English

• Taking advantage of the low electrochemical potential (−3.04 V), low density (0.534 g/cm³), and high theoretical specific capacity (3860 mAh/g) of Li-metal anode and the multielectron conversion chemistry of sulfur cathode, Lithium-sulfur batteries (LSBs) are considered one of the most promising ways to meeting future energy storage requirements [1-4]. A typical LSB undergoes multiphase redox reactions, with reactants, intermediates, and products ranging from solid sulfur to soluble lithium polysulfides (LiPSs). However, such a redox process has multiple issues, including the most critical and well-known "shuttle effect" of LiPSs, resulting in poor Coulombic efficiency and cycle life [5-9].

  To date, extensive efforts have been made to immobilize soluble LiPSs through various strategies including the design of cathode host materials [10-13], developing polarity anchoring materials [14-18], and enhancing separator functionality [19-23]. Designing interlayer structure has been validated as a versatile strategy to tackle the above issues by taking the following advantages: (1) Creating an independent three-dimensional (3D) conductive network; (2) providing a highly porous structure; and (3) accommodating diverse polar materials and catalysts [24-29]. However, conventional interlayer structures often lead to the accumulation of 'dead sulfur' near the separator surface, resulting in poor sulfur utilization efficiency. Thus, optimizing the sulfur reduction reaction process is crucial in addressing these challenges.

  Various catalysts, including oxides, nitrides, sulfides, and selenides, have been employed to enhance the sulfur reduction reaction in LSBs [30-34]. Recently, defect engineering strategies have been developed for these catalysts, focusing on modifying unsaturated coordination structures and electronic structures via oxygen or sulfur vacancies. These modifications significantly enhance catalytic efficacy and improve the conversion kinetics of LiPSs [35-39]. In contrast to the d-band energy levels of conventional catalysts, rare earth compounds exhibit unique properties in magneto-optical devices, catalysis, and energy storage, owing to their distinctive 4f electronic structure and magnetic characteristics [40,41]. Recent research indicates that the high electronegativity of rare earth metal ions and their Lewis acidic nature facilitate strong interactions with LiPSs, enhancing LiPSs capture [42]. Among rare earth materials, europium(Ⅱ) oxide (Eu₂O₃) stands out due to its special 4f electronic structure [41-43]. The half-filled 4f orbitals with a spin characteristic of S = 7/2 and an isotropic magnetic moment effectively shielded by 5s² and 5p⁶ electron shells confer properties akin to Heisenberg magnetic materials [44]. While previous studies have highlighted the potential of rare earth compound catalysts for high-performance LSBs, comprehensive investigations into how the electron spin evolution of Eu₂O₃ influences the sulfur reduction reaction process are lacking. In particular, detailed analyses on the role of oxygen vacancies in Eu₂O₃ for catalyzing sulfur reduction and immobilizing LiPSs remain unexplored.

  Herein, we propose to use oxygen-deficient Eu₂O_3−δ nanoparticles as catalysts to design a multifunctional interlayer structure for LSBs. Magnetic hysteresis (M-H) and electron paramagnetic resonance (EPR) tests show a typical spin evolution of Eu₂O_3−δ, which is expected to enhance the adsorption and catalytic capability of the interlayer material. Further analysis reveals that Eu₂O_3−δ/carbon nanotube (CNT) interlayer synchronizes the strong adsorption capability and catalytic sites, thus significantly promoting the catalytic process of LiPSs to Li₂S and regulating cycling performance. Therefore, engineering europium-based compounds with spin evolution are crucial in developing versatile interlayer materials for high-performance LSBs.

  In LSBs, nanosized catalysis particles are usually required to enable high catalytic activities during redox reactions. However, it is challenging to directly prepare Eu₂O_3−δ nanoparticles by a conventional way. Therefore, we first synthesized EuSe nanowire through a hydrothermal reaction at 180 ℃. The diagram of the synthesis route is shown in Fig. 1a. Then, this EuSe nanowire can be further decomposed to form Eu₂O_3−δ nanoparticles after carbonization at 600 ℃, which also improves the electronic and ionic conductivity of the catalyst. Subsequently, CNT and Eu₂O_3−δ nanoparticles were combined to construct the interlayer structure (Eu₂O_3−δ/CNT) between the separator and active materials in LSBs. In the Eu₂O_3−δ/CNT composite interlayer, CNT plays two key roles: (1) The CNT provides a 3D conducting network for the Eu₂O_3−δ catalyst, which ensures the rapid electron transfer during the reaction and thus unleashes the LiPSs catalytic functionality of the Eu₂O_3−δ catalyst. (2) The composite interlayer is prepared by pumping filtration of a single-walled CNT composite with the rare-earth catalyst. The single-walled CNT can ensure the formation of the interlayer film, and the self-supporting interlayer structure enhances the LiPSs diffusion. The schematic diagram of the designed interlayer structure and its function in LSBs is shown in Fig. 1b. Here, the Eu₂O_3−δ/CNT interlayer structure is expected to suppress the shuttle of LiPSs by benefiting from the high LiPSs immobilization capability and catalytic activity of Eu₂O_3−δ, thus promoting the LiPSs redox process before shuttling to the anode side.

