with the reagents, making the process faster and less expensive compared to traditional extraction methods. Pyrometallurgical tests were conducted in a muffle furnace, varying the molar ratio between chlorides (MgCl₂:CaCl₂) at 1:0, 0:1, 1:1, 2:1, and 1:2 and the mass ratio of spodumene to chlorides at 1:4, 1:6, and 1:8. The best lithium extraction result was approximately 95%, the conditions for obtaining the result were a spodumene:chloride ratio of 1:6 and a molar ratio between chlorides of 2:1. The results provide a better understanding of the chlorination roasting process and demonstrate the potential of the chlorination technique as a viable alternative to conventional lithium extraction methods.

1. Introduction

The Brazilian government has classified lithium ore as a strategic resource for the development of high-tech products and processes [1]. One of its main applications is the production of rechargeable lithium-ion batteries, widely used in electronic devices and electric vehicles. This metal possesses unique properties that ensure superior performance in battery cathodes, such as its low density—being the lightest of all metals—and its high electrochemical potential (−3.04 V), making it essential for the energy efficiency of these devices [2,3].

The largest lithium deposits are concentrated in South America in the Andean Cordillera region known as the “Lithium Triangle.” This area spans three countries and contains significant estimated mineral resources: Bolivia (23 million tons of Li), Argentina (22 million tons of Li), and Chile (11 million tons of Li). Regarding lithium mineral production, in 2023, the leading producers were Australia (86 thousand tons), Chile (44 thousand tons), and China (33 thousand tons). Brazil ranked as the fifth-largest global producer, producing 4.9 thousand tons in 2023, representing an 86.31% increase compared to 2022. This growth was driven by the authorization of lithium mineral and derivative commercialization, granted by Decree No. 11,120, dated 5 July 2022 [4,5].

Therefore, to reach the quantities required by the lithium market, it is important to expand lithium production beyond brine deposits and develop processes to extract lithium from lithium minerals resources as well [6,7,8,9].

In this context, the main sources of lithium in the world include brines and pegmatites. Each has advantages and challenges: brines have lower operating costs but long extraction processes and water supply problems. Pegmatites, on the other hand, have a more stable supply as well as require fewer extraction and purification steps but contain large volumes of ore with no viable commercial value. In addition, pegmatite ores guarantee have higher-security supply as they do not depend on seasonal or geographical conditions, making them a stable alternative for the industry. However, the growing demand for lithium has driven the exploration of both brines and hard rock in order to ensure a continuous and sustainable supply of the metal [10,11,12].

The lithium-bearing minerals reserves are widely distributed around the world (Australia, Chile, Argentina, China, the EUA, Brazil, and other countries) and have high potential for extraction. At present, the extraction of lithium from lithium minerals such as spodumene, lepidolite, and zinnwaldite is an important way of obtaining lithium resources [13,14,15]. Among these lithium minerals, spodumene (LiAlSi₂O₆) is the most important for the strategic development of the battery industry due to its high theoretical lithium content (8.03% Li₂O). The use of spodumene has a relatively lower process cost in comparison with the other minerals mentioned due to spodumene being the main raw material developed and utilized in industrial processes [16,17,18].

In nature, spodumene exists in the α-phase form. With most extraction methods, the α phase is resistant to the action of chemical reagents, either gaseous or liquid. Therefore, α-spodumene is usually first subjected to calcination at 1000–1100 °C for 1–2 h to convert it to β-spodumene, which is much more reactive and less resistant to common chemical agents [17,19,20,21].

Several studies have been conducted with the aim of identifying more efficient ways of extracting lithium from spodumene in order to reduce operational costs [20,22,23]. Among these investigations, different methods stand out, such as the acid process [19,20,24,25], the alkaline process [15,26,27], and chlorination [16,28,29]. The research on various methods employing different reagents and their respective experimental profiles for lithium extraction are summarized in

Table 1. Main reported methods for lithium recovery from minerals.

Method	Mine Li	Additive	Reaction Time	Temperature	Lithium Extraction (%)
Acid roasting [20,21,22,23,24,25]	β-spodumene (>1050 °C)	H2SO4, HF, HCl	20–30 min	75–225 °C	90–97
Alkaline process [3,15,19,26,27]	β-spodumene (>1050 °C)	Na2CO3 CaO, NaOH	60–360 min	225–250 °C	90.7–96
Chlorination roasting [16,28]	α-spodumene (>25 °C)	CaCl2, Cl2	120 min	900 °C	90.2

Source: authors.

.

