CuCo2S4 Nanoparticles Embedded in Carbon Nanotube Networks as Sulfur Hosts for High Performance Lithium-Sulfur Batteries

Rational design of sulfur hosts for lithium-sulfur (Li-S) batteries is essential to address the shuttle effect and accelerate reaction kinetics. Herein, the composites of bimetallic sulfide CuCo2S4 loaded on carbon nanotubes (CNTs) are prepared by hydrothermal method. By regulating the loading of CuCo2S4 nanoparticles, it is found that when Cu2+ and CNT are prepared in a 10:1 ratio, the CuCo2S4 nanoparticles loaded on the CNT are relatively uniformly distributed, avoiding the occurrence of agglomeration, which improves the electrical conductivity and number of active sites. Through a series of electrochemical performance tests, the S/CuCo2S4-1/CNT presents a discharge specific capacity of 1021 mAh g−1 at 0.2 C after 100 cycles, showing good cycling stability. Even at 1 C, the S/CuCo2S4-1/CNT cathode delivers a discharge capacity of 627 mAh g−1 after 500 cycles. This study offers a promising strategy for the design of bimetallic sulfide-based sulfur hosts in Li-S batteries.


Introduction
Lithium-sulfur (Li-S) batteries are the up-and-coming next-generation rechargeable batteries because of the merits of being environment-friendly, their high energy density (2600 Wh kg −1 ) and theoretical capacity (1675 mAh g −1 ) [1][2][3]. However, soluble lithium polysulfides (LiPSs) are dissolved into the electrolyte during the charge-discharge process, which induces the shuttle effect and rapid capacity decay, limiting the exploitation of highperformance Li-S batteries [4][5][6][7][8]. Therefore, various solutions, including the design of sulfur host, separator and electrolyte modification, are committed to solving the above problem. Among them, the design and preparation of suitable sulfur carriers play an essential role in boosting the performance of Li-S batteries.
In previous studies, various carbon materials including carbon spheres, carbon nanofibers and carbon nanotubes (CNTs) were used as sulfur hosts in Li-S batteries by virtue of physical adsorbing LiPSs. This method presents the effect of sulfur fixation to a certain extent; however, it still has some limitations [9][10][11][12]. Some studies reported that polar materials including metal oxides, metal sulfides and metal phosphides, etc., could mitigate the shuttle effect effectively by chemical adsorption and catalysis [11][12][13][14], such as SiO 2 [15], MnO 2 /TiO 2 [16], nickel-plated [17] and CoP-CNT@C [18]. Among them, transition metal sulfides not only interact strongly with LiPSs but also show excellent catalytic activity in the electrochemical reaction. In addition, it can stabilize the electrochemical performance and enhance the energy efficiency of Li-S batteries [19,20]. For example, CoS 2 [21,22], NiS [23,24] and Co 3 S 4 [25,26] were reported to improve the electrochemical performance by a synergistic role of adsorption and catalysis. Compared to monometallic sulfides, bimetallic sulfides 2 of 11 possess lower band gap energy and improved electrical conductivity [27]. Simultaneously, bimetallic sulfides can provide more reactive sites than monometallic sulfides. Therefore, extensive research has been devoted to the development of new bimetallic sulfides catalyst. Huang et al. prepared Co-Fe bimetallic sulfides with robust chemical adsorption and catalytic activity, it exhibited a high reversible capacity of 1126.5 mAh g −1 at 0.2 C [28]. Lu et al. fabricated the NiCo 2 S 4 @CNTs/S for Li-S batteries. CNTs were found to promote the electronic transportation capacity and conductivity of the cathode material effectively, while NiCo 2 S 4 showed strong adsorption toward the LiPSs, effectively suppressing the diffusion of LiPSs [29]. Previous work has proved that bimetallic sulfide/carbon composite can show a strong effect in inhibiting the shuttle effect. However, the development of different polysulfide/carbon composites is still lacking at present, and the conductivity, electrochemical stability, and conversion kinetics need to be further improved.
