Facile Synthesis of High Purity Li₂S by Titanothermic Reduction

Abstract

Lithium sulfide (Li₂S) is indispensable for lithium–sulfur batteries and sulfide solid-state batteries. However, its high preparation cost and strict process conditions represent core bottlenecks restricting large-scale commercial application. To address this issue, a novel process featuring a low-cost, high-safety, and controllable reaction is proposed in this work. Compared with the commercial H₂S-based route for Li₂S production, the developed process presents distinct advantages, including accessible raw materials, high safety, low overall cost, and low environmental load. Using Li₂SO₄·H₂O as the raw material and Ti as the reducing agent, high-purity T-Li₂S (>99.9%) is successfully synthesized via solid-state sintering and purification, yielding a higher purity level than that of commercial C-Li₂S (>99.7%). Furthermore, sulfide all-solid-state electrolytes T-Li_5.3PS_4.3ClBr_0.7 and C-Li_5.3PS_4.3ClBr_0.7 are prepared using the as-obtained T-Li₂S and commercial C-Li₂S as precursors, respectively. The room-temperature Li-ion conductivities are determined to be 14.5 mS/cm and 11.0 mS/cm, revealing faster ion migration and efficient ion transport in T-Li_5.3PS_4.3ClBr_0.7 without high-temperature assistance, which fully validates the feasibility of the proposed strategy. Overall, this work provides a new technical route for the preparation of high-purity Li₂S, showing promising application prospects.

1. Introduction

The rapid development of digital consumer batteries, electric vehicles, and large-scale energy storage has posed higher requirements for the energy density and safety of batteries [1,2,3,4,5,6,7]. Lithium–sulfur batteries (including solid-state lithium–sulfur batteries) and sulfide all-solid-state batteries represent two advanced battery systems that offer higher specific energy and enhanced safety compared to conventional lithium-ion batteries, demonstrating significant market potential [8,9,10,11]. Lithium sulfide (Li₂S) serves as the critical material for these cutting-edge battery systems. In lithium–sulfur batteries, Li₂S functions as the cathode material with an exceptionally high theoretical mass energy density, which is five to six times that of conventional lithium batteries. When paired with non-lithium anodes such as graphite or silicon carbon, this design fundamentally eliminates the risks of dendrite formation and explosion associated with active metallic lithium anodes [12,13]. Furthermore, Li₂S is the essential raw material for synthesizing sulfide solid-state electrolytes, which exhibit ion conductivity comparable to or surpassing liquid electrolytes, enabling rapid charging and discharging. By replacing the flammable electrolyte in traditional lithium batteries with non-flammable solid-state electrolytes, sulfide-based systems completely resolve thermal runaway and fire hazards [14,15,16].

However, the current high market price of high-purity Li₂S (2–5 million RMB/ton) has become a major obstacle to the large-scale application of all-solid-state batteries [17,18]. Driven by the booming new energy sector and advanced manufacturing, the demand for high-purity Li₂S with low impurity content is growing rapidly. Consequently, exploring safe, efficient, and cost-effective synthetic routes for Li₂S has emerged as a key research focus in the battery field [19,20,21].

Currently, the synthetic strategies for Li₂S primarily include direct sulfurization of lithium metal, liquid-phase reactions, and carbothermal reduction. Although the direct reaction between elemental lithium and sulfur can yield Li₂S in a straightforward manner, metallic lithium is highly chemically reactive and readily undergoes violent reactions with air and moisture. Consequently, the preparation must be performed under stringent inert or vacuum conditions, leading to high operational risks and prohibitive production costs [22,23]. Liquid-phase approaches typically use lithium carbonate (Li₂CO₃), lithium hydroxide (LiOH), lithium oxide (Li₂O), or lithium hydride (LiH) as precursors, which react with toxic gases such as hydrogen sulfide (H₂S) and carbon disulfide (CS₂) in organic media. However, these routes suffer from the strong corrosivity of raw materials, challenging exhaust-gas treatment, and significant environmental burdens [24,25]. In contrast, the carbothermal reduction method relies on lithium sulfate (Li₂SO₄) as the lithium source, which contains sulfur in a high oxidation state. This induces complex side reactions during high-temperature reduction, resulting in poor controllability over product purity and low overall conversion efficiency [26]. Owing to the flammable and explosive nature of hydrogen at high temperatures, a H₂-based reduction method also presents significant safety challenges [27]. Collectively, these technical bottlenecks severely impede the scalable production of Li₂S. Therefore, the development of safe, stable, and precisely controllable synthetic routes is critically important for accelerating the industrial application of Li₂S.