  Figure 1

  

  Figure 1.  (a) The synthesis route of Eu₂O_3−δ/CNT interlayer. (b) Schematic diagram of Eu₂O_3−δ polarity anchoring LiPSs and catalytic conversion of LiPSs to Li₂S.

  DownLoad: Full-Size Img PowerPoint

  The crystal phase of prepared Eu₂O_3−δ nanoparticles was confirmed by powder X-ray diffraction (XRD). As shown in Fig. S1 (Supporting information), the XRD pattern is consistent with the standard card (JCPDS No. 34-0392), and all the diffraction peaks can be well indexed to the (211), (222), (400), (440) crystal planes of Eu₂O_3−δ [45]. It is noted that the full width at half maximum of the diffraction peaks is wide compared to the reported results, suggesting defects exist in Eu₂O_3−δ particles. This is not surprising because the low thermolysis temperature is designed to enable the formation of oxygen vacancies. The TEM image shows that the particle size of Eu₂O_3−δ is 20–30 nm (Fig. 2a), which is expected to enable high catalytic activities in the redox process. Further high-resolution TEM (Fig. 2b) combined with the planar space counting result (Fig. 2c) reveal that the diffraction fringes can be indexed to the (110) crystal plane of Eu₂O_3−δ, which is consistent with the XRD results [46]. The STEM image and corresponding element distribution characterization (Figs. 2d-g) show the homogeneous distribution of Eu and O elements within Eu₂O_3−δ nanoparticles.

  Figure 2

  

  Figure 2.  (a) The TEM image and (b) corresponding HRTEM image of Eu₂O_3−δ. (c) Interplanar spacing of Eu₂O_3−δ along the direction shown in (b). (d) STEM image and EDS elemental mappings of (e) Eu/O, (f) Eu and (g) O of Eu₂O_3−δ. Scale bar in (d) refers to 5 nm. (h) The thickness measurements of Eu₂O_3−δ/CNT interlayer. (i, j) Surface morphology of Eu₂O_3−δ/CNT interlayer.

  DownLoad: Full-Size Img PowerPoint

  The Eu₂O_3−δ/CNT interlayer structure was then further characterized by SEM, as shown in Fig. 2h, the thickness of the interlayer structure was around 43 µm. Figs. 2i and j show the surface morphology of the interlayer structure, from which it can be seen that the interlayer structure consists of interwoven CNT, and the Eu₂O_3−δ catalyst particles are dispersed in the CNT network. The CNT network enables rapid electron transfer during the reaction of the Li-S batteries, and the Eu₂O_3−δ catalyst promotes the rapid conversion of LiPSs. To further obtain insights into the spin-related electron occupation, we utilized the physical property measurement system (PPMS) to characterize the whole magnetic properties of the sample. Fig. S2a (Supporting information) shows the recorded magnetization curves in the magnetic field from −20 kOe to 20 kOe. The magnetization curve shows no hysteresis feature, suggesting the paramagnetic behaviors under ambient conditions. We also performed the temperature-dependent magnetization measurements (Fig. S2b in Supporting information) with a magnetic field of H = 1 kOe under field-colling procedures to further investigate the spin structures of M ions. Above 200 K, the susceptibilities derived from the magnetization (χ = M/H) obey a paramagnetic Curie-Weiss law: χ = C/(T − Θ), where C is Curie constant, and Θ is Curie-Weiss temperature. Therefore, charge transport during redox process can be facilitated owing to the delocalized electrons in the paramagnetic materials.

