Carbon nanotube (99.9%) was purchased from Aladdin. CoCl₂·6H₂O (99%), NiCl₂·6H₂O (99%), NaH₂PO₂·2H₂O (99%), Na₃C₆H₅O₇·2H₂O (99%), and NaOH (98%) were purchased from InnoChem. Sulfur powder (99.5%) and ethanol (AR) were purchased from Kermel. Polyvinylidene fluoride (PVDF), polytetrafluoroethylene (PTFE), and the commercial membrane (Celgard 2500) were purchased from Canrd Technology Co. Ltd. Reduced graphene oxide (99.9%) was purchased from Carmery. The Li₂S₆ electrolyte (0.2 M) and Li₂S₈ electrolyte (0.2 M) were sourced from DODOCHEM. Deionized water was produced in-house.

Nickel-cobalt (Ni-Co) nanosheets were synthesized via a typical hydrothermal method. In a standard procedure, Na₃C₆H₅O₇·2H₂O (10 mmol), CoCl₂·6H₂O (2 mmol), NiCl₂·6H₂O (0.5 mmol), and NaH₂PO₂·2H₂O (10 mmol) were dissolved in 80 mL of deionized water under magnetic stirring at 400 rpm and room temperature (~25 ^oC). The total time for complete dissolution is about 10 min. Subsequently, 10 mL of NaOH solution (10 M) was added dropwise, yielding final precursor concentrations of 22.22 mM Co²⁺ and 5.56 mM Ni²⁺. After 10 min of stirring, the mixture was hydrothermally treated in a sealed 100 mL Teflon-lined autoclave at 150 °C for 24 h and then naturally cooled to room temperature. The resulting grey precipitate was collected by centrifugation, washed repeatedly with ethanol and deionized water, and dried under vacuum at 60 °C for 5 h. Co nanosheets were synthesized under identical conditions, but without adding NiCl₂·6H₂O.

Ni-CoTe was synthesized by placing Ni-Co and Te powder (in a mass ratio of 1:3) into separate quartz boats. The boats were positioned in the first and second temperature zones of a quartz tube furnace. A mixed gas of Ar/H₂ (volume ratio 9:1) was introduced into the tube. The tube was ramped to 200 °C at 5 °C·min^-1 and subsequently annealed at this temperature for 4 h to yield Ni-CoTe. CoTe was prepared following the same procedure.

The Ni-CoTe interlayer was fabricated using a straightforward surface coating technique. Typically, a composite consisting of 70 wt.% Ni-CoTe, 20 wt.% Gr, and 10 wt.% PVDF was homogenized in N-methyl-2-pyrrolidone (NMP) via vigorous mechanical stirring to yield a well-dispersed slurry. The slurry was coated onto a clean Celgard 2500 separator and subsequently vacuum-dried at 55 °C for 8 h. After that, the separator was punched into 16-mm-diameter circular pieces. The areal loading of the interlayer was approximately 0.34 mg cm^-2. The CoTe and Gr interlayers were synthesized using the same method.

The carbon nanotubes/sulfur (CNTs/S) composite was synthesized using a simple melt-diffusion method. Specifically, CNTs and sulfur were combined in a mass ratio of 3:7, ground evenly, and then annealed at 155 °C for 12 h. The CNTs/S cathode was prepared by mixing CNTs/S, CNTs, and PVDF (8:1:1 w/w) in NMP. The resulting slurry was coated onto carbon-coated Al foil and vacuum-dried at 55 °C for 12 h.

The composite was obtained by mixing CNT/S, CNTs, and PTFE in ethanol, using a weight ratio of 80:15:5. The ethanol was subsequently evaporated using a dryer until the mixture achieved a consistency similar to silly putty. The resultant slurry was then rolled into a thin sheet and subsequently cut into 10 mm diameter circular electrodes. These electrodes were dried at 55 °C for 12 h.

