Introduction

Lithium–sulfur (Li–S) batteries have recently attracted great interest as promising electrochemical devices for energy conversion and storage applications because of the abundance, low cost, environmental friendliness and high theoretical capacity (1,675 mA h g−1) of sulfur1,2,3,4. Despite these advantages, the practical application of Li–S batteries is still handicapped by the following problems: (1) the low electrical conductivities of sulfur (5 × 10−30 S cm−1 at 25 °C), intermediate polysulphides and Li2S; (2) the dissolution of lithium polysulphides, which results in a shuttling effect and in the deposition of insoluble lithium sulfide on the anode in each of the charge/discharge cycles and eventually the complete loss of capacity of the sulfur cathode; and (3) severe volume changes in the active electrode materials during the lithiation/delithiation processes1,4,5,6,7, resulting in the pulverization of the electrode materials.

To overcome these problems, various carbon materials, including graphene8,9,10,11,12,13,14,15, carbon nanotubes16,17, porous carbon18,19,20,37,38,Fig. 3g–i), as well as the homogenous distribution of sulfur nanoparticles in the PGC framework (Fig. 3i, sulfur map**). A nanosized distribution is extremely important for the application of the sulfur particles as a cathode material for Li–S batteries, as the utilization rate of sulfur is higher for smaller sulfur particles because of the short diffusion path of the electrons and lithium ions3. Compared with previously reported hybrid structures that have a comparably high sulfur content (∼70 wt%) (refs 10, 11, 30), the sulfur particles in the 3D S@PGC (90% S) were much smaller and had a much more uniform distribution. These features can be attributed to the simultaneous formation of the porous structures and the in situ deposition of sulfur nanoparticles through the oxidation of Na2S.

Figure 3: Morphology of the 3D S@PGC (90% S) composite.
figure 3

(a–c) SEM images and (d,e) TEM images of the 3D S@PGC (90% S) composite at different magnifications. (f) HRTEM image of a sulfur nanoparticle in the composite. (g) TEM image of the 3D S@PGC (90% S) composite. (h,i) EDS elemental maps of (h) carbon and (i) sulfur, which were collected from the entire area shown in g. Scale bars in a,b and c: 20; 1; and 0.5 μm. Scale bars in d,e,f and g: 500; 50; 2; and 200 nm. The SEM and TEM images indicate that the composite possesses a 3D network consisting of interconnected submicron-sized macropores. From the SEM images, the sulfur nanoparticles anchored to the walls of the PGC network were calculated to have a size distribution of 18–54 nm. The EDS results indicate that the sulfur is uniformly distributed in the composite. HRTEM, high-resolution TEM.

Full size image

To further investigate the interactions between sulfur and 3D PGC, X-ray photoelectron spectroscopy (XPS) was performed. For comparison, XPS data were also collected for pure 3D PGC, which was prepared by immersing NaCl–Na2S@GC in water to remove NaCl and Na2S (Fig. 4a). The C 1s XPS spectrum of pure 3D PGC has a major peak at 284.7 eV, corresponding to sp2 hybridized carbon, as well as three weak peaks at 286.4, 287.2 and 288.9 eV, which can be ascribed to C–O, C=O and O–C=O species, respectively14. The survey XPS spectrum of the 3D S@PGC (90% S) composite confirms the presence of sulfur in 3D PGC (Fig. 4b). In contrast to the C 1s XPS spectrum of pure 3D PGC, that of the 3D S@PGC (90% S) composite has an additional peak at 285.5 eV, which is ascribed to C–S bonds (Fig. 4c)14. This finding reveals the presence of covalent bonding between sulfur and PGC. The S 2p XPS peaks, that are characterized by an S 2p3/2 and 2p1/2 doublet with an energy separation of 1.2 eV, reconfirm the presence of C–S bonds (Fig. 4d), as the binding energy of the S 2p3/2 peak (163.5 eV) is lower than that of elemental sulfur (164.0 eV)14,44. The weak peak at 168.6 eV is due to sulfate species formed by the oxidation of sulfur in air14. The presence of C–S bonds is also supported by Fourier transform infrared (FTIR) spectroscopy because the vibration characteristic of C–S bonds was detected at 671 cm−1 (Supplementary Fig. 6)45. The C–S bonds could be formed through the addition of various reactive intermediates, including free radicals (for example, HS·) and radical anions (for example, S·− and SX·−)46, to the unsaturated carbon–carbon double bonds of the PGC as well as through the nucleophilic attack of transient negatively charged polysulphides (for example, SX2−) with residual oxygen-containing functional groups present in the PGC (see Supplementary Note 1). Therefore, the C−S bonds were formed during the oxidation of Na2S by Fe(NO3)3 because free radicals, radical anions and negatively charged polysulphides were the intermediate products of the oxidation reaction46.