  X-ray photoelectron spectroscopy (XPS) was performed to characterize the valence state of Eu₂O_3−δ, as shown in Fig. 3. High-resolution XPS spectrum of Eu 3d (Fig. 3a) shows two characteristic peaks at 1162.8 and 1133.4 eV, ascribing to the spin orbits of Eu 3d_3/2 and Eu 3d_5/2 of Eu₂O_3−δ, respectively. Another Eu 3d_3/2 fitting peak at lower binding energy was detected, suggesting the appearance of oxygen vacancies in the Eu₂O_3−δ material. To further confirm the oxygen vacancy, high-resolution O 1s spectrum is fitted and shown in Fig. 3b. There are two characteristic peaks detected at 528.6 eV and 531.1 eV, which corresponds to the lattice oxygen and oxygen vacancies in Eu₂O_3−δ, respectively [47,48]. To demonstrate the valence evolution during thermolysis, the XPS of EuSe after thermal treatment for 30 min was also tested. As shown in Fig. S3a (Supporting information), four elements including Eu, O, C, and Se can be observed in the XPS survey spectrum of pre-processed EuSe, suggesting the phase evolution from EuSe to Eu₂O_3−δ after calcination. In Fig. S3b and c (Supporting information), mixed valence states of Eu³⁺ and Eu²⁺ are detected according to the 3d_5/2 and 3d_3/2 binding energy of the Eu element [49]. These results indicate that Eu²⁺ on the surface of EuSe is first oxidized to Eu³⁺ to generate Eu₂O_3−δ, which can be further confirmed by the fitted peaks in the O 1s spectrum (Fig. S3d in Supporting information).

  Figure 3

  

  Figure 3.  High-resolution XPS spectra (a) Eu 3d and (b) O 1s. (c) Electron paramagnetic resonance (EPR) spectra of Eu₂O_3−δ. (d) Optical photos of CNT and Eu₂O_3−δ adsorption on LiPSs and XPS spectroscopy of S 2p after Eu₂O_3−δ adsorption of LiPSs.

  DownLoad: Full-Size Img PowerPoint

  The thermolysis of EuSe can form Eu₂O_3−δ with abundant oxygen vacancies. Although the XPS result confirmed the phase evolution of EuSe, electron paramagnetic resonance (EPR) measurement was performed to further reveal the magnetic properties and oxygen vacancy characteristics of Eu₂O_3−δ. As shown in Fig. 3c, the Eu₂O_3−δ sample shows a major g-value of about 2.004, confirming the existence of oxygen vacancy [50]. The abundance of oxygen vacancy has been shown help to promote the electronic conductivity of metal oxides. As we mentioned in the former discussion, oxygen vacancies in the Eu₂O_3−δ sample are expected to show both high catalytic activities and immobilization capability to suppress the LiPSs shuttling. The visualization experiment of LiPSs adsorption was performed to demonstrate the adsorption capability of Eu₂O_3−δ (inset of Fig. 3d). The LiPSs solution with Eu₂O_3−δ becomes clearer after only 2 h, while the one with CNT remains dark yellow color after 12 h, demonstrating that Eu₂O_3−δ has a strong adsorption effect on LiPSs. Such evidence of strong LiPSs immobilization capability of Eu₂O_3−δ was further demonstrated by the ultraviolet-visible spectroscopy (UV–vis) (Fig. S4 in Supporting information) and XPS measurement of the Eu₂O_3−δ sample after adsorption of LiPSs (Fig. 3d), showing S⁰B and S⁻¹T spectrum of S at 163.0–166.0 eV and 159.0–161.0 eV that related to sulfur at different positions in LiPSs [51,52].

  A series of electrochemical performance tests are taken to evaluate the catalytic activity of Eu₂O_3−δ. The CV test is carried out in the voltage range of −1.0 V~1.0 V, and the scanning speed is 1 mV/s. To study the LiPSs conversion process and the deposition behavior of Li₂S, CNT, and Eu₂O_3−δ/CNT are used as catalysts in batteries. The Li₂S deposition curves of the CNT and Eu₂O_3−δ/CNT interlayer materials are shown in Figs. 4a and b, respectively. The result shows that the current response of Eu₂O_3−δ/CNT LSBs appears earlier than that of the CNT, suggesting its higher Li₂S deposition capacity. The decomposition process of Li₂S with different catalysts is shown in Fig. 4c, during which the Eu₂O_3−δ catalyst significantly promoted the reversible conversion process of Li₂S and rapid reaction kinetics. As shown in Fig. 4d and Fig. S5 (Supporting information), compared with CNT, the CV curves of the Li₂S₆ symmetrical battery with Eu₂O_3−δ/CNT show no distinct peak shift after the initial cycle test. These CV curves almost overlapped during the first two cycles, indicating decent electrochemical stability. In addition, from the CV oxidation–reduction peaks characteristics, it is found that the corresponding exchange current density of the Eu₂O_3−δ/CNT battery is higher than that of the CNT, further proving the strong LiPSs catalytic ability of the Eu₂O_3−δ catalyst [53-55].

