English

• In order to meet the challenge of the increasingly severe energy crisis, the development and utilization of renewable energy such as secondary batteries have been significantly developed [1-4]. Among them, lithium-sulfur batteries (LSBs) are considered one of the most promising secondary battery systems due to their high theoretical specific capacity of 1675 mAh/g and theoretical energy density of 2600 Wh/kg, as well as the abundant sources of sulfur, its non-toxicity, and environmental friendliness [5,6]. However, LSBs also face many problems that have limited their large-scale commercial production: (1) Sulfur is an electronic insulator, with its electronic conductivity as low as 5 × 10⁻³⁰ S/cm, which would suppress the electrochemical reactions at the cathode and limit the utilization of active materials, resulting in a decrease of discharge capacity for the battery [7]. (2) During the mutual conversion process from S to Li, there is a volume expansion of about 80%, which leads to the formation of "dead sulfur", resulting in capacity decay [8]. (3) The shuttle effect. During the cycling process, soluble lithium polysulfides (Li₂S_n, 4 ≤ n ≤ 8) are produced, and the dissolved polysulfides diffuse towards the anode, reacting directly with metallic lithium, which leads to a decrease in battery life and low Coulombic efficiency. To address these issues, numerous experiments have been conducted by researchers. Many efficient polar materials and electrocatalysts have been developed to immobilize lithium polysulfides and accelerate the redox kinetics of sulfur, aiming to suppress the shuttle effect [9-11]. By employing solution-free solid-state electrolytes, optimizing the current collector design, and introducing artificial protective films, the shuttle effect of polysulfides and lithium dendrite growth can be effectively suppressed, thereby addressing the stability issues of lithium metal anodes and enhancing the overall battery performance [12]. A hierarchical porous structure was successfully fabricated via a solvent exchange method, resulting in a separator with high porosity and mechanical strength. This structure, characterized by a macroporous polymer framework and interconnected nanopores, provides an ideal design to address the dendrite formation issue in lithium metal batteries [13]. In addition to these methods, an advanced and efficient modified separator has been designed, which can significantly enhance the redox reaction kinetics while suppressing the shuttle effect of lithium polysulfides. Therefore, the design of advanced composite material-modified separators provides a viable solution for building Li-S batteries with excellent performance. Besides, direct modification of the separator presents excellent feasibility and practicality, which can be used in large-scale manufacturing processes [14].

  In summary, the utilization of multifunctional interlayers presents an effective strategy to mitigate polysulfide shuttling and enhance the cycling stability of lithium-sulfur batteries [15]. Multifunctional interlayers provide both chemical adsorption of polysulfides through polar compounds and physical barriers to prevent the diffusion of LiPSs [16]. Moreover, the catalytic active sites within the multifunctional interlayers can accelerate the kinetics of polysulfide conversion at various stages, particularly the transformation from long-chain polysulfides to short-chain polysulfides and the final formation of Li₂S [17]. By rationally designing the composition and structure of multifunctional interlayers, such as introducing defects, constructing heterojunctions, or employing single-atom doping, the catalytic performance of the interlayer can be further optimized, thereby improving the overall performance of lithium-sulfur batteries [18]. Xia et al. [19] reported a novel VC_0.75/Co-CoO heterostructure as a sulfur host for lithium-sulfur batteries. By tuning the electronic structure and constructing a heterojunction, VC_0.75 significantly enhanced the chemical adsorption and electrocatalytic conversion of polysulfides by Co-CoO, effectively suppressing the polysulfide shuttle effect. This work paves the way for the development of high-performance lithium-sulfur batteries and provides new insights into addressing the critical issues of short cycle life and low Coulombic efficiency. Luo et al. [20] successfully synthesized a novel FeNi@NC composite via a bimetallic synergistic catalysis strategy. The abundant porous structure and high specific surface area enable efficient adsorption of polysulfides, while the catalytic effect accelerates the conversion of polysulfides, effectively suppressing the polysulfide shuttle effect. This novel composite, when employed as a separator modification material for lithium-sulfur batteries, significantly enhances the cycling stability and rate performance, opening up broad prospects for alloy catalysts in lithium-sulfur battery applications.

  So far, transition metal-based materials have been extensively explored as efficient catalysts for lithium-sulfur batteries due to their excellent catalytic activity, electrical conductivity, and affinity for lithium, and sulfur [21-23]. These transition metal nanomaterials can prevent the dissolution of lithium polysulfides and promote their transformation, thereby improving the performance of lithium-sulfur batteries [24,25]. In CoNi bimetallic alloys, nickel atoms exhibit a stronger tendency to adsorb and catalyze the conversion of long-chain polysulfides to short-chain polysulfides, while cobalt atoms demonstrate superior catalytic activity in the deposition of Li₂S, significantly accelerating the reaction kinetics. This synergistic effect endows CoNi alloys with enhanced adsorption and conversion capabilities towards polysulfides, thereby effectively suppressing the shuttle effect and improving the cycling stability and rate performance of lithium-sulfur batteries [26]. Zhu et al. [27] reported a material in which Co/Ni bimetallic phosphides were encapsulated into a nitrogen-doped dual carbon conductive network. The encapsulated Ni/Co phosphide particles can significantly enhance the adsorption and catalytic conversion of lithium polysulfides, thereby enhancing battery performance. Luo et al. [28] reported a mesoporous carbon-supported cobalt-nickel bimetallic composite material that demonstrated excellent catalytic conversion capabilities for lithium polysulfides, enabling the battery to exhibit outstanding electrochemical performance and superior rate capabilities. Ouyang et al. [29] have designed a novel 3D multi-structured flower-like CoNi@ZnV₂O₄/ZnO—N, C heterojunction. Its unique structure and abundant active sites enable efficient conversion and stable deposition of lithium polysulfides, effectively suppressing the shuttle effect and significantly improving the cycling stability and rate performance of lithium-sulfur batteries. Cobalt-nickel bimetallic catalysts with high activity and stability can be prepared through doping with other substances or encapsulation, thus significantly enhancing the cycle stability and sulfur utilization of lithium-sulfur batteries [30].

