English

• The high cost of solid-state electrolytes (SSEs) poses a significant barrier to the industrial adoption of all-solid-state lithium-ion batteries (ASSLBs). This cost is primarily driven by two factors: the use of expensive raw materials and energy-intensive synthesis processes [1]. For instance, elements such as La in Li₇La₃Zr₂O₁₂ [2–6], Ge in Li₁₀GeP₂S₁₂ [7–9], and Sc/In Li₃ScCl₆/Li₃InCl₆ [10–12] are scarce and significantly increase material costs. Although earth-abundant elements like P, S, and Cl are used in sulfide SSEs (e.g., Li₆PS₅Cl), the complex preparation process of Li₂S still elevates the overall cost [13]. Similarly, oxide electrolytes with garnet and NASICON structures require sintering temperatures exceeding 1000 ℃, leading to high energy consumption [4,14–17]. Halide SSEs, such as Li₂ZrCl₆, often involve inefficient synthesis methods like long time high-energy ball milling (500 rpm for 45 h) [18–23]. These challenges underscore the urgent need to develop low-cost SSEs composed of earth-abundant elements and scalable synthesis methods.

  Aluminum (Al), the most abundant metallic element in the Earth's crust (Figs. S1a and b in Supporting information), offers a promising solution. Aluminum chloride (AlCl₃), a low-cost precursor with a melting point of 194 ℃, enables the synthesis of SSEs (e.g., LiAlCl₄) at relatively low temperatures (< 200 ℃) [24–27]. However, the ionic conductivity of solid LiAlCl₄ remains low (10^-6 S/cm at 25 ℃), limiting its applicability in ASSBs [24,28].

  The ionic conductivity of SSEs is closely tied to their structural properties. Short distances between adjacent charge carriers, particularly in face-sharing polyhedra, are critical for achieving rapid ion conduction. Notable examples include body-centered sulfides [29–33] and LaCl₃-based Li⁺/Na⁺ conductors [34,35]. In LiAlCl₄, however, all Li⁺ ions occupy octahedral sites (4e, O site), which are either corner-sharing or edge-sharing, with a Li-Li distance exceeding 4.22 Å (Fig. 1a). The migration of Li⁺ ions between adjacent octahedral sites involves passing through one or two tetrahedral sites (T site), following an O-T-O or O-T-T-O pathway. The higher site energy of tetrahedral sites increases the activation energy for ion diffusion, resulting in low ionic conductivity (Fig. 1b and Fig. S2 in Supporting information) [25]. Previous studies have shown that structural distortion induced by high-energy ball milling can reduce the energy of Li⁺ ions in tetrahedral sites, enabling partial relocation of Li⁺ ions to these sites [24]. This approach increases the ionic conductivity of LiAlCl₄ by approximately 20 times, reaching 2.1 × 10⁻⁵ S/cm [24]. Nevertheless, this value remains insufficient for practical applications.

  Figure 1

  

  Figure 1.  (a) The structural model of LiAlCl₄. The light green, light blue, and green balls represent Li, Al, and Cl atoms, respectively. (b) Li⁺ ion diffusion path between adjacent octahedral sites. (c) Solid-state ⁶Li NMR spectra of Li_xAlCl_3-xO_x. (d-f) 2D exchange NMR spectra of Li_0.8AlCl_2.2O_0.8 with different mixing times.

  DownLoad: Full-Size Img PowerPoint

  Anion mixing has emerged as an effective strategy for tuning the site energy and distribution of Li⁺ ions in SSEs. Since anions directly coordinate with Li⁺ ions, they exert a strong influence on Li⁺ site energy. Anion mixing (e.g., Cl⁻/O²⁻, S²⁻/Cl⁻, and S²⁻/O²⁻) enables Li⁺ ions to occupy a wider range of sites, shortening the migration distance between charge carriers and enhancing ionic conductivity [18,36–45]. In this study, we employ Cl⁻/O²⁻ mixing to significantly improve the ionic conductivity of Li_xAlCl_3-xO_x. This strategy increases the number of Li⁺-occupied sites and reduces the distance between adjacent sites, leading to a lower ion migration activation energy (0.30 eV) and a remarkable enhancement in ionic conductivity (1.5 × 10⁻³ S/cm).