The chlorination roasting process could be an alternative method of extracting metallic elements as chlorides from minerals as lithium chloride. Chlorination roasting is generally applied as a heat treatment for solid materials such as oxides, silicates, sulfides, and other compounds with a chlorinating agent at a suitable temperature [27,28,29,30,31].

As mentioned by Barrios et al. (2022) [31] and Fosu et al. (2022) [29], the chlorination process is efficient in some aspects such as its high rate of chlorination because of the high reactivity of chlorinating agents as well as the beneficial physical and chemical properties of certain metals such the low melting points of its chlorides; chlorination agents are available at reasonable prices. Nevertheless, this process faces some operational challenges such as fugitive emissions and waste. Some solutions have been applied to solve these issues, for example, treating and recovering the waste generated by the chlorination process so that the residual chlorinating agent can be regenerated and reused in the process. In addition, the waste generated can be treated with alkaline reagents and disposed of with minimal environmental impact. For this reason, emission reduction and reagent recovery in the chlorination process are important research development directions [16,29,30,31,32,33,34].

Chlorination roasting followed by water leaching was used by Yan et al. (2012) to extract lithium from lepidolite [35]. The results indicated that the chlorination roasting temperature, time, and mass ratio of the chlorinating agent (CaCl₂ and NaCl) affected the lithium leaching efficiency significantly. The conditional experiments indicated that the best mass ratio of lepidolite to NaCl to CaCl₂ was 1:0.6:0.4 with a chlorination roasting time of 30 min at a temperature 880 °C. A lithium leaching efficiency of 92.86% was achieved with these parameters.

El-Naggar and Medina (1984) [36] and El-Naggar et al. (1988) [37] showed that it is possible to extract lithium from spodumene via its reaction with mineral tachyhydrite (2MgCl₂.CaCl₂.12H₂O) to produce lithium chloride as the final product. The results indicated that the reactions that occur are as follows: 1. The decomposition of tachyhydrite, followed by MgC1₂ hydrolyzing to MgO; 2. The transformation of α-spodumene into β-spodumene; and 3. the reaction between β-spodumene, MgO, and CaCl₂ of tachyhydrite, according to the chemical reactions presented in Equations (1)–(3).

MgCl₂ + H₂O → MgO + 2HCl

(1)

α-Li₂O.Al₂O₃.4SiO₂ → β-Li₂O.Al₂O₃.4SiO₂ ∆G°: −10 KJ/mol at 1100 °C

(2)

β-Li₂O.Al₂O₃.4SiO₂ + CaCl₂ + 4MgO → 2LiCl + CaO.MgO.SiO₂ + 3SiO₂.2MgO + Al₂O₃.MgO

(3)

The thermodynamic effects of the extraction of lithium from mineral α-spodumene by CaCl₂ and MgCl₂ were determined within the temperature range of 25 to 1200 °C, and the results are presented in Figure 1.

The chlorination reaction with MgCl₂ + CaCl₂ was thermodynamically favorable in the whole temperature range examined (25–1200 °C) using a molar ratio of 1:1 (α-spodumene: 4MgO + CaCl₂). However, it is known that the phase transformation begins above 750 °C, and, at 925 °C, the α phase does not complete the transform to the β phase. Therefore, above 750 °C, there is a mixture of both polymorphs; consequently, lithium chloride (LiCl) could be obtained from both phases, which could increase the recovery of lithium. But, temperatures above 925 °C are necessary to obtain high lithium recovery rates [29]. According to the model generated by HSC software, the product of this reaction is LiCl, which is a solid or liquid product according to the temperature [16,27], calcium and magnesium silicate (CaO.MgO.SiO₂), magnesium aluminate (MgO.Al₂O₃), and silica (SiO₂). Based on these results, it is possible to infer that the reaction that occurs during calcination is as shown in Equation (4) [37].

$${\beta -\mathrm{L}\mathrm{i}}_2\mathrm{O}.{\mathrm{A}\mathrm{l}}_2{\mathrm{O}}_3.4\mathrm{S}\mathrm{i}{\mathrm{O}}_2+{\mathrm{C}\mathrm{a}\mathrm{C}\mathrm{l}}_2+4\mathrm{M}\mathrm{g}\mathrm{O}\to 2\mathrm{L}\mathrm{i}\mathrm{C}\mathrm{l}+\mathrm{C}\mathrm{a}\mathrm{O}.\mathrm{M}\mathrm{g}\mathrm{O}.{\mathrm{S}\mathrm{i}\mathrm{O}}_2+3{\mathrm{S}\mathrm{i}\mathrm{O}}_2.2\mathrm{M}\mathrm{g}\mathrm{O}+{\mathrm{A}\mathrm{l}}_2{\mathrm{O}}_3.\mathrm{M}\mathrm{g}\mathrm{O}$$

(4)

Our results corroborate those of a study of the chlorination of spodumene with tachyhydrite carried out by El-Naggar et al. (1988) [36] as well as the findings of Barbosa et al. (2015) [28].