  Figure 4

  

  Figure 4.  (a) Li₂S deposition curve of CNT interlayer. (b) Li₂S deposition curve of Eu₂O_3−δ/CNT interlayer. (c) Decomposition curves of Li₂S in different interlayers. (d) CV curve of Li₂S₆ symmetrical battery of Eu₂O_3−δ/CNT interlayer.

  DownLoad: Full-Size Img PowerPoint

  Then, CV was further used to study the reaction process of LSBs in a voltage range of 1.7–2.6 V with a scanning speed of 0.1 mV/s (Figs. 5a–c). All CV curves exhibit two typical reduction peaks at around 2.1 V and 2.3 V and one oxidation peak at around 2.5 V, which are related to the two-step conversion process of sulfur octagonal Li₂S (S₈²⁻ → S₄²⁻, S₄²⁻ → Li₂S₂/Li₂S and Li₂S₂/Li₂S → S₈ reactions, respectively) [56,57]. Compared with blank and CNT samples, the Eu₂O_3−δ/CNT delivers the highest peak current and the lowest potential, indicating the accelerated reaction kinetics for the oxidation of Li₂S/Li₂S₂ to S₈ [58]. To further investigate the catalytic effect of Eu₂O_3−δ on boosting the multi-step conversion of LiPSs, the CV curves under different sweep rates (0.1–1 mV/s) were displayed in Figs. 5d–f. The significantly higher peak current and smaller polarization of Eu₂O_3−δ/CNT LSBs (0.10 V and 0.01 V) confirm the accelerated redox kinetics, which would be beneficial for the LiPSs conversion. In Figs. 5g–i, the current pairs of the CV curves of different LSBs exhibit the linear relationship with the square root of the scan rate, indicating that the conversion process of LiPSs is diffusion-controlled. According to the Randles–Sevcik equation [59-61], the diffusion coefficients of lithium ions at the redox peak can be calculated separately, and the slope of the fitted curve reflects the situation of the diffusion coefficient of lithium ions. It can be seen that peak current varies linearly with the scanning rate, so that the diffusion coefficient is proportional to the slope of the fitted line. The Eu₂O_3−δ/CNT LSBs exhibit steeper slopes at various sweep rates during the two-step reduction process, suggesting the promoted conversion process of S₈ to LiPSs in the LSBs. The improvement in the conversion kinetics of LiPSs and lithium ion transport kinetics by utilizing Eu₂O_3−δ/CNT interlayer ensures a swift and uninterrupted conversion of LiPSs, preventing their accumulation, which ensures efficient utilization of catalytic sites.

  Figure 5

  

  Figure 5.  (a) CV curves of blank battery under the 1^st, 2^nd, 5^th, and 10^th cycles. (b) CV curves of CNT interlayer battery under the 1^st, 2^nd, 5^th, and 10^th cycles. (c) CV curves of Eu₂O_3−δ/CNT interlayer battery under the 1^st, 2^nd, 5^th, and 10^th cycles; Under different scanning rates (0.1, 0.2, 0.5, and 1.0 mV/s), the (d) blank battery (e) CNT interlayer battery and (f) Eu₂O_3−δ/CNT interlayer battery, as shown in Plot of the logarithm of current peak versus logarithm of scan rate at (g) Ⅰ, (h) Ⅱ and (i) Ⅲ & Ⅳ.

  DownLoad: Full-Size Img PowerPoint

  Furthermore, we studied the effect of different Eu₂O_3−δ loading on the reaction kinetics of LSBs. As shown in Fig. S6 (Supporting information), CV curves of Eu₂O_3−δ/CNT interlayer with 2.5 wt%, 5 wt%, 10 wt%, and 20 wt% Eu₂O_3−δ loading were compared. Combined with the CV result in Fig. 5c (10 wt% Eu₂O_3−δ), the polarization gradually decreases when the loading amount of Eu₂O_3−δ increases from 2.5 wt% to 10 wt%, and the charge-discharge platform becomes longer. In addition, Fig. S7 (Supporting information) shows CV curves measured at different scanning rates and corresponding kinetic analysis. It can be seen that, with the increase of scanning rate, both the low and high amounts of Eu₂O_3−δ loadings displayed a large polarization. The result indicates that an optimized amount of catalyst is required to enable high catalytic activities because a large amount of catalyst leads to an agglomeration effect, thus decreasing catalytic sites in the Eu₂O_3−δ/CNT interlayer structure.