In this work, CNTs loaded with CuCo 2 S 4 bimetallic sulfides (CuCo 2 S 4 /CNT) were prepared and used as the sulfur host for Li-S batteries. By further regulating the loading amount of CuCo 2 S 4 nanoparticles on CNT materials, it is explored that the appropriate loading amount of CuCo 2 S 4 nanoparticles can effectively improve the kinetics of LiPSs conversion, inducing a good electrochemical performance. The as-obtained S/CuCo 2 S 4 -1/CNT can sustain a specific capacity of 627 mAh g −1 after 500 cycles, with a capacity decay rate of only 0.08% per cycle.

Materials and Methods
Fabrication of CuCo 2 S 4 /CNT and CuCo 2 S 4 A total of 15 mg slightly oxidized carbon nanotubes were ultrasonically dispersed into 30 mL ethylene glycol, and the suspension was sonicated for 2 h with stirring. Then, 0.15 g Cu(CH 3 COO) 2 -H 2 O (A reagent) and 0.0265 g Co(CH 3 COO) 2 -4H 2 O (B reagent) were dissolved in the mixture and stirred magnetically for 1 h. Afterwards, 0.117 g thiourea was added and stirred for 40 min. The mixture was poured into a 50 mL autoclave, sealed and reacted at 180 • C for 24 h. After cooling, the mixture was cleaned by centrifugation with anhydrous ethanol four times. The product was gathered and dried under vacuum at 70 • C to obtain CuCo 2 S 4 -1/CNT. Holding all other parameters constant, CuCo 2 S 4 -2/CNT was also obtained by adding 0.3 g A reagent and 0.053 g B reagent, while CuCo 2 S 4 -0.5/CNT can be obtained by adding 0.075 g A reagent and 0.01325 g B reagent. CuCo 2 S 4 nanoparticles were obtained without adding slightly oxidized carbon nanotubes and ethylene glycol under the same fabrication conditions as CuCo 2 S 4 /CNT above.
Further details about the fabrication of the S/CuCo 2 S 4 /CNT and S/CuCo 2 S 4 composites, preparation of Li 2 S 6 solution, material characterization, electrochemical measurements and symmetric cells measurement, can be obtained from Supporting S0.

Results and Discussion
The schematic of the synthesis process and structure of S/CuCo 2 S 4 /CNT is shown in Figure 1. In brief, CuCo 2 S 4 /CNT is first synthesized by the hydrothermal method. Then, S/CuCo 2 S 4 /CNT can be obtained by heating of S and CuCo 2 S 4 /CNT mixture. The detailed process can be found in Supporting S0. The final product S/CuCo 2 S 4 /CNT was used as a cathode in this work for Li-S batteries application. By adjusting the content of Cu(CH 3 COO) 2 -H 2 O and Co(CH 3 COO) 2 -4H 2 O, the ratio of CuCo 2 S 4 particles loaded on CNTs can be regulated. The products are marked as CuCo 2 S 4 -0.5/CNT, CuCo 2 S 4 -1/CNT and CuCo 2 S 4 -2/CNT, respectively, with the increase in contents of raw materials. As shown in Figure 2a, when CuCo 2 S 4 particles were synthesized by hydrothermal method, the particle size was about 30-55 nm. However, severe particle agglomeration occurs which reduces the specific surface area of the material. As shown in Figure S1, although the loading of CuCo 2 S 4 on CNT (CuCo 2 S 4 -0.5/CNT) inhibits CuCo 2 S 4 agglomeration, the loading is too sparse ( Figure S1a), which limits the adsorption ability toward polysulfides. While the loading of CuCo 2 S 4 on CNTs is too dense for CuCo 2 S 4 -2/CNT ( Figure S1c Figure S1b) exhibits uniform loading of CuCo 2 S 4 particles on the CNTs' surface, which most possibly enhances the performance of Li-S batteries. Transmission electron microscope (TEM) images of CuCo 2 S 4 -1/CNT in Figure 2d also confirm that CNTs are closely covered by CuCo 2 S 4 with a granular diameter of 8-15 nm. Furthermore, it can be found from the above images that the CNTs are multi-walled. The average diameter and lengths of CNTs are 34 nm and 2 µm, respectively. In addition, when CNTs are exposed to air, they are inevitably oxidized. Some oxygen-containing groups, such as epoxide (C−O−C), hydroxyl (−OH), carboxyl (−COOH), and carbonyl (C=O), may be produced on the CNTs' surface [30]. The presence of these oxygen-containing groups may affect the loading of CuCo 2 S 4 , as well as the electrochemical performance of Li-S batteries. Therefore, related tests need to be further explored in the future. The corresponding element mapping demonstrates the uniform distribution of S, Co, Cu (CuCo 2 S 4 particle) on CNTs (Figure 2e-i).
