English

• It is urgently required to develop the new energy system for resolving the energy needs and environmental pollution issues. As an important encourage for the new energy system, advanced energy storage technologies have received ever-increasing attention [1–3]. It has been recognized that the current lithium-ion battery technology is no longer able to meet the ever-increasing need of social development [4]. Therefore, people have begun to pay renewed attention to high energy density lithium batteries with high-capacity lithium metal as the anode [5–8]. Especially, Li-S batteries with high theoretical energy density 2567 Wh/kg are considered as one of the promising next-generation secondary batteries [9–12]. Since its birth, the large-scale commercial viability of Li-S batteries is largely restricted through the lithium polysulfides (LiPSs) shuttling. The notorious shuttle effect refers to the dissolution of the solution LiPSs into the organic electrolyte and through the cathode side to anode side during the battery cycling, which finally causes low active sulfur utilization, sluggish reaction kinetics, rapid capacity fading, poor Coulombic efficiency (CE) and short cycle life [13–15].

  Hence, it is crucial to improve the utilization of sulfur species via restraining polysulfides in the cathode and realize their instant conversion during the charge-discharge process. Nowadays, researchers are working hard to address this problem through sulfur cathode design, electrolyte engineering, and separator modification [16–19]. Constructing interlayers between commercial separators and positive electrode is believed to be an authentic and simple method [20]. Unique microstructure and surface chemistry are often designed elaborately to achieve effective adsorption for polysulfides physically and chemically, as well as to facilitate the conversion of polysulfides [21,22].

  Thanks to their high specific surface area, large porosity, and easily-tunable structure, porous carbon spheres can accommodate large amount of sulfur species and alleviate cathode volume expansion, leading to improved sulfur utilization, which have been extensively studied to fabricate the sulfur cathode hosts and the interlayers for inhibiting the shuttle effect of polysulfides [16,23,24]. For example, ordered micro-mesoporous carbon nanospheres embedded with ultrafine iron carbide (Fe₃C/OMMCNSs) [25], nitrogen-doped hollow carbon sphere loaded with atomic cobalt (ACo@HCS) [26], hollow indented carbon spheres (HICS) [27], FeNi₃@hollow porous carbon spheres (FeNi₃@HC) [28], and CoP nanoparticle-doped hollow ordered mesoporous carbon spheres (CoP-OMCS) [29] have been designed as sulfur hosts to improve the battery performance. Porous microstructure is beneficial for performing adsorption function, which is very important for effectively inhibiting the polysulfide shuttling [29–31]. It has been realized that the capture ability of porous carbon materials with LiPSs is closely related to the porous microstructure and porosity. Interlinked porous channels and rich micropores are highly desirable [20]. Researchers have found that micropores can trap elemental sulfur and polysulfides by capillary forces [32–34]. Interconnected porous structure contributes to shortening ion transport paths and thereby improves the rate property of batteries. However, porous carbon materials often suffer from poor electronic conductivity and thus is not conducive to obtaining fast reaction kinetics. In addition, pure carbon materials have low polarity that is disadvantageous to the adsorption for polysulfides, and therefore heteroatom doping is often employed [35–37]. It is widely known that nitrogen doping is beneficial for increasing both electronical conductivity and polarity of carbon materials.

  Herein, nitrogen-doped carbon spheres with cross-linked porous core and lamellar carbon shell were developed as the interlayer in the Li-S cells. The cross-linked porous structure of the core shortens the ion transport paths and meanwhile the lamellar shell improves the electronic conductivity of the carbon spheres. Furthermore, the rich micropores with pore size of ~1 nm provide the core-shell spheres with large specific surface area and strengthen the adsorption for active sulfur species through capillary force. Finally, promoted polysulfide translation and efficacious suppression toward the shuttle effect are achieved. The PCS-PP based Li-S batteries achieve high initial specific capacity up to 1002 mAh/g at 2 C with 574 mAh/g holding after 600 cycles, strengthen rate performance with a discharge specific capacity of 740 mAh/g at a high rate of 4 C. The Li-S cells still achieve high areal capacity of 5.48 mAh/cm² with a high loading of 4.67 mg/cm² at 0.1 C. The corresponding pouch cells also have stable cycling stability with an initial discharge specific capacity of 1082 mAh/g at 0.1 C and high CE of 99.34%.

  The carbon spheres (PCS) with interconnected porous core and lamellar shell are prepared through a double template method as schematically illustrated in Fig. S1 (Supporting information). For comparison, a porous carbon spheres (PC) without the lamellar shell are also prepared. The preparation details can be found in the Experimental Section (Supporting information). The SEM images (Fig. S2 in Supporting information) show the diameter of the PCS ranges from 500 nm to 800 nm, and higher magnification SEM images (Fig. 1a and Fig. S2) clearly show that the PCS is comprised of interconnected porous core and shell. Further, TEM pattern (Fig. 1b) proves the porous microstructure of the core with several macropores with size of 70–140 nm and many mesopores, being consistent with the SEM characterization results. High resolution TEM (HRTEM) pattern (Fig. 1c) of the PCS exhibits that the randomly-arranged lattice fringes and the shell thickness is about 14.5 nm, and the faint diffraction ring in the SAED (inset of Fig. 1c) indicates its amorphous microstructure. The elemental mapping (Fig. 1d) exhibits the uniform distribution of carbon (C) and nitrogen (N) elements in the PCS.