The composition of the products was analyzed using X-ray diffraction (XRD) (Bruker D8 ADVANCE, Bruker AXS GmbH, Germany) and Raman spectroscopy (LabRAM Odyssey, HORIBA France SAS, France). The chemical composition and valence states of the elements were examined by X-ray photoelectron spectroscopy (XPS) (SCIENTIFIC ESCALAB 250Xi, Thermo Fisher Scientific, USA). The microstructure and distribution of the materials were observed with a field emission scanning electron microscopy (SEM) (JSM-7001F, JEOL Ltd., Japan). Transmission electron microscopy (FEI Tecnai G2 F20, FEI Company, USA) was used to further observe nanoscale size, shape, and structure, as well as any potential crystal defects. Electron paramagnetic resonance (Bruker EMX PLUS, Bruker BioSpin GmbH, Germany) spectroscopy was employed to detect the formation of Te vacancies, with the magnetic field set between 2,000 and 5,000 G and a scan rate of 10 G s^-1.

Battery fabrication was carried out inside an argon-purged glove box, employing a conventional Li-S electrolyte solution and Li foil as the counter electrode. The battery performance was characterized by galvanostatic cycling with the Neware test system (Neware Technology Limited, China) within the voltage range of 1.7-2.7 V, by cyclic voltammetry (CV) with the Ivium workstation (Ivium Technologies B.V., Netherlands), and by electrochemical impedance spectroscopy (EIS) with the DH7000C workstation (Jiangsu Donghua Analytical Instrument Co., Ltd., China) over a frequency range of 0.1 Hz to 10⁴ Hz.

An H-type glass cell (5 mL) was employed to evaluate the polysulfide blocking capability of different modified separators. The cell consisted of two chambers separated by the modified separator. The left chamber was filled with a 20 mM Li₂S₆ solution (in LiTFSI electrolyte), while the right chamber contained pure LiTFSI electrolyte without polysulfides. The color change or concentration variation of the solution in the right chamber was monitored after different standing durations inside an argon-filled glovebox.

The adsorption capability of the different modified separators for polysulfides was further characterized using ultraviolet-visible (UV-vis) absorption spectroscopy. The UV-vis spectra were recorded to analyze the absorption intensity of the Li₂S₆ solution before and after adsorption, enabling quantitative evaluation of polysulfide removal efficiency.

The battery was constructed with Ni-CoTe or CoTe composites deposited onto 10-mm-diameter carbon paper as the cathode and Li foil as the anode. 20 µL of 0.2 M Li₂S₈ solution was used as the catholyte, and 20 µL of conventional Li-S electrolyte was used as the anolyte. The assembled cells were discharged via a galvanostatic step at 0.112 mA until the voltage decreased to 2.06 V, and then held potentiostatically at 2.05 V for Li₂S deposition until the current dropped to 10^-5 A.

A Li-S battery incorporating a quartz window for in situ Raman spectroscopy analysis at a 532 nm laser wavelength was assembled using Ni-CoTe or CoTe as the interlayer. The battery was evaluated at 0.2 C.

All density functional theory (DFT) computations were carried out with the Vienna Ab-initio Simulation Package (VASP) code, while the exchange-correlation functional was described via the generalized gradient approximation (GGA) using the Perdew-Burke-Ernzerhof (PBE) parametrization. The (103) crystal surface of CoTe was selected for its low surface energy and structural stability. The surface slab was constructed comprising 24 Co and 24 Te atoms, using the optimized bulk lattice constants. A vacuum layer of 15 Å was applied along the surface-normal direction to eliminate periodic interactions. Structural relaxations were performed using a 2 × 2 × 1 Monkhorst-Pack k-point mesh and a 300 eV plane-wave cutoff. Convergence thresholds were set to 10^-5 eV for total energy and 0.01 eV Å^-1 for the maximum atomic force.

The adsorption energy ∆E_ad was calculated as^[29]:

(1) $$ \Delta E_{a d}=E_{(s u r f+a d)}-E_{s u r f}-E_{a d} $$

In which the symbols E_{(surf + ad)}, E_surf, and E_ad denote the energies of the LiPS-adsorbed surface, the clean substrate, and the free LiPSs, respectively.