Barbosa et al. (2014) [16] extracted lithium from β-spodumene with Cl₂ gas at temperatures of 1000 to 1100 °C in a fixed-bed reactor. The results of the isothermal chlorination roasting of β-spodumene showed that the sample underwent mass losses, which increased markedly with the increase in the chlorination temperature, which was due to the volatilization of both lithium chloride and the chlorides formed by the impurities present in the spodumene ore. Complete lithium extraction was achieved with a reaction time of 150 min.

According to Jena and Brocchi (1997) [33], CaCl₂ is the most effective chloridizing agent in the presence of silica. Barbosa et al. (2015) [28] also carried out experiments involving the chlorination of β-spodumene with CaCl₂, finding the optimal extraction conditions to be a temperature of 900 °C and 120 min of chlorination roasting, and they achieved a conversion degree of 90.2% and a LiCl production of 47.2 mg using a molar ratio 1:2 of β-LiAlSi₂O₆:CaCl₂. According to the authors, the reaction strengthens with increasing time and temperature, and the products of the chlorination reaction are lithium chloride, anorthite, and amorphous silica.

Here, we propose a preliminary process to extract lithium chloride from mineral spodumene from Brazil. In this process, lithium chloride is treated in its α-phase form with chlorination torrefaction using calcium chloride and/or magnesium chloride as the chlorinating agent. This study aimed to optimize the lithium extraction process, mainly by reducing energy consumption by joining two stages of the process. The products volatilized during the chlorination reaction were not studied in this phase of this project.

2. Materials and Methods

2.1. Procedures and Methods

The spodumene concentrate used in this study was supplied by Companhia Brasileira de Lítio, located in the state of Minas Gerais, Brazil. Initially, the spodumene was subjected to a grinding process in a ball mill, followed by sieving for particle size classification using Tyler series sieves. The grinding process was carried out until the particle diameter corresponded to the point where 80% of the distribution volume (P80) had a size smaller than 74 µm (200 mesh).

The tests for lithium chloride production were conducted on a bench scale. The methodology of Medina and El-Naggar (1984) [36] was modified by replacing the taquidrite ore (CaMg₂Cl₆·12H₂O) with a combination of salts: magnesium chloride hexahydrate (MgCl₂·6H₂O) with 99% purity (Sigma-Aldrich, Darmstadt, Germany) and calcium chloride dihydrate (CaCl₂·2H₂O) with 99% purity (P.A., Vetec, Speyer, Germany). The block diagram in Figure 2 illustrates the operational procedure adopted.

All chlorination tests were performed in triplicate and on a bench scale. In cases where the variation in the obtained results exceeded 5%, the tests were repeated to ensure the reproducibility of the experimental data.

After the homogenization stage, the spodumene and chloride mixtures were transferred to a porcelain crucible and subjected to heat treatment in a muffle furnace at a temperature of 1150 °C for a period ranging from 30 to 120 min. The clinker resulting from the thermal process was ground and leached with water at 95 °C. After leaching, the residue was sent for X-ray diffraction (XRD) analysis to identify the mineral phases, while the rich liquor was characterized by inductively coupled plasma–optical emission spectroscopy (ICP-OES) and X-ray fluorescence (XRF).

2.2. Experimental Conditions of Chlorination

The preliminary lithium extraction tests were conducted to evaluate three variables: the molar ratio of MgCl₂:CaCl₂ (tests 1–5), the mass ratio between spodumene and chlorides (tests 4–9), and the chlorination roasting time (tests 9–12), as presented in

Table 2. Operating conditions of the lithium extraction tests.

Test	Spodumene: Chlorides	Molar Ratio	Chlorination Roasting
		MgCl2:CaCl2	t (min)
1	1:8	0:1	120
2	1:8	1:0	120
3	1:8	1:2	120
4	1:8	1:1	120
5	1:8	2:1	120
6	1:4	1:1	120
7	1:4	2:1	120
8	1:6	1:1	120
9	1:6	2:1	120
10	1:6	2:1	90
11	1:6	2:1	60
12	1:6	2:1	30

Source: authors.

. The goal was to optimize lithium extraction by reducing reagent as well as energy consumption and, consequently, process costs.