  Figure 1

  

  Figure 1.  (a) SEM, (b) TEM, and (c) HRTEM images of PCS. (d) The elemental mapping results of PCS. (e) XRD patterns and (f) Raman spectra of the PC and PCS. (g) N 1s XPS spectra of PCS. (h) Four probe test results of PCS and PC, showing the much higher electrical conductivity of PCS than PC. (i) N₂ adsorption-desorption isotherm of PCS. Inset: The corresponding pore diameter distribution. (j) Schematic diagram showing the excellent constraint of PCS toward the shuttle effect of LiPSs.

  DownLoad: Full-Size Img PowerPoint

  The crystallographic structures of PC and PCS are comparative through XRD. As shown in Fig. 1e, PC shows two broad peaks located at 23.8° and 43.5°, suggesting that the PC has low degree of crystallinity. The diffraction peaks of PCS are more intensive than that of PC in the XRD patterns, implying that the PCS has a relatively higher degree of graphitization as the result of the lamellar shell. The high degree of graphitization in the PCS would endow the carbon material with good electrical conductivity and thus contributes to improving the electrochemical properties [38]. This is further confirmed by Raman spectroscopy analysis. The Raman spectra (Fig. 1f) of both PC and PCS show two typical peaks at 1342 and 1598 cm⁻¹, corresponding to the D peak (A_1g mode for the disordered carbon atom or defective graphitic structure) and G peak (sp²-hybridized ordered graphite lattice with a Raman active E_2g in-plane vibration mode), respectively [38,39]. Therefore, the intensity ratio of the two peaks (I_D/I_G) is bound up with graphitization degree. Normally, the high intensity ratio represents low graphitization degree. The I_D/I_G values of PC and PCS are counted to be 1.104 and 1.013, respectively, implying the higher crystallinity of PCS, which may be attributed to the randomly-arranged lattice fringes of the carbon shell in PCS. The higher degree of crystallinity of PCS implies its higher electronic conductivity, which would be beneficial for achieving better (Fig. 1g and Fig. S3 in Supporting information) reveal the prosperous doping of N in those carbon spheres, and the doped nitrogen content is 2.78 and 2.22 at% for PCS and PC, respectively. As is well known, N doping helps improve electronic conductivity and increases the polarity of the carbon materials, and thereby contributes to facilitating reaction kinetics and enhancing chemical affinity to LiPSs [40]. Fig. 1h exhibits electrical conductivity of PCS is up to 4.91 × 10³ S/m, four orders of magnitude higher than that (0.31 S/m) of PC. This result stems from the existence of the lamellar carbon shell that can provide fast electron transport pathways, which is consistent with the previous XRD, Raman, and XPS characterization results.

  The pore structure of PCS was investigated by nitrogen adsorption-desorption measurement. The isotherm (Fig. 1i of PCS belongs to type H1, indicative of the coexistence of micropores and mesopores [38], and has a satisfactory specific surface area (S_BET) of 526 m²/g. The pore size distribution (PSD) curve (inset in Fig. 1i) clearly display that PCS mainly contains micropores with pore size of ~1 nm and a small amount of mesopores. The rich micropores indicates that there are abundant capillary hollow nanochannels in the PCS, which would enable capillary forces to absorb polysulfides and provide space for the polysulfide uptake and confinement. In contrast, the isotherm of PC exhibits a permutation of type Ⅰ and Ⅳ, and has a large S_BET of 947.06 m²/g. The PSD curve reveals that the PCS mainly contains large amount of mesopores. The comparison of PCS toward the polysulfide shuttling are schematically illustrated in Fig. 1j.