  Mxene is an emerging class of two-dimensional layered nanomaterial with a unique structure and exceptional physical and chemical properties. Owing to its high electrical conductivity, strong adsorption capacity for lithium polysulfides, and structural stability, it has been widely applied in LSBs and is regarded as one of the most promising materials in this field [31,32]. Li et al. [33] reported a multifunctional separator composed of deoxyribonucleic acid-carbon nanotube and MXene hybrid-modified polypropylene, which demonstrated excellent adsorption capacity for lithium polysulfide and improved wettability for the electrolyte. Its unique 3D interconnected network structure further provided a new strategy for high-safety Li-S batteries. Yang et al. [34] reported a novel nitrogen-doped MXene-CoS₂ nanohybrid material, where the doped nitrogen sites and CoS₂ nanoparticles demonstrated high adsorption capacity for lithium polysulfide, and the layered MXene accelerated the redox reaction kinetics of sulfur, enabling Li-S batteries to exhibit excellent rate performance and cycle stability. As research on MXene advances, effective solutions have been provided for some of the challenges faced by Li-S batteries [35,36]. In addition, through simple heat treatment, Ti₃C₂T_x can be transformed into TiO₂ and carbon [37]. Compared to Ti₃C₂T_x, MXene-derived TiO₂ delivers superior adsorption capabilities for lithium polysulfides and more effectively suppresses the shuttle effect, as well as improves the redox reactions kinetics [38]. Therefore, the research on the incorporation of MXene-derived TiO₂ and CoNi bimetallic materials shows great developmental potential in addressing the shortcomings of Li-S batteries.

  Herein, we anchored CoNi-MOFs in situ growth on few-layer Ti₃C₂T_x via a simple liquid-phase deposition. Subsequently, a novel CoNi bimetallic alloy and MXene-derived TiO₂ heterostructure (CoNi@TiO₂/C) were prepared via high-temperature annealing (Fig. S1 in Supporting information). Among them, TiO₂ derived from MXene inherits some properties of MXene, such as high specific surface area and good electrical conductivity. Its two-dimensional layered structure effectively facilitates rapid electron transfer within the material, ensuring efficient electron transfer during the charging and discharging processes of the battery. In addition, ultrasmall CoNi alloy particles are uniformly distributed on the matrix of MXene-derived TiO₂, which is beneficial for the full exposure of active sites. The unique point-to-surface model, combined with the properties of CoNi bimetallic characteristics, exhibits strong chemical interaction and catalytic ability towards lithium polysulfides, more effectively suppressing the shuttle effect and accelerating the redox reaction kinetics. The prepared CoNi@TiO₂/C powder was coated onto a commercial polypropylene (PP) separator. The assembled battery exhibited a high initial capacity of 1206.81 mAh/g at 0.5 C and excellent cyclic stability, with a capacity decay rate as low as 0.048% per cycle for 919 cycles at 2 C. Besides, the battery also exhibited excellent rate capabilities, delivering a capacity of 698.89 mAh/g at the high current density of 3 C. Furthermore, as the sulfur loading increased from 1.856 mg to 3.328 mg, the battery continued to demonstrate excellent cyclic stability. These results indicate that in-situ decoration of CoNi alloy material on layered TiO₂ to modify the separator is a very promising strategy to improve the electrochemical performance of lithium-sulfur batteries.

  The microstructures of the various materials at different scales are depicted through SEM images in Fig. 1. In Figs. 1a and b, the microscopic morphologies of MAX and few-layer MXene are displayed, confirming the successful synthesis of Ti₃C₂T_x. Figs. 1c and d displays the microstructures of CoNi-MOFs and CoNi@C. It is evident from Fig. 1c that the CoNi-MOFs exhibit uniformly sized rhombic dodecahedra. After high-temperature decomposition, the morphology of the CoNi-MOFs is disrupted, transitioning from rhombic dodecahedra to nanospheres (Fig. 1d). Fig. 1e illustrates that CoNi-MOFs have been successfully anchored onto Ti₃C₂T_x nanosheets (CoNi-MOFs/MXene). However, in comparison to Fig. 1c, the morphology and size of the anchored CoNi-MOFs nanoparticles are irregular. In Fig. 1f, after high-temperature annealing in an argon atmosphere, the layered structure of CoNi-MOFs/MXene remains unchanged. The CoNi-MOFs transform into CoNi bimetallic nanoparticles with a uniform size distribution, which is markedly different from those in CoNi@C (Fig. 1d). Figs. 1g-n presents the EDS elemental mapping of the CoNi@TiO₂/C composite material obtained via scanning electron microscopy, revealing a uniform distribution of elements including C, N, O, Ti, Co, and Ni. Through the analysis of Figs. 1l-m, Ti is observed to be distributed within a layered structure, while Co is found to be present in small particles on the surface of the layered structure.

  Figure 1

  

  Figure 1.  SEM of (a) Ti₃AlC₂, (b) few-layer Ti₃C₂T_x, (c) CoNi-MOFs, (d) CoNi@C, (e) CoNi-MOFs/MXene, (f) CoNi@TiO₂/C. (g, h) SEM of image and the corresponding elemental mapping images of (i) C, (j) N, (k) O, (l) Ti, (m) Co, and (n) Ni elements of CoNi@TiO₂/C.