For a long time, titanium has been recognized as a structurally stable metal whose excellent properties are only manifested under strongly corrosive or high-temperature oxidizing conditions. Its industrial applications have been predominantly focused on structural materials (such as aerospace alloys) [28] or catalytic materials (such as TiO₂ photocatalysts) [29], while it has rarely been considered as a reducing agent, and its high-temperature reducibility remains underexplored.

In this study, the high-temperature reducibility of titanium was exploited to precisely reduce lithium sulfate (Li₂SO₄), successfully yielding the target product Li₂S and the by-product dilithium titanate (Li₂TiO₃). The specific chemical reaction involved is expressed as follows:

4Ti + 5Li₂SO₄ → Li₂S + 4Li₂TiO₃ + 4SO₂↑

Notably, no other elements participate in the above reaction process. After the reaction is completed, the solid products exclusively include Li₂S and Li₂TiO₃, which can be separated by means of dissolution. Subsequent removal of the solvent enables the acquisition of high-purity and low-carbon Li₂S product. In addition, Li₂TiO₃ can be further utilized as an anode material for lithium-ion batteries or as a tritium breeder in nuclear fusion reactors. Meanwhile, the single gaseous product SO₂ can be recycled into concentrated sulfuric acid or sulfur through existing mature processes, realizing the cyclic utilization of all products and achieving a multi-benefit effect. Moreover, the core raw materials are easily accessible, non-toxic, and low-cost, with unstringent requirements for storage environments. This study thus provides a novel technical pathway for the industrial-scale preparation of Li₂S.

2. Experimental Section

Synthesis of Li₂S: 55 g of anhydrous lithium sulfate (Li₂SO₄, >99%, Aladdin, Shanghai, China; prepared by vacuum drying lithium sulfate monohydrate Li₂SO₄·H₂O at 200 °C for 8 h) and 19.2 g of titanium powder (Ti, >99.5%, Aladdin) were added into a ball milling jar and mixed at 350 rpm for 1 h under argon protection. The mixture was collected in a glove box and transferred into a corundum crucible, followed by thermal reduction in a tube furnace under argon atmosphere at 800 °C for 3 h to obtain crude Li₂S. The crude Li₂S was dissolved in 400 mL of anhydrous ethanol under stirring. After filtration, the ethanol solution of Li₂S was collected, and the filter residue was dried to obtain Li₂TiO₃. High-purity Li₂S precursor was obtained by spray drying the Li₂S ethanol solution. The precursor was then heated stepwise in a nitrogen atmosphere furnace to produce high-purity Li₂S powder. The stepwise heating procedure was as follows: first heated to 150 °C and held for 5 h, then further heated to 800 °C and maintained for 3 h. A schematic illustration of the corresponding Li₂S synthesis process is shown in Figure 1.

Characterization of Li₂S: The phase composition of the samples was characterized by X-ray diffraction (XRD, Smartlab, Rigaku, Tokyo, Japan) with a Cu target radiation source at a scanning rate of 3 °/min. The microstructure of the powder was observed by field emission scanning electron microscopy (SEM, Sigma 300, Carl Zeiss, Oberkochen, Germany). In addition, the residual carbon content in Li₂S was determined by a high-frequency infrared carbon-sulfur analyzer (HCS-140, Dekai Instruments, Shanghai, China). The contents of trace metallic impurity elements were measured by inductively coupled plasma optical emission spectrometry (ICP-OES, Avio 200, PerkinElmer, Waltham, MA, USA). Particle size distribution was determined by a Malvern Mastersizer 3000(Malvern Panalytical, Malvern, UK) particle size analyzer (UK). Specific surface area was measured by a JW-BK222 nitrogen adsorption BET (JWGB, Beijing, China) surface area analyzer.