  The electrolyte developed in this study uses AlCl₃ and LiOH as raw materials and can be prepared with heating at 200 ℃ for only 0.5 h (Fig. S1c in Supporting information). In contrast to crystalline LiAlCl₄, the synthesized Li_xAlCl_3-xO_x exhibits a pure amorphous structure (Fig. S3 in Supporting information). The incorporation of O²⁻ into Li_xAlCl_3-xO_x likely facilitates the formation of highly stable Al−O−Al fragments. During cooling from the molten state, the presence of these Al−O−Al fragments restricts atomic rearrangement (crystallization), resulting in the retention of an amorphous structure at room temperature [43]. However, when x exceeds 0.8, a LiCl phase emerges in the product, and its content increases with higher O²⁻ concentrations (Fig. S3). This is because the enthalpy (H) and entropy (S) of Li_xAlCl_3-xO_x increases upon the incorporation of O²⁻ (Li⁺) and formation of defects (O²⁻ substituted Cl⁻, interstitial Li⁺). When ∆G = ∆H -T∆S < 0, the structure of Li_xAlCl_3-xO_x would be thermal-dynamically stable (x ≤ 0.8). Otherwise, the structure would be unstable and ultimately decompose to impurities such as LiCl (x > 0.8). In the main glass phase, more anion vacancies and fewer Li⁺ interstitial defects would be formed (Li_x-yAlCl_3-x-yO_x). This indicates that the high energy of interstitial Li⁺ ions is the main defect leading to instability of Li_xAlCl_3-xO_x.

  The Cl⁻/O²⁻ mixing in the amorphous phase is further confirmed by X-ray photoelectron spectroscopy (XPS) (Fig. S4 in Supporting information). The Li 1s and Al 2p spectral peaks for Li_0.8AlCl_2.2O_0.8 are situated between those of LiAlCl₄ and LiAlO₂, providing clear evidence of Cl⁻/O²⁻ anion mixing in Li_0.8AlCl_2.2O_0.8.

  Solid-state nuclear magnetic resonance (ssNMR) spectroscopy is a powerful tool for probing the local structure and distribution of Li⁺ ions [46–48], providing critical insights into ion conduction mechanisms in LiAlCl₄ and Li_xAlCl_3-xO_x (Figs. 1c-f). In LiAlCl₄, a single ⁷Li resonance signal is observed, confirming that Li⁺ ions exclusively occupy octahedral sites (4e sites) coordinated by six Cl⁻ ions (Fig. 1a and Fig. S5 in Supporting information) [43]. The introduction of O²⁻ significantly alters the local Li⁺ environment in the halide SSE. The ⁶Li NMR spectra of Li_xAlCl_3-xO_x reveal three distinct resonance signals: one corresponding to the LiCl impurity (~−1.2 ppm) and two others (~−0.88 and −0.65 ppm) arising from different Li⁺ sites within the Li_xAlCl_3-xO_x structure (Fig. 1c). The presence of cross-peaks in the 2D exchange spectroscopy (EXSY) NMR spectra further confirms the rapid exchange of Li⁺ ions between these two distinct sites (Figs. 1d-f). In the 1D spectra, all signals representing different Li⁺ environments are clearly resolved. At a short mixing time (0.05 ms), the exchange of Li⁺ ions between the two sites is negligible (Fig. 1d). However, as the mixing time is extended to 1 ms and 100 ms, distinct cross-peaks emerge between the two Li⁺ chemical environments (−0.88 and −0.65 ppm), as highlighted in the dotted boxes in Figs. 1e and f. This observation demonstrates that Li⁺ ions dynamically migrate between neighboring sites, creating efficient conduction pathways.

  The incorporation of O²⁻ into Li_xAlCl_3-xO_x significantly enhances its ionic conductivity, reaching values on the order of 10⁻³ S/cm at 25 ℃. The composition is optimal at x = 0.8, when it is phase-pure amorphous without significant impurities (based on XRD), and the ionic conductivity reaches an optimum of 1.5 × 10⁻³ S/cm (Figs. 2a and c). For x < 0.8, the improvement in ionic conduction with increasing O²⁻ content is attributed to the rising defect density resulting from O²⁻/Cl⁻ mixing. Conversely, for x > 0.8, the decline in ionic conductivity is due to the formation of impurities (Fig. S3). Additionally, the amorphous structure induced by O₂^- doping should also facilitates the distribution of Li⁺ ions across multiple sites, further contributing to the enhanced ionic conductivity [24].

  Figure 2

  

  Figure 2.  (a) Nyquist plots, (b) Arrhenius plots, and (c) ionic conductivities and activation energies of Li_xAlCl_3-xO_x. (d) Electronic conductivity of Li_0.8AlCl_2.2O_0.8.

  DownLoad: Full-Size Img PowerPoint

  Conductivity measurements also reveal a significant reduction in activation energy to 0.30 eV for Li_0.8AlCl_2.2O_0.8 (Figs. 2b and c, Fig. S6 in Supporting information). Furthermore, the electronic conductivity of Li_0.8AlCl_2.2O_0.8, determined from DC polarization curves (Fig. 2d and Fig. S7 in Supporting information), is calculated to be 6.98 × 10⁻¹⁰ S/cm. This value is substantially lower than that of many widely studied solid-state electrolytes, highlighting the material's potential for minimizing electronic leakage in solid-state battery applications [49–51].