The Gibbs free energy is evaluated via the following equation^[29]:

(2) $$ \Delta G\left(S_{8}-L i_{2} S_{8}\right): E_{\left(L i_{2} S_{8}\right)}-E_{S_{8}}-2 E_{\left(L i^{+}\right)} $$

(3) $$ \Delta G\left(L i_{2} S_{8}-L i_{2} S_{6}\right): E_{\left(L i_{2} S_{6}\right)}-E_{L i_{2} S_{8}}+0.25 E_{S_{8}} $$

(4) $$ \Delta G\left(L i_{2} S_{6}-L i_{2} S_{4}\right): E_{\left(L i_{2} S_{4}\right)}-E_{L i_{2} S_{6}}+0.25 E_{S_{8}} $$

(5) $$ \Delta G\left(L i_{2} S_{4}-L i_{2} S_{2}\right): E_{\left(L i_{2} S_{2}\right)}-E_{L i_{2} S_{4}}+0.25 E_{S_{8}} $$

(6) $$ \Delta G\left(L i_{2} S_{2}-L i_{2} S\right): E_{\left(L i_{2} S\right)}-E_{L i_{2} S_{2}}+0.25 E_{S_{8}} $$

Interfacial interactions involving surface adsorption and charge transfer in Li-S batteries critically impact the sulfur conversion process^[30,31]. Thus, regulating interfacial interactions at the substrate surface is essential for sulfur redox reactions^[32,33]. In analogy to the d-band center theory for metals, the p-band center of nonmetallic anions is a key descriptor that significantly influences the adsorption energy of reaction intermediates, similar to how the d-band center affects proton adsorption in electrocatalysis^[29,34].

To investigate the relationship between the Te p-band electronic structure and sulfur redox kinetics, a series of M-CoTe (M= V, Mn, Fe, Ni, Cu, Zn) compounds were designed to regulate the Te electronic structure. The calculated projected density of states (PDOS) in Figure 1A shows that the Te p-band center exhibits the most significant upward shift toward the Fermi energy level (E_F) upon Ni doping, which strengthens interfacial electronic interactions by modulating the Li-Te orbital coupling, thereby optimizing the adsorption strength between M-CoTe and LiPSs. Simultaneously, the d-band center of the Co site shifts toward the E_F upon Ni doping, from -2.937 eV in pure CoTe to -2.858 eV in Ni-CoTe, which tunes the occupancy of the antibonding orbitals between Co and LiPSs, adjusting the adsorption strength to an optimal level [Supplementary Figure 1]. The strategy of concurrently regulating Co d-band and Te p-band through Ni doping prevents excessive binding of reaction intermediates, thus preserving the availability of active catalytic sites^[35]. Furthermore, the adsorption configurations of ^*S₈ and Li₂S_x (x = 1, 2, 4, 6, 8) on Ni-CoTe suggest that the Co, Ni, and Te sites form Co-S, Ni-S, and Li-Te bonds with the S and Li atoms in the polysulfides, whereas only Co-S and Li-Te bonds are observed on the CoTe catalyst [Supplementary Figures 2 and 3]. The adsorption energies of ^*S₈, ^*Li₂S₈, ^*Li₂S₆, ^*Li₂S₄, ^*Li₂S₂, and ^*Li₂S on the Ni-CoTe surface are -0.95, -1.54, -1.49, -1.84, -2.08, and -3.01 eV, respectively, which are higher than those on the CoTe surface [Figure 1B]. The enhanced adsorption of LiPSs on Ni-CoTe significantly optimized the sulfur redox kinetics.