the particle size was about 30-55 nm. However, severe particle agglomeration occurs which reduces the specific surface area of the material. As shown in Figure S1, although the loading of CuCo2S4 on CNT (CuCo2S4-0.5/CNT) inhibits CuCo2S4 agglomeration, the loading is too sparse ( Figure S1a), which limits the adsorption ability toward polysulfides. While the loading of CuCo2S4 on CNTs is too dense for CuCo2S4-2/CNT ( Figure S1c), restraining the exposure of active sites. The scanning electron microscope (SEM) images of CuCo2S4-1/CNT (Figures 2b,c and S1b) exhibits uniform loading of CuCo2S4 particles on the CNTs' surface, which most possibly enhances the performance of Li-S batteries. Transmission electron microscope (TEM) images of CuCo2S4-1/CNT in Figure 2d also confirm that CNTs are closely covered by CuCo2S4 with a granular diameter of 8-15 nm. Furthermore, it can be found from the above images that the CNTs are multi-walled. The average diameter and lengths of CNTs are 34 nm and 2 μm, respectively. In addition, when CNTs are exposed to air, they are inevitably oxidized. Some oxygen-containing groups, such as epoxide (C−O−C), hydroxyl (−OH), carboxyl (−COOH), and carbonyl (C=O), may be produced on the CNTs' surface [30]. The presence of these oxygen-containing groups may affect the loading of CuCo2S4, as well as the electrochemical performance of Li-S batteries. Therefore, related tests need to be further explored in the future. The corresponding element mapping demonstrates the uniform distribution of S, Co, Cu (CuCo2S4 particle) on CNTs (Figure 2e    the particle size was about 30-55 nm. However, severe particle agglomeration occurs which reduces the specific surface area of the material. As shown in Figure S1, although the loading of CuCo2S4 on CNT (CuCo2S4-0.5/CNT) inhibits CuCo2S4 agglomeration, the loading is too sparse ( Figure S1a), which limits the adsorption ability toward polysulfides. While the loading of CuCo2S4 on CNTs is too dense for CuCo2S4-2/CNT ( Figure S1c), restraining the exposure of active sites. The scanning electron microscope (SEM) images of CuCo2S4-1/CNT (Figures 2b,c and S1b) exhibits uniform loading of CuCo2S4 particles on the CNTs' surface, which most possibly enhances the performance of Li-S batteries. Transmission electron microscope (TEM) images of CuCo2S4-1/CNT in Figure 2d also confirm that CNTs are closely covered by CuCo2S4 with a granular diameter of 8-15 nm. Furthermore, it can be found from the above images that the CNTs are multi-walled. The average diameter and lengths of CNTs are 34 nm and 2 μm, respectively. In addition, when CNTs are exposed to air, they are inevitably oxidized. Some oxygen-containing groups, such as epoxide (C−O−C), hydroxyl (−OH), carboxyl (−COOH), and carbonyl (C=O), may be produced on the CNTs' surface [30]. The presence of these oxygen-containing groups may affect the loading of CuCo2S4, as well as the electrochemical performance of Li-S batteries. Therefore, related tests need to be further explored in the future. The corresponding element mapping demonstrates the uniform distribution of S, Co, Cu (CuCo2S4 particle) on CNTs (Figure 2e-i).   