  To elucidate the local structure of amorphous Li_xAlCl_3-xO_x and gain deeper insights into Li⁺ ion conduction pathways, density functional theory (DFT) calculations were performed (Figs. 3a-d). The initial structure of Li_0.75AlCl_2.25O_0.75 was constructed by replacing 0.75 Cl⁻ with O²⁻ in AlCl₃ and adding 0.75 Li⁺ ions to maintain charge neutrality. The hypothesized structure was heated to 800 K for 10 ps with a time step of 1 fs to obtain a fully melted configuration. Subsequently, the amorphous structure was annealed at lower temperatures by adjusting the model size to achieve the appropriate pressure. The final structure of Li_0.75AlCl_2.25O_0.75 was energy-minimized (Fig. 3a and Table S1 in Supporting information). The simulated structure reveals that Al³⁺ ions remain in tetrahedral sites, with partial replacement of Cl⁻ by O²⁻ leading to the formation of Al-O-Al fragments (Fig. 3a and Fig. S8 in Supporting information). The framework of Li_0.75AlCl_2.25O_0.75 consists of AlCl_nO_4-n units, while Li⁺ ions occupy various interstitial sites, coordinated by different combinations of Cl⁻ and O²⁻ ions (e.g., 3Cl, 1O2Cl, 4Cl, 2O3Cl). Compared to LiAlCl₄, the coordination environments of Li⁺ ions in Li_0.75AlCl_2.25O_0.75 are more complex, resulting in two broad peaks in the NMR spectra (Fig. 2b). The substitution of Cl⁻ with the more electronegative O²⁻ reduces the electron cloud density around the Li⁺ nucleus, leading to a larger chemical shift (toward the left) [36,52]. Consequently, the left-shifted ⁶Li peak (−0.65 ppm) likely corresponds to Li⁺ ions coordinated with a higher number of O²⁻ anions.

  Figure 3

  

  Figure 3.  (a) Simulated structure of Li_0.75AlCl_2.25O_0.75. (b) Microstructural fragments of Li_0.75AlCl_2.25O_0.75 with the shortest Li⁺ transition distances. (c) The RDF of Li–Li in Li_0.75AlCl_2.25O_0.75 at different temperatures. R represents the distance between different Li⁺ ions. (d) The Arrhenius plots of Li-ion diffusivity as a function of temperature for Li_0.75AlCl_2.25O_0.75. (e) Neutron pair distribution function (nPDF) analysis of Li_0.8AlCl_2.2O_0.8 from total neutron scattering data. (f) Expanded structure (3 × 3 × 3) of Li_0.75AlCl_2.25O_0.75 after refinement with nPDF results.

  DownLoad: Full-Size Img PowerPoint

  Compared to LiAlCl₄, where Li⁺ ions predominantly occupy octahedral sites, the increased distribution of Li⁺ sites in Li_xAlCl_3-xO_x suggests a reduction in the distance between adjacent Li⁺ sites. A segment of the simulated Li_0.75AlCl_2.25O_0.75 structure (Fig. 3b) confirms that the Li–Li distance (3.03–3.80 Å) is significantly shorter than that in LiAlCl₄ (4.22–4.88 Å). This observation is further supported by the radial distribution function (RDF) of Li–Li (Fig. 3c), which reveals shorter Li-Li distances in Li_0.75AlCl_2.25O_0.75. Additionally, the peaks corresponding to short Li-Li distances gradually diminish with increasing temperature, indicating structural collapse at elevated temperatures. At 600 K, the loss of short-range order confirms that Li_0.75AlCl_2.25O_0.75 transitions to a molten state.

  In solid-state electrolytes (SSEs), the distribution of Li⁺ ions is governed by their site energy at different interstitial positions, which is primarily determined by the bond energy between Li⁺ ions and surrounding anions. This bond energy can be described by the equation E(r) =−ZiZj4πɛ0r+Brn, where ε₀ is the dielectric constant, B and n are constant, Z_i and Z_j are ion charge number, r is the distance between Li⁺ ions and surrounding anions [44]. When the anion type changes, the site energy and Li⁺ distribution in SSEs undergo significant modifications. In chloride-based SSEs, Li⁺ ions predominantly occupy octahedral interstices, where the distance between adjacent sites is relatively large. However, upon the introduction of O²⁻ ions, the site energy of Li⁺ in tetrahedral and interstitial sites decreases, increasing the number of favorable distribution sites. Consequently, the distance between adjacent Li⁺ sites is substantially reduced.

  The Li⁺ ion conduction in Li_0.75AlCl_2.25O_0.75 was also simulated (Fig. 3d). The Li⁺ diffusion coefficients range from 10⁻⁶ cm²/s to 10⁻⁵ cm²/s between 300 K and 800 K, demonstrating rapid ion migration across this temperature range [38]. The calculated activation energy (0.16 eV) and ionic conductivity (26.6 mS/cm at 300 K) further confirm the presence of fast ion-diffusion channels and efficient ion conduction in Li_0.75AlCl_2.25O_0.75. Notably, the Li⁺ ion probability densities reveal potential diffusion pathways (Fig. S9 in Supporting information). At 300 K, Li⁺ ions exhibit partial long-range diffusion through low-energy pathways. As the temperature increases to 600 K and 800 K, the more uniform probability densities suggest a flatter energy landscape in Li_0.75AlCl_2.25O_0.75.

