Understanding the improved performance of sulfur‐doped interconnected carbon microspheres for Na‐ion storage

Abstract

As one of the low‐cost energy storage systems, Na‐ion batteries (NIBs) have received tremendous attention. However, the performance of current anode materials still cannot meet the requirements of NIBs. In our work, we obtain sulfur‐doped interconnected carbon microspheres (S‐CSs) via a simple hydrothermal method and subsequent sulfurizing treatment. Our S‐CSs exhibit an ultrahigh reversible capacity of 520 mAh g^–1 at 100 mA g^–1 after 50 cycles and an excellent rate capability of 257 mAh g^–1, even at a high current density of 2 A g^–1. The density functional theory calculations demonstrate that sulfur doping in carbon favors the adsorption of Na atom during the sodiation process, which is accountable for the performance enhancement. Furthermore, we also utilize operando Raman spectroscopy to analyze the electrochemical reaction of our S‐CSs, which further highlights the sulfur doping in improving Na‐ion storage performance.

INTRODUCTION

As one of the most popular energy storage systems, Li-ion batteries (LIBs) have been widely applied in portable electronic devices and electric vehicles on account of their high energy density and long cycle life. ^1–3 However, further development of LIBs suffers great restrictions due to the rareness and inhomogeneous distribution of Li resources. ⁴ To address this concern, numerous efforts have been made to explore cost-effective rechargeable battery systems beyond LIBs. ^5,6 Therefore, Na-ion batteries (NIBs) have received much attention due to the abundant source and the similar properties of Na to Li. However, many host materials only exhibited a finite performance for NIBs due to the larger diameter of the Na ion, compared with Li ion. ^7,8 Seeking an appropriate accommodator with excellent reversibility for Na-ion storage is still a challenge.

In reported studies, carbonaceous materials have been regarded as a category of promising anode material for NIBs due to their low cost, superior conductivity, and excellent stability, compared with other candidates. ^9,10 Among all of the carbonaceous materials, hard carbon is demonstrated as one of the most appropriate anodes for NIBs. ^11,12 However, due to the low diffusion coefficient of Na ion, hard carbon always exhibits a poor reversible capability and inferior rate performance, which hinder its practical applications for NIBs. ¹³ To address this issue, developing hard carbon with a smart structure is one of the effective strategies to improve its electrochemical performance for NIBs, which is due to the enhancement of electrode/electrolyte contact. ^14–16 Gaddam et al. ¹⁷ designed surface-modified hard carbon for NIBs, delivering a specific capacity of 203 mAh g^–1 at 100 mA g^–1 after 50 cycles. Suo et al. ¹⁸ obtained hard carbon spheres interconnected by carbon nanotubes for NIBs, achieving a reversible specific capacity of 151.7 mAh g^–1 at 100 mA g^–1 after 160 cycles. Despite these excellent works, the Na-ion storage performance is insufficient for the demand. It is still urgent to develop high-performance hard carbon for Na-ion storage.

To further improve Na-ion storage performance, another valid strategy is doping heteroatom to decorate the surface and structure of hard carbon, which can effectively enhance its electrical conductivity. ^19–24 Among all the heteroatoms, sulfur doping in hard carbon could lead to an increase of lattice distance, which is in favor of the effective insertion and extraction of Na ion. Li et al. ²⁵ designed sulfur-doped disordered carbon by pyrolysis of precursor in sulfur steam, leading to excellent reversibility for Na-ion storage (271 mAh g^–1 at 1 A g^–1 after 1000 cycles). Wang et al. ²⁶ obtained sulfur-doped carbon microtubes via pyrolysis of biomass, which exhibited a superior cycling performance (281 mAh g^–1 after 1000 cycles at 1 A g^–1). Jin et al. ¹³  synthesized sulfur-doped carbon nanofibers, which presented a high reversible capacity of 460 mAh g^–1 at 0.05 A g^–1. ¹³ Li et al. ²⁷ prepared sulfur-doped porous carbons derived from conjugated microporous polymers, which received a capacity of 440 mAh g⁻¹ at 50 mA g⁻¹ after 50 cycles. These works have demonstrated that doping sulfur in hard carbon would display a superior activity with high reversibility for Na-ion storage. Therefore, the hard carbon is expected to present a high performance for NIBs if favorable nanostructured hard carbon could be sulfur-doped. Moreover, most works about sulfur-doped hard carbon mainly relied on ex-situ characterization techniques to analyze the electrochemical reaction mechanism for Na-ion storage. ^13,27 However, for the aforementioned ex-situ characterization techniques, there still inevitably exist some drawbacks, including that the reaction processes cannot be detected in real time for the electrode. To reflect the real reaction conditions of battery, some operando techniques have been performed to understand the electrochemical reaction mechanism for battery. Liang et al. ²⁸  utilized operando X-ray diffraction to discover the insights into Na-ion de-/intercalation model evolution. Wu et al. ²⁹ exploited operando visualization to explore the structural and morphological evolution of material in all-solid-state batteries. Chen et al. ³⁰ used operando surface-enhanced Raman technique to investigate the dynamic evolution of the cathode–electrolyte interfaces. On the basis of the investigation of operando techniques, it is possible to further explore the sodiation–desodiation mechanism of sulfur-doped hard carbon.