carbon. The intensity ratio of D peak to G peak of CuCo 2 S 4 -1/CNT (I D /I G , 0.69) is lowest in the experimental materials, indicating that the graphitization degree and electric conductivity of CuCo 2 S 4 -1/CNT are higher than that of CuCo 2 S 4 -0.5/CNT (0.81), CuCo 2 S 4 -2/CNT (0.77) and CNT (0.85) ( Figure S2) [31,32]. In addition, Figure 3c displays the thermogravimetric analysis (TGA) plots of different composites. It could be seen that S/CuCo 2 S 4 -1/CNT presents higher sulfur loading up to 76.3%. The specific surface area and pore size characteristics of CuCo 2 S 4 -1/CNT, CuCo 2 S 4 -2/CNT, CuCo 2 S 4 -0.5/CNT and CuCo 2 S 4 were studied by N 2 adsorption-desorption experiments (Figures 3d and S3a,c). It displays typical type III isotherms with H3 type hysteresis loop, indicating the existence of mesopores. CuCo 2 S 4 -1/CNT (152.7 m 2 g −1 ) shows a higher surface area than CuCo 2 S 4 -2/CNT (138.6 m 2 g −1 ), CuCo 2 S 4 -0.5/CNT (102.5 m 2 g −1 ) and CuCo 2 S 4 (85.4 m 2 g −1 ). Pore distribution reveals that there exists a large proportion of micropores in CuCo 2 S 4 -1/CNT compared with the other three materials (Figures 3e,f and S3b,d). This is beneficial to enhance the sulfur limitation by physical role. Higher surface area also facilitates the exposure of active sites and provides a rich electrode/electrolyte interface for LiPSs conversion. marked as the D peak reflecting disordered and defective carbon, while the peak at 1587 cm -1 is attributed to the G peak of carbon, relating to the presence of sp 2 -hybridized carbon. The intensity ratio of D peak to G peak of CuCo2S4-1/CNT (ID/IG, 0.69) is lowest in the experimental materials, indicating that the graphitization degree and electric conductivity of CuCo2S4-1/CNT are higher than that of CuCo2S4-0.5/CNT (0.81), CuCo2S4-2/CNT (0.77) and CNT (0.85) ( Figure S2) [31,32]. In addition, Figure 3c displays the thermogravimetric analysis (TGA) plots of different composites. It could be seen that S/CuCo2S4-1/CNT presents higher sulfur loading up to 76.3%. The specific surface area and pore size characteristics of CuCo2S4-1/CNT, CuCo2S4-2/CNT, CuCo2S4-0.5/CNT and CuCo2S4 were studied by N2 adsorption-desorption experiments (Figures 3d and S3a,c). It displays typical type III isotherms with H3 type hysteresis loop, indicating the existence of mesopores. CuCo2S4-1/CNT (152.7 m 2 g −1 ) shows a higher surface area than CuCo2S4-2/CNT (138.6 m 2 g −1 ), CuCo2S4-0.5/CNT (102.5 m 2 g −1 ) and CuCo2S4 (85.4 m 2 g −1 ). Pore distribution reveals that there exists a large proportion of micropores in CuCo2S4-1/CNT compared with the other three materials (Figures 3e,f and S3b,d). This is beneficial to enhance the sulfur limitation by physical role. Higher surface area also facilitates the exposure of active sites and provides a rich electrode/electrolyte interface for LiPSs conversion. In Li-S batteries, X-ray photoelectron spectrometry (XPS) is usually used to determine the composition, structure and element content of the material. Therefore, in order to identify the composition and valence of the CuCo2S4-1/CNT, we conducted XPS measurement. It can be concluded that Cu, Co, S, C, and O elements exist in CuCo2S4-1/CNT (Figure 4a). The Co XPS spectrum (Figure 4b) shows six peaks at 794.8 eV for Co 3+ 2p1/2, 779.5 eV for Co 3+ 2p3/2, 798.8 eV for Co 2+ 2p1/2, 781.4 eV for Co 2+ 2p3/2, 805.1 and 785.2 eV for satellite peaks [33]. In the Cu XPS spectrum (Figure 4c), the binding energy values at 952.5 eV and 932.5 eV correspond to Cu + 2p1/2 and Cu + 2p3/2, respectively. While 954.0 eV and In Li-S batteries, X-ray photoelectron spectrometry (XPS) is usually used to determine the composition, structure and element content of the material. Therefore, in order to identify the composition and valence of the CuCo 2 S 4 -1/CNT, we conducted XPS measurement. It can be concluded that Cu, Co, S, C, and O elements exist in CuCo 2 S 4 -1/CNT (Figure 4a). The Co XPS spectrum (Figure 4b) shows six peaks at 794.8 eV for Co 3+ 2p 1/2 , 779.5 eV for Co 3+ 2p 3/2 , 798.8 eV for Co 2+ 2p 1/2 , 781.4 eV for Co 2+ 2p 3/2 , 805.1 and 785.2 eV for satellite peaks [33]. In the Cu XPS spectrum (Figure 4c), the binding energy values at 952.5 eV and 932.5 eV correspond to Cu + 2p 1/2 and Cu + 2p 3/2 , respectively. While 954.0 eV and 933.5 eV can be contributed to Cu 2+ 2p 1/2 and Cu 2+ 2p 3/2 , and 943.7 eV and 963.2 eV for satellite peaks [34]. In addition, two characteristic peaks in the S 2p XPS spectra at 163.9 eV (2p 1/2 ) and 162.2 eV (2p 3/2 ) correspond to S 2− species (Figure 4d) [35,36]. The lower intensity characteristic peak at 168.8 eV suggests the presence of small amounts of sulphate or sulfite species and the presence of thin oxide layers on the surface. The peak at 165.1 eV probably corresponds to an M-S bond (M = Cu or Co), where the sulfur presents in the form of polysulfides (S n 2− , 2 ≤ n < 8) [37]. In addition, the present type of polysulfides in different charge-discharge states can be detected by XPS, which can provide a better understanding of the charge-discharge mechanism of lithium-sulfur batteries. These in-depth analyses and discussions will be carried out and published in the future. 933.5 eV can be contributed to Cu 2+ 2p1/2 and Cu 2+ 2p3/2, and 943.7 eV and 963.2 eV for satellite peaks [34]. In addition, two characteristic peaks in the S 2p XPS spectra at 163.9 eV (2p1/2) and 162.2 eV (2p3/2) correspond to S 2-species (Figure 4d) [35,36]. The lower intensity characteristic peak at 168.8 eV suggests the presence of small amounts of sulphate or sulfite species and the presence of thin oxide layers on the surface. The peak at 165.1 eV probably corresponds to an M-S bond (M=Cu or Co), where the sulfur presents in the form of polysulfides (Sn 2− , 2 ≤ n < 8) [37]. In addition, the present type of polysulfides in different charge-discharge states can be detected by XPS, which can provide a better understanding of the charge-discharge mechanism of lithium-sulfur batteries. These in-depth analyses and discussions will be carried out and published in the future.  