  The simulated structure was further validated by neutron pair distribution function (nPDF) analysis. The PDF profile features a peak at 2.165 Å, corresponding to Al-Cl pairs, while strong peaks at 3.595 Å and 6.425 Å are attributed to Al-Al pairs. Below 6 Å, Li_0.8AlCl_2.2O_0.8 exhibits multiple strong peaks, indicative of its characteristic short-range order. Beyond 10 Å, the peak intensity diminishes, reflecting the material's long-range disorder and amorphous nature, consistent with XRD results. To improve refinement accuracy, the structural unit of Li_0.75AlCl_2.25O_0.75 was expanded to a 3 × 3 × 3 supercell. The refined nPDF spectrum yields an good fit (χ² = 4.8), validating the simulated 3 × 3 × 3 structure (Fig. 3e) [53]. The refined structure is shown in Fig. 3f, which closely matches the initial simulation.

  Like other chloride solid electrolytes, Li_0.8AlCl_2.2O_0.8 is susceptible to deliquescence in natural air [54], but the electrolyte is well stabilized in dry air. After exposing Li_0.8AlCl_2.2O_0.8 to a dry room atmosphere (dew point: −40 ℃) for 24 h, the ionic conductivity decreased slightly from 1.5 mS/cm to 1.1 mS/cm, while the amorphous structure remained intact (Fig. S10 in Supporting information). A minor mass increase from 0.3346 g to 0.3449 g further confirms its stability in dry air, attributed to the strong Al-O and Li-O bonds. Linear sweep voltammetry (LSV) of Li_0.8AlCl_2.2O_0.8-Super P||Li_0.8AlCl_2.2O_0.8||Li reveals an electrochemical stability window ranging from 1.12 V to 3.96 V vs. Li⁺/Li (Fig. S11 in Supporting information), comparable to other halide solid-state electrolytes [34,55].

  To assess the practical applicability of Li_0.8AlCl_2.2O_0.8 as a solid-state electrolyte (SSE), an all-solid-state battery (ASSB) was assembled using a commercial LiCoO₂ composite cathode and a homemade Li-Si anode. A transition layer of homemade Li₆PS₅Cl was incorporated to isolate the Li_0.8AlCl_2.2O_0.8 SSE from the Li-Si anode (Fig. 4a). Given the electrochemical stability window of Li_0.8AlCl_2.2O_0.8, it was employed as a catholyte in the composite cathode. To enhance stability, Li-Si (with a higher potential than Li metal) was selected as the anode [56], while Li₆PS₅Cl (Fig. S12 in Supporting information), known for its compatibility with anodes, was used as the anolyte to separate Li_0.8AlCl_2.2O_0.8 from the Li-Si anode. To minimize oxidation and decomposition of Li_0.8AlCl_2.2O_0.8 above its oxidation potential (3.96 V vs. Li⁺/Li), conductive carbon was excluded from the composite cathode, enabling LiCoO₂ to be charged to 4.2 V vs. Li-Si [57].

  Figure 4

  

  Figure 4.  Electrochemical performance of the LiCoO₂||Li_0.8AlCl_2.2O_0.8||Li₆PS₅Cl||Li-Si cell. (a) Schematic illustration of the ASSBs. (b) SEM image of cross-sectional ASSBs. (c) EDS image of Li_0.8AlCl_2.2O_0.8. (d) The initial charge/discharge curves at 0.1 C, with the Coulombic efficiency η_Coulomb denoted. (e) Discharge profiles at different current densities. (f) Rate capability at 0.1, 0.2, 0.3, 0.5 and 1 C. (g) Long-term cycling performance at 0.3 C. (h) Spider plot comparing the performance of sulfide, oxide, conventional halide SSEs and Li_0.8AlCl_2.2O_0.8 [58–66].

  DownLoad: Full-Size Img PowerPoint

  Cross-sectional SEM images of the ASSB (Figs. 4b and c) reveal layer thicknesses of ~24 µm (LiCoO₂ composite cathode), 300 µm (Li_0.8AlCl_2.2O_0.8 SSE), 110 µm (Li₆PS₅Cl SSE), and 130 µm (Li-Si anode). High-magnification SEM images (Fig. S13 in Supporting information) confirm the dense morphology of the Li_0.8AlCl_2.2O_0.8 layer. Under a pressure of 150 MPa, Li_0.8AlCl_2.2O_0.8 appears compact and pinhole-free (Fig. S13b), while Li₆PS₅Cl exhibits a porous structure (Fig. S13c). The total resistance of the ASSB before cycling exceeds 400 Ω (Fig. S14 in Supporting information), primarily attributed to the composite cathode (Fig. S15 in Supporting information). This high resistance stems from the high content (75 wt%) of poorly conductive LiCoO₂ in the cathode layer.