  The adsorption effect of materials for polysulfide is closely related to the polysulfide shuttle inhibition. Therefore, visualized adsorption test was firstly performed by immersing 30 mg of PCS into 3 mmol/L Li₂S₆ solution (inset in Fig. 2a). For comparison, the Li₂S₆ adsorption of PC was also investigated. After 3 h, the Li₂S₆ solution with PCS becomes colorless, while the color of PC changes faint yellow, such phenomenon can also be observed in Li₂S₄ solution (inset in Fig. S5 in Supporting information), indicating that PCS has better adsorption ability for polysulfides than PC, which is owing to its abundant microporous structure of PCS. The ultraviolet-visible (UV–vis) spectra of all these Li₂S₆ (Fig. 2a) solutions have evident absorption band at ~280 nm (Fig. 2a) that is corresponding to S₆²⁻ species. This S₆^2- signal is the lowest for the Li₂S₆ solution with PCS, indicating the strongest polysulfide adsorption of PCS as the result of its abundant micropores. Notably, although the PC with larger specific surface area (Fig. S4 in Supporting information), its adsorption ability for Li₂S₆ is weaker than PCS, which indicates that the polysulfide adsorption ability is tightly related to the pore structure in the pace of the specific surface area and micropores play a critical role in trapping polysulfides. The same conclusion can also be obtained from the UV–vis spectra of the Li₂S₄ solutions (Fig. S5 in Supporting information). Then, XPS was used to analyze the chemical interaction between LiPSs and PCS. The Li 1s spectra of pristine Li₂S₆ shows a single Li-S bond at 55.4 eV (Fig. 2b). Whereas upon contact with PCS, the Li-S bond of as-obtained sample (denoted as PCS-Li₂S₆) shifts to a higher binding energy (Fig. 2b), and a supplementary peak at 55.68 eV attributed to the Li-N bond appears, which is formed as the result of strong electrostatic interaction between the electron-rich pyridinic N in the N-doped carbon spheres and the element lithium in the LiPSs, leading to the hydrogen bond-like Li bond (N···Li-S) that plays a role in anchoring LiPSs [41–45]. Fig. 2c compares the S 2p XPS spectrum of pristine Li₂S₆ and the PCS-Li₂S₆. The S 2p XPS spectra of Li₂S₆ can be deconvoluted into the terminal (S_T⁻¹) at 163.0 and 163.9 eV as well as the bridging sulfur (S_B⁰) at 164.9 and 165.8 eV [13,14,46,47]. These characteristic peaks shift to high binding energies after Li₂S₆ being absorbed by PCS, indicating the decreased electron cloud density around element S in the LiPSs [41]. Furthermore, two new broad peaks at 168.36 and 169.65 eV appear in the spectrum of PCS-Li₂S₆, which is due to the polythionate complex and sulfate species formed through the redox reaction between Li₂S₆ and the oxidized carbon surface [10,41]. In the N 1s XPS spectra of PC (Fig. S6 in Supporting information) and PCS (Fig. 2d), the downshift of the N characteristic peaks in the PCS-Li₂S₆ indicate the increased electron cloud around the doped N atoms and further verifies the electron transfer from the element S in the polysulfide to the element N in the carbon spheres [48]. The XPS characterization results confirm the chemical interaction between the LiPSs and the carbon spheres as well as the importance of the N doping in enhancing the polysulfide adsorption ability of carbon materials.

  Figure 2

  

  Figure 2.  (a) Compared UV–vis spectra and optical photos (inset) of the Li₂S₆ solutions after adsorption by different samples. (b) Li 1s, (c) S 2p, (d) N 1s XPS spectra of Li₂S₆ and PCS-Li₂S₆. (e, f) Li₂S nucleation curves and (g, h) dissolution profiles at 2.35 V of the cells based on PCS and PC as the electrodes. The insets in (e-h) SEM images of surface morphology of the carbon fiber electrodes after the Li₂S nucleation and dissolution tests.

  DownLoad: Full-Size Img PowerPoint

  Li₂S₆ symmetric cells were assembled by using PCS or PC coated carbon fiber papers as the electrodes and Li₂S₆ as the electrolyte. Then, cyclic voltammetry (CV) measurement was conducted to evaluate the LiPSs conversion kinetics. As shown in Fig. S7 (Supporting information), the curves have two pairs of redox peaks at −0.11 V (S₈ → Li₂S_n, 4 ≤ n), 0.11 V (Li₂S₂/Li₂S → Li₂S_n, 4 ≤ n) and −0.37 V (Li₂S_n → Li₂S₂/Li₂S, 4 ≤ n), 0.37 V (Li₂S_n → S₈, 4 ≤ n), respectively. The cell with PCS exhibits much larger peak current density, indicating the higher LiPSs liquid-liquid conversion dynamics, which is likely due to fact that PCS has higher electronic conductivity than PC.

  The transformation experiment of soluble to solid LiPSs was carried out on carbon fiber paper with PCS and PC coatings to understand the kinetics of Li₂S nucleation [23]. The Li₂S precipitation profile of the PCS based cell (Fig. 2e) shows the current peak appearing at 3100s, much earlier than that (4550s) of the PC based cell (Fig. 2f). Besides, the capacity of the precipitated Li₂S on the PCS is 263.67 mAh/g, much higher than that of PC (250.10 mAh/g), indicating that PCS effectively accelerates the liquid-solid transformation kinetics from Li₂S₈ to Li₂S, which is attributed to the higher electronic conductivity in PCS. Then, SEM is used to observe surface structure of the electrodes after the Li₂S deposition. Thin layer of solid deposits is observed on the PC surface (inset in Fig. 2e), while much thicker deposit layer is found on the PCS (inset in Fig. 2f), which reveals that PCS with high electrical conductivity and strong polysulfide adsorption capacity can accelerate polysulfide conversion for more efficient Li₂S deposition, being consistent with above Li₂S deposition test results. The decomposition curves of deposited Li₂S were studied based on potentiostatic charging using carbon paper as the conductive network (Figs. 2g and h). The PCS based electrode exhibits higher oxidation peak current and shorter oxidation time (1.42 mA, 0.26 h) than PC (0.55 mA, 0.36 h), revealing the effective decomposition of Li₂S on PCS, and few Li₂S deposits can be found after this reaction (inset in Figs. 2g and h), indicating PCS is more conducive to the dissolution of Li₂S. The rapid dissolution of Li₂S can alleviate the passivation of Li₂S on the electrode surface and thus improve the sulfur utilization. It is suggested that the higher electronic conductivity of PCS and its stronger adsorption ability for polysulfides facilitate the charge transfer and thus enhance the LiPSs conversion reaction kinetics.