Hence, we prepared sulfur-doped interconnected carbon microspheres (S-CSs) by the hydrothermal method and subsequent thermal treatment in sulfur steam. After optimization, our S-CSs exhibited an ultrahigh reversible capacity of 520 mAh g^–1 at 100 mA g^–1 after 50 cycles and an excellent rate capability of 257 mAh g^–1, even at a high current density of 2 A g^–1, outperforming most of the sulfur-doped carbonaceous materials. According to the density functional theory (DFT) calculations, we found that sulfur doped in carbon would lead to a stronger adsorption ability of Na atom, which is conducive to electron transfer during the sodiation–desodiation process. Furthermore, we also developed operando Raman technique to investigate our S-CSs for NIBs, which revealed the electrochemical enhancement mechanism of sulfur doping during the sodiation–desodiation process in the real state.

RESULTS AND DISCUSSION

Figures 1a and 1b show the scanning electron microscopy (SEM) images of CS-600 and S-CS-600. Both of them display the carbon sphere structure. To further observe the morphology of S-CSs, an enlarged SEM image of S-CS-600 is provided, as shown in Figure 1c. It is found that the carbon spheres are interconnected to form a cross-linked network, which can accelerate electron transport. Figure 1d presents the transmission electron microscopy (TEM) image of S-CS-600, which confirms the interconnected structure of carbon spheres. According to the high-resolution TEM images of S-CS-600 shown in Figure 1e, the inordinate lattice fringe can be detected, indicating that the obtained interconnected structure of carbon spheres exhibits a disordered structure. Figure 1f shows the element mapping of S-CS-600. It can be observed that the sulfur element is distributed uniformly, which indicates that sulfur element has been doped in interconnected carbon spheres.

Figure 1. SEM images of (a) CS-600 and (b) S-CS-600, (c) the enlarged SEM image of S-CS-600, (d) TEM and (e) high-resolution TEM images of S-CS-600, (f) elemental mapping of S-CS-600. S-CS, sulfur-doped interconnected carbon microsphere; SEM, scanning electron microscopy; TEM, transmission electron microscopy