  The LiCoO₂||Li_0.8AlCl_2.2O_0.8||Li₆PS₅Cl||Li-Si ASSB was cycled at various current densities between 2.5 V and 4.2 V under an operating pressure of 120 MPa at 25 ℃ (Figs. 4d-g). At 0.1 C (25 ℃), the battery delivers a discharge capacity of 150 mAh/g with a high initial Coulombic efficiency (CE) of 97%, indicating highly reversible reactions (Fig. 4d). This performance underscores the effectiveness of the battery configuration and the excellent oxidation stability of Li_0.8AlCl_2.2O_0.8, which minimizes side reactions during cycling. As the current density increases, the discharge capacities decrease to 138 mAh/g (0.2 C), 128 mAh/g (0.3 C), 108 mAh/g (0.5 C), and 73 mAh/g (1 C). The rate capability is currently limited by the high cathode resistance, necessitating further optimization of the cathode composition [67]. Remarkably, the ASSB retains 94.4% of its discharge capacity after 180 cycles at 0.3 C (Fig. 4g). Even after 1000 cycles at 1 C, the capacity retention reaches 70.6% (Fig. S16 in Supporting information), demonstrating exceptional cycling stability. Post-cycling resistance measurements (Fig. S14) reveal a slight reduction in total resistance after 180 cycles at 0.3 C or 1000 cycles at 1 C, further confirming the outstanding oxidation stability of Li_0.8AlCl_2.2O_0.8 SSE.

  In conclusion, we have developed a low-cost Li_xAlCl_3-xO_x solid-state electrolyte (SSE) through an innovative anion mixing strategy, which offers a compelling combination of low cost, high oxidation stability, excellent interfacial compatibility, and superior ionic conductivity (Fig. 4h). The use of earth-abundant elements significantly reduces material costs, while the short sintering time (30 min) at a low temperature (200 ℃) further minimizes production expenses. The O²⁻/Cl⁻ dual-anion design enables Li⁺ ions to occupy multiple sites within the Li_xAlCl_3-xO_x lattice, reducing Li⁺ migration distances and enhancing ionic conductivity. The optimized Li_xAlCl_3-xO_x SSE achieves a room-temperature ionic conductivity of 1.5 mS/cm, representing a three-order-of-magnitude improvement over LiAlCl₄. The incorporation of O²⁻ into the AlCl₃-based halide not only improves oxidation stability but also enhances air stability, owing to the stronger bonding between O²⁻ and Li⁺/Al³⁺. The dense morphology of Li_xAlCl_3-xO_x SSE ensures excellent interfacial contact with the cathode, further supporting its practical application. The exceptional oxidation stability and compactability of Li_xAlCl_3-xO_x make it a promising catholyte candidate for all-solid-state batteries (ASSBs). In a proof-of-concept demonstration, a LiCoO₂/Li-Si ASSB incorporating Li_0.8AlCl_2.2O_0.8 as the catholyte delivered outstanding electrochemical performance, achieving a high initial Coulombic efficiency of 97% and retaining 94.4% of its discharge capacity after 180 cycles at 0.3 C. These results highlight the significant potential of Li_xAlCl_3-xO_x SSEs for practical applications in next-generation solid-state batteries. However, the absence of conductive carbon and the high weight ratio of low conductive LiCoO₂ (75%) at the cathode make the ASSB have high resistance which affects the rate performance. Further research could be enhancement of oxidation stability of SSE, adding conductive carbon if stability allows, or surface modify the LiCoO₂ cathode to improve the battery performance.

  CRediT authorship contribution statement

  Liming Zhang: Writing – original draft, Methodology, Formal analysis, Data curation. Chenjie Lou: Formal analysis, Data curation. Shuaika Liang: Formal analysis, Data curation. Wei Xia: Investigation, Formal analysis. Rui Wang: Writing – original draft, Resources, Funding acquisition. Longfei Li: Data curation. Chengyu Fu: Data curation. Mingxue Tang: Resources, Investigation, Formal analysis, Data curation. Rong Xiang: Writing – review & editing, Funding acquisition. Wei Wang: Writing – review & editing, Formal analysis. Yichun Xu: Writing – original draft, Formal analysis, Data curation. Hongfa Xiang: Writing – review & editing, Supervision, Resources, Funding acquisition, Conceptualization. Xuyong Feng: Writing – review & editing, Supervision, Investigation, Funding acquisition, Conceptualization.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  This study was supported by the National Natural Science Foundation of China (Nos. U2330101, 52072105), the Major Science and Technology Projects in Anhui Province (Nos. 202203a05020032 and 2022e03020004), Innovation R & D Program of Anhui Province (No. 202423i08050026), Ministry of Science and Technology of China (No. 2023YFE0101300) and Zhejiang Province (No. 2022R01001), XXX-Project (No. 2020-XXXX-XX-246–00) and the Fundamental Research Funds for the Central Universities (No. JZ2024HGTG0292).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.111558.