  DownLoad: Full-Size Img PowerPoint

  The morphology and microstructure of CoNi@TiO₂/C were further investigated through transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM) (Fig. S2 in Supporting information). Figs. S2a and c clearly depicts the morphological features of the CoNi@TiO₂/C composite, where the CoNi alloy is relatively uniformly distributed on TiO₂/C nanosheets. In the HR-TEM images, the diverse interplanar spacings confirm the presence of the TiO₂ phase and the CoNi phase. In Figs. S2b, S4a and b (Supporting information), the lattice spacing of 0.32 nm is shown to correspond to the (110) plane of TiO₂, while the spacing of 0.203 nm is associated with the (111) plane of the CoNi alloy. Fig. S3 (Supporting information) further confirms this, exhibiting diffraction rings corresponding to the TiO₂ (110) and CoNi (111) planes [39]. The EDS mapping of the product (Figs. S2d-i) indicates a uniform distribution of C, O, N, Ti, Co, and Ni elements in the CoNi@TiO₂/C sample, confirming the successful preparation of the CoNi/TiO₂ heterostructure.

  Further analysis of the material's chemical composition is performed using XRD and XPS. The XRD pattern of Ti₃C₂T_x (Fig. S5a in Supporting information) indicates that the characteristic peak of Ti₃AlC₂ around 39° has disappeared, showing that the Al layer is completely etched. Additionally, the (002) crystal plane of Ti₃C₂T_x shifts to a smaller angle and broadens, demonstrating that the interlayer spacing is increased after exfoliation. The material composition of CoNi@C and CoNi@TiO₂/C is revealed through the XRD pattern (Fig. S5b in Supporting information). For the CoNi@C sample, only characteristic peaks of Co and Ni are detected. Several main characteristic peaks consistent with MXene-derived TiO₂ (Rutile: PDF #21–1276), Co (PDF #15–0806), and Ni (PDF #04–0850) are detected in the XRD pattern of the CoNi@TiO_2/C composite material, demonstrating the successful co-existence of Co, Ni, and TiO₂ [37]. The chemical state and composition of CoNi@TiO_2/C are analyzed using XPS (Figs. S6a-g in Supporting information). Fig. S6a displays the XPS full spectrum of the CoNi@TiO₂/C composite material, where the survey curve shows the obvious characteristic peaks of Ni, Ti, Co, C, N, and O six elements. Fig. S6b displays the XPS curve of C 1s, where peaks centered at 284.8, 284.5, 286.18, and 288.91 eV are respectively represented as C—C/C=C, C—O-C/C—OH, and O—C=O groups. In the N 1s curve (Fig. S6c), the three main peaks displayed are respectively represented as pyridinic-N (397.67 eV), pyrrolic-N (398.91 eV), and graphitic-N (400.82 eV), indicating the incorporation of nitrogen atoms into the carbon frameworks [40]. As depicted in Fig. S6d, the peaks of O 1s at 529.90, 530.88, and 532.16 eV are respectively attributed to Ti-O, C—O, and C=O, respectively. In the Ti 2p spectrum (Fig. S6e), the characteristic peaks situated at 458.75 and 464.47 eV are respectively attributed to the Ti-O 2p_3/2 and Ti-O 2p_1/2 of MXene [41]. For the Co 2p curve (Fig. S6f), the characteristic peaks situated at 770.84 and 795.15 eV are attributed to Co metal particles, while the satellite peaks located at 781.44 and 797.16 eV are assigned to Co²⁺, indicating the partial oxidation of Co nanocrystals. In Fig. S6g, the peak positions of Ni 2p_3/2 and Ni 2p_1/2 are observed, with the peaks at 853.71 and 870.82 eV corresponding to the metallic state of Ni, and the peaks at 857.23 and 874.34 eV corresponding to the oxidized state of Ni²⁺. The test results indicate that the target material has been successfully synthesized.

  The physical properties of the modified separator were analyzed more intuitively from the following experiments (Fig. S7 in Supporting information). In Fig. S7a, the surface microscopy of a commercial PP separator is displayed, showing an interwoven fishnet-like structure. The CoNi@TiO₂/C composite material is uniformly coated onto the surface of the PP separator (Fig. S7b). The cross-section of the modified separator (CoNi@TiO₂/C//PP) is characterized through SEM images, revealing a coating thickness of approximately 18 µm (Fig. S7c). The diameter of the modified separator is 19 mm (Fig. S7d). As shown in Fig. S7e, the modified separator exhibits excellent mechanical toughness, recovering its original appearance after folding without any powder falling off. The contact angles of the three separators are shown in Figs. S8a-c (Supporting information), 23.9° for the PP separator, 20.6° for the CoNi@C separator, and 17.5° for the CoNi@TiO₂/C separator, the results indicate that the CoNi@TiO₂/C separator is best at wettability by the electrolyte, followed by the CoNi@C separator, with the PP separator being the worst. A drop of electrolyte (~20 µL) was added to the surface of each separator, and the distribution of the electrolyte is shown in Fig. S7f after 2 s, the results, observed to be consistent with the conclusions from the contact angles, indicate that the affinity of the modified separators for the electrolyte has been enhanced. These results fully demonstrate the significant potential of CoNi@TiO₂/C as a modified separator material for lithium-sulfur batteries.