  The electrolyte wettability of a separator is the key to the electrochemical performance of Li-S cells. Contact angle tests were performed to compare the electrolyte wettability of the commercial polypropylene (PP), PCS coated PP (PCS-PP), and PC coated PP (PC-PP). As shown in Figs. 3a–c, the PCS-PP has the smallest contact angle of 13.7°, while the PC-PP and PP show higher contact angles of 18.9° and 37.5°, respectively, which indicates that the PCS has the best electrolyte wettability, which is attributed to its abundant micropores that trap polysulfides by capillary forces and would be beneficial for enhancing the rate capability of batteries. The good wettability could facilitate Li-ion migration [49].

  Figure 3

  

  Figure 3.  Contact angles of electrolytes on the surface of (a) PP, (b) PC-PP, and (c) PCS-PP. (d) Ionic conductivity of PP, PC-PP, and PCS-PP. (e) Chronoamperometry curve of PCS-PP. Inset: The Nyquist plots before and after polarization. (f) CV curves at 0.1 mV/s and Tafel plots for the (g) A2 peak and (h) A3 peak in the PP, PC-PP, and PCS-PP based Li-S batteries. (i) CV files of the PCS-PP based Li-S cell at different scan rates.

  DownLoad: Full-Size Img PowerPoint

  Then, the ionic conductivity of these modified separators is measured by electrochemical impedance spectroscopy (EIS) (Fig. S8 in Supporting information) based on the thickness of the interlayers (Fig. S9 in Supporting information). As shown in Fig. 3d, the PCS-PP has the highest ionic conductivities of 1.52 mS/cm, which is ascribed to the interconnected microporous core that shortens the ion transport distance. The high ionic conductivity can accelerate diffusion ability of Li ions and adjust the Li ion flux, and thereby enhance the electrochemical reaction kinetices of polysulfide conversion. The Li ion transference number (t_Li) was then assessed by EIS and DC polarization measurements based on the Li//Li cells with different separator. According to the corresponding chronoamperometry profiles and Nyquist plots (Fig. 2e and Fig. S10 in Supporting information), the t_Li are calculated to be 0.45, 0.81, and 0.91 for PP, PC-PP, and PCS-PP, respectively. The highest t_Li of the PSC-PP is probably due to the higher concentration of N element that provide lithium ions with additional transport pathways as the result of the as-formed Li-N bond and the shortened ion transport distance endowed by the interconnected porous structure [24]. The high t_Li of PSC-PP may be due to that the higher concentration of electron-rich N atoms in PCS facilitate dissociation of the lithium salt and establish rich lithium ion transport paths through electrostatic interaction [50]. The high t_Li is conducive to reducing the concentration polarization during charging and discharging and promoting the conversion of polysulfides, which makes high-rate Li-S batteries possible. To investigate the electrochemical redox reaction for Li-S batteries with different separator, CV experiments with 0.1 mV/s are conducted (Fig. 3f). The PP, PC-PP, and PCS-PP based cells show two typical cathodic and anodic peaks. Among them, the reduction peaks A₁ and A₂ corresponding to the conversion of S element to Li₂S_n (Li₂S_n, n = 4, 6, and 8) and then to Li₂S₂/Li₂S precipitates. While, two oxidation peaks A₃ and A₄ corresponding to the conversion of Li₂S/Li₂S₂ to soluble Li₂S_n and then to solid S element. Obviously, the peak potential separation between peak A₂ and peak A₄ is narrower for the cell with PCS-PP (0.394 V) than those with PP (0.439 V) and PC-PP (0.409 V), indicating the lowest electrochemical polarization in the PCS-PP based cell, which is the result of the faster chemical reaction proceeding in the PCS-PP cell, implying the rapid conversion of polysulfides. The cell with PCS-PP shows highest current density than PP and PC-PP, indicating the fastest polysulfide conversion kinetics enabled by PCS. The corresponding Tafel curves of the peak A₂ at 2.0 V and peak A₃ at 2.3 V are calculated and plotted in Figs. 3g and h, respectively. The slopes of fitted curves are 57.47, 66.88, and 109.12 mV/dec, respectively, for the batteries with PCS-PP, PC-PP, and PP during the reduction, while the slopes are 50.06, 55.87, and 72.12 mV/dec, respectively, during the oxidation process. The smallest slopes of the cell with PCS-PP suggests its fastest polysulfide conversion kinetics between LiPSs and Li₂S, which is attributed to much higher electronic conductivity of PCS and its stronger adsorption ability for LiPSs. The Li⁺ diffusion coefficients (D_Li) of the PCS-PP, PC-PP, and PP based Li-S cells were calculated by the Randles-Sevcik equation and CV curves (Fig. 3i and Fig. S11 in Supporting information) at different scan rates. The as-calculated D_Li are compared in Table S1 (Supporting information). The largest D_Li of the PCS-PP based cell, hinting the enhanced lithium ion transport in the PCS-PP based cell. The above results show that the PCS is beneficial for realizing fast electrochemical redox reaction kinetics, which is owe to the cooperative effect of high electronic conductivity, strong adsorption for LiPSs, shortened ion transport distance in PCS.