  
  
• 1. [1]

     S. Xin, X. Zhang, L. Wang, et al., Sci. China Chem. 67 (2024) 13–42. doi: 10.1007/s11426-023-1908-9

  2. [2]

     X. Yao, S. Chen, C. Wang, et al., Adv. Energy Mater. 14 (2023) 2303422.

  3. [3]

     M.J. Wang, E. Carmona, A. Gupta, P. Albertus, J. Sakamoto, Nat. Commun. 11 (2020) 5201. doi: 10.1038/s41467-020-19004-4

  4. [4]

     Y. Li, X. Chen, A. Dolocan, et al., J. Am. Chem. Soc. 140 (2018) 6448–6455. doi: 10.1021/jacs.8b03106

  5. [5]

     X. Wang, Y. Chang, Y. Zhang, et al., Chin. Chem. Lett. 37 (2026) 110626. doi: 10.1016/j.cclet.2024.110626

  6. [6]

     Y. Feng, L. Yang, Z. Yan, et al., Energy Storage Mater. 63 (2023) 103053. doi: 10.1016/j.ensm.2023.103053

  7. [7]

     N. Kamaya, K. Homma, Y. Yamakawa, et al., Nat. Mater. 10 (2011) 682–686. doi: 10.1038/nmat3066

  8. [8]

     F. Du, X. Ren, J. Yang, J. Liu, W. Zhang, J. Phys. Chem. C 118 (2014) 10590–10595. doi: 10.1021/jp5000039

  9. [9]

     F. Han, Y. Zhu, X. He, Y. Mo, C. Wang, Adv. Energy Mater. 6 (2016) 1501590. doi: 10.1002/aenm.201501590

  10. [10]

      J. Liang, X. Li, S. Wang, et al., J. Am. Chem. Soc. 142 (2020) 7012–7022. doi: 10.1021/jacs.0c00134

  11. [11]

      X. Li, J. Liang, J. Luo, et al., Energy Environ. Sci. 12 (2019) 2665–2671. doi: 10.1039/c9ee02311a

  12. [12]

      Z. Jiang, C. Liu, J. Yang, et al., Chin. Chem. Lett. 36 (2025) 109741. doi: 10.1016/j.cclet.2024.109741

  13. [13]

      F. Tu, Z. Zhao, X. Zhang, et al., ACS Sustain. Chem. Eng. 10 (2022) 15365–15371. doi: 10.1021/acssuschemeng.2c02238

  14. [14]

      T. Thompson, J. Wolfenstine, J.L. Allen, et al., J. Mater. Chem. A 2 (2014) 13431–13436. doi: 10.1039/C4TA02099E

  15. [15]

      Y. Zeng, B. Ouyang, J. Liu, et al., Science 378 (2022) 1320–1324. doi: 10.1126/science.abq1346

  16. [16]

      Q. Wang, Z. Jiang, C. Yu, L. Li, G. Li, Chin. Chem. Lett. 36 (2025) 110006. doi: 10.1016/j.cclet.2024.110006

  17. [17]

      D. Zuo, L. Yang, Z. Zou, et al., Adv. Energy Mater. 13 (2023) 2301540. doi: 10.1002/aenm.202301540

  18. [18]

      L. Hu, J. Wang, K. Wang, et al., Nat. Commun. 14 (2023) 3807. doi: 10.1038/s41467-023-39522-1

  19. [19]

      K. Wang, Q. Ren, Z. Gu, et al., Nat. Commun. 12 (2021) 4410. doi: 10.1038/s41467-021-24697-2

  20. [20]

      H. Kwak, D. Han, J. Lyoo, et al., Adv. Energy Mater. 11 (2021) 2003190. doi: 10.1002/aenm.202003190

  21. [21]

      S. Zhang, F. Zhao, L.Y. Chang, et al., J. Am. Chem. Soc. 146 (2024) 2977–2985. doi: 10.1021/jacs.3c07343

  22. [22]

      S. Chen, C. Yu, S. Chen, et al., Chin. Chem. Lett. 33 (2022) 4635–4639. doi: 10.1016/j.cclet.2021.12.048

  23. [23]

      Y. Ye, Z. Gu, J. Geng, et al., Nano Lett. 25 (2025) 3747–3755. doi: 10.1021/acs.nanolett.4c05460

  24. [24]

      N. Tanibata, S. Takimoto, K. Nakano, et al., ACS Mater. Lett. 2 (2020) 880–886. doi: 10.1021/acsmaterialslett.0c00127

  25. [25]

      N. Flores-González, N. Minafra, G. Dewald, et al., ACS Mater. Lett. 3 (2021) 652–657. doi: 10.1021/acsmaterialslett.1c00055