In tests 1–5, the influence of the molar ratio between MgCl₂ and CaCl₂ was investigated, varying it from 0:1, 1:0, 1:2, 2:1, to 1:1, while the ratio between chlorides and spodumene was kept constant at 1:8, with a residence time of 120 min. In tests 4–9, the relationship between spodumene and chlorides was analyzed, adopting mass ratios of 1:8, 1:6, and 1:4, while keeping the chloride ratio (MgCl₂/CaCl₂) fixed at 2:1 and 1:1, and the residence time in the muffle furnace at 120 min.

To assess the influence of roasting time, tests 9–12 were conducted, for which the duration was adjusted to 30, 60, 90, and 120 min. The other parameters were kept constant, with a ratio of 2:1 for chlorides (MgCl₂/CaCl₂) and 1:6 for the spodumene:chloride ratio.

2.3. Experimental Conditions for Extraction of Lithium

The resulting clinker was crushed with an agate mortar and pestle and leached in water for 240 min at 95 °C. We used a solid–liquid ratio of 10% (w/w) for the aqueous leaching. At the end of the leaching, the liquor and residue were separated by filtration and the chemicals were identified. All the experiments were carried out in triplicate. Based on Fosu et al. (2022) [29], after the lithium extraction tests, the liquor products and residues were analyzed to evaluate the extraction according to Equation (5).

Li extraction (%) = 100.(lithium in pregnant liquor (g))/Lithium in pregnant liquor (g) + Lithium in residue (g)

(5)

2.4. Chemical and Physical Analyses

The experiment was monitored with lithium chemical analysis using inductively coupled plasma–optical emission spectroscopy (ICP-OES, Ultima 2, Horiba, Kyoto, Japan), X-ray fluorescence (XRF), flame emission spectroscopy (FES, 910, Analyser, Surrey, UK), atomic absorption spectrometry (AAS, SpectrAA 55B, Varian, Australia), and X-ray diffraction (XRD, Bruker-D4 Endeavor diffractometer, Billerica, MA, USA).

The chemical identification was carried out by inductively coupled plasma–optical emission spectroscopy (ICP-OES) using a Horiba Jobin Yvon spectrometer, with radial plasma observation used for direct determination. For quantitative chemical analysis by X-ray fluorescence (XRF), the samples were fused with lithium tetraborate (Li₂B₄O₇) at 1150 °C in a 1:10 ratio using a PANalytical EAGON 2 fusion machine, Malvern, UK). Lithium iodide (LiI) (0.1 g) was used as the releasing agent, and the samples were analyzed by wavelength dispersive X-ray fluorescence (WDXRF) using a PANalytical Axios Max instrument, Malvern, UK.

Mineral identification was performed using X-ray diffraction (XRD) with the powder method on a Bruker-D4 Endeavor diffractometer under the following operating conditions: CoKa radiation (40 kV/40 mA), a step size of 0.02° 2θ, a counting time of 184 s per step, and data collection from 5° to 105° 2θ. The qualitative spectrum interpretation was conducted by comparing the reference patterns from the PDF04 database using Diffrac.Plus software V.7

3. Results and Discussion

3.1. Characterization of Sample

The spodumene sample was analyzed to identify the compounds present, and the results are shown in

Table 3. Composition (wt/wt%) of the spodumene sample. Source: authors.

Component	Al2O3	Li2O	K2O	MgO	Mn2O3	Fe2O3	Na2O	P2O5	CaO
wt/wt%	23.26	5.67	1.40	0.17	0.14	1.07	0.79	0.65	0.04

. It was observed that Al₂O₃ (23.265%) and Li₂O (5.667%) were the predominant compounds, which was expected since spodumene is a lithium aluminosilicate. These results are similar to those presented by Krishan and Gopan (2024) [38], who found that the spodumene composition was Al₂O₃ (20.79%) and Li₂O (2.14%). Considering that the present study used a spodumene concentrate, it was expected that the concentrated lithium ore would have higher lithium percentages than those found in nature (Li₂O 1.9–3.3%) [2,39].

The X-ray diffraction analysis of the spodumene sample is shown in Figure 3. The X-ray diffraction indicated that the sample was initially composed of spodumene (S) and associated minerals, such as albite (NaAlSi₃O₈), quartz (SiO₂), and muscovite (KAl₂(Si,Al)₄O₁₀(OH)₂. This result is similar to that observed by Timich (2021) [40], who characterized a spodumene concentrate and identified the phases of spodumene, quartz, albite, and muscovite.