  To further clarify the changes in the redox reaction kinetics of lithium polysulfides caused by CoNi@TiO₂/C, cyclic voltammetry (CV) tests were performed. As shown in Fig. 2a, the CV curves of the Li-S batteries equipped with functional separators at a scan rate of 0.1 mV/s exhibit three distinct redox peaks, corresponding to the oxidation–reduction reactions of sulfur at different voltages. During the discharge phase, there are two noticeable reduction peaks at ~2.3 V and ~2.0 V, corresponding to the transformation of S₈ into soluble Li₂S_x (4 ≤ x ≤ 8) and then into Li₂S. During the charge phase, the oxidation peak at ~2.4 V corresponds to the oxidation of Li₂S into Li₂S_x (4 ≤ x ≤ 8) and finally into S₈ [42]. To evaluate the lithium-ion diffusion rate of the separators, the diffusion coefficient was determined via cyclic voltammetry (CV) tests at various scan rates (Figs. 2b and c). Batteries equipped with a CoNi@TiO₂/C//PP are observed to exhibit the highest current response and the smallest peak separation, indicating that the reaction kinetics are enhanced. The diffusion coefficient of Li⁺ ions was estimated using the Randles-Sevcik equation:

  $ I_p=2.69 \times 10^5 n^{3 / 2} A D^{1 / 2} v^{1 / 2} C_A $

  (1)

  Figure 2

  

  Figure 2.  (a) The CV curve of three cells at 0.1 mV/s. (b, c) The CV curve of CoNi@TiO₂/C and CoNi@C with the scan rate of 0.1–0.5 mV/s. (d) The corresponding I_p-v^1/2 linear fitting plot. (e) Diffusion coefficients of lithium ions in various processes. (f) The CV curve of the symmetric cell at 10 mV/s. (g) Electrochemical impedance spectra of the Li₂S₆ symmetric cells. (h, i) Potentiostatic nucleation tests on various modified materials.

  DownLoad: Full-Size Img PowerPoint

  In the equation: I_p represents the peak current, n is the number of electrons transferred, A is the effective area of the electrode, D is the diffusion coefficient, v is the scan rate, and C_A is the concentration of the analyte. A good linear relationship between the peak currents of the oxidation and reduction peaks and the square root of the scan rate is observed in Fig. 2d, indicating that the process is diffusion-controlled [43,44]. The diffusion coefficient of Li⁺ is calculated from the slope of the fitted line, with the results plotted as shown in Fig. 2e. This indicates that the CoNi@TiO₂/C//PP separator significantly facilitates the diffusion of lithium ions, accelerating the reaction kinetics during the charge-discharge process. In Fig. 2f, the CV curves of symmetric batteries are displayed. At a scan rate of 10 mV/s, symmetric cells using CoNi@TiO₂/C//PP as the separator exhibit high current peak values and large peak areas, indicating the presence of more electrochemical active sites, which implies excellent electrochemical performance [45]. The Nyquist plot of the symmetric cell, as shown in Fig. 2g, indicates that the CoNi@TiO₂/C electrode possesses a smaller electrochemical reaction impedance, which is beneficial for rapid charge transfer. The electrochemical deposition from liquid polysulfides to solid Li₂S₂ and Li₂S was studied through potentiostatic nucleation tests. In Figs. 2h and i, it is demonstrated that the electrochemical deposition capability of CoNi@TiO₂/C at a static potential of 2.05 V is superior to that of CoNi@C, which may be attributed to the synergistic effect of the bimetallic components and the catalytic properties of TiO₂/C, together facilitating the conversion of polysulfides and the formation of Li₂S [46].

  In Fig. 3a, the Nyquist plots of different batteries are displayed, where it is shown that the electrochemical reaction impedance (R_ct) of the battery with the CoNi@TiO₂/C-modified separator is lower than that of CoNi@C, which is beneficial for the rapid transfer of charge at the interface between the CoNi@TiO₂/C modified layer and the positive electrode [47]. In Figs. 3b and c, Figs. S9a and b (Supporting information), the rate performance of three batteries from 0.2 C to 3 C is illustrated, where it is shown that the discharge capacity of the battery with the CoNi@TiO₂/C-modified separator at 3 C is 698.98 mAh/g. When the current density is restored to 0.2 C, its specific capacity still reaches a high value of 1015 mAh/g. In Fig. 3d and Fig. S10 (Supporting information), the initial charge-discharge curves of different materials at 0.5 C and 1 C are displayed, where it is shown that the battery with the CoNi@TiO₂/C-modified separator exhibits the highest capacity and excellent polarization voltage. In Fig. 3e and Figs. S12a-h (Supporting information), the static charge-discharge curves for the 1^st, 50^th, 100^th, 150^th, and 200^th cycles at the same coulombic efficiency are displayed, illustrating the catalytic capability of CoNi@TiO₂/C towards lithium polysulfides. The charge-discharge curves consist of low and high discharge platforms, which respectively represent the conversion of Li₂S₂ and S₈ to long-chain lithium polysulfides. In Fig. 3f and Figs. S12i and j (Supporting information), the corresponding Q_H and Q_L values are compared, where it is observed that the Q_H capacity of the battery with the CoNi@TiO₂/C-modified separator is higher than that of other batteries, indicating that CoNi@TiO₂/C accelerates the conversion of S₈ to long-chain lithium polysulfides [48]. In Figs. 3g and h and Fig. S11 (Supporting information), the long cycling stability under 0.5, 2, and 1 C conditions is presented. At 0.5 C, the battery with the CoNi@TiO₂/C-modified separator exhibits an initial discharge specific capacity of 1206.81 mAh/g, which remains at 856.09 mAh/g after 300 charge-discharge cycles, with a capacity decay rate of 0.09% per cycle and a coulombic efficiency close to 100%. At 1 C, the initial discharge specific capacity is 1020.33 mAh/g, maintaining 603.98 mAh/g after 405 cycles, with a capacity decay rate of 0.10% per cycle. At 2 C, the initial discharge specific capacity is 872.34 mAh/g, which remains at 483.22 mAh/g after 919 cycles, with a capacity decay rate of 0.04% per cycle. This indicates that the CoNi@TiO₂/C composite material significantly enhances the conductivity of the separator. In contrast, the batteries with CoNi@C-modified separators exhibit faster capacity decay rates, demonstrating that the batteries with CoNi@TiO₂/C-modified separators have superior cycling stability and rate performance. An in-depth investigation into the cycling stability of batteries with CoNi@TiO₂/C-modified separators, as depicted in Fig. 3i, is conducted. At a sulfur loading of 1.856 mg, a high initial capacity of 952.91 mAh/g is exhibited at 0.5 C, with a high capacity of 768.1 mAh/g maintained even after 150 cycles. Even with an increased sulfur loading of 3.328 mg, a capacity of 549.82 mAh/g is retained after 150 cycles. As shown in Fig. S13 (Supporting information), compared to similar batteries reported in the literature, the Li-S battery fabricated in this study effectively suppressed the shuttle effect of polysulfides, leading to significantly improved cycling stability.