Figure 5a shows the Nyquist plots of Li-S batteries of different cathodes. The electrochemical impedance spectroscopy (EIS) curves contain a semicircle and a slope line, in line with the charge transfer resistance and the Warburg bulk impedance, respectively. The charge-transfer resistance of S/CuCo2S4-1/CNT is smaller than other electrodes, indicating it has the smallest charge-transfer resistance [38]. As shown in Figure 5b, the cyclic In order to investigate the feasibility of S/CuCo 2 S 4 /CNT composites as Li-S batteries cathodes, a series of electrochemical performance tests were carried out. As shown in Figure S4, the red lines and blue lines correspond to the standard PDF cards of sulfur (JCPDS 08-0247) and CuCo 2 S 4 (JCPDS 42-1450), respectively. The XRD results of S/CuCo 2 S 4 /CNT composites also show characteristic diffraction peaks of S and CuCo 2 S 4 , indicating a successful sulfur loading. The final mass ratios of CuCo 2 S 4 to CNT in S/CuCo 2 S 4 -0.5/CNT, S/CuCo 2 S 4 -1/CNT, and S/CuCo 2 S 4 -2/CNT composites are calculated by combining XPS, EDS and inductively coupled plasma mass spectrometry (ICP-MS) results, showing 4.92:1, 9.81:1 and 18.53:1, respectively, which are close to the theoretical materials input ratios of 5:1, 10:1 and 20:1. Figure 5a shows the Nyquist plots of Li-S batteries of different cathodes. The electrochemical impedance spectroscopy (EIS) curves contain a semicircle and a slope line, in line with the charge transfer resistance and the Warburg bulk impedance, respectively. The charge-transfer resistance of S/CuCo 2 S 4 -1/CNT is smaller than other electrodes, indicating it has the smallest charge-transfer resistance [38]. As shown in Figure 5b, the cyclic voltammetry (CV) curves at 0.1 mV s −1 show two distinct reduction peaks during discharge at 2.02 V and 2.31 V. The reduction peak at 2.31 V represents the reduction of S 8 to soluble LiPSs (Li 2 S n , n = 4, 6, 8). The peak at 2.02 V represents the conversion reaction of LiPSs to Li 2 S 2 /Li 2 S. During charging, the oxidation peak splits into two peaks, which are attributed to the oxidation from solid Li 2 S to LiPSs and eventually to S 8 [39,40]. Furthermore, the first three cycles of CV curves of the S/CuCo 2 S 4 -1/CNT composite are well overlapped, reflecting excellent cycle reversibility. In addition, the first cycle CV curves of S/CuCo 2 S 4 -1/CNT, S/CuCo 2 S 4 -2/CNT and S/CuCo 2 S 4 -0.5/CNT cathodes at the scan rate of 0.1 mV s −1 are shown in Figure S5. It is obvious that S/CuCo 2 S 4 -1/CNT has the largest current response, indicating that it has superior catalytic performance. At a low current, the charging-discharging process of Li-S battery is relatively slow, it tends to produce more LiPSs, which dissolve in the electrolyte, causing the shuttle effect. In this way, we can verify the limitation of the shuttle effect by different types of CuCo 2 S 4 and CNT composites [41]. Moreover, a lot of works have also examined electrochemical performance at 0.2 C so that we can fully compare the electrochemical data of this work with previously published works. Therefore, we perform measurements at 0.2 C based on the above considerations. Figure 5c shows the cycling property of different materials at 0.2 C. The S/CuCo 2 S 4 -1/CNT cathode shows the best electrochemical performance with a first discharge capacity of 1104.5 mAh g −1 and a very low cycle decay rate. In fact, each type was prepared for three samples. One battery of S/CuCo 2 S 4 -1/CNT presents an initial capacity of 1364.5 mAh g −1 .