  To further verify the effects of Eu₂O_3−δ catalysts and assess the optimal regulation design, the electrochemical performance of LSBs was investigated. As shown in Fig. 6a, rate cycling tests were conducted at 0.2, 0.5, 1, 2, 3, and 5 C. The reversible capacities of Eu₂O_3−δ/CNT LSBs are 1172.5, 873.1, 765.8, 692.2, 667.6, and 620.7 mAh/g, respectively. On the contrary, in blank battery and CNT interlayer battery, lower battery capacities are observed under the same current density. When the current density was switched back to 0.2 C, the capacity recovered to a high capacity of 991.9 mAh/g for the Eu₂O_3−δ, indicating the suppressed shuttle effect of LiPSs and good rate performance. Figs. 6b, c and Fig. S8 (Supporting information) show the charge/discharge curves of LSBs tested at different current densities. It shows that the charge/discharge curve of Eu₂O_3−δ/CNT LSBs has a longer and flatter voltage plateau with lower polarization. These results further demonstrate that the Eu₂O_3−δ/CNT interlayer can inhibit the LiPSs diffusion and enhance the conversion kinetics in LSBs. To further testify the long-term cycling stability, the cycling performances at 1 C and 2 C were compared. As shown in Fig. 6d and Fig. S9 (Supporting information), the Eu₂O_3−δ/CNT LSB delivers an impressive specific capacity of around 600 mAh/g after 300 cycles. More importantly, at a high current density of 2 C, the capacity of around 500 mAh/g remains after 200 cycles (Fig. 6e) for the Eu₂O_3−δ/CNT LSB. Comparatively, the electrochemical performance of Eu₂O_3−δ/CNT LSB loaded with different amounts of Eu₂O_3−δ was further studied. As shown in Fig. S10a (Supporting information), when the amount of Eu₂O_3−δ is increased from 2.5% to 20% at current densities of 0.2, 0.5, 1, 2, 3, and 5 C, the capacity retention and stability are lower than that of the optimized Eu₂O_3−δ amount. Through the analysis of charge-discharge curves under different current densities, when the amount of Eu₂O_3−δ is low (Fig. S10b in Supporting information), the polarization becomes larger and the battery capacity decays fast owing to the disability of immobilizing LiPSs species during electrochemical reactions. Meanwhile, for the high amount of Eu₂O_3−δ sample (Fig. S10d in Supporting information), the polarization becomes the worst, which is well consistent with the former analysis result of the catalytic effect. At last, the cycling performance of LSBs with different amounts of Eu₂O_3−δ at 1 C and 2 C current densities was tested (Figs. S11 and S12 in Supporting information). As expected, the (10 wt%) Eu₂O_3−δ/CNT LSB shows the best long-cycle performance, indicating that the shuttle effect can be effectively suppressed by the optimized amount of catalysts, and the sulfur loss is significantly reduced, which results in the reversible and durable capacities of the LSBs. As shown in Fig. 6f, charge/discharge cycles were realized with a lean electrolyte of 9.5 µL/mg and a high loading of 6.63 mg/cm². The capacity is maintained at 6.29 mAh/cm² after 20 cycles, exceeding the current requirement of 4.0 mAh/cm² for battery applications. As shown in Fig. 6g, a high surface capacity of 7.94 mAh/cm² and a high specific capacity of 1197.24 mAh/g can be delivered in the first cycle, further demonstrating the great potential of Eu₂O_3−δ/CNT interlayer in LSBs (Fig. 6g).

  Figure 6

  

  Figure 6.  (a) Rate cycle performance of different LSBs. (b) Charge and discharge curve of blank battery at 0.2–5 C. (c) Charge and discharge curve of Eu₂O_3−δ/CNT interlayer battery at 0.2–5 C. (d) Long cycle charge and discharge test of different batteries at 1 C. (e) Cyclic charge and discharge test of different batteries at high rate 2 C. (f) Charge-discharge cycle test of Eu₂O_3−δ/CNT interlayer under 6.63 mg/cm² high loading. (g) First charge-discharge curve at 0.05 C under 6.63 mg/cm².

  DownLoad: Full-Size Img PowerPoint

  In summary, we report a Eu₂O_3−δ/CNT interlayer structure to suppress the LiPSs shuttle effect in LSBs. The study demonstrates a new conceptual strategy for precisely tuning in catalysts to achieve superior LSB performance. The oxygen-deficient Eu₂O_3−δ nanoparticles showed a high catalytic effect during LiPSs redox process in LSBs, thus regulating the LiPSs redox chemistry and immobilizing LiPSs through spin interaction. A series of electrochemical measurements demonstrates that the Eu₂O_3−δ also enables fast ion diffusion and facilitates the rapid and continuous transfer of LiPSs, preventing their accumulation and facilitating the efficient utilization of active sites, thereby reducing the energy barrier for LiPSs conversion and the undesired shuttle process. This work provides new insights into the precise regulation of rare earth metal oxides as catalysts and highlights the positive role of the rich vacancy structure of rare earth metal oxides in a rational design principle for the development of high-performance LSB technologies.