  Figure 3

  

  Figure 3.  (a) The Nyquist plots curve of different separators. (b) Rate capabilities of the CoNi@TiO₂/C//PP, CoNi@C//PP, and PP cells. (c) Charge/discharge curves of the CoNi@TiO₂/C//PP cell at various rates. (d) Charge and discharge curves of different batteries at 0.5 C current. (e) Charge/discharge curves of the CoNi@TiO₂/C//PP cell at 0.5 C. (f) Q_H and Q_L values at 0.5 C. Cycling stability of the CoNi@TiO₂/C//PP cell at (g) 0.5 C and (h) 2 C. (i) Cyclic performance of CoNi@TiO₂/C//PP at 0.5C with sulfur loading of 1.856 and 3.328 mg/cm².

  DownLoad: Full-Size Img PowerPoint

  To further investigate the internal resistance of CoNi@TiO₂/C//PP during battery operation, galvanostatic intermittent titration technique (GITT) measurements were conducted at 0.1 C. As shown in Figs. 4a and b and Fig. S14 (Supporting information), the ΔIR of the CoNi@TiO₂/C//PP battery is the smallest. At the same depth of discharge, the relaxation voltage of CoNi@TiO₂/C//PP (40.9 mV) is less than that of CoNi@C//PP (49.3 mV). At the same depth of charge, the relaxation voltage of CoNi@TiO₂/C//PP (62.9 mV) is also less than that of CoNi@C//PP (81.2 mV). It is demonstrated by CoNi@TiO₂/C//PP having lower polarization, indicating that the heterostructure is beneficial for electron transfer during the charging and discharging processes [49]. In Fig. 4c, a comparison of the relaxation voltages at the same charge-discharge depth is illustrated. Self-discharge tests are displayed in Figs. 4d and e and Fig. S15 (Supporting information). After being discharged to 2.1 V in the 9^th cycle and left to stand for 72 h, the capacity retention rates for the PP and CoNi@C//PP batteries are 79.7% and 91.7%, respectively. A low capacity loss of 7.4% (88 mAh/g) is exhibited by the CoNi@TiO₂/C//PP battery, indicating excellent anti-self-discharge performance. In Fig. 4f, the rate performance of symmetric batteries was assessed by stripping/plating at capacities of 1 mAh/cm² under different current densities. At each current density, the voltage response of the CoNi@TiO₂/C Li||Li symmetric battery is lower than that of the CoNi@C Li||Li symmetric battery. A stable lithium deposition process on the anode is also a primary guarantee for the battery to withstand long cycle performance. The electrochemical stability of symmetric batteries under repeated stripping/plating processes at 1 mA/cm² and 1 mAh/cm² is displayed in Fig. 4h. After 50 h of cycling, the voltage of the CoNi@TiO₂/C Li||Li symmetric battery is consistently lower than that of the CoNi@C@Li and PP symmetric batteries (Fig. 4g) and remains stable for nearly 400 h (Fig. 4i), while the voltage of the CoNi@C Li||Li symmetric battery fluctuates sharply during all time [50]. The results indicate that the CoNi@TiO₂/C separator possesses a more stable Li⁺ transport capability, promoting uniform nucleation in the battery and enhancing its safety performance.

  Figure 4

  

  Figure 4.  (a, b) GITT plots of CoNi@TiO₂/C separator and CoNi@C separator batteries. (c) A columnar comparison of potential difference. (d, e) Self-discharge curves of CoNi@TiO₂/C separator and CoNi@C separator batteries. (f) Rate performance and (g-i) cycling stability of Li||Li symmetric cells with CoNi@TiO₂/C and CoNi@C separators.