While the other two samples of S/CuCo 2 S 4 -1/CNT cathode show the first discharge capacity of 1100.3 mAh g −1 and 1108.9 mAh g −1 at 0.2 C ( Figure S6). Considering that one of the values is abnormally high, we conservatively choose the other two similar values to report. Therefore, the average discharge capacity with an error is 1104.6 ± 4.3 mAh g −1 . After 100 cycles, it can maintain a high cycle capacity (1021 mAh g −1 ) and its coulomb efficiency closes to 100%, demonstrating the excellent reversibility of the reaction. In contrast, the S/CuCo 2 S 4 -0.5/CNT and S/CuCo 2 S 4 -2/CNT cathodes exhibited rapid capacity decay and low cycling capacity. In addition, we also compare the S/CuCo 2 S 4 samples without CNT, which exhibit the lowest cycling performance. This can be attributed to the fact that it lacks the CNT's hollow structure and three-dimensional conducting framework. For charge-discharge curves of different samples (Figures 5d and S7), there are two obvious reductive plateaus and a slope, which are related to the reduction and oxidation of LiPSs. The voltage profiles of the S/CuCo 2 S 4 -1/CNT cathode exhibit slower capacity decay and smaller polarization, demonstrating it has excellent catalytic activity. The rate performance of different electrode materials is exhibited in Figure 5e. The specific discharge capacities of S/CuCo 2 S 4 -1/CNT at 0.2, 0.5, 1, 2 and 3 C are 1138 mAh g −1 , 943 mAh g −1 , 887 mAh g −1 , 741 mAh g −1 and 656 mAh g −1 , respectively, which is higher than the other three electrode materials. Even when the current density reverts to 0.2 C, the capacity of S/CuCo 2 S 4 -1/CNT can reach 1072 mAh g −1 , demonstrating the efficient and reversible use of the active sulfur. Moreover, the charge-discharge curves of S/CuCo 2 S 4 -1/CNT at different current densities (Figure 5f) can maintain the characteristic discharge plateau of Li-S batteries compared with S/CuCo 2 S 4 -0.5/CNT, S/CuCo 2 S 4 -2/CNT and S/CuCo 2 S 4 ( Figure S8) [42].
To further investigate the effect of S/CuCo 2 S 4 -1/CNT on the electrochemical performance, we also carried out the EIS test and morphology analysis after cycling for 100 cycles. As shown in Figure 6a, the impedance diagram is composed of two semicircles and an oblique line. The first semicircle represents the formation of the Li 2 S 2 -Li 2 S interface (R SEI ). It can be concluded that S/CuCo 2 S 4 -1/CNT has the lowest impedance, indicating its superior electrochemical kinetics [43,44]. Moreover, the morphology of CuCo 2 S 4 -1/CNT after cycling remains relatively intact. The carbon nanotubes retain their original conductive skeleton structure (Figure 6b,c). Based on the above results, long-term cycling performance at 1 C was also carried out. As exhibited in Figure 6d, the specific capacity of S/CuCo 2 S 4 -1/CNT can maintain at 627 mAh g −1 after 500 cycles, and the capacity decay rate is only 0.08%/cycle. In contrast, S/CuCo 2 S 4 -2/CNT, S/CuCo 2 S 4 -0.5/CNT and S/CuCo 2 S 4 decayed to 441, 389 and 236 mAh g −1 after 500 cycles, respectively. This can be ascribed to the good catalytic effect of the CuCo 2 S 4 -1/CNT composite on the conversion of LiPSs. To further investigate the effect of S/CuCo2S4-1/CNT on the electrochemical performance, we also carried out the EIS test and morphology analysis after cycling for 100 cycles. As shown in Figure 6a, the impedance diagram is composed of two semicircles and an oblique line. The first semicircle represents the formation of the Li2S2-Li2S interface (RSEI). It can be concluded that S/CuCo2S4-1/CNT has the lowest impedance, indicating its superior electrochemical kinetics [43,44]. Moreover, the morphology of CuCo2S4-1/CNT after cycling remains relatively intact. The carbon nanotubes retain their original conductive skeleton structure (Figure 6b,c). Based on the above results, long-term cycling performance at 1 C was also carried out. As exhibited in Figure 6d, the specific capacity of S/CuCo2S4-1/CNT can maintain at 627 mAh g −1 after 500 cycles, and the capacity decay rate is only 0.08%/cycle. In contrast, S/CuCo2S4-2/CNT, S/CuCo2S4-0.5/CNT and S/CuCo2S4 decayed to 441, 389 and 236 