Ni-CoTe nanosheets were synthesized through a hydrothermal method, followed by a chemical vapor deposition (CVD) step, with detailed preparation protocols provided in Supplementary Figure 4. For comparative purposes, CoTe and V-CoTe nanosheets were synthesized using the same procedure. SEM images reveal that the initial Ni-Co nanosheets exhibit a hexagonal structure [Supplementary Figure 5]. Subsequent SEM analyses [Figure 2A] confirm that the nanosheets retain their hexagonal morphology after the CVD treatment. The high-resolution TEM (HRTEM) image of Ni-CoTe [Figure 2B] and its inverse fast-Fourier-transform (IFFT) pattern [Figure 2C] present a clear crystalline structure, with a lattice spacing of 0.158 nm corresponding to the (103) facet of CoTe. XRD confirmed the successful synthesis and crystal phase of the as-prepared Ni-CoTe. As shown in Figure 2D, the predominant signals match well with the characteristic peaks of CoTe (PDF#34-0402), consistent with the HRTEM results. The clear diffraction peaks indicate good crystallinity, with peaks at 31.3º, 42.3º, 46.6º, and 58.2° corresponding to the (101), (102), (110), and (103) planes of orthorhombic CoTe. It is worth noting that introducing a small amount of Ni into CoTe slightly shifted the main XRD peak without creating new phases, indicating successful intercalation into the crystal lattice^[36].

XPS was then employed to analyze the composition, elemental valence states, and electron transfer in Ni-CoTe [Figure 2E and F, Supplementary Figure 6]. The Ni-CoTe spectra exhibit Co 2p peaks that are shifted by +0.8 eV to higher binding energy in comparison to those in CoTe, while the Te 3d spectra demonstrate a slight downshift of -0.4 eV to lower binding energy. These alterations in binding energy indicate substantial electronic coupling between Ni and CoTe, suggesting electron transfer from Ni to Te and modifying the electronic structure of CoTe, further corroborating the successful incorporation of Ni into CoTe^[37]. Furthermore, in contrast to the flat Electron paramagnetic resonance (EPR) signal observed in CoTe, the Ni-CoTe sample exhibited an unpaired-electron peak with a g-value of 2.000, attributed to the removal of Te atoms [Supplementary Figure 7]^[38]. The enhanced signal intensity in Ni-CoTe indicates a higher concentration of unpaired electrons, confirming the introduction of Te vacancies induced by Ni doping. This result also suggests possible electron transfer between the defect sites and metal centers in Ni-CoTe, which plays a key role in boosting the catalytic activity for optimizing sulfur redox kinetics.

Li-S batteries were assembled using CNTs/S cathodes and different interlayers [Ni-CoTe, V-CoTe, or CoTe-modified polypropylene (PP) separators] with lithium foil as the reference anode. The rate performance of the three batteries was systematically compared [Supplementary Figure 8]. The Ni-CoTe battery consistently achieved the highest capacities at all rates and displayed a reduced overpotential (ΔE = 151.5 mV) compared with the V-CoTe (177.9 mV) and CoTe (173.6 mV) batteries. This experimental trend corroborates the theoretical prediction that the doping of Ni causes Te p-band center to exhibit the most significant upward shift, which significantly optimizes the sulfur redox kinetics. Based on the theoretical calculations and electrochemical results, this study investigates the catalytic activity of Ni-CoTe for sulfur redox electrocatalysis in the subsequent discussion.

The catalytic activity of the as-prepared Ni-CoTe nanosheets was evaluated by performing three consecutive CV measurements on the Ni-CoTe, CoTe, and Gr electrodes [Supplementary Figure 9]. The nearly identical CV plots for the three batteries indicate reversible electrochemical properties. Figure 3A shows the first-cycle CV plots for the three batteries. In the SRR process, two cathodic peaks at ≈2.30 V (C₁) and 2.05 V (C₂) indicate the conversion of S₈ to soluble Li₂S_n and then to Li₂S. During the sulfur evolution reaction, two anodic peaks at ≈2.30 V (A₁) and 2.36 V (A₂) correspond to the conversion of Li₂S back to Li₂S_n, and finally to S₈. Notably, the Ni-CoTe battery exhibited the highest current response and largest peak areas for both cathodic and anodic peaks, suggesting enhanced electrochemical reaction kinetics^[39]. This enhancement is attributed to Ni doping-induced Te vacancies and the regulation of the p-band center of CoTe.