  26. [26]

      W. Weppner, R.A. Huggins, J. Electrochem. Soc. 124 (1977) 35–38. doi: 10.1149/1.2133238

  27. [27]

      S. Zhang, Y. Xu, H. Wu, et al., Angew. Chem. Int. Ed. (2024) e202401373.

  28. [28]

      L. Xu, X. Hu, S. Zhou, et al., Chin. Chem. Lett. 35 (2024) 109103. doi: 10.1016/j.cclet.2023.109103

  29. [29]

      Y. Wang, W.D. Richards, S.P. Ong, et al., Nat. Mater. 14 (2015) 1026–1031. doi: 10.1038/nmat4369

  30. [30]

      Y. Li, S. Song, H. Kim, et al., Science 381 (2023) 50–53. doi: 10.1126/science.add7138

  31. [31]

      Y. Kato, S. Hori, T. Saito, et al., Nat. Energy 1 (2016) 16030. doi: 10.1038/nenergy.2016.30

  32. [32]

      W.D. Richards, Y. Wang, L.J. Miara, et al., Energy Environ. Sci. 9 (2016) 3272–3278. doi: 10.1039/C6EE02094A

  33. [33]

      X. Feng, P.H. Chien, S. Patel, et al., Chem. Mater. 32 (2020) 3036–3042. doi: 10.1021/acs.chemmater.0c00025

  34. [34]

      Y.C. Yin, J.T. Yang, J.D. Luo, et al., Nature 616 (2023) 77–83. doi: 10.1038/s41586-023-05899-8

  35. [35]

      C. Fu, Y. Li, W. Xu, et al., Nat. Commun. 15 (2024) 4315. doi: 10.1038/s41467-024-48712-4

  36. [36]

      J. Zhang, C. Lou, L. Dong, et al., Small Struct. 5 (2024) 2300565. doi: 10.1002/sstr.202300565

  37. [37]

      X. Wang, R. Xiao, H. Li, L. Chen, Phys. Chem. Chem. Phys. 18 (2016) 21269–21277. doi: 10.1039/C6CP03179J

  38. [38]

      X. Feng, P.H. Chien, Y. Wang, et al., Energy Storage Mater. 30 (2020) 67–73. doi: 10.1016/j.ensm.2020.04.042

  39. [39]

      S. Zhang, F. Zhao, L.Y. Chang, et al., J. Am. Chem. Soc. 146 (2024) 2977–2985. doi: 10.1021/jacs.3c07343

  40. [40]

      S. Zhang, F. Zhao, J. Chen, et al., Nat. Commun. 14 (2023) 3780. doi: 10.1038/s41467-023-39197-8

  41. [41]

      Y. Tanaka, K. Ueno, K. Mizuno, et al., Angew. Chem. Int. Ed. (2023) e202217581.

  42. [42]

      Z. Liu, S. Ma, J. Liu, et al., ACS Energy Lett. 6 (2021) 298–304. doi: 10.1021/acsenergylett.0c01690

  43. [43]

      T. Dai, S. Wu, Y. Lu, et al., Nat. Energy 8 (2023) 1221–1228. doi: 10.1038/s41560-023-01356-y

  44. [44]

      Y. Tan, J. Gatts, C. Fu, et al., Small 21 (2025) 2504677. doi: 10.1002/smll.202504677

  45. [45]

      Y. Wang, W.D. Richards, S.P. Ong, et al., Nat. Mater. 14 (2015) 1026–1031. doi: 10.1038/nmat4369

  46. [46]

      Y. Gao, J. Liu, C. Lou, et al., Chin. Chem. Lett. 34 (2023) 108268. doi: 10.1016/j.cclet.2023.108268

  47. [47]

      Y. Shi, W. Zhang, L. Xu, et al., Chin. Chem. Lett. 37 (2026) 111077. doi: 10.1016/j.cclet.2025.111077

  48. [48]

      G. Yang, Z. Hao, C. Fang, et al., Chin. Chem. Lett. 36 (2025) 111185. doi: 10.1016/j.cclet.2025.111185

  49. [49]

      F. Han, A.S. Westover, J. Yue, et al., Nat. Energy 4 (2019) 187–196. doi: 10.1038/s41560-018-0312-z

  50. [50]

      X. Fan, X. Ji, F. Han, et al., Sci. Adv. 4 (2018) eaau9245. doi: 10.1126/sciadv.aau9245

  51. [51]

      X. Feng, P. Chien, Z. Zhu, et al., Adv. Funct. Mater. 29 (2019) 1807951. doi: 10.1002/adfm.201807951

  52. [52]

      D. Wang, G. Zhong, W.K. Pang, et al., Chem. Mater. 27 (2015) 6650–6659. doi: 10.1021/acs.chemmater.5b02429

  53. [53]