Figure 4: XPS results for pure 3D PGC and the 3D S@PGC (90% S) composite.
figure 4

(a) C 1s XPS spectrum of pure PGC. (b) XPS survey spectrum of the 3D S@PGC (90% S) composite. (c) C 1s and (d) S 2p XPS spectra of the 3D S@PGC (90% S) composite. The data indicate the presence of C–S bonds in the 3D S@PGC (90% S) composite.

Full size image

To reconfirm the presence of C−S bonds, the 3D S@PGC (90% S) composite was subjected to Soxhlet extraction using CS2. The TGA curve of the extracted sample revealed a continual weight loss up to 700 °C (Supplementary Fig. 7); such a weight loss could be assigned to the removal of bonded sulfur47, and the percentage of bonded sulfur was estimated to be ∼48 wt%. By comparing the sulfur content of the as-prepared 3D S@PGC (90% S) composite to that of the Soxhlet-extracted sample, the bonded and unbonded sulfur content were calculated to be ∼9 and 81 wt% in the 3D S@PGC (90% S) composite (Supplementary Table 1). The X-ray diffraction pattern of the extracted sample did not have any sulfur peaks (Supplementary Fig. 8), which is in line with the formation of C−S bonds48. Compared with previously reported physical strategies for confining sulfur37,52,59,60,61,62. Although some capacities reported at higher current densities (for example, 4 and 5 C) were relatively high (650–750 mA h g−1), the sulfur content of their composites was relatively low (40–65 wt%)37,63,69. Therefore, 3D S@PGC is a high-sulfur-content (up to 90%) cathode material that exhibits excellent cycling stability at a high current density. The pure sulfur cathode (the control) used under the same conditions exhibited a much lower specific capacity and worse cycling stability than those of the 3D S@PGC (90% S) composite (Supplementary Fig. 18).

Discussion

The electrochemical performance of the 3D S@PGC (64% S) and 3D S@PGC (70% S) composites as cathodes in Li–S batteries were also evaluated. As summarized in Supplementary Fig. 19 and Supplementary Tables 3 and 4, 3D S@PGC composites with a lower sulfur content exhibited better performance: specifically, higher specific capacities, higher capacity retention and higher rate performance (Supplementary Note 3). These results are attributable to the smaller sizes of the sulfur nanoparticles in the 3D S@PGC composites with a relatively low sulfur content. Indeed, smaller sulfur nanoparticles have larger specific contact areas with the 3D PGC framework, which helps alleviate the shuttling effect and improves the cycle stability. Smaller particle sizes also facilitate electron and Li+ diffusion and lead to better sulfur utilization and a higher specific capacity. The larger specific surface areas effectively reduce the discharging current densities and the Li+ flux, thereby limiting the formation of a Li2S blocking layer at high charge/discharge rates51. Although the 3D S@PGC composites with a lower sulfur content exhibited higher specific capacities calculated on the basis of sulfur, the relatively low sulfur content reduced the overall volumetric capacity and energy density of the corresponding cathodes. Therefore, 3D S@PGC composites with relatively high sulfur content may be promising candidates for use in practical applications.

The excellent overall electrochemical performance of the 3D S@PGC composites can be attributed to the following factors that stem from the design of the materials. First, the in situ chemical deposition method allows access to composites with high sulfur content (up to 90 wt%) and affords the nanoscale distribution of the sulfur particles in the resultant 3D PGC network. As described above, nanosized sulfur particles facilitate a high sulfur utilization rate (82.5% for 3D S@PGC (90% S), 84.5% for 3D S@PGC (70% S) and 86% for 3D S@PGC (64% S) at 0.5 C). Second, the C–S bonds formed between the sulfur nanoparticles and 3D PGC can effectively prevent agglomeration of the sulfur nanoparticles, minimize the loss of lithium polysulphides to the electrolyte and suppress the shuttling effect during the charge/discharge cycles. Third, the 3D PGC networks that display high electrical conductivities, large surface areas and high mechanical flexibility confer high electrical conductivity and structural integrity to the electrodes. The numerous walls between the interconnected macropores may function as multilayered barriers to further mitigate the dissolution of polysulphides into the electrolyte. Finally, the unique interconnected hierarchical pores in the 3D PGC network facilitate access to the sulfur nanoparticles by the electrolyte and preserve the rapid transport of Li+ to the active material.