Tafel slopes obtained from CV curves were further fitted to confirm the high electrocatalytic activity of the Ni-CoTe catalysts [Figure 3B and Supplementary Figure 10]. The Ni-CoTe battery exhibited a lower Tafel slope than the CoTe and Gr batteries in both the reduction (Li₂S_n → Li₂S) and the oxidation (Li₂S → soluble Li₂S_n) processes. To evaluate the reproducibility and reliability of the Tafel slope analysis, the average Tafel slopes from the three cycles are shown in Figure 3C alongside the cycle-to-cycle variation represented by error bars. The less-than-4% bias indicated by these bars confirms the strong reliability of the measured data. The reduction in Tafel slopes observed for both the oxidation and reduction processes in the Ni-CoTe battery confirms its bidirectional catalytic capability toward LiPS conversion^[40,41].

Li₂S precipitation/dissolution tests were further employed to assess the liquid-solid and solid-liquid conversion efficiency facilitated by the Ni-CoTe catalyst^[42]. Coin cells were assembled utilizing CP@Ni-CoTe, CP@CoTe, and CP@Gr (where Ni-CoTe nanosheets, CoTe nanosheets, or graphene were loaded onto carbon fiber paper, respectively denoted as CP@Ni-CoTe, CP@CoTe, and CP@Gr) as cathodes, lithium foil as the anode, and a Li₂S₈/tetraglyme solution as the catholyte. The Li₂S precipitation capacity was quantitatively determined based on Faraday’s law [Supplementary Figure 11]. As shown in Figure 3D-F and Supplementary Figure 12, the CP@Ni-CoTe cathode demonstrated the highest capacity of 246 mAh g^-1 and the earliest nucleation time (1947 s). Furthermore, in contrast to the pronounced Li₂S aggregation observed on the CP@CoTe and CP@Gr electrodes, SEM images in Supplementary Figure 13 show a uniform Li₂S deposition on the CP@Ni-CoTe electrode. This finding provides direct evidence for the accelerated conversion of LiPSs to Li₂S enhanced by the Ni-CoTe catalyst^[17]. Additionally, the CP@Ni-CoTe electrode demonstrated the highest Li₂S dissolution capacity and the earliest onset of Li₂S dissolution, relative to those of the CP@CoTe and CP@Gr electrodes [Supplementary Figure 14]. These results confirm that the Ni-CoTe nanosheets function as bidirectional catalysts, accelerating both the reduction of LiPSs and the oxidation of Li₂S^[43].

The SRR kinetics (E_a) at a given voltage were experimentally determined by fitting the charge transfer resistance (R_ct), measured at different temperatures using EIS, to the Arrhenius equation [Figure 3G and H, Supplementary Figures 15-17]^[44,45]. To ensure voltage stability during the specified redox reaction, the battery was initially discharged to a preset potential and subsequently maintained at that level via chronoamperometric control. Following this, EIS measurements were carried out across a range of temperatures, with the thermal environment precisely controlled by a variable-temperature thermotank. As shown in Figure 3I, the Ni-CoTe battery exhibited a lower E_a of 0.59 eV, in contrast to the CoTe battery (0.68 eV), suggesting a more efficient conversion of S₈ to Li₂S_n facilitated by the Ni-CoTe catalyst^[44]. When the voltage decreased to 2.0 V, corresponding to the conversion of RDS (Li₂S_n to Li₂S₂/Li₂S), the E_a for the Ni-CoTe battery increased marginally to 0.65 eV. However, the CoTe battery demonstrated a higher E_a of 0.73 eV at this voltage, resulting in the accumulation of LiPSs and thus the shuttle effect, ultimately diminishing sulfur utilization. The substantial reduction in E_a of RDS, facilitated by the Ni-CoTe catalyst, significantly mitigates the accumulation of Li₂S_n and their diffusion to the lithium anode.