Figure 2a displays the Raman spectra of CS-600 and S-CS-600. Both of the Raman spectra exhibit two strong vibration modes at about 1320 and 1579 cm^–1, resulting from the D and G bands of carbon. Generally, D band comes from the disordered state and the G band is due to the graphite state. ³¹ Thus, the peak intensity ratio (I _D/I _G) of D and G band is always used to evaluate the disordered level of carbon. ^32,33 From the Raman spectra, the I _D/I _G values can be calculated to be 0.955 and 1.536 for CS-600 and S-CS-600, respectively. This result indicates that sulfur doping can improve the disordered structure of carbon, which is helpful to promote the Na-ion storage performance. To confirm the pore structure of S-CSs, nitrogen adsorption–desorption isotherms of CS-600 and S-CS-600 are provided in Figure 2b. According to the fitting results, the specific surface areas of CS-600 and S-CS-600 are 549 and 440 m² g^–1, respectively. Figure 2c shows the pore size distributions of CS-600 and S-CS-600. It is found that CS-600 exhibits more porous structure than S-CS-600. This result indicates that sulfur doping would slightly damage the porous structures of S-CSs. However, this slight damage of porous structures does not affect the Na-ion storage performance due to the volume phase reaction (especially the insertion reaction) in the battery system. ^34,35 To better investigate the change of structure, the X-ray photoelectron spectroscopy (XPS) spectra of S-CSs are provided. From the result of XPS, the sulfur contents of S-CS-400, S-CS-500, S-CS-600, S-CS-700, and S-CS-800 are 13.31, 14.94, 12.46, 9.96, and 7.88 at%, respectively, indicating that the sulfur content in S-CSs decreases with the increase of thermal treatment temperature. Figures 2c and 2d show the C 1s spectra of CS-600 and S-CS-600, respectively. Both of the C 1s spectra can be deconvoluted into three peaks at 284.8, 285.5, and 289.2 eV, which can be indexed to C–C/C═C, C–S/C–O, and O–C═O bonds, respectively. ^36,37 Figure 2f shows the XPS S 2p spectrum of S-CS-600. After deconvolution, we can observe that the doped sulfur is mainly constituted by thiophene S, implying that sulfur has been doped in S-CSs.

Figure 2. (a) Raman spectra, (b) nitrogen sorption isotherms, and (c) pore size distributions of CS-600 and S-CS-600; XPS C 1s spectra of (d) CS-600 and (e) S-CS-600, (f) XPS S 2p spectrum of S-CS-600. S-CS, sulfur-doped interconnected carbon microsphere; XPS, X-ray photoelectron spectroscopy

To evaluate the electrochemical performance of S-CSs, the initial four cyclic voltammetry (CV) curves of CS-600 and S-CS-600 are provided, as shown in Figures 3a and 3b. Both of them exhibit an irreversible capacity contribution in the first cycle, caused by the formation of the solid electrolyte interphase (SEI) layer. Different from the CV curves of CS-600, S-CS-600 exhibits a redox couple located at about 1.0/1.7 V, which can be attributed to the doped sulfur in S-CSs. Figure 3c,d shows the galvanostatic charge–discharge (GCD) profiles of CS-600 and S-CS-600. A couple of plateaus around 1.0/1.7 V can be detected, which are corresponding to their CV curves. To further investigate the electrodes, electrochemical impedance spectroscopy (EIS) spectra of CS-600 and S-CS-600 are also provided, as shown in Figure 3e. It can be seen that S-CS-600 exhibits a smaller charge-transfer resistance than CS-600, indicating that sulfur doping can facilitate the electron transfer, which is helpful to improve its Na-ion storage performance. ³⁸

Figure 3. CV curves of (a) CS-600 and (b) S-CS-600; GCD profiles of (c) CS-600 and (d) S-CS-600; (e) EIS of CS-600 and S-CS-600; (f) cycling performance and (g) corresponding CEs of CS-600 and S-CS-600; (h) rate performance and (i) long-term cycling performance of CS-600 and S-CS-600. CE, coulombic efficiency; CV, cyclic voltammetry; EIS, electrochemical impedance spectroscopy; GCD, galvanostatic charge–discharge; S-CS, sulfur-doped interconnected carbon microsphere