  DownLoad: Full-Size Img PowerPoint

  Fig. 5 shows the test of the adsorption capacity of CoNi@TiO₂/C composite material for Li₂S₆. Fig. 5a illustrates the UV–vis tests conducted on the supernatant liquid after adsorption experiments are performed in CoNi@C and CoNi@TiO₂/C. In the UV–vis experiments, the peak around 260 nm is primarily associated with Li₂S₆, and the intensity at 260 nm for different samples is observed to decrease compared to the Li₂S₆ control solution, with the intensity after adsorption by CoNi@TiO₂/C being the weakest [51]. The illustration reflects the color of the solution after adsorption for 8 h, indicating that the CoNi@TiO₂/C sample is most effective in adsorbing Li₂S₆. Fig. 5b displays the XPS survey spectrum of the CoNi@TiO₂/C precipitate after the adsorption experiment, where two additional characteristic peaks are observed: S 2p and F 1s. This confirms the successful absorption of Li₂S₆ by the CoNi@TiO₂/C composite material. The high-resolution S 2p, N 1s, Co 2p, and Ti 2p XPS spectra of CoNi@TiO₂/C before and after Li₂S₆ adsorption are shown in Figs. 5c-f, respectively. Fig. 5c displays the S 2p XPS curve, where four peaks are observed at 161.52, 163.83, 165.05, and 168.93 eV, corresponding to metal sulfide (M-S), S 2p_3/2, S 2p_1/2, and sulfate (SO_x²⁻) species, respectively. For the N 1s spectrum (Fig. 5d), the absorption peaks are observed to shift towards lower binding energy, indicating possible N-S bonding between CoNi@TiO₂/C and LiPSs. For the O 1s spectrum (Fig. S16b in Supporting information), a slight shift in the peaks is observed after adsorption, which may be due to the formation of O-S bonds. Additionally, in the S 2p XPS spectrum of CoNi@TiO₂/C-Li₂S₆, an S-O bond can also be observed at 168.93 eV (Fig. 5c). It is worth noting that in Fig. 5e, the Co 2p peaks are observed to shift slightly towards higher binding energy, which may be due to the formation of M-S bonds. For the Ni 2p spectrum (Fig. S16c in Supporting information), the peaks are observed to shift towards lower binding energy, indicating that electrons are transferred from Li₂S₆ to Ni atoms. Compared to CoNi@TiO₂/C, a new peak is observed at 458.60 eV in the Ti 2p XPS spectrum of CoNi@TiO₂/C-Li₂S₆ (Fig. 5f), corresponding to the Ti-S bond. Simultaneously, M-S bonds can also be observed at 161.52 eV in the S 2p XPS spectrum of CoNi@TiO₂/C-Li₂S₆ (Fig. 5c). These results confirm the strong interaction between CoNi@TiO₂/C and Li₂S₆, indicating that CoNi@TiO₂/C has a strong adsorption capacity for LiPSs. The permeability of polysulfides through the barrier layers on the PP separators was further examined (Fig. 5g). In three H-shaped glass bottles, a quantitative 0.2 mol/L Li₂S₆ dissolved in a DOL/DME mixed solution and pure DOL/DME mixed solution (with a DOL to DME volume ratio of 1:1) was separated by the PP separator and two modified PP separators, respectively [52]. Over time, the diffusion of Li₂S₆ into the pure DOL/DME solution was clearly observed in three H-shaped glass bottles. In contrast, only trace amounts of Li₂S₆ were passed through the CoNi@TiO₂/C//PP separator before 12 h. At 24 h, superior adsorption capabilities were also demonstrated by the CoNi@TiO₂/C//PP separator compared to the CoNi@C//PP separator. These results confirm that the CoNi@TiO₂/C//PP separator effectively suppresses the shuttle effect, indicating that the CoNi@TiO₂/C//PP separator has strong application potential in lithium-sulfur batteries.

  Figure 5

  

  Figure 5.  (a) UV–vis spectra and optical photographs (inset) of Li₂S₆ adsorption for different samples. (b) The XPS full spectra before and after adsorption. (c-f) The high resolution of XPS before and after adsorption presents S 2p, N 1s, O 1s, Ti 2p, and Co 2p, respectively. (g) Diffusion tests of Li₂S₆ solution with PP, CoNi@C//PP, and CoNi@TiO₂/C//PP separators.

  DownLoad: Full-Size Img PowerPoint

  In summary, a novel CoNi@TiO₂/C heterostructure was successfully prepared through high-temperature annealing of its CoNi-MOF/MXene precursor. These heterostructures provide a large specific surface area to ensure the full exposure of active sites. While used as a functional layer for separator modification, CoNi@TiO₂/C heterostructure suppresses the "shuttle effect" and accelerates the redox kinetics in lithium-sulfur batteries. Compared to the CoNi bimetallic alloy, the heterostructure incorporation of CoNi bimetallic alloy and layered TiO₂ exhibits more excellent chemisorption and catalytic conversion abilities towards LiPSs, significantly improving the electrochemical performance of lithium-sulfur batteries with the CoNi@TiO₂/C modified separator. This study presents a rational strategy for designing layered TiO₂ derived from MXene and bimetallic alloys as the catalytic conversion for polysulfides in lithium-sulfur batteries. This work also advances the progress on high-performance and long-life rechargeable Li-S batteries through separator modification.

  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 work was supported by National Natural Science Foundation of China (Nos. 52472194, 52101243), Natural Science Foundation of Guangdong Province, China (No. 2023A1515012619) and the Science and Technology Planning Project of Guangzhou (No. 202201010565).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110732.

  
  
• 1. [1]

     S.M. Wu, Y.N. Li, L.T. Chen, et al., Chin. Chem. Lett. 36 (2025) 109796. doi: 10.1016/j.cclet.2024.109796

  2. [2]

     R. Sun, F. Xu, C.H. Wang, et al., Rare Met. 43 (2024) 1906–1931. doi: 10.1007/s12598-023-02563-y

  3. [3]

     L. Chen, Z. Liu, W. Yang, et al., J. Colloid Interface Sci. 666 (2024) 416–423. doi: 10.1016/j.jcis.2024.04.044

  4. [4]

     S.J. Lu, J.Y. Lin, C.H. Wang, et al., Rare Met. 43 (2024) 3713–3723. doi: 10.1007/s12598-024-02650-8

  5. [5]

     Z.L. Li, P.Y. Li, X.P. Meng, Z. Lin, R.H. Wang, Adv. Mater. 33 (2021) 2102338. doi: 10.1002/adma.202102338