The rate performance of Li-S batteries was compared at rates ranging from 0.2 to 5.0 C (1 C = 1,675 mA g^-1) to evaluate the sulfur redox chemistry [Figure 4A] under conditions of an areal sulfur loading of ~1.0 mg cm^-2 and 20 μL of electrolyte on the cathode side. The Ni-CoTe battery delivered a significantly higher initial capacity of 1,293.2 mAh g^-1 compared to the CoTe (1,224.0 mAh g^-1) and the Gr (1,106.5 mAh g^-1) batteries. Upon increasing the rate to 0.5, 1.0, 2.0, 3.0, 4.0, and 5.0 C, the Ni-CoTe battery respectively demonstrated elevated capacities of 1,048.6, 911.3, 821.5, 773.1, 735.9, and 688.0 mAh g^-1, much higher than those of other batteries, suggesting efficient utilization of active sulfur even at high current density^[46,47].

The overpotential (∆E) of the Ni-CoTe battery, calculated from the first-cycle charge/discharge plateaus, was 142 mV, notably lower than the values for the CoTe (152 mV) and Gr (165 mV) batteries [Figure 4B and C]. Furthermore, the differential plots corresponding to the charge/discharge profiles of Ni-CoTe, CoTe, and Gr batteries have been analyzed [Supplementary Figure 18], further revealing that the Ni-CoTe battery exhibits a lower overpotential compared to the CoTe battery and the Gr battery. The lower overpotential suggests that the electrochemical kinetics are more favorable with the Ni-CoTe catalyst^[48]. Besides, the shuttle constant (k_s) obtained from the charge/discharge profiles is much lower for the Ni-CoTe battery (0.054 h^-1) than that for the CoTe (0.070 h^-1) and Gr (0.126 h^-1) batteries [Supplementary Figures 19 and 20], which indicates that the shuttle effect was effectively inhibited with the Ni-CoTe catalyst, enhancing the utilization of sulfur^[6].

To directly assess the impact of catalysis on battery lifespan, cycling performance at 2.0 C was compared. As shown in Figure 4D, the Ni-CoTe battery achieved a high initial specific capacity of 921.9 mAh g^-1 at 2.0 C, with capacity retention reflected in a low decay rate of 0.049 % per cycle over 1,000 cycles. Notably, this long-term stability test was conducted under conditions of low-sulfur loading (0.88 mg cm^-2) and excess electrolyte (20 μL on the cathode side), which were intentionally chosen to evaluate the intrinsic catalytic activity of the Ni-CoTe catalyst. In sharp contrast, the CoTe battery suffered from rapid capacity fading, declining to 461.8 mAh g^-1 with a decay rate of 0.21% per cycle after 200 cycles. Furthermore, the Ni-CoTe battery exhibited more stable Coulombic efficiency than the CoTe battery, indicating that polysulfide shuttling was significantly suppressed. It is widely considered that both the increase in areal sulfur loading and the decrease in electrolyte dosage are essential for the prospective application of Li-S batteries. Therefore, batteries with a sulfur loading as high as 10.0 mg cm^-2 and an E/S ratio of only 5.0 µL mg^-1 were constructed. As shown in Figure 4E, with an initial capacity of 1,350.5 mAh g^-1 (13.5 mAh cm^-2) at 0.02 C, the Ni-CoTe battery also showed a high areal capacity of 10.9 mAh cm^-2 when tested at 0.05 C. Moreover, a capacity of 8.36 mAh cm^-2 was maintained over 100 cycles, affording a retention of 76.7%; the initial capacity and its retention capability at 0.05 C were much superior to those of the CoTe battery, indicating improved sulfur utilization with the Ni-CoTe catalyst. The electrochemical performance surpassed most high-sulfur-loaded Li-S batteries reported in previous literature [Supplementary Table 1], demonstrating promising prospects for practical application^[40,48-56]. Furthermore, compared with the CoTe battery, the Ni-CoTe battery exhibited stable discharge-charge plateaus and Coulombic efficiency during cycling under such a high sulfur mass loading, confirming the high catalytic activity of the Ni-CoTe catalyst in promoting polysulfide conversion [Figure 4E and F, Supplementary Figure 21].