  The PCS-PP, PC-PP, and PP based coin cells were used to evaluate their electrochemical performance. Fig. 4a shows the galvanostatic charge-discharge (GCD) curves of these cells at 0.2 C. The initial discharge capacity of the PCS-PP based cell is 1242 mAh/g, much higher than PC-PP (1166 mAh/g) and PP (847 mAh/g). Moreover, the PCS-PP based cell shows the smallest liquid-solid conversion reaction polarization (0.143 V) as shown in Fig. 4a, which indicates that the PCS effectively promotes the redox reaction from LiPSs to Li₂S₂/Li₂S. The PCS-PP based cell shows the smallest overpotential for the liquid-solid phase conversion (Fig. 4b).

  Figure 4

  

  Figure 4.  (a) Compared GCD profiles at 0.2 C. (b) The overpotentials for the transformation reaction from solution polysulfide to Li₂S₂/Li₂S. (c) Nyquist plots, (d) rate of the PCS-PP, PC-PP and PP based cells. The inset in (c): The corresponding equivalent circuit of the Nyquist plots. (e) GCD curves of the PCS-PP at different rates. (f) Charge-discharge curves at 4 C. cycle performance of the three cells at (g) 0.2 C, and (h) at 2 C.

  DownLoad: Full-Size Img PowerPoint

  The EIS was conducted to research the interface kinetics of these cells (Fig. 4c). The Nyquist plots contain Ohmic resistance (R_s) and charge transfer resistance (R_ct). Clearly, the PCS-PP based cell has the smaller R_ct (30 Ω) than PC-PP (38 Ω), and PP (56 Ω), indicating the most enhanced electron transfer, being conducive to the fast transformation reaction of polysulfides, which is believed to be directly related to the excellent conductivity of PCS. Fig. 4d compares the rate performance of PCS-PP, PC-PP, and PP based cells. The PCS-PP cell exhibits the best rate property with capacity of 740 mAh/g at 4 C, while the cells using PC-PP and PP show much lower discharge capacities of 595 and 407 mAh/g, respectively. While the rate shifts back to 0.2 C, a capacity of 1074 mAh/g is recovered for the PCS-PP based cell, indicating the quick reaction kinetics and excellent reversibility enabled by the PCS. The PCS-PP based cell exhibits two well-defined discharge plateaus at all rates (Fig. 4e) and the apparent platforms still remain even at 4 C, which is due to the strong trapping ability for LiPSs, shortened ion transport distance, and excellent conductivity of PCS. In contrast, the lack of discharge plateaus and severe polarization of the PP and PC-PP based cells (Fig. S12 in Supporting information) can be attributed to their sluggish redox kinetics. The GCD profiles of these cells at 4 C (Fig. 4f) shows that the PCS-PP has the smallest polarization even at a high rate 4 C, which demonstrates the fastest polysulfide conversion kinetics, being ascribed to the high electrical conductivity and abundant microporous structure of PCS.

  Fig. 4g compares the cycling stability of PCS-PP, PC-PP, and PP based cell at 0.2 C. The PCS-PP based battery exhibits a higher initial capacity of 1219 mAh/g, much higher than that of PC-PP (1184 mAh/g) and PP (851 mAh/g). The PCS-PP based cell still has a higher initial discharge capacity of 1002 mAh/g at 2 C, while the cells with PC-PP, and PP deliver much lower initial discharge capacities of 895 and 660 mAh/g with 353 and 292 mAh/g remaining, respectively, after 600 cycles (Fig. 4h).

  As we know, the shuttling of polysulfides would cause side reaction between the positive and negative electrodes. For this reason, SEM was used to measure the surface structure of the lithium metal anodes in the cycled Li-S cells. The lithium metal surface morphology was compared after 50 cycles at 2 C (Figs. 5a–c). The cycled lithium metal is relatively smooth in the PCS-PP based cell (Fig. 5a), while uneven surface is observed for the lithium anode surfaces with PC-PP (Fig. 5b) and PC-PP (Fig. 5c). In addition, the photos of lithium metal (insets in Figs. 5a–c) discover that the surface of cycled lithium metal in the Li-S battery with PCS-PP is silver color. In contrast, the PC-PP and PP based batteries present black color. The result reveals that the shuttle effect is well inhibited by PCS that provides abundant micropores to trap LiPSs and improves the redox reaction kinetics of LiPSs mainly as the result of its high electronic conductivity.