  6. [6]

     W.L. Xia, Y.F. Chen, M.Y. Han, et al., Adv. Funct. Mater. 34 (2024) 2400262. doi: 10.1002/adfm.202400262

  7. [7]

     H. Wang, N.P. Deng, Y.Y. Lu, et al., Appl. Catal. B 357 (2024) 124289. doi: 10.1016/j.apcatb.2024.124289

  8. [8]

     C. Ma, Y.Q. Zhang, Y.M. Feng, et al., Adv. Mater. 33 (2021) 2100171. doi: 10.1002/adma.202100171

  9. [9]

     X.Y. Liu, L. Yuan, X.L. Peng, et al., Chin. Chem. Lett. 36 (2025) 110369. doi: 10.1016/j.cclet.2024.110369

  10. [10]

      J. Wu, T. Ye, Y.C. Wang, et al., ACS Nano 16 (2022) 15734–15759. doi: 10.1021/acsnano.2c08581

  11. [11]

      X.Y. Jin, X.C. Guo, S.Y. Dong, et al., Chin. Chem. Lett. 36 (2025) 110604. doi: 10.1016/j.cclet.2024.110604

  12. [12]

      T. Tao, S. Lu, Y. Fan, et al., Adv. Mater. 29 (2017) 1700542. doi: 10.1002/adma.201700542

  13. [13]

      A. Jung, M.J. Lee, S.W. Lee, et al., Small 18 (2022) 2205355. doi: 10.1002/smll.202205355

  14. [14]

      J. Balach, T. Jaumann, M. Klose, et al., Adv. Funct. Mater. 25 (2015) 5285–5291. doi: 10.1002/adfm.201502251

  15. [15]

      W.L. Xia, M.Y. Han, Y.F. Chen, et al., ACS Sustain. Chem. Eng. 11 (2023) 1742–1751. doi: 10.1021/acssuschemeng.2c05009

  16. [16]

      D. Zhang, T.F. Duan, Y.Z. Luo, et al., Adv. Funct. Mater. 33 (2023) 2306578. doi: 10.1002/adfm.202306578

  17. [17]

      H. Wang, N. Deng, S. Wang, et al., J. Mater. Chem. A 10 (2022) 23433–23466. doi: 10.1039/d2ta05576g

  18. [18]

      T.T. Sun, X.M. Zhao, B. Li, et al., Adv. Funct. Mater. 31 (2021) 2101285. doi: 10.1002/adfm.202101285

  19. [19]

      W.L. Xia, Y. Chen, W.X. Wang, et al., Chem. Eng. J. 458 (2023) 141477. doi: 10.1016/j.cej.2023.141477

  20. [20]

      Y.X. Luo, D. Zhang, Y.Q. He, et al., Chem. Eng. J. 474 (2023) 145751. doi: 10.1016/j.cej.2023.145751

  21. [21]

      N. Gao, Y.J. Zhang, C. Chen, et al., J. Mater. Chem. A 10 (2022) 8378–8389. doi: 10.1039/d2ta00406b

  22. [22]

      P. Xia, S.L. Li, L. Yuan, et al., J. Membr. Sci. 694 (2024) 122395. doi: 10.1016/j.memsci.2023.122395

  23. [23]

      Z.Q. Ye, Y. Jiang, L. Li, F. Wu, R.J. Chen, Adv. Mater. 33 (2021) 2101204. doi: 10.1002/adma.202101204

  24. [24]

      R.R. Chu, T.T. Nguyen, Y.Q. Bai, N.H. Kim, J.H. Lee, Adv. Energy Mater. 12 (2022) 2102805. doi: 10.1002/aenm.202102805

  25. [25]

      X.Y. Ren, Q. Wang, Y.L. Pu, et al., Adv. Mater. 35 (2023) 2304120. doi: 10.1002/adma.202304120

  26. [26]

      Z. Lian, H. Wu, W. Yang, et al., Chem Catal. 3 (2023) 100824.

  27. [27]

      H.Y. Zhu, S.Y. Dong, J. Xiong, et al., J. Colloid Interface Sci. 641 (2023) 942–949. doi: 10.1016/j.jcis.2023.03.083

  28. [28]

      R. Luo, Z.C.Y. Zhang, J. Zhang, et al., Small 17 (2021) 2100414. doi: 10.1002/smll.202100414

  29. [29]

      Z.Y. Ouyang, B. Zheng, C.C. Shi, et al., J. Energy Storage 94 (2024) 112405. doi: 10.1016/j.est.2024.112405

  30. [30]

      Y.L. Ren, L.B. Hu, S.Z. Chang, et al., Small 20 (2024) 2400068. doi: 10.1002/smll.202400068

  31. [31]

      C.Y. Zhang, J.A. Feng, X. Guo, et al., Appl. Phys. Lett. 122 (2023) 193901. doi: 10.1063/5.0147522

  32. [32]

      G.Y. Jiang, N. Zheng, X. Chen, et al., Chem. Eng. J. 373 (2019) 1309–1318. doi: 10.1016/j.cej.2019.05.119

  33. [33]

      Y. Li, M. Li, Y.C. Zhu, et al., Compos. Pt. B: Eng. 251 (2023) 110465. doi: 10.1016/j.compositesb.2022.110465

  34. [34]

      C.Y. Yang, Y. Li, W.C. Peng, F.B. Zhang, X.B. Fan, Chem. Eng. J. 427 (2022) 131792. doi: 10.1016/j.cej.2021.131792

  35. [35]

      Y.Z. Wang, T.C. Guo, E. Alhajji, et al., Adv. Energy Mater. 13 (2023) 2202860. doi: 10.1002/aenm.202202860