Preparation and electrochemical performance characterization of electrolyte materials: The purified Li₂S, LiCl, P₂S₅, and LiBr were weighed according to the stoichiometry of Li_5.3PS_4.3ClBr_0.7 (molar ratio of Li₂S: LiCl: P₂S₅: LiBr = 1.8: 1: 0.5: 0.7), and a total of 10 g of raw materials was used for ball milling amorphization. The ball milling was performed at 500 rpm for 20 h with a mass ratio of zirconia balls to powder of 20:1. The ball-milled powder was sintered at 450 °C for 10 h under argon atmosphere. After grinding, the as-obtained powder was packed into a stainless steel testing mold for cold pressing and electrochemical measurement under a pressure of 400 MPa. The lithium-ion conductivity was measured using an electrochemical workstation (PMC-200, Ametek-Princeton, Oak Ridge, TN, USA) with a DC bias of 20 mV and a frequency range of 1 Hz to 1 MHz.

3. Results and Discussion

To investigate the phase composition of products at various stages during the preparation of high-purity Li₂S via the titanothermic reduction, X-ray diffraction (XRD) was employed. Figure 2a,b show the XRD patterns of the raw materials Li₂SO₄ and Ti, respectively. Both present pure phases without impurities, confirming the successful dehydration of Li₂SO₄·H₂O. In the XRD pattern of the mixed product after titanothermic reduction (Figure 2c), diffraction peaks located at 27.1°, 31.3°, 44.7°, and 53.1° are assigned to the (111), (200), (220), and (331) crystal planes of Li₂S, respectively [30]. Meanwhile, diffraction peaks at 18.5°, 35.9°, 43.6°, and 63.2° correspond to the (002), (−131), (−133), and (312) crystal planes of Li₂TiO₃, respectively [31]. The hump at the low angle region originates from the high-temperature adhesive tape used for sealing. After separation and purification, the target product high-purity Li₂S was obtained. Its XRD pattern is displayed in Figure 2d, in which no impurity peaks can be observed. All diffraction peaks match well with those of Li₂S in Figure 2c, verifying the excellent purification effect.

The micro-morphologies of the raw materials and products were characterized by scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS).

Figure 3a,b show the SEM images of the raw materials Li₂SO₄ and Ti, respectively. It can be seen that both raw materials consist of small particles with sizes ranging from several micrometers to tens of micrometers. The micro-morphology of the mixture after the titanothermic reduction is shown in Figure 3c. After high-temperature sintering, the agglomeration of the sample is further enhanced, the particle size is increased, and slight agglomeration can be observed. Figure 3d displays the SEM image of the target product Li₂S, which exhibits a similar morphology to Li₂SO₄, presenting microscale spheres formed by the accumulation of fine particles. Figure 3e–h show the SEM image and corresponding EDS elemental mappings of the mixture after the titanothermic reduction. The Ti, O, and S elements exhibit a high degree of spatial overlap, indicating that Li₂S and Li₂TiO₃ are closely distributed, and thus further purification is required to obtain pure Li₂S.

To evaluate the purity of Li₂S prepared by titanothermic reduction, multiple characterization methods were employed in this work. ICP results for metallic impurity elements and carbon content of titanothermic-reduced Li₂S (T-Li₂S) and commercial Li₂S (C-Li₂S) are summarized. As shown in

Table 1. Results of trace metallic impurity elements by ICP and carbon content by carbon-sulfur analyzer.

Content of Main Impurities /ppm							Carbon Content /wt%
Element	Mg	Cu	Fe	Ni	Al	Ca	C
T-Li2S	68	26	18	5	397	67	0.15
C-Li2S	38	28	19	6	383	168	0.04

, the target product T-Li₂S exhibits a carbon content of only 0.04 wt% and a total metallic impurity content of 581 ppm, corresponding to a purity higher than 99.9%. In contrast, commercial C-Li₂S has a carbon content of 0.15 wt% and a total metallic impurity content of 642 ppm, with a purity above 99.7%, which is lower than that of T-Li₂S. This indicates that the preparation method adopted in this study can effectively control the impurity content of Li₂S at a low level.

