A highly conductive and antioxidative MoO₂-doped Li argyrodite electrolyte for all-solid-state Li batteries

Abstract

A MoO₂-doped Li_5.5PS_4.5Cl_1.5 solid electrolyte with ionic conductivity of 12 mS cm⁻¹ and an electrochemical window of 4.3 V vs. Li/Li⁺ was prepared, which enables a LiNi_0.8Co_0.1Mn_0.1O₂-based full cell to deliver a specific capacity of 194 mA h g⁻¹ at 0.1C and retain 80% capacity after 3500 cycles at 1C.

---

The rapid development of electric vehicles has aroused demand for high-energy-density, high-safety all-solid-state Li batteries (ASSLBs).¹ These batteries are designed by employing safer solid electrolytes (SEs) to replace hazardous liquid electrolytes and separators found in traditional Li-ion batteries.² As key components, SEs are categorized into polymer-, oxide-, sulfide-, and halide-based electrolytes.³ Among these, sulfide-based SEs (SSEs) have garnered the most attention due to their particularly high ionic conductivity (σ_i) and favorable mechanical properties, such as high elasticity and ductility. These characteristics offer the potential to improve both the safety and the energy-power performance of ASSLBs.^4,5

Of the various SSEs, Li argyrodite of Li₆PS₅Cl is the most widely used due to its high σ_i (>1 mS cm⁻¹), excellent ductility (low grain boundary resistance), and cost-effective raw materials that avoid rare elements.⁶ However, Li₆PS₅Cl suffers from poor electrochemical stability and requires enhanced σ_i.⁷ Substituting P, S, and Cl sites in Li₆PS₅Cl with other elements can modify the Li sublattice and increase S²⁻/Cl⁻ site disorder, facilitating Li⁺ diffusion and improving σ_i.^8,9 Additionally, doping elements such as In, Sn, Mg, F, and I can aid in the formation of a favorable solid electrolyte interphase (SEI), stabilizing the Li metal anode.^7,10–12 Despite these advances, little research has been dedicated to improving the oxidation stability of Li₆PS₅Cl for compatibility with uncoated high-voltage cathodes.

In this study, we introduce dual-functional MoO₂ into the crystal lattice of Cl-rich Li_5.5PS_4.5Cl_1.5 SSE by substituting a portion of P₂S₅ raw material with MoO₂ (1–3%). The optimized Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 SSE exhibits higher σ_i (12.0 vs. 9.1 mS cm⁻¹) and a wider electrochemical stability window (4.3 vs. 3.1 V vs. Li/Li⁺) compared to the Li_5.5PS_4.5Cl_1.5 matrix. As a result, a full cell using this modified SSE and an uncoated single-crystal LiNi_0.8Co_0.1Mn_0.1O₂ (NCM811) cathode achieves a high specific capacity (194 mA h g⁻¹ at 0.1C) and an ultralong cycling life (80% capacity retention after 3500 cycles at 1C). In situ Raman spectroscopy and ex situ X-ray photoelectron spectroscopy (XPS) confirm the excellent oxidation resistance of the MoO₂-doped Li argyrodite.

A series of MoO₂-doped Li_5.5+xP_1−xMo_xS_4.5−2xO_2xCl_1.5 (x = 0.01, 0.02, 0.03) and control Li_5.5PS_4.5Cl_1.5 SSEs were synthesized through a solid-state process involving initial ball milling, followed by cold pressing, and finally annealing. The prepared Li_5.5+xP_1−xMo_xS_4.5−2xO_2xCl_1.5 SSEs exhibit an aggregation morphology with primary particle sizes ranging from a few microns to ten microns, with all elements distributed homogeneously. The MoO₂ doping does not alter the morphology and elemental distribution of the SSEs (Fig. S1–S5, ESI†). The crystalline structures of the synthesized SSEs were analyzed using X-ray diffraction (XRD). As shown in Fig. 1a, all samples exhibit similar XRD patterns that align with Li₇PS₆ (PDF#34-0688), indicating a cubic F4̄3m space group. As the MoO₂ content increases, impurity peaks from LiCl and Li₃PO₄ become more pronounced due to the limited solubility of Mo and O dopants. Additionally, the main peaks in the XRD patterns of Li_5.5+xP_1−xMo_xS_4.5−2xO_2xCl_1.5 shift slightly to smaller 2θ angles as the dopant concentration increases (Fig. 1b), indicating unit cell expansion.¹⁰ This expansion can be attributed to the substitution of smaller P⁵⁺ ions with larger Mo⁴⁺ ions (34 vs. 65 pm). The doping of lower-valence Mo⁴⁺ also increases the Li⁺ concentration within the lattice, improving σ_i.⁷

Fig. 1 (a) and (b) XRD patterns of Li_5.5+xP_1−xMo_xS_4.5−2xO_2xCl_1.5 (x = 0, 0.01, 0.02, 0.03). Rietveld XRD pattern refinement (c) and crystal structural diagram (d) of Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5.