  Declaration of competing interest

  The authors declare that they have no conflict of interest.

  CRediT authorship contribution statement

  Ming Xu: Writing – review & editing, Writing – original draft, Validation, Formal analysis, Data curation, Conceptualization. Teng Deng: Writing – original draft, Formal analysis. Chenzhaosha Li: Writing – original draft, Formal analysis. Hongyang Zhao: Writing – review & editing. Juan Wang: Resources. Yatao Liu: Resources. Jianan Wang: Validation, Resources. Guodong Feng: Resources. Na Li: Resources, Formal analysis, Data curation. Shujiang Ding: Validation, Supervision, Software, Resources. Kai Xi: Writing – review & editing, Validation, Supervision, Resources.

  Acknowledgments

  The authors deeply acknowledge the financial support from the National Natural Science Foundation of China (Nos. 52104312, 22278329, 22271229, 22105153), Qin Chuangyuan Talent Project of Shaanxi Province (Nos. 2021QCYRC4-43, QCYRCXM-2022-308) and the State Key Laboratory for Electrical Insulation and Power Equipment (No. EIPE23125) and the authors deeply acknowledged teachers at Instrumental Analysis Center of Xi'an Jiaotong University for Characterizations tests.

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110372.

  
  
• 1. [1]

     S. Zhou, J. Shi, S. Liu, et al., Nature 621 (2023) 75–81. doi: 10.1038/s41586-023-06326-8

  2. [2]

     S. Yu, Y. Zhang, S. Yang, et al., Chin. Chem. Lett. 34 (2023) 107911.

  3. [3]

     L. Chen, G. Cao, Y. Li, et al., Nano-Micro Lett. 16 (2024) 97. doi: 10.1007/s40820-023-01299-9

  4. [4]

     Y. Zhao, C. Liu, C. Zha, et al., Chin. Chem. Lett. 34 (2023) 108189.

  5. [5]

     Q. Pang, X. Liang, C.Y. Kwok, L.F. Nazar, Nat. Energy 1 (2016) 16132.

  6. [6]

     H. Wang, X. Lai, C. Chen, et al., Chin. Chem. Lett. 35 (2024) 108473.

  7. [7]

     C. Zhao, G.L. Xu, T. Zhao, K.J. Amine, Angew. Chem Int. Ed. 59 (2020) 17634–17640. doi: 10.1002/anie.202007159

  8. [8]

     R. Liu, Z. Wei, L. Peng, et al., Nature 626 (2024) 98–104.

  9. [9]

     W. Yao, J. Xu, L. Ma, et al., Adv. Mater. 35 (2023) 2212116. doi: 10.1002/adma.202212116

  10. [10]

      Y. Mo, L. Liao, D. Li, et al., Chin. Chem. Lett. 34 (2023) 107130.

  11. [11]

      Y. Wang, H. Chen, F. Yu, et al., Chin. Chem. Lett. 35 (2024) 109001.

  12. [12]

      X. Sun, S. Liu, W. Sun, C. Zheng, Chin. Chem. Lett. 34 (2022) 107501.

  13. [13]

      C. Wang, R. Liu, W. Liu, et al., Adv. Funct. Mater. 34 (2024) 2316221.

  14. [14]

      Y. Guo, R. Khatoon, J. Lu, et al., Carbon Energy 3 (2021) 841–855. doi: 10.1002/cey2.145

  15. [15]

      P. Geng, M. Du, X. Guo, et al., Energy Environ. Mater. 5 (2022) 599–607. doi: 10.1002/eem2.12196

  16. [16]

      M.L. Wang, D. Yin, Y.D. Cao, et al., Chin. Chem. Lett. 33 (2022) 4350–4356.