To further evaluate the electrochemical performance of S-CSs, the cycling performance and corresponding Coulombic efficiencies (CEs) of CS-600 and S-CS-600 are also provided, as shown in Figure 3f,g. It is found that CS-600 carries an initial reversible specific capacity of 110 mAh g^–1 and maintains a reversible specific capacity of 138 mAh g^–1 after 50 cycles with a capacity retention rate of 125%. The increased capacity of CS-600 should be attributed to the activated process during sodiation–desodiation process. However, when sulfur is doped into S-CSs, the electrochemical performance of the S-CS-600 electrode is remarkably enhanced, which delivers an initial reversible specific capacity of 640 mAh g^–1 and holds the reversible specific capacity of 520 mAh g^–1 after 50 cycles with a capacity retention rate of 81%. Meanwhile, the initial CE of 33% for CS-600 also improves to 72% for S-CS-600 after sulfur doping. Figure 3h shows the rate performances of CS-600 and S-CS-600. It is found that the S-CS-600 delivers reversible specific capacities of 656, 483, 436, 390, and 257 mAh g^–1 at the current densities of 100, 200, 500, 1000, and 2000 mA g^–1, respectively (with specific capacity retention rates of 39% at 2 A g^–1), which are much higher than that of CS-600 (reversible specific capacities of 102, 90, 66, 49, and 29 mAh g^–1 at the current densities of 100, 200, 500, 1000, and 2000 mA g^–1, respectively [with specific capacity retention rates of 28% at 2 A g^–1]). This result demonstrates that our S-CSs exhibit an excellent Na-ion reversible capacity. To further confirm the electrochemical performance of S-CSs, we also present the long-term cycling performance of CS-600 and S-CS-600 for NIBs, as shown in Figure 3i. It is found that the S-CS-600 electrode managed to hold a high specific capacity of 453 mAh g^–1 at 200 mA g^–1 after 315 cycles, much higher than that of CS-600 (only 103 mAh g^–1 at 200 mA g^–1 after 315 cycles). This result also indicates that our S-CS-600 electrode exhibits remarkable stability for Na-ion storage.

To understand the electrochemical enhancement mechanism of our S-CSs for NIBs after sulfur doping, we investigate their adsorption energies and electronic structures to explore the effect of doping sulfur in carbon (graphene is selected for denoting carbon layer in our work) as anode material for NIBs, based on the DFT calculations. ^39,40 Two types of carbon defects after sulfur doping, including the doping one sulfur atom (S1) and doping three sulfur atoms with one vacancy (S3-V), are considered, as shown in Figure 4. ^41,42 According to the calculations, the defect formation energies (E _f) of S1 and S3-V sites are 1.78 and −3.91 eV, respectively. This result indicates that the doping sulfur in carbon is inclined to the S3-V defect in real experimental conditions. ⁴³ We consider three possible adsorption sites (top site, bridge site, and hollow site) for Na atom on perfect carbon (P-carbon), as shown in Figure 4a. The detailed adsorption energies (E _a) are listed in Table S1. According to the positive E _a values of P-carbon, the hollow site (0.455 eV) is more favored to adsorb Na atom. ⁴⁴ Further explorations show that the doping sulfur atom can increase the adsorption ability of Na atom on the carbon layer, as shown in Figure 4b–d and Table S1. ⁴⁵ From the results of calculation, the values of E _a are about −1.46 to −1.84 eV and −1.76 to −2.33 eV for the S1 carbon and S3-V carbon, respectively, depending on the different adsorption sites of Na atom. The corresponding values reflect that the adsorption ability for Na atom on S3-V carbon is stronger than that on S1 carbon. Furthermore, both sulfur doping states in carbon (S1 site and S3-V site) exhibit more negative E _a than that of P-carbon, indicating that sulfur-doped carbon can greatly enhance its Na-ion storage performance. ⁴⁶ To explore the electronic properties of carbon layer for NIBs, we also discuss the p-band center and electron transfer for the three slabs with the most stable site adsorbed by Na atom, as shown in Figure 4e. It is observed that the adsorption ability of the Na atom exhibits a negative and nonlinear relationship with the p-band center of the slab, which is induced by the different electron transfer between the adsorbed Na atom and the slabs. ^47–49 According to the calculation, the transferred electrons from adsorbed Na to the slab are 0.79, 0.98, and 0.99 e for P-carbon, S1 carbon, and S3-V carbon, respectively. Obviously, the sulfur-doped carbon exhibits a larger electronegativity, resulting in an increase of interaction between the slab and Na atom. This result will lead to much larger transferred electron values of two types of sulfur-doped carbon atoms, as compared with P-carbon. In addition, the electron of the Na atom mainly transfers to carbon atoms (0.98 e) in S1 carbon and sulfur atom (0.99 e) in S3-V carbon. The most stable adsorption site for the Na atom on the three slabs is picked up for comparison (detailed atomic configurations are shown in Figure S2). On the basis of the aforementioned calculations, we find that sulfur doping in carbon could facilitate electron transfer and increase the p-band center and adsorption ability of the Na atom, resulting in the remarkable improvement of its Na-ion storage performance.