To further evaluate the structural changes in the SSE following MoO₂ incorporation, Rietveld refinement of the XRD pattern of Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 was performed (Fig. 1c and Table S1, ESI†). The calculated pattern matches the experimental data, confirming the successful doping of MoO₂, forming a cubic argyrodite-type Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 SSE with the F4̄3m space group. Fig. 1d illustrates the crystalline structure of the MoO₂-doped SSE, where Mo atoms substitute P atoms at the 4b site, and O atoms replace S atoms at the 16e site.¹⁰ Furthermore, the ⁷Li magic angle spinning nuclear magnetic resonance spectra for both Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 and Li_5.5PS_4.5Cl_1.5 show a single peak without a shift (Fig. S6, ESI†), indicating similar Li chemical environments in both samples. This further confirms the successful substitution of P and S atoms by Mo and O atoms, consistent with the XRD results.

The chemical composition of the synthesized Li_5.5+xP_1−xMo_xS_4.5−2xO_2xCl_1.5 (x = 0, 0.01) SSEs was examined using XPS. The P 2p XPS spectra (Fig. S7a, ESI†) show characteristic peaks for PS₄³⁻ at 131.8 and 132.7 eV in both samples.¹³ Additionally, a peak at 133.7 eV corresponds to PO₄³⁻, likely due to minor oxidation of PS₄³⁻.¹⁴ The S 2p (Fig. S7b, ESI†) and Mo 3d (Fig. S7c, ESI†) XPS spectra of the Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 SSE confirm the presence of Mo–S bonds and the absence of Mo–O bonds,¹⁵ further verifying successful doping of Mo and O atoms into the lattice, replacing P and S atoms, respectively. Raman spectra of all prepared SSEs show characteristic peaks for PS₄³⁻ at 198, 265, 428, 576, and 598 cm⁻¹.¹⁶ Additionally, the Mo–S peak at 490 cm⁻¹ is observed in the MoO₂-doped SSEs, with its intensity increasing as the MoO₂ content rises (Fig. S8, ESI†).¹⁵

The σ_i of the synthesized Li_5.5+xP_1−xMo_xS_4.5−2xO_2xCl_1.5 (x = 0, 0.01, 0.02, 0.03) SSEs was measured using electrochemical impedance spectroscopy (EIS). The resulting Nyquist plots and corresponding Arrhenius curves are displayed in Fig. S9 (ESI†) and Fig. 2a, respectively. Based on the calculations, the room temperature (RT, 25 °C) σ_i and activation energy (E_a) of all the SSEs were determined (Fig. 2b). The RT σ_i of the Li_5.5PS_4.5Cl_1.5 matrix is calculated to be 9.1 mS cm⁻¹, consistent with previously reported values.¹⁷ Notably, doping with 1% MoO₂ significantly enhances the RT σ_i of the SSE to 12.0 mS cm⁻¹. To the best of our knowledge, this is the highest reported RT σ_i, surpassing those of previously reported O-doped Li argyrodites (Fig. S10 and Table S2, ESI†), as O doping typically has a negative impact on the σ_i of SSEs.¹¹ The enhanced RT σ_i of the Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 SSE can be attributed to (i) the inherently high σ_i of the Li_5.5PS_4.5Cl_1.5 matrix, and (ii) moderate Mo doping, which increases Li⁺ concentration and expands Li⁺ transport pathways.¹⁰ However, with higher MoO₂ doping levels, the detrimental effects of O doping become predominant, resulting in a decrease in RT σ_i. The E_a of the SSEs exhibits an inverse relationship with their RT σ_i, with the Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 SSE showing the lowest E_a of 0.28 eV, indicating the fastest Li⁺ transfer kinetics.

Fig. 2 Electrochemical properties of Li_5.5+xP_1−xMo_xS_4.5−2xO_2xCl_1.5 (x = 0, 0.01, 0.02, 0.03). (a) Arrhenius curves, (b) RT σ_i and E_a, (c) σ_e, and (d) ESWs.