  36. [36]

      C.L. Wei, B.J. Xi, P. Wang, et al., Adv. Mater. 35 (2023) 2303780. doi: 10.1002/adma.202303780

  37. [37]

      Z.H. Chang, W.D. Liu, J.N. Feng, et al., Batteries Supercaps 7 (2024) 202300516. doi: 10.1002/batt.202300516

  38. [38]

      X.T. Gao, Y. Xie, X.D. Zhu, et al., Small 14 (2018) 1802443. doi: 10.1002/smll.201802443

  39. [39]

      Z.K. Liu, J. Chen, M.D. Que, et al., Chem. Eng. J. 450 (2022) 138442. doi: 10.1016/j.cej.2022.138442

  40. [40]

      F.C. Zheng, Y. Yang, Q.W. Chen, Nat. Commun. 5 (2014) 5261. doi: 10.1038/ncomms6261

  41. [41]

      M.Q. Wang, B.Y. Qin, F. Xu, et al., J. Colloid Interface Sci. 650 (2023) 446–455. doi: 10.1016/j.jcis.2023.07.007

  42. [42]

      P. Xia, X.L. Peng, L. Yuan, et al., J. Colloid Interface Sci. 678 (2025) 619–629.

  43. [43]

      C.Y. Zha, X.R. Zhu, J. Deng, et al., Energy Storage Mater. 26 (2020) 40–45. doi: 10.1016/j.ensm.2019.12.032

  44. [44]

      X.C. Guo, P.F. Wan, P. Xia, et al., J. Colloid Interface Sci. 678 (2025) 393406.

  45. [45]

      Y.Z. Huang, L. Lin, Y.G. Zhang, et al., Nano Micro Lett. 15 (2023) 67. doi: 10.1007/s40820-023-01037-1

  46. [46]

      Y.S. He, M.J. Li, Y.G. Zhang, et al., Adv. Funct. Mater. 30 (2020) 2000613. doi: 10.1002/adfm.202000613

  47. [47]

      S.F. Zhang, J. Luo, F.J. Zhang, et al., J. Membr. Sci. 652 (2022) 120461. doi: 10.1016/j.memsci.2022.120461

  48. [48]

      Y.J. Li, S.Y. Lin, D.D. Wang, et al., Adv. Mater. 32 (2020) 1906722. doi: 10.1002/adma.201906722

  49. [49]

      M. Li, X.X. Zhou, X.T. Ma, et al., Chem. Eng. J. 409 (2021) 128164. doi: 10.1016/j.cej.2020.128164

  50. [50]

      T. Shi, Q.S. Tu, Y.S. Tian, et al., Adv. Energy Mater. 10 (2020) 1902881. doi: 10.1002/aenm.201902881

  51. [51]

      C. Yan, Z. Jiang, C. Yang, et al., Chem. Eng. J. 495 (2024) 153826. doi: 10.1016/j.cej.2024.153826

  52. [52]

      J. Liu, J.P. Zhang, J. Zhu, et al., Nano Res. 16 (2023) 12601–12607. doi: 10.1007/s12274-023-5683-1

• 

  Figure 1  SEM of (a) Ti₃AlC₂, (b) few-layer Ti₃C₂T_x, (c) CoNi-MOFs, (d) CoNi@C, (e) CoNi-MOFs/MXene, (f) CoNi@TiO₂/C. (g, h) SEM of image and the corresponding elemental mapping images of (i) C, (j) N, (k) O, (l) Ti, (m) Co, and (n) Ni elements of CoNi@TiO₂/C.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) The CV curve of three cells at 0.1 mV/s. (b, c) The CV curve of CoNi@TiO₂/C and CoNi@C with the scan rate of 0.1–0.5 mV/s. (d) The corresponding I_p-v^1/2 linear fitting plot. (e) Diffusion coefficients of lithium ions in various processes. (f) The CV curve of the symmetric cell at 10 mV/s. (g) Electrochemical impedance spectra of the Li₂S₆ symmetric cells. (h, i) Potentiostatic nucleation tests on various modified materials.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) The Nyquist plots curve of different separators. (b) Rate capabilities of the CoNi@TiO₂/C//PP, CoNi@C//PP, and PP cells. (c) Charge/discharge curves of the CoNi@TiO₂/C//PP cell at various rates. (d) Charge and discharge curves of different batteries at 0.5 C current. (e) Charge/discharge curves of the CoNi@TiO₂/C//PP cell at 0.5 C. (f) Q_H and Q_L values at 0.5 C. Cycling stability of the CoNi@TiO₂/C//PP cell at (g) 0.5 C and (h) 2 C. (i) Cyclic performance of CoNi@TiO₂/C//PP at 0.5C with sulfur loading of 1.856 and 3.328 mg/cm².

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a, b) GITT plots of CoNi@TiO₂/C separator and CoNi@C separator batteries. (c) A columnar comparison of potential difference. (d, e) Self-discharge curves of CoNi@TiO₂/C separator and CoNi@C separator batteries. (f) Rate performance and (g-i) cycling stability of Li||Li symmetric cells with CoNi@TiO₂/C and CoNi@C separators.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) UV–vis spectra and optical photographs (inset) of Li₂S₆ adsorption for different samples. (b) The XPS full spectra before and after adsorption. (c-f) The high resolution of XPS before and after adsorption presents S 2p, N 1s, O 1s, Ti 2p, and Co 2p, respectively. (g) Diffusion tests of Li₂S₆ solution with PP, CoNi@C//PP, and CoNi@TiO₂/C//PP separators.

  下载: 全尺寸图片 幻灯片
•