  17. [17]

      X. Tian, Y. Zhou, B. Zhang, N.B. Selabi, G. Wang, J. Energy Chem. 74 (2022) 239–251.

  18. [18]

      P. Feng, W. Hou, Z. Bai, et al., Chin. Chem. Lett. 34 (2023) 107427.

  19. [19]

      A. Kim, S.H. Oh, A. Adhikari, et al., J. Mater. Chem. A 11 (2023) 7833–7866. doi: 10.1039/d2ta09266b

  20. [20]

      S. Xia, J. Song, Q. Zhou, et al., Adv. Sci. 10 (2023) 2301386.

  21. [21]

      Y. Song, M. Zhou, Z. Chen, et al., Chin. Chem. Lett. 35 (2024) 109200.

  22. [22]

      J. Zhou, S. Sun, X. Zhou, et al., Chem. Eng. J. 487 (2024) 150574.

  23. [23]

      X. Zhou, X. Li, Z. Li, et al., Mater. Today Energy 26 (2022) 100990.

  24. [24]

      H. Wang, C. Xu, X. Du, et al., Chem. Eng. J. 471 (2023) 144338.

  25. [25]

      B.J. Lee, C. Zhao, J. Yu, et al., Nat. Commun. 13 (2022) 4629.

  26. [26]

      X. Dai, G. Lv, Z. Wu, et al., Adv. Energy Mater. 13 (2023) 2300452. doi: 10.1002/aenm.202300452

  27. [27]

      D.W. Kim, C. Senthil, S.M. Jung, et al., Energy Storage Mater. 47 (2022) 472–481.

  28. [28]

      D. Son, Park H, W. Lim, et al., ACS Nano 17 (2023) 25507–25518. doi: 10.1021/acsnano.3c09333

  29. [29]

      X. Men, T. Deng, X. Jiao, et al., Electrochim. Acta 431 (2022) 141100.

  30. [30]

      L. Ma, Y. Wang, Z. Wang, et al., ACS Nano 17 (2023) 11527–11536. doi: 10.1021/acsnano.3c01469

  31. [31]

      W. Sun, S. Liu, Y. Li, et al., Adv. Funct. Mater. 32 (2022) 2205471.

  32. [32]

      S. Wang, H. Li, G. Zhao, et al., Rare Met. 42 (2023) 515–524. doi: 10.1007/s12598-022-02140-9

  33. [33]

      X. Zhang, M. Lei, S. Tian, J.G. Wang, Rare Met. 43 (2024) 624–634.

  34. [34]

      J. He, A. Manthiram, Adv. Energy Mater. 10 (2020) 2002654.

  35. [35]

      K. Zou, X. Chen, W. Jing, et al., Energy Storage Mater. 48 (2022) 133–144.

  36. [36]

      C. Zhao, B. Jiang, Y. Huang, et al., Energy Environ. Sci. 16 (2023) 5490–5499. doi: 10.1039/d3ee01774e

  37. [37]

      Y. Wang, R. Zhang, J. Chen, H. Wu, S. Ding, Adv. Energy Mater. 9 (2019) 1900953.

  38. [38]

      L. He, X. Zhang, D. Yang, et al., Nano Lett. 23 (2023) 7411–7418. doi: 10.1021/acs.nanolett.3c01838

  39. [39]

      L. Sun, X. Meng, J. Zhang, et al., J. Power Sources 575 (2023) 233173.

  40. [40]

      E. Zhang, X. Hu, L. Meng, et al., J. Am. Chem. Soc. 144 (2022) 18995–19007. doi: 10.1021/jacs.2c07655

  41. [41]

      R. Si, Y. Zhang, L. You, C. Yan, Angew. Chem Int. Ed. 44 (2010) 3256–3260.

  42. [42]

      P. Zeng, M. Chen, J. Luo, et al., ACS Appl. Mater. Interfaces 11 (2019) 42104–42113. doi: 10.1021/acsami.9b13533

  43. [43]

      S. Shi, C. Ouyang, Q. Fang, et al., EPL 83 (2008) 69001. doi: 10.1209/0295-5075/83/69001

  44. [44]

      W. Sllinger, W. Heiss, R.T. Lechner, K. Rumpf, G.J. Springholz, Phys. Rev. B 81 (2009) 2149.

  45. [45]

      L. Peng, Z. Yu, M. Zhang, et al., Nanoscale 13 (2021) 16696–16704. doi: 10.1039/d1nr04855d

  46. [46]

      C. Chen, J. Long, K. Shen, X. Liu, W. Zhang, ACS Appl. Mater. Interfaces 14 (2022) 38677–38688. doi: 10.1021/acsami.2c07373

  47. [47]

      J.P. Baltrus, M.J. Keller, Surf. Sci. Spectra 26 (2019) 014001.