  Figure 5

  

  Figure 5.  Surface morphology of lithium metal from (a) PCS-PP, (b) PC-PP, and (c) PP based Li-S cells over 50 cycles at 2 C. (d) Cycle performance of the cells based on PCS-PP with sulfur loadings of 4.69 mg/cm² at 0.1 C. (e) Cycling performance of the PCS-PP based pouch cell at 0.1 C. (f) Photos of the PCS-PP pouch cell showing a stable OCV of 2.769 V. Photos of the PCS-PP pouch cell lightening the LED lamps under (g) normal and (h) folded states.

  DownLoad: Full-Size Img PowerPoint

  The PCS-PP based cell with high sulfur area loadings is also assembled and tested. As shown in Fig. 5d, the PCS-PP based cells with 4.69 mg/cm² deliver favorable capacities of 5.48 mAh/cm² at 0.1 C, exceeding to area capacities of commercial lithium-ion batteries. The pouch cells are assembled (see details in Experimental section in Supporting information) to demonstrate the potential application of PCS-PP in real conditions. The pouch cell displays stable cycling performance (Fig. 5e) and steady open circuit voltage (OCV) (Fig. 5f), and the light-emitting-diode (LED) lamps can be lightened by the normal (Fig. 5g) and even folded (Fig. 5h) pouch cell, showing the potential application of PCS-PP in realizing high-energy-density Li-S batteries.

  In summary, pure carbon spheres with interconnected porous core and lamellar shell are developed to reduce the shuttle effect of LiPSs. The interconnected porous channels in the core shorten the ion transfer distance, which helps accelerate the electrochemical reaction kinetics. In addition, high-efficient adsorption toward LiPSs is realized due to the capillary force of rich micropores in PCS and the strong chemical interaction between polysulfide anions and the doped N atoms. Importantly, the lamellar carbon shell greatly promotes the electronic conductivity of the porous carbon spheres, further promoting the polysulfide conversion kinetics. As a result, the Li-S battery with PCS exhibit excellent capacity of 1002 mAh/g at 2 C and excellent rate capacity of 740 mAh/g at 4 C. At a sulfur area loading up to 4.69 mg/cm², the Li-S cells still exhibit high initial gravimetric capacity of 1033 mAh/g and high capacity of 5.48 mAh/cm² at 0.1 C. This work provides a good strategy for designing a barrier layer based on pure carbon materials with both fast ionic/electronic conductivity and strong polysulfide adsorption to effectively suppress the shuttle effect and obtain high-energy-density lithium sulfur batteries.

  Declaration of competing interest

  There are no conflicts of interest to declare.

  Acknowledgments

  This work was financially supported by National Natural Science Foundation of China (Nos. 51972070 and 52062004), Guizhou Provincial High Level Innovative Talents Project (No. QKHPTRC-GCC[2022]013–1), Innovation Team for Advanced Electrochemical Energy Storage Devices and Key Materials of Guizhou Provincial Higher Education Institutions (No. QianJiaoJi[2023]054), Guizhou Provincial Science and Technology Projects (No. QKHJC[2020]1Z042), Cultivation Project of Guizhou University (No. GDPY[2019]01).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2023.109200.

  
  
• 1. [1]

     Y. Tang, H. Zhou, K. Zhang, et al., Chem. Eng. J. 262 (2015) 260–267. doi: 10.1016/j.cej.2014.09.095

  2. [2]

     B. Zhang, C. Luo, Y. Deng, et al., Adv. Energy Mater. 10 (2020) 2000091. doi: 10.1002/aenm.202000091

  3. [3]

     Q. Liang, S. Wang, X. Jia, et al., J. Mater. Sci. Technol. 151 (2023) 89–98. doi: 10.1016/j.jmst.2022.11.059

  4. [4]

     J. Tang, F. Wu, X. Dai, et al., Chem. Eng. J. 452 (2023) 139139. doi: 10.1016/j.cej.2022.139139

  5. [5]

     X.Y. Li, Q. Zhang, J. Energy Chem. 65 (2022) 302–303. doi: 10.1016/j.jechem.2021.05.039

  6. [6]

     M. Zhao, H.J. Peng, Z.W. Zhang, et al., Angew. Chem. Int. Ed. 58 (2019) 3779–3783. doi: 10.1002/anie.201812062

  7. [7]

     Y. Lin, S. He, Z. Ouyang, et al., J. Mater. Sci. Technol. 134 (2023) 11–21.

  8. [8]

     X.Z. Fan, M. Liu, R. Zhang, et al., Chin. Chem. Lett. 33 (2022) 4421–4427. doi: 10.1016/j.cclet.2021.12.064

  9. [9]

     Y. Song, S. Zhu, X. Long, et al., Electrochim. Acta 441 (2023) 141875. doi: 10.1016/j.electacta.2023.141875

  10. [10]