The electronic conductivity (σ_e) of SEs is typically used to assess their ability to suppress the growth of Li/Li-alloy dendrites.⁷ As shown in Fig. S11 (ESI†) and Fig. 2c, the σ_e of the Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 SSE is calculated to be 1.27 × 10⁻⁸ S cm⁻¹, which is significantly lower than that of the Li_5.5PS_4.5Cl_1.5 matrix (4.33 × 10⁻⁸ S cm⁻¹). This suggests that the MoO₂-doped SSE forms a more stable interface with Li/Li-alloy anodes, which also can be clarified in the Li|Li (Fig. S12, ESI†) and Li|Cu cells (Fig. S13, ESI†). The electrochemical stability window (ESW), which reflects the oxidative stability of SEs, was measured using linear sweep voltammetry (LSV). As shown in Fig. 2d, the MoO₂-doped SSEs exhibit a much wider ESW compared to Li_5.5PS_4.5Cl_1.5 SSE (4.3 vs. 3.1 V vs. Li/Li⁺), indicating that the MoO₂-doped SSEs are more stable when paired with high-voltage Li-layered oxide cathodes.

The performance of the synthesized SSEs was further evaluated in Li–In|NCM811 full cells, where the SSEs functioned as both the separator and a component of the composite cathode. All tests were conducted at RT under 35 MPa, with a cathode active material (CAM) loading of 5.6 mg cm⁻². Fig. 3a presents the initial voltage profiles of the full cells using the four different SSEs at 0.1C (1C = 200 mA g⁻¹). Notably, the cell using the Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 SSE delivers the highest specific capacity of 194 mA h g⁻¹ with a coulombic efficiency (CE) of 80.5%. In comparison, the cell with the Li_5.5PS_4.5Cl_1.5 SSE achieves an initial specific capacity of 173 mA h g⁻¹ and a CE of 75.5%. At a higher rate of 1C, the cell with 1% MoO₂-doped SSE still exhibits the highest specific capacity of 134 mA h g⁻¹ (Fig. 3b).

Fig. 3 Electrochemical performance of Li–In|NCM811 full cells with the as-prepared SSEs. (a) Voltage profiles at 0.1C, (b) voltage profiles at 1C, (c) performance comparison, (d) rate capability, and (e) long-term cycling stability.

The long-term electrochemical performance of the cells with the four SSEs was further assessed through repeated galvanostatic charge–discharge cycling (Fig. 3c). All cells demonstrate high stability over 200 cycles. Among them, the cell with 1% MoO₂-doped SSE maintains the highest specific capacity, followed by the cell with 2% MoO₂-doped SSE. The undoped SSE-based cell comes next, while the cell with 3% MoO₂-doped SSE shows the lowest specific capacity. This suggests a positive correlation between the electrochemical performance of the full cells and the RT σ_i of the adopted SSEs.

The rate capability of the full cells with the synthesized SSEs was also evaluated (Fig. 3d and Fig. S14, ESI†). The cell with Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 SSE demonstrates the best rate performance, achieving specific capacities of 194, 181, 160, 134, 103, and 78 mA h g⁻¹ at rates of 0.1C, 0.2C, 0.5C, 1C, 2C, and 3C, respectively. Additionally, the specific capacity recovered when the rate returns to 0.1C. Long-term cycling stability was also compared for the full cells using the four SSEs. As shown in Fig. 3e, the cell with Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 SSE operates for over 5000 cycles at 1C and maintains 80% capacity retention after 3500 cycles. In contrast, the cells with 2% MoO₂-doped, undoped, and 3% MoO₂-doped SSEs achieve 80% capacity retention after 2100, 1600, and 600 cycles, respectively. Overall, the full cell with 1% MoO₂-doped SSE exhibits significantly better electrochemical performance compared to previously reported O-doped Li argyrodite-based full cells (Table S3, ESI†).