Figure 4. The possible adsorption sites of Na atom on (a) P-carbon layer, (b) S1 carbon, and (c) S3-V carbon; (d) Ea of Na atom on P-carbon, S1 carbon, and S3-V carbon; (e) the p-band center and the Ea for each slab with Na adsorption, and the transferred electron from Na atom to the slab is also listed. The “T,” “B,” and “H” represent the top, bridge, and hollow sites, respectively. The gray, yellow, and violet balls indicate the carbon, sulfur, and sodium atoms, respectively. Some unstable adsorption sites of Na atom on S1 carbon are also listed in Figure S1

To further investigate the electrochemical behavior of our S-CSs, the CV curves of the S-CS-600 electrode at different scan rates are provided, as shown in Figure 5a. It can be observed that all of the CV curves exhibit a similar shape, indicating that our electrode exhibits a stable electrochemical reaction. According to the relationship of scan rates (v) and current (i) shown below, ⁵⁰ [Image Omitted. See PDF]which can be rearranged to[Image Omitted. See PDF]

Figure 5. (a) CV curves of S-CS-600 at different scan rates between 0.2 and 1.2 mV–1 after 10 cycles; (b) relationship between log (i) and log (v) of S-CS-600; (c) capacitive contribution area at 1.0 mV s–1 and (d) capacitive contribution ratios at different scan rates of S-CS-600. CV, cyclic voltammetry; S-CS, sulfur-doped interconnected carbon microsphere

For the constant in Equation (2), the b value is determined by the slope. Generally, the b value close to 0.5 represents diffusion-controlled behavior and the b value close to 1 is capacitive behavior. According to the linear fitting of peaks 1, 2, 3, and 4 in Figure 5b, we can obtain the corresponding b values of 1.12, 0.73, 0.94, and 0.70, respectively, which indicate that the specific capacity of our S-CSs is mainly constituted by capacitive behavior. To further quantify the capacitive contribution of our S-CSs, the following equation is utilized for investigation ⁵¹ :[Image Omitted. See PDF]which can be converted to[Image Omitted. See PDF]where k ₁ v and k ₂ v ^1/2 represent contributions of capacitive and diffusion-controlled behaviors, respectively. Therefore, the contributions of capacitive and diffusion-controlled behaviors can be quantified from the values of k ₁ and k ₂. After calculation, we can confirm the capacitive contribution of 72% at a scan rate of 1.0 mV s^–1, as shown in Figure 5c. Figure 5d presents the capacitive contributions of different scan rates, which exhibit the contribution ratios of 54%, 60%, 65%, 69%, 72%, and 75% at the scan rates of 0.2, 0.4, 0.6, 0.8, 1.0, and 1.2 mV s^–1, respectively. These results indicate that capacitive storage of the Na ion accounts for the main contribution. Such high capacitive contribution indicates that our S-CSs electrode can withstand the impact of higher density currents, thus enhancing the electrochemical performance of the electrode material.

To confirm the electron transfer characteristic of our S-CSs for Na-ion storage, we also present the galvanostatic intermittent titration technique (GITT) curves of CS-600 and S-CS-600 for comparison. Figures 6a and 6b display the GITT profiles in the initial two cycles. The corresponding diffusion coefficients of Na ion can be calculated according to the equation below ⁵⁰ :[Image Omitted. See PDF]where m _B, M _B, V _m, S, L, and τ represent the mass, molecular weight, the molar volume, surface area, thickness, and the time for pulse galvanostatic current, respectively. ΔE _s and ΔE _τ involve the steady-state voltage change after pulse current and the voltage change during the pulse current. According to Equation (5), the diffusion coefficients of Na ion of CS-600 and S-CS-600 can be obtained, as shown in Figure 6c–f. From the GITT curves, we can observe that the average diffusion coefficient of the S-CS-600 electrode is significantly higher than that of the CS-600 electrode, indicating that the doping sulfur can remarkably decrease diffusion barrier energy of S-CSs during the sodiation–desodiation process, which also confirms the result of our calculation.