      X. Liang, C. Hart, Q. Pang, et al., Nat. Commun. 6 (2015) 5682. doi: 10.1038/ncomms6682

  11. [11]

      K. Zou, X. Chen, W. Jing, et al., Energy Storage Mater. 48 (2022) 133–144. doi: 10.1016/j.ensm.2022.03.003

  12. [12]

      S. Xin, T. Liu, J. Li, et al., Small 19 (2023) e2207924. doi: 10.1002/smll.202207924

  13. [13]

      L. Huang, S. Shen, Y. Zhong, et al., Adv. Mater. 34 (2022) e2107415. doi: 10.1002/adma.202107415

  14. [14]

      Y. Zhang, G. Li, J. Wang, et al., Adv. Energy Mater. 11 (2021) 2100497. doi: 10.1002/aenm.202100497

  15. [15]

      Z.X. Chen, Y.T. Zhang, C.X. Bi, et al., J. Energy Chem. 82 (2023) 507–512. doi: 10.1016/j.jechem.2023.03.015

  16. [16]

      P. Wang, Z. Zhang, X. Yan, et al., J. Mater. Chem. A 6 (2018) 14178–14187. doi: 10.1039/c8ta04214d

  17. [17]

      F. Pei, L. Lin, A. Fu, et al., Joule 2 (2018) 323–336. doi: 10.1016/j.joule.2017.12.003

  18. [18]

      X. Long, Z.H. Luo, W.H. Zhou, et al., Energy Storage Mater. 52 (2022) 120–129. doi: 10.1016/j.ensm.2022.07.041

  19. [19]

      Y. Liao, L. Yuan, X. Liu, et al., Energy Storage Mater. 48 (2022) 366–374. doi: 10.1016/j.ensm.2022.03.035

  20. [20]

      Y. Song, X. Long, Z. Luo, et al., ACS Appl. Mater. Interfaces 14 (2022) 32183–32195. doi: 10.1021/acsami.2c09331

  21. [21]

      J.L. Feng Wu, Y.F. Su, J. Wang, et al., Nano Lett. 16 (2016) 5488–5494. doi: 10.1021/acs.nanolett.6b01981

  22. [22]

      G. Zhou, Y. Zhao, A. Manthiram, Adv. Energy Mater. 5 (2015) 1402263. doi: 10.1002/aenm.201402263

  23. [23]

      J.S. Yeon, Y.H. Ko, T.H. Park, et al., Energy Environ. Mater. 5 (2021) 555–564.

  24. [24]

      H. Tian, T. Wang, F. Zhang, et al., J. Mater. Chem. A 6 (2018) 12816–12841. doi: 10.1039/c8ta02353k

  25. [25]

      Z.S. Jina, M. Zhao, T.N. Lin, et al., Chem. Eng. J. 388 (2020) 124315. doi: 10.1016/j.cej.2020.124315

  26. [26]

      Z. Wang, J. Shen, X. Xu, et al., Small 18 (2022) 2106640. doi: 10.1002/smll.202106640

  27. [27]

      Y. Zhong, S. Wang, Y. Sha, et al., J. Mater. Chem. A 4 (2016) 9526–9535. doi: 10.1039/C6TA03187K

  28. [28]

      L. Wang, H. Wang, S. Zhang, et al., ACS Nano 15 (2021) 15218–15228. doi: 10.1021/acsnano.1c05778

  29. [29]

      F. Wang, Y. Han, R. Xu, et al., Small (2023) e2303599.

  30. [30]

      Q. Sun, B. He, X.Q. Zhang, et al., ACS Nano 9 (2015) 8504–8513. doi: 10.1021/acsnano.5b03488

  31. [31]

      J. Zheng, G. Guo, H. Li, et al., ACS Energy Lett. 2 (2017) 1105–1114. doi: 10.1021/acsenergylett.7b00230

  32. [32]

      B. Zhang, X. Qin, G.R. Li, et al., Energy Environ. Sci. 3 (2010) 1531–1537. doi: 10.1039/c002639e

  33. [33]

      Z. Li, L. Yuan, Z. Yi, et al., Adv. Energy Mater. 4 (2014) 1301473. doi: 10.1002/aenm.201301473

  34. [34]

      H. Chen, C. Wang, W. Dong, et al., Nano Lett. 15 (2014) 798–802.

  35. [35]

      G. Zhou, E. Paek, G.S. Hwang, et al., Nat. Commun. 6 (2015) 7760. doi: 10.1038/ncomms8760

  36. [36]

      L.C. Yin, J. Liang, G.M. Zhou, et al., Nano Energy 25 (2016) 203–210. doi: 10.1016/j.nanoen.2016.04.053

  37. [37]

      X. Zhang, F. Ma, K. Srinivas, et al., Energy Storage Mater. 45 (2022) 656–666. doi: 10.1016/j.ensm.2021.12.010

  38. [38]