To assess the interface evolution within the full cells, in situ EIS measurements were conducted on Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 and Li_5.5PS_4.5Cl_1.5 SSE-based full cells during the initial charge–discharge cycles. Nyquist plots (Fig. S15a, ESI†) and the corresponding equivalent circuit fitting values (Table S4, ESI†) reveal that the cathodic interfacial resistance (R_CI) of the Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 SSE-based cell increases from 20.8 to 108.7 Ω over the first cycle. In comparison, the Li_5.5PS_4.5Cl_1.5 SSE-based cell exhibits a larger initial R_CI of 70.4 Ω, which increases to 239.8 Ω after the first cycle (Fig. S16a, ESI†). These results demonstrate that the MoO₂-doped SSE provides better stability with NCM811 than the undoped SSE. To gain further insight, the distribution of the relaxation time (DRT) diagrams was transformed from the Nyquist plots, providing detailed impedance information. As shown in Fig. S15b, c and S16b, c (ESI†), the DRT spectra exhibit five distinct peaks corresponding to different resistances: grain boundary (10⁻⁶ s), SEI/cathode electrolyte interphase (CEI) (10⁻⁵ s), charge transfer at the anode and cathode (10⁻⁴–10⁻³ and 10⁻²–10⁻¹ s), and solid-state diffusion (10⁰–10¹ s).¹⁸ The results confirm that the R_CI, including CEI and cathodic charge transfer resistance, increases continuously during charging and discharging.

In situ Raman spectroscopy was employed to detect the compositional evolution of the cathodic interface during the first charge–discharge cycle. The Raman spectra collected at the Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 SSE–NCM811 interface (Fig. 4a) show only the characteristic peaks of the PS₄³⁻ tetrahedron and the Mo–S bond.¹⁶ Furthermore, the intensity of the PS₄³⁻ main peak at 428 cm⁻¹ remains constant during cycling (Fig. S17a, ESI†), indicating the exceptional stability of the MoO₂-doped SSE under high voltage and oxidation conditions. In sharp contrast, the in situ Raman spectra for the Li_5.5PS_4.5Cl_1.5 SSE–NCM811 interface (Fig. 4b) show a distinct S–S bond peak at 472 cm⁻¹ in addition to the PS₄³⁻ peaks.¹⁹ Moreover, the intensity of the PS₄³⁻ peak decreases during cycling (Fig. S17b, ESI†), suggesting that the undoped SSE is prone to oxidation, forming high-valence S species.

Fig. 4 In situ Raman spectra collected at the cathodic interface with (a) Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5, and (b) Li_5.5PS_4.5Cl_1.5 SSEs. P 2p (c), S 2p (d), and Mo 3d (e) XPS spectra of the cycled composite cathodes.

To further evaluate the antioxidative durability of the synthesized SSEs, full cells cycled for 200 cycles were disassembled, and the composite cathodes were analyzed. The cycled cathode with Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 SSE retains an intact and dense morphology, whereas the cathode with Li_5.5PS_4.5Cl_1.5 SSE exhibits voids and cracks (Fig. S18, ESI†). The P 2p XPS spectrum of the cycled Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 SSE (Fig. 4c and Table S5, ESI†) shows peaks corresponding to PS₄³⁻ and PO₄³⁻,¹⁴ with the PO₄³⁻ content remaining close to that in the fresh SSE. In contrast, the cycled Li_5.5PS_4.5Cl_1.5 SSE exhibits an increased PO₄³⁻ content and the formation of a new oxidized product, P₂O₅ (at 134.6 eV).¹⁴ The S 2p XPS spectrum of the cycled Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 SSE (Fig. 4d and Table S5, ESI†) shows, in addition to the PS₄³⁻ and S–Mo bond, a small amount of sulfate (5.44%).²⁰ In contrast, around 40% of the Li_5.5PS_4.5Cl_1.5 SSE is oxidized to form S–S bonds (163.5 and 164.7 eV) and sulfates after prolonged cycling.²⁰ The Mo 3d XPS spectrum of the cycled Li_5.51P_0.99Mo_0.01S_4.48O_0.02Cl_1.5 SSE (Fig. 4e) further confirms the high stability of the Mo–S bond under high-voltage conditions. In conclusion, MoO₂ doping significantly enhances the antioxidative stability of Li argyrodite, improving their performance in high-voltage applications.

In summary, a series of MoO₂-doped Li argyrodites were successfully synthesized via the conventional solid-state reaction method. The optimized SSE, with 1% MoO₂ doping, exhibits a high RT σ_i of 12.0 mS cm⁻¹ and a broad ESW of 4.3 V vs. Li/Li⁺. These properties enable the corresponding Li–In|NCM811 full cell to achieve an impressive specific capacity of 194 mA h g⁻¹ at 0.1C, along with an ultra-long cycling life, retaining 80% of its capacity after 3500 cycles at 1C. In situ EIS, in situ Raman, and ex situ XPS analyses further confirm that the MoO₂-doped Li argyrodite demonstrates superior resistance to oxidation compared to the undoped counterpart.