mAh g −1 after 500 cycles, respectively. This can be ascribed to the good catalytic effect of the CuCo2S4-1/CNT composite on the conversion of LiPSs. In order to explore its potential mechanism in improving the electrochemical performance of Li-S batteries, the adsorption experiments were performed firstly by immersing the different materials in Li2S6 solution. Equal amounts of samples of CuCo2S4-1/CNT, CuCo2S4-2/CNT, CuCo2S4-0.5/CNT and CuCo2S4 were added to the same volume of Li2S6 solution, and the mixed solutions stand for 24 h. Then, as shown in Figure 7a, the Li2S6 In order to explore its potential mechanism in improving the electrochemical performance of Li-S batteries, the adsorption experiments were performed firstly by immersing the different materials in Li 2 S 6 solution. Equal amounts of samples of CuCo 2 S 4 -1/CNT, CuCo 2 S 4 -2/CNT, CuCo 2 S 4 -0.5/CNT and CuCo 2 S 4 were added to the same volume of Li 2 S 6 solution, and the mixed solutions stand for 24 h. Then, as shown in Figure 7a, the Li 2 S 6 solution with CuCo 2 -1/CNT material became clear, demonstrating the significant adsorption effect of CuCo 2 S 4 -1/CNT material. Simultaneously, the ultraviolet-visible (UV-Vis) spectrum also confirms the results (Figure 7b) [45,46]. In addition, to further investigate the electrocatalytic performance, symmetric cells were also assembled toward different materials. In Figure 7c, the EIS curve shows that the CuCo 2 S 4 -1/CNT electrode has the lowest resistance, confirming its excellent electrochemical reaction kinetics. The CV curves of the CuCo 2 S 4 -1/CNT electrode clearly show the sharpest redox peaks at −0.215/0.215 V and −0.454/0.454 V and the smallest polarization, proving the most excellent catalyzing behavior of the LiPSs conversion (Figure 7d). In addition, as shown in Figure S9, the first three cycles of CV curves of the CuCo 2 S 4 -1/CNT electrode have a relatively high degree of overlap, demonstrating relatively good reversibility [47,48]. Based on the above electrochemical data, CNTs improve the overall conductivity of composites and promote efficient ion/electron transport. At the same time, the highly interconnected 3D conductive network frameworks provide adequate space to buffer volume changes during the charging-discharging cycle. In addition, the uniform loading of CuCo 2 S 4 particles on CNTs surface guarantee abundant active sites on CuCo 2 S 4 -1/CNT, which further ensures that the material possesses a high loading of active sulfur. The CuCo 2 S 4 -1/CNT composite presents strong adsorption and catalytic conversion ability for LiPSs. In conclusion, the excellent electrochemical performance of the S/CuCo 2 S 4 -1/CNT cathode can be attributed to the synergistic effect of CuCo 2 S 4 and CNTs.

Conclusions
In summary, bimetallic sulfide CuCo2S4 nanoparticles loaded with CNT composites were synthesized by the hydrothermal method in this work. By modulating the different loadings of the CuCo2S4 nanoparticles, it is found that the CuCo2S4-1/CNT composites ef-

Conclusions
In summary, bimetallic sulfide CuCo 2 S 4 nanoparticles loaded with CNT composites were synthesized by the hydrothermal method in this work. By modulating the different loadings of the CuCo 2 S 4 nanoparticles, it is found that the CuCo 2 S 4 -1/CNT composites effectively improved the property of Li-S batteries, which can be attributed to the improved overall electrical conductivity of the CNT, promoting efficient ion/electron transport. Moreover, the bimetallic sulfide CuCo 2 S 4 nanoparticles can provide rich adsorption sites for anchoring LiPSs and improve the conversion kinetics of LiPSs. Thus, the S/CuCo 2 S 4 -1/CNT cathode can achieve a first discharge capacity of 1104.6 ± 4.3 mAh g −1 at 0.2 C with a coulombic efficiency close to 100%. After 100 cycles, the discharge specific capacity can maintain 1021 mAh g −1 . In addition, a reversible capacity of 627 mAh g −1 is demonstrated at 1 C after 500 cycles. This work provides a promising strategy for the design of a bimetallic sulfide-CNT network as a sulfur host for Li-S batteries.

Conflicts of Interest:
The authors declare no conflict of interest.