3.2. Tests of MgCl₂/CaCl₂ Molar Ratio

In the lithium extraction tests (Figure 4), the influence of the molar ratio of MgCl₂:CaCl₂ on the recovery of lithium contained in the spodumene was studied. The results (Figure 4) show that the presence of magnesium at a molar quantity greater than or equal to that of calcium (ratios of 2:1 or 1:1) afforded high extraction rates (>96%) of the lithium present in the mineral. This result is in accordance with the reaction in Equation (5), in which 4 mol of MgO to 1 mol of CaCl₂ is necessary. According to El-Naggar et al. (1988) [37], this is due to the reaction of β-spodumene with MgO (formed from the hydrolysis of MgCl₂, which occurs in the temperature range between 400 and 700 °C) and to the stability of the calcium chloride at elevated temperatures as well as the possible volatilization of this reagent. The tests in which a single chloride species (MgCl₂ or CaCl₂) was used, low lithium extraction values were observed.

The X-ray diffraction analysis of the clinker produced in the test with a spodumene:chloride mass ratio of 1:8 and a MgCl₂:CaCl₂ molar ratio of 2:1 showed the presence of the minerals forsterite (Mg₂SiO₄), spinel (MgAl₂O₄), cordierite (Mg₂Al₄Si₅O₁₈), tachyhydrite (CaMg₂Cl₆·12H₂O), and sinjarite (CaCl₂.2H₂O) in the leaching residue (Figure 5). The presence of β-spodumene was not observed. This indicated its transformation into other species.

3.3. Studies on Spodumene:Chloride Mass Ratio

The objective of the complementary tests was to determine the influence of the spodumene:chloride mass ratio on the recovery of lithium. Figure 6 shows that it was possible to maintain high lithium extraction (97.46%) with a spodumene:chloride mass ratio of 1:6 and a MgCl₂:CaCl₂ molar ratio of 2:1. For the test with a spodumene:chloride mass ratio of 1:4 and a MgCl₂:CaCl₂ molar ratio of 2:1, there was a significant reduction in the lithium extraction (34.72%), probably due to the lower quantity of calcium chloride available for the reaction, even in the presence of magnesium. In this case, there was insufficient chloride in the reaction to form lithium chloride (LiCl) according to the stoichiometry presented in Equation (5).

For lithium extraction with a spodumene:chloride mass ratio of 1:8 and a MgCl₂:CaCl₂ molar ratio of 2:1, the X-ray diffractogram in Figure 7 shows the presence of periclase minerals (MgO), forsterite (Mg₂SiO₄), spinel (MgAl₂O₄), cordierite (Mg₂Al₄Si₅O₁₈), brucite (Mg(OH)₂), and calcium aluminum oxide chloride ((Al(H₂O)₆)Cl₃). The mineral species tachhydrite (CaMg₂Cl₆·12H₂O) and sinjarite (CaCl₂.2H₂O) were not identified in the results that are presented in Figure 6. Therefore, it was inferred that these soluble minerals were leached.

3.4. Studies on Roasting Chlorination Time

Figure 8 shows the results of the roasting tests (1150 °C) with different heating durations. It was observed that with increasing time, the extraction of lithium also increased [10,16,27]. However, after 30 min, lithium extraction was 96.89%, and after 60 and 90 min, it was 97.52% and 97.42%, respectively. The 0.63% increase in lithium extraction may not justify energy costs for a duration of 60 min. These tests were carried out with a spodumene:chloride mass ratio of 1:6 and a MgCl₂:CaCl₂ molar ratio of 2:1.

4. Conclusions

According to the model created in HSC software, it was possible to determine that the reaction products between the spodumene and chlorides (MgCl₂:CaCl₂) were lithium chloride (LiCl), calcium and magnesium silicate (CaO.MgO.SiO₂), magnesium aluminate (MgO.Al₂O₃), and silicon dioxide (SiO₂).

In the XRD diffraction pattern of the chlorination roasting product (clinker), α- or β-spodumene was not observed, indicating the transformation into other species. LiCl had an amorphous form; therefore, it was not detected by XRD. LiCl was detected by AAS in the aqueous leach liquor.

Based on the preliminary tests, the lithium extraction rates were greater than 96%, with a mixture of chlorides, proving that magnesium was present in a molar ratio greater than or equal to that of calcium.

The complementary tests showed that it was possible to optimize the process by using a spodumene:chloride mass ratio of 1:6 and a MgCl₂:CaCl₂ molar ratio of 2:1 for a 30 min duration at a roasting temperature of 1150 °C, achieving lithium extraction of higher than 95%.