  48. [48]

      F. Mercier, C. Alliot, L. Bion, et al., J. Electron. Spectrosc. Relat. Phenom. 150 (2006) 21–26.

  49. [49]

      N. Li, L. Qin, H. Zhao, et al., Chem. Mater. 28 (2016) 2507–2510. doi: 10.1021/acs.chemmater.6b00120

  50. [50]

      X. Wang, X. Zhang, Y. Zhao, et al., Angew. Chem Int. Ed. 62 (2023) e202306901.

  51. [51]

      L. Ni, J. Gu, X. Jiang, et al., Angew. Chem Int. Ed. 62 (2023) e202306528.

  52. [52]

      L. Ni, G. Yang, Y. Liu, Z. Wu, Y. Wei, ACS Nano 15 (2021) 12222–12236. doi: 10.1021/acsnano.1c03852

  53. [53]

      L. Chen, Y. Sun, X. Wei, et al., Adv. Mater. 35 (2023) 2300771.

  54. [54]

      L. Ren, K. Sun, Y. Wang, et al., Adv. Mater. 36 (2024) 2310547.

  55. [55]

      S. Fu, C. Hu, J. Li, et al., J. Energy Chem. 88 (2024) 82–93.

  56. [56]

      K. Xi, D. He, C. Harris, et al., Adv. Sci. 6 (2019) 1800815.

  57. [57]

      T. Jeong, D. Choi, H. Song, et al., ACS Energy Lett. 2 (2017) 327–333. doi: 10.1021/acsenergylett.6b00603

  58. [58]

      X. Zhang, T. Yang, J. Liu, et al., Small 15 (2024) 2311086.

  59. [59]

      D. He, X. Liu, X. Li, P. Lyu, Z. Rao, Chem. Eng. J. 419 (2021) 129509.

  60. [60]

      B. Wang, L. Wang, Y. Kong, et al., Adv. Energy Mater. 13 (2023) 2300590.

  61. [61]

      L. Liang, L. Niu, T. Wu, D. Zhou, Z. Xiao, ACS Nano 16 (2022) 7971–7981. doi: 10.1021/acsnano.2c00779

• 

  Figure 1  (a) The synthesis route of Eu₂O_3−δ/CNT interlayer. (b) Schematic diagram of Eu₂O_3−δ polarity anchoring LiPSs and catalytic conversion of LiPSs to Li₂S.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) The TEM image and (b) corresponding HRTEM image of Eu₂O_3−δ. (c) Interplanar spacing of Eu₂O_3−δ along the direction shown in (b). (d) STEM image and EDS elemental mappings of (e) Eu/O, (f) Eu and (g) O of Eu₂O_3−δ. Scale bar in (d) refers to 5 nm. (h) The thickness measurements of Eu₂O_3−δ/CNT interlayer. (i, j) Surface morphology of Eu₂O_3−δ/CNT interlayer.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  High-resolution XPS spectra (a) Eu 3d and (b) O 1s. (c) Electron paramagnetic resonance (EPR) spectra of Eu₂O_3−δ. (d) Optical photos of CNT and Eu₂O_3−δ adsorption on LiPSs and XPS spectroscopy of S 2p after Eu₂O_3−δ adsorption of LiPSs.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Li₂S deposition curve of CNT interlayer. (b) Li₂S deposition curve of Eu₂O_3−δ/CNT interlayer. (c) Decomposition curves of Li₂S in different interlayers. (d) CV curve of Li₂S₆ symmetrical battery of Eu₂O_3−δ/CNT interlayer.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) CV curves of blank battery under the 1^st, 2^nd, 5^th, and 10^th cycles. (b) CV curves of CNT interlayer battery under the 1^st, 2^nd, 5^th, and 10^th cycles. (c) CV curves of Eu₂O_3−δ/CNT interlayer battery under the 1^st, 2^nd, 5^th, and 10^th cycles; Under different scanning rates (0.1, 0.2, 0.5, and 1.0 mV/s), the (d) blank battery (e) CNT interlayer battery and (f) Eu₂O_3−δ/CNT interlayer battery, as shown in Plot of the logarithm of current peak versus logarithm of scan rate at (g) Ⅰ, (h) Ⅱ and (i) Ⅲ & Ⅳ.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  (a) Rate cycle performance of different LSBs. (b) Charge and discharge curve of blank battery at 0.2–5 C. (c) Charge and discharge curve of Eu₂O_3−δ/CNT interlayer battery at 0.2–5 C. (d) Long cycle charge and discharge test of different batteries at 1 C. (e) Cyclic charge and discharge test of different batteries at high rate 2 C. (f) Charge-discharge cycle test of Eu₂O_3−δ/CNT interlayer under 6.63 mg/cm² high loading. (g) First charge-discharge curve at 0.05 C under 6.63 mg/cm².

  下载: 全尺寸图片 幻灯片
•