      J. Xu, G. Du, L. Xie, et al., ACS Sustain. Chem. Eng. 8 (2020) 8024–8036. doi: 10.1021/acssuschemeng.0c02677

  39. [39]

      K. Gao, R. Xu, Y. Chen, et al., Chin. Chem. Lett. 34 (2023) 107229. doi: 10.1016/j.cclet.2022.02.034

  40. [40]

      J. Wang, Z. Zhang, X. Yan, et al., Nano-micro. Lett. 12 (2019) 1–14.

  41. [41]

      Q. Pang, J. Tang, H. Huang, et al., Adv. Mater. 27 (2015) 6021–6028. doi: 10.1002/adma.201502467

  42. [42]

      T.Z. Hou, W.T. Xu, X. Chen, et al., Angew. Chem. Int. Ed. 56 (2017) 8178–8182. doi: 10.1002/anie.201704324

  43. [43]

      X. Chen, Y.K. Bai, C.Z. Zhao, et al., Angew. Chem. Int. Ed. 59 (2020) 11192–11195. doi: 10.1002/anie.201915623

  44. [44]

      H. Li, B. Cai, Y. Song, et al., Chin. Chem. Lett. 34 (2023) 107811. doi: 10.1016/j.cclet.2022.107811

  45. [45]

      X. Zhao, T. Gao, W. Ren, et al., J. Energy Chem. 75 (2022) 250–259. doi: 10.1016/j.jechem.2022.08.033

  46. [46]

      S. Hu, M. Yi, X. Huang, et al., J. Mater. Chem. A 9 (2021) 2792–2805. doi: 10.1039/d0ta10607k

  47. [47]

      L. Ji, X. Wang, Y. Jia, et al., Chin. Chem. Lett. 34 (2023) 107123. doi: 10.1016/j.cclet.2022.01.016

  48. [48]

      Q. Pang, D. Kundu, M. Cuisinier, et al., Nat. Commun. 5 (2014) 4759. doi: 10.1038/ncomms5759

  49. [49]

      J. Li, Y. Huang, S. Zhang, et al., ACS Appl. Mater. Interfaces 9 (2017) 7499–7504. doi: 10.1021/acsami.7b00065

  50. [50]

      Y. Wen, X. Wang, Y. Yang, et al., J. Mater. Chem. A 7 (2019) 26540–26548. doi: 10.1039/c9ta09570e

• 

  Figure 1  (a) SEM, (b) TEM, and (c) HRTEM images of PCS. (d) The elemental mapping results of PCS. (e) XRD patterns and (f) Raman spectra of the PC and PCS. (g) N 1s XPS spectra of PCS. (h) Four probe test results of PCS and PC, showing the much higher electrical conductivity of PCS than PC. (i) N₂ adsorption-desorption isotherm of PCS. Inset: The corresponding pore diameter distribution. (j) Schematic diagram showing the excellent constraint of PCS toward the shuttle effect of LiPSs.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) Compared UV–vis spectra and optical photos (inset) of the Li₂S₆ solutions after adsorption by different samples. (b) Li 1s, (c) S 2p, (d) N 1s XPS spectra of Li₂S₆ and PCS-Li₂S₆. (e, f) Li₂S nucleation curves and (g, h) dissolution profiles at 2.35 V of the cells based on PCS and PC as the electrodes. The insets in (e-h) SEM images of surface morphology of the carbon fiber electrodes after the Li₂S nucleation and dissolution tests.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Contact angles of electrolytes on the surface of (a) PP, (b) PC-PP, and (c) PCS-PP. (d) Ionic conductivity of PP, PC-PP, and PCS-PP. (e) Chronoamperometry curve of PCS-PP. Inset: The Nyquist plots before and after polarization. (f) CV curves at 0.1 mV/s and Tafel plots for the (g) A2 peak and (h) A3 peak in the PP, PC-PP, and PCS-PP based Li-S batteries. (i) CV files of the PCS-PP based Li-S cell at different scan rates.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Compared GCD profiles at 0.2 C. (b) The overpotentials for the transformation reaction from solution polysulfide to Li₂S₂/Li₂S. (c) Nyquist plots, (d) rate of the PCS-PP, PC-PP and PP based cells. The inset in (c): The corresponding equivalent circuit of the Nyquist plots. (e) GCD curves of the PCS-PP at different rates. (f) Charge-discharge curves at 4 C. cycle performance of the three cells at (g) 0.2 C, and (h) at 2 C.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  Surface morphology of lithium metal from (a) PCS-PP, (b) PC-PP, and (c) PP based Li-S cells over 50 cycles at 2 C. (d) Cycle performance of the cells based on PCS-PP with sulfur loadings of 4.69 mg/cm² at 0.1 C. (e) Cycling performance of the PCS-PP based pouch cell at 0.1 C. (f) Photos of the PCS-PP pouch cell showing a stable OCV of 2.769 V. Photos of the PCS-PP pouch cell lightening the LED lamps under (g) normal and (h) folded states.

  下载: 全尺寸图片 幻灯片
•