To further verify the high purity of Li₂S synthesized via titanothermic reduction, X-ray photoelectron spectroscopy (XPS) was performed on T-Li₂S and C-Li₂S. The binding energy of XPS peaks is closely related to the valence state of the corresponding elements, and higher valence states generally correspond to higher binding energies. As displayed in Figure 4, the S 2p spectra of T-Li₂S and C-Li₂S within 158.0–170.0 eV can be well fitted into two doublets. The lower energy doublet located at 2p_3/2 (160.2 eV) and 2p_1/2 (161.4 eV) is assigned to S^2- in Li₂S. The doublet at slightly higher binding energy (2p_3/2: 161.8 eV; 2p_1/2: 162.9 eV) corresponds to S⁻¹ in Li₂S₂ [32]. No additional peaks associated with high valence sulfur species (such as SO₄²⁻and SO₃²⁻) are observed at higher binding energies for either sample, indicating the absence of such oxide impurities. Quantitative analysis based on peak area ratios reveals that T-Li₂S is composed of 98.5 mol% Li₂S and 1.5 mol% Li₂S₂, whereas C-Li₂S consists of 88.3 mol% Li₂S and 11.7 mol% Li₂S₂. The significantly higher fraction of the target Li₂S phase in T-Li₂S further confirms its superior purity.

Compared with the previously reported metallothermic reduction strategies, such as magnesothermal reduction [33] and aluminothermal reduction [26], both strategies induce small-scale explosions during the high-temperature heating of reactants, presenting significant challenges in operational safety and process control. Additionally, the target product Li₂S obtained via these two reduction routes contains trace impurities such as Li₂SO₄ or LiAlO₂, respectively, while the by-products (MgO and Al₂O₃) exhibit low application value and limited reusability. In contrast, the titanothermic reduction route proposed in this study features a milder reaction process, free from uncontrollable phenomena such as explosions. XRD and XPS characterizations confirm that the target product Li₂S is free of any impurities, with a lower content of Li₂S₂ and higher purity. Moreover, the reaction by-product (Li₂TiO₃) can be utilized with high added value.

Overall, the titanothermic reduction route proposed in this study exhibits mild reaction conditions, easy separation and reuse of by-products, high purity of the target product Li₂S, and excellent operational stability and scalability. It effectively addresses the safety hazards and impurity issues associated with traditional metallothermic reduction strategies, demonstrating superior practical application value and promotion potential.

As depicted in Figure 5 (SEM images of T-Li₂S and C-Li₂S), both samples show a similar particulate morphology with particle sizes in the range of 3–5 μm. Further particle size distribution analysis reveals that T-Li₂S exhibits a D10 of 0.53 μm, D50 of 3.1 μm, and D90 of 5.9 μm, whereas C-Li₂S shows a D10 of 0.48 μm, D50 of 2.6 μm, and D90 of 5.3 μm, indicating a negligible difference in particle size distribution between the two materials. Moreover, BET-specific surface area measurements yield comparable values for T-Li₂S (2.8 m²/g) and C-Li₂S (3.0 m²/g). These results collectively confirm that T-Li₂S and C-Li₂S possess highly similar morphological and surface characteristics.

To verify the feasibility of replacing commercial Li₂S with the as-prepared material in practical applications, T-Li₂S was employed as the raw material to synthesize Li_5.3PS_4.3ClBr_0.7 sulfide-based all-solid-state electrolytes (ASSSEs) following an identical synthetic procedure. For comparison, electrolytes prepared from T-Li₂S and commercial C-Li₂S are denoted as T-Li_5.3PS_4.3ClBr_0.7 and C-Li_5.3PS_4.3ClBr_0.7, respectively. As displayed in Figure 6a, all diffraction peaks of both electrolytes can be well indexed to the standard PDF card, confirming the successful preparation of phase-pure argyrodite-type sulfide electrolytes.

Furthermore, electrochemical impedance spectroscopy (EIS) was performed at room temperature to analyze the ionic transport properties, and the corresponding Nyquist plots are presented in Figure 6b. The Nyquist plots show a straight line over the entire frequency range, demonstrating that lithium-ion diffusion dominates the overall transport process. Quantitative calculations reveal that the ionic conductivity of T-Li_5.3PS_4.3ClBr_0.7 reaches 14.5 mS/cm, while that of C-Li_5.3PS_4.3ClBr_0.7 is 11.0 mS/cm. Both values are remarkably high and represent a substantial improvement. Given the negligible differences in morphology and surface properties between T-Li₂S and C-Li₂S, the superior ionic conductivity of the T- Li_5.3PS_4.3ClBr_0.7 electrolyte can be reasonably attributed to the higher purity of T-Li₂S relative to commercial C-Li₂S.