Figure 6. GITT measurement of (a) CS-600 and (b) S-CS-600; the Na-ion diffusion coefficients of CS-600 and S-CS-600 from GITT curves after calculation during (c) initial discharge, (d) initial charge, (e) second discharge, and (f) second charge processes after normalization of times. GITT, galvanostatic intermittent titration technique; S-CS, sulfur-doped interconnected carbon microsphere

To further investigate the Na-ion storage behavior of our S-CSs, we also present the operando Raman spectroscopies of S-CS-600 during the sodiation–desodiation process. Figures 7a and S3 show the measurement schematic and photograph of operando Raman system. As shown in Figure 7b, we can observe that the D and G bands of S-CS-600 in the Raman spectra are changed in the different sodiation–desodiation states. In the initial discharge process from 2.5 to 1.0 V, the position of D bands exhibits a right shift with the decrease of peak intensity, indicating the occurrence of surface passivation. ⁵² Meanwhile, the Na-ion surface adsorption and preinsertion process in NIBs also occur in this stage. ⁵³ An interesting phenomenon while discharging to 0.5 V is that the G-band intensity gradually enhances and the D-band intensity remains unchanged, indicating that there is a decrease in the disorder structure of our S-CSs. We suggest that this phenomenon should be attributed to the breathing motion of sp² carbon atoms with the adsorption of Na ions. ^54,55 This result could be due to the formation of an SEI layer in this stage. Though further discharging to 0.01 V, we also observe that the intensity of G band weakens again, which is regarded as the Na-ion intercalation process of S-CSs. In the charging process from 0.01 to 1.0 V, the G-band intensity gradually increases, which is attributed to the deintercalation process of Na ion. While further charging to 3 V, we can observe a left shift of G and D band, indicating that the structure of S-CSs recovers, which also proves that the S-CSs gradually return to the original state after the Na-ion deintercalation process.

Figure 7. (a) A schematic of the operando Raman system, (b) operando Raman spectra, and (c) selected Raman spectra of S-CS-600 in the initial cycle. S-CS, sulfur-doped interconnected carbon microsphere

After investigating the effect of sulfur doping in S-CSs for NIBs, we also compared the Na-ion storage performance of S-CSs after different temperature treatments, as shown in Figure 8a. As observed, the S-CS-400 and S-CS-500 deliver reversible specific capacities of 322 and 465 mAh g^–1 at 100 mA g^–1 after 50 cycles, respectively. With the increase of treatment temperature, S-CS-600 exhibits an optimal Na-ion storage performance of 520 mAh g^–1 after 50 cycles. However, with the further increase of treatment temperature, S-CS-700 and S-CS-800 show decreasing reversible specific capacities of 470 and 382 mAh g^–1 after 50 cycles, respectively. Although the reversible specific capacities changed, all of the S-CSs treated under different temperatures exhibit a similar initial CE (Figure 8b), which is supposed to be the similar surface structure after sulfur doping. To investigate the electrochemical improvement of S-CSs treated at a suitable temperature, the EIS spectra of S-CSs treated at different temperatures are provided, as shown in Figure 8c. It is found that the electron transfer impedance decreases with the increase of treatment temperature, which can facilitate the Na-ion storage. However, the high treatment temperature of our S-CSs also leads to a decrease in the disordered structure, as shown in Figure S4. This results in the reduction of the lattice distance, which goes against the Na-ion storage. Furthermore, we also provide Na-ion storage performance of all the sulfur-doping carbon materials for comparison, as shown in Figure 8d. ^{13,25,27,37,40,56–60} It is found that our S-CSs surpass all the sulfur-doping carbon materials, which demonstrates that our S-CSs exhibit superior performance for Na-ion storage.

Figure 8. (a) Cycling performance, (b) corresponding CEs, and (c) EIS spectra of S-CS-400, S-CS-500, S-CS-600, S-CS-700, and S-CS-800; (d) Na-ion storage performance of other sulfur-doping carbon-based materials. CE, coulombic efficiency; EIS, electrochemical impedance spectroscopy; S-CS, sulfur-doped interconnected carbon microsphere

CONCLUSION

In summary, we reasonably obtain S-CSs by a hydrothermal process with subsequent thermal treatment, which exhibit an excellent Na-ion storage performance. After sulfur doping, the S-CSs deliver a high reversible capacity of 520 mAh g^–1 at 100 mA g^–1 after 50 cycles, and a capacity of 453 mAh g^–1 at 200 mA g^–1 is maintained after 315 cycles. Meanwhile, we utilize operando Raman spectroscopy to further analyze the electrochemical reaction of our S-CSs during sodiation–desodiation process. To reveal the electrochemical enhancement of our S-CSs for Na-ion storage, we also provide DFT calculations. The results show that sulfur doping in carbon can facilitate electron transfer and increase the p-band center and adsorption ability of the Na atom, resulting in the remarkable improvement of its Na-ion storage performance.