This work was financially supported by the National Natural Science Foundation of China (NSFC: 52173205).

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. X. Hu, Z. Zhang, X. Zhang, Y. Wang, X. Yang, X. Wang, H. A. Yehezkel, S. Langford, D. Zhou, B. Li, G. Wang and D. Aurbach, Nat. Rev. Mater., 2024, 9, 305–320.
2. X. Yang, Q. Yin, C. Wang, K. Doyle-Davis, X. Sun and X. Li, Prog. Mater. Sci., 2023, 140, 101193.
3. L. Jia, J. Zhu, X. Zhang, B. Guo, Y. Du and X. Zhuang, Electrochem. Energy Rev., 2024, 7, 12.
4. J. Wu, S. Liu, F. Han, X. Yao and C. Wang, Adv. Mater., 2020, 32, 2000751.
5. N. Zhang, Q. He, L. Zhang, J. Zhang, L. Huang and X. Yao, Adv. Mater., 2024, 36, 2408903.
6. L. Zhou, K.-H. Park, X. Sun, F. Lalère, T. Adermann, P. Hartmann and L. F. Nazar, ACS Energy Lett., 2018, 4, 265–270.
7. C. Liu, B. Chen, T. Zhang, J. Zhang, R. Wang, J. Zheng, Q. Mao and X. Liu, Angew. Chem., Int. Ed., 2023, 62, e202302655.
8. S. Li, J. Lin, M. Schaller, S. Indris, X. Zhang, T. Brezesinski, C. W. Nan, S. Wang and F. Strauss, Angew. Chem., Int. Ed., 2023, 62, e202314155.
9. Y. Ma, Z. Yang and G. Zhang, Acta Chim. Sin., 2023, 81, 1387–1393.
10. C. Wang, J. Hao, J. Wu, H. Shi, L. Fan, J. Wang, Z. Wang, Z. Wang, L. Yang, Y. Gao, X. Yan and Y. Gu, Adv. Funct. Mater., 2024, 34, 2313308.
11. G. Li, S. Wu, H. Zheng, Y. Yang, J. Cai, H. Zhu, X. Huang, H. Liu and H. Duan, Adv. Funct. Mater., 2022, 32, 2211805.
12. G. Liu, J. Zhang, J. Yang, J. Chen, X. Xiao and X. Yao, J. Energy Chem., 2025, 100, 50–58.
13. X. Zhang, X. Li, S. Weng, S. Wu, Q. Liu, M. Cao, Y. Li, Z. Wang, L. Zhu, R. Xiao, D. Su, X. Yu, H. Li, L. Chen, Z. Wang and X. Wang, Energy Environ. Sci., 2023, 16, 1091–1099.
14. J. H. Zhu, J. Y. Luo, J. Y. Li, S. H. Huang, H. Z. Geng, Z. Y. Chen, L. N. Jia, Y. Z. Fu, X. Zhang and X. D. Zhuang, Adv. Mater., 2024, 36, e2407128.
15. R. Luo, W. W. Xu, Y. Zhang, Z. Wang, X. Wang, Y. Gao, P. Liu and M. Chen, Nat. Commun., 2020, 11, 1011.
16. R. Zhang, Y. Wu, Z. Chen, Y. Wang, J. Zhu and X. Zhuang, J. Mater. Chem. A, 2023, 11, 19195–19209.
17. W. D. Jung, J.-S. Kim, S. Choi, S. Kim, M. Jeon, H.-G. Jung, K. Y. Chung, J.-H. Lee, B.-K. Kim, J.-H. Lee and H. Kim, Nano Lett., 2020, 20, 2303–2309.
18. Y. Lu, C.-Z. Zhao, J.-Q. Huang and Q. Zhang, Joule, 2022, 6, 1172–1198.
19. D. Cao, X. Sun, F. Li, S. M. Bak, T. Ji, M. Geiwitz, K. S. Burch, Y. Du, G. Yang and H. Zhu, Angew. Chem., Int. Ed., 2023, 62, e202302363.
20. D. H. S. Tan, E. A. Wu, H. Nguyen, Z. Chen, M. A. T. Marple, J.-M. Doux, X. Wang, H. Yang, A. Banerjee and Y. S. Meng, ACS Energy Lett., 2019, 4, 2418–2427.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4cc04771k

---