      C.L. Farrow, P. Juhas, J.W. Liu, et al., J. Phys. : Condens. Matter 19 (2007) 335219. doi: 10.1088/0953-8984/19/33/335219

  54. [54]

      Q. Wang, X. Ma, Q. Liu, D. Sun, X. Zhou, J. Alloy. Compd. 969 (2023) 172479. doi: 10.1016/j.jallcom.2023.172479

  55. [55]

      Y. Ye, J. Geng, D. Zuo, et al., ACS Nano 18 (2024) 18368–18378. doi: 10.1021/acsnano.4c02678

  56. [56]

      H. Pan, M. Zhang, Z. Cheng, et al., Sci. Adv. 8 (2022) eabn4372. doi: 10.1126/sciadv.abn4372

  57. [57]

      M.A. Kraft, S. Ohno, T. Zinkevich, et al., J. Am. Chem. Soc. 140 (2018) 16330–16339. doi: 10.1021/jacs.8b10282

  58. [58]

      X. Li, J. Liang, X. Yang, et al., Energy Environ. Sci. 13 (2020) 1429–1461. doi: 10.1039/c9ee03828k

  59. [59]

      H. Kwak, S. Wang, J. Park, et al., ACS Energy Lett. 7 (2022) 1776–1805. doi: 10.1021/acsenergylett.2c00438

  60. [60]

      J. Lau, R.H. DeBlock, D.M. Butts, et al., Adv. Energy Mater. 8 (2018) 1800933. doi: 10.1002/aenm.201800933

  61. [61]

      L. Xu, J. Li, W. Deng, et al., Adv. Energy Mater. 11 (2021) 2000648. doi: 10.1002/aenm.202000648

  62. [62]

      C. Wang, J.T. Kim, C. Wang, X. Sun, Adv. Mater. 35 (2023) 2209074. doi: 10.1002/adma.202209074

  63. [63]

      K.N. Gao, F. Bai, Z. Sun, T. Zhang, Energy Storage Mater. 70 (2024) 103444. doi: 10.1016/j.ensm.2024.103444

  64. [64]

      A. Manthiram, X. Yu, S. Wang, Nat. Rev. Mater. 2 (2017) 16103. doi: 10.1038/natrevmats.2016.103

  65. [65]

      K. Tuo, C. Sun, S. Liu, Energy Rev. 6 (2023) 17.

  66. [66]

      P. Lu, D. Wu, L. Chen, H. Li, F. Wu, Electrochem. Energy Rev. 5 (2022) 3. doi: 10.1007/s41918-022-00149-3

  67. [67]

      Y.G. Lee, S. Fujiki, C. Jung, et al., Nat. Energy 5 (2020) 299–308. doi: 10.1038/s41560-020-0575-z

• 

  Figure 1  (a) The structural model of LiAlCl₄. The light green, light blue, and green balls represent Li, Al, and Cl atoms, respectively. (b) Li⁺ ion diffusion path between adjacent octahedral sites. (c) Solid-state ⁶Li NMR spectra of Li_xAlCl_3-xO_x. (d-f) 2D exchange NMR spectra of Li_0.8AlCl_2.2O_0.8 with different mixing times.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) Nyquist plots, (b) Arrhenius plots, and (c) ionic conductivities and activation energies of Li_xAlCl_3-xO_x. (d) Electronic conductivity of Li_0.8AlCl_2.2O_0.8.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) Simulated structure of Li_0.75AlCl_2.25O_0.75. (b) Microstructural fragments of Li_0.75AlCl_2.25O_0.75 with the shortest Li⁺ transition distances. (c) The RDF of Li–Li in Li_0.75AlCl_2.25O_0.75 at different temperatures. R represents the distance between different Li⁺ ions. (d) The Arrhenius plots of Li-ion diffusivity as a function of temperature for Li_0.75AlCl_2.25O_0.75. (e) Neutron pair distribution function (nPDF) analysis of Li_0.8AlCl_2.2O_0.8 from total neutron scattering data. (f) Expanded structure (3 × 3 × 3) of Li_0.75AlCl_2.25O_0.75 after refinement with nPDF results.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Electrochemical performance of the LiCoO₂||Li_0.8AlCl_2.2O_0.8||Li₆PS₅Cl||Li-Si cell. (a) Schematic illustration of the ASSBs. (b) SEM image of cross-sectional ASSBs. (c) EDS image of Li_0.8AlCl_2.2O_0.8. (d) The initial charge/discharge curves at 0.1 C, with the Coulombic efficiency η_Coulomb denoted. (e) Discharge profiles at different current densities. (f) Rate capability at 0.1, 0.2, 0.3, 0.5 and 1 C. (g) Long-term cycling performance at 0.3 C. (h) Spider plot comparing the performance of sulfide, oxide, conventional halide SSEs and Li_0.8AlCl_2.2O_0.8 [58–66].

  下载: 全尺寸图片 幻灯片
•