EXPERIMENTAL SECTION Synthesis

Briefly, 60 mL of 0.75 M d-glucose was poured into a 100 mL Teflon-lined autoclave and heated to 180°C for 24 h. Then, the precipitate was cleaned by deionized water and dried in an oven at 80°C for 24 h, which yielded the carbon sphere precursor. Subsequently, the carbon sphere precursor was mixed with sulfur powder with a ratio of 1:1 and then annealed at different temperatures for 2 h under the flowing nitrogen. After cooling down, the S-CSs were obtained. Among the samples, the obtained S-CSs treated at 400°C, 500°C, 600°C, 700°C, and 800°C were named as S-CS-400, S-CS-500, S-CS-600, S-CS-700, and S-CS-800, respectively. For comparison, the carbon sphere precursor annealed at 600°C without mixing sulfur powder was obtained, which was named CS-600. To confirm the yield of S-CS-600, we also compared with the mass of S-CS-600 before and after thermal treatment and found that the yield of carbon was about 66%.

Characterization

The morphologies of electrode materials were analyzed by field-emission SEM (Ultra 55; Zeiss) with an Oxford Aztec energy dispersive spectrometer and TEM (JEOL-2100). The structure and the surface of samples were measured by a Raman spectrometer (T64000; Horiba) and XPS (K-Alpha; Thermo Fisher Scientific). Brunauer–Emmett–Teller specific surface area was obtained from nitrogen adsorption isotherms at 77 K by a nitrogen adsorption apparatus (×1000; Biaode-Kubo). The detailed DFT computational methods are described in Supporting Information.

Electrochemical measurement

For the preparation of the electrode, the active material, super P, and sodium carboxymethyl cellulose were mixed with a mass ratio of 7:1:2 in water to form sizing. Then the sizing was coated on the Cu foil and dried at 100°C in a vacuum oven overnight. Next, the obtained polar piece was cut into a disc with a diameter of 14 mm as a working electrode (all the areal loadings of our samples are about 1 mg cm^–1). The CR2032-type half batteries were packaged in a glovebox (Lab2000; Etelux) under argon atmosphere. Na foil and glass fiber filter (Whatman) were used as a counter and separator, respectively. Also, 0.2 mL of 1 M NaClO₄ in ethylene carbonate/propylene carbonate (volume/volume) mixed solvent was used as an electrolyte in each battery. The GCD curves and cycling performance of batteries were recorded by a battery test system (BTS-4000; Neware) in a voltage range of 0.01–3.0 V at 100 mA g^–1, unless otherwise specified. CV curves were measured by an electrochemical workstation (C1030; Shanghai Chenhua) in a voltage range of 0.01–3.0 V at a scan rate of 0.2 mV s^–1, unless otherwise specified. EIS was performed by an electrochemical workstation (STAT 3400; Princeton Veras) with a frequency range of 0.01–100 kHz. GITT was measured by a battery test system (CT3001A; Land Electronic) under the condition of a pulse current density of 50 mA g^–1 for 0.5 h and relaxation time of 2 h.

ACKNOWLEDGMENTS

The authors thank the financial support from the National Natural Science Foundation of China (51702056, 51772135, 21703081), the Ministry of Education of China (6141A02022516), the Shenzhen Science and Technology Program (JCYJ20200109113606007), and the Fundamental Research Funds for the Central Universities (21617330). They also acknowledge the computational support of the High-Performance Super Computing Platform of Jinan University.

CONFLICT OF INTERESTS

The authors declare no conflict of interests. [Correction added on 31 August 2021, after first online publication: Conflict of Interest section has been added.]