As shown in Table 2, which summarizes representative solid electrolytes reported in recent years and their lithium-ion conductivities, the as-synthesized T-Li_5.3PS_4.3ClBr_0.7 in this work exhibits outstanding lithium-ion transport performance.

Overall, these characterizations confirm that T-Li₂S prepared via the titanothermic reduction method exhibits superior comprehensive performance to commercial C-Li₂S, and the as-prepared sulfide solid-state electrolyte achieves significantly improved lithium-ion conductivity compared with previous studies, fully meeting the stringent requirements for the commercialization of high-performance sulfide-based all-solid-state battery materials. The high lithium-ion conductivity of bromine-doped argyrodite originates from lattice expansion, increased lithium vacancies, disordered anion framework, and reduced grain-boundary resistance. The Cl/Br co-substitution establishes a disordered structure to accelerate Li-ion diffusion, while lithium defects provide sufficient charge carriers. These factors collectively contribute to superionic conduction [34].

Table 2. Comparison of lithium-ion conductivities of different types of solid electrolytes.

Material	Electrolyte Type	Lithium-Ion Conductivity /mS/cm	Ref.
Li6.4La3Zr1.4Ta0.6O12	Oxide Solid Electrolyte	0.662	[35]
KBZ-PEO	Polymer Solid Electrolyte	0.140	[36]
sc-HfO2@LCB	Composite Solid Electrolyte	1.23	[37]
Li3YBr5.7F0.3	Halide Solid Electrolyte	1.80	[38]
Li5.3PS4.3ClBr0.7	Sulfide Solid Electrolyte	5.20	[39]
T-Li5.3PS4.3ClBr0.7	Sulfide Solid Electrolyte	14.5	This Work

Table 2. Comparison of lithium-ion conductivities of different types of solid electrolytes.

Material	Electrolyte Type	Lithium-Ion Conductivity /mS/cm	Ref.
Li6.4La3Zr1.4Ta0.6O12	Oxide Solid Electrolyte	0.662	[35]
KBZ-PEO	Polymer Solid Electrolyte	0.140	[36]
sc-HfO2@LCB	Composite Solid Electrolyte	1.23	[37]
Li3YBr5.7F0.3	Halide Solid Electrolyte	1.80	[38]
Li5.3PS4.3ClBr0.7	Sulfide Solid Electrolyte	5.20	[39]
T-Li5.3PS4.3ClBr0.7	Sulfide Solid Electrolyte	14.5	This Work

Although the titanothermic reduction method enables low-cost and safe synthesis of high-purity Li₂S, it has certain limitations. The stability of Li₂S during large-scale continuous production has not been fully validated, and the recovery and utilization of the by-product Li₂TiO₃ remain underexplored. Nevertheless, these issues are beyond the scope of the present study. Future work will focus on scaling up T-Li₂S production, verifying process stability, and developing efficient strategies for Li₂TiO₃ recovery and reuse.

4. Conclusions

In this study, high-purity Li₂S with low carbon content and low metallic impurities was successfully prepared via solid-state sintering and purification, using Li₂SO₄·H₂O as the lithium and sulfur source, and Ti as the reducing agent. Compared with the H₂S-based preparation method [40], the titanothermic reduction process exhibits distinct advantages, including non-toxic and readily available raw materials, simple process flow, high product purity, and remarkable environmental benefits. Li₂TiO₃, the by-product of solid-state sintering, can be efficiently separated from the target product Li₂S and possesses multiple application values. It can not only serve as an anode material for lithium-ion batteries but also be further resourcefully utilized as a tritium breeder in nuclear fusion reactors. Meanwhile, SO₂, the single gaseous product generated from the reaction, can be recovered and converted into concentrated sulfuric acid or sulfur by relying on the existing mature processes.

The above process realizes the recycling of all reaction products, achieving multiple benefits and reducing the overall production cost. Furthermore, the as-prepared T-Li₂S exhibits higher purity and lower contents of metallic impurities and residual carbon than commercial C-Li₂S. The electrolyte synthesized from T-Li₂S delivers superior lithium-ion conductivity (14.5 mS/cm), which is attributed to the disordered structure induced by Cl/Br co-substitution that promotes lithium-ion diffusion. Overall, this study provides a novel technical route for the green production of Li₂S, with broad application prospects.