To further verify that the shuttle effect was inhibited with the Ni-CoTe catalyst, an in situ Raman test was performed during cycling at 0.2 C. A Li-S battery was assembled using a polypropylene separator modified with Ni-CoTe or CoTe catalyst, a quartz window, and a small aperture in the lithium anode, allowing confocal Raman signals to be captured at the separator/lithium interfaces [Figure 5A-F]^[57,58]. When using the CoTe-modified separator, characteristic Raman peaks at 218 and 472 cm^-1, assigned to S₈^2-, were observed around 2.5 V, indicating the liberation and migration of high-order LiPSs from the conductive substrate toward the lithium anode^[48]. As discharge proceeded, the S₈^2- peaks diminished, while peaks at 398 and 534 cm^-1 corresponding to Li₂S₆/Li₂S₄ and Li₂S₃ were detected at 2.2 V, indicating the conversion from long-chain LiPSs to short-chain LiPSs. At the end of the discharge, the disappearance of these LiPS peaks suggested further reduction to Li₂S₂/Li₂S, either on the anode or upon returning to the cathode, leading to reduced sulfur utilization and a consequent decline in discharge capacity^[59]. Upon charging, the S₆^2-/S₄^2- and S₃^2- peaks were redetected, indicating substantial shuttling of soluble LiPSs. In sharp contrast, when the Ni-CoTe-modified separator was employed, the observed Raman spectra showed distinct differences. Only small LiPS signals were detected in the Ni-CoTe battery during both discharge and charge processes [Figure 5D-F], indicating effective suppression of polysulfide shuttling during cycling.

To further investigate suppression of polysulfide shuttling, H-shaped electrolytic cells were assembled, with Ni-CoTe-coated, CoTe-coated, and uncoated separators positioned between two electrolytes. As shown in Figure 5G-H, the left chamber contained 5 mL of 20 mM Li₂S₆ solution, while the right chamber held 5 mL of blank electrolyte, and the two compartments were separated by either a Ni-CoTe-coated, CoTe-coated, or uncoated separator. Experiments conducted with uncoated separators revealed significant Li₂S₆ permeation within 4 h, whereas both the Ni-CoTe and CoTe separators kept the right chamber colorless. After 12 h of standing, the uncoated separator caused the solution in the right chamber to turn yellow. In contrast, the CoTe separator allowed only a limited amount of Li₂S₆ to cross over this period, resulting in a faint yellow, while the Ni-CoTe separator maintained the right chamber almost colorless. This result indicates that both CoTe and Ni-CoTe coatings significantly suppress LiPS permeation compared to the uncoated PP separator, with Ni-CoTe providing substantially superior suppression. Extending the standing time to 24 h resulted in nearly identical coloration on both sides of the cell with an uncoated PP separator, while the CoTe-coated separator gave rise to a light yellow in the right chamber. By sharp contrast, the right chamber exhibited a considerably paler shade when the Ni-CoTe separator was employed. The Li₂S₆ concentration in the right-chamber electrolyte was monitored via ultraviolet-visible (UV/Vis) spectroscopy. As shown in Figure 5H, after 24 h, the S₆^2- peak intensity at 265 nm for the solution exposed to the Ni-CoTe separator was lower than that for the solution exposed to the CoTe separator and the uncoated PP separator. These results confirm that Ni-CoTe possesses a strong adsorption capacity for LiPSs, thereby effectively suppressing the polysulfide shuttling.

Recently, novel electrolytes such as sparingly solvating electrolytes (SSEs) have shown great promise in preventing polysulfide shuttling but faced challenges in catalyzing sulfur conversion^[60]. The Ni-CoTe catalyst, with its abundant Te vacancies and optimized p-band center, provided strong adsorption sites that facilitate the direct capture and subsequent conversion of LiPSs at the catalyst surface. Therefore, the combined effect of SSEs and Ni-CoTe is anticipated to significantly improve battery performance, particularly under practical conditions of high sulfur loading and lean electrolyte.