In conclusion, we report a new methodology that is facile and scalable and allows the in situ preparation of 3D S@PGC composites with a high sulfur content. The strategy utilizes Na2S as a sulfur precursor and NaCl and Na2S as a template for the porous structure of the resultant composite. The sulfur nanoparticles were homogenously distributed and covalently bonded to 3D PGC, as confirmed by various spectroscopic and microscopic techniques. Li–S batteries prepared using the composites as cathodes exhibited excellent performance; specifically, high sulfur utilization, high specific capacities, good cycling stabilities and high rate capabilities were observed. Notably, Li–S batteries prepared using 3D S@PGC (90% S) as a cathode displayed a long cycling stability, with a capacity decay of only 0.039% per cycle over 1,000 cycles at a high charge/discharge current (2 C). Overall, the methodology described herein offers a new avenue for the fabrication of cathode materials based on carbon–sulfur hybridized nanostructures for use in high-performance Li–S batteries. We believe that the strategy may also inspire the preparation of other 3D porous structures for use in other areas, including applications in catalysis, selective adsorption, separations and sensing.

Methods

Materials

All reagents were purchased from commercial sources and used without further purification. All solvents used were purified using standard procedures.

Representative synthesis of 3D S@PGC composites

In a typical synthesis, Na2S·9H2O (2.0 g), NaCl (5.0 g) and glucose (0.8 g) were dissolved in DI water (15 ml). The resultant solution was frozen in liquid nitrogen, and the water in the mixture was removed via freeze-drying. The resultant gel was ground into a fine powder and then heated at 750 °C for 2 h under an atmosphere of argon. A black powder was obtained and subsequently stirred in an aqueous solution of Fe(NO3)3 (20 g Fe(NO3)3·9H2O in 150 ml DI water) for 40 h to dissolve the residual NaCl crystals and to deposit the sulfur. Afterwards, the black powder product was washed several times with DI water and centrifuged to afford the desired composite. Composites with various sulfur contents were synthesized by using different amounts of glucose (0.9 and 1.0 g) in the aforementioned procedure.

Characterization

X-ray diffraction data were collected on a Bruker D8 Focus diffractometer using an incident wavelength of 0.154 nm (Cu Kα radiation) and a Lynx-Eye detector. Raman spectra were recorded on a Renishaw inVia-Reflex confocal Raman microscope at an excitation wavelength of 532 nm. TGA measurements were carried out using a TGA Q50 at a scanning rate of 10 °C min−1. SEM observations were performed on a field-emission SEM (Hitachi S-4800) equipped with EDS. TEM images were obtained using a JEOL-2100F microscope operated under an accelerating voltage of 200 kV. EDS analysis was also performed on Tecnai F20 scanning transmission electron microscope operated at 200 keV using an Oxford detector with a beam current of ∼1 nA. N2 adsorption–desorption isotherms and pore-size distribution were obtained at 77 K using a QuadraSorb SI MP apparatus. The total specific surface areas of the samples were calculated via the Brunauer–Emmett–Teller method. The pore-size distribution was calculated via the density functional theory model. XPS spectra were recorded on a PHI Quantera Scanning X-ray Microprobe using monochromated Al Kα radiation (1486.7 eV). FTIR spectra were recorded on an Excalibur 3100 spectrometer with a resolution of 0.2 cm−1 using KBr pellets.

Electrochemistry

The 3D S@PGC composites were combined with conductive carbon and poly(vinylidene fluoride) as a binder with a mass ratio of 80:10:10 and milled into a slurry with N-methylpyrrolidone. The slurry was then blade cast onto a carbon-coated Al foil and dried at 50 °C for 10 h in a vacuum oven. The loading density of sulfur was ca. 2.36 mg cm−2. CR2032 coin cells were assembled in an argon-filled glove box employing the 3D S@PGC-coated Al foil as the cathode, a porous membrane (Celgard 3501) as the separator, and lithium foil as the reference/counter electrode. The electrolyte used was lithium bis(trifluoromethane)sulphonimide (0.38 M) and lithium nitrate (0.31 M) in a solvent mixture of 1,3-dioxolane and 1,2-dimethoxy ethane (1:1 v/v). Pristine sulfur electrodes were fabricated under similar conditions. Cyclic voltammetry curves were collected using a CHI 660E electrochemical workstation at a scan rate of 0.1 mV s−1 from 3.0 to 1.5 V. Cycling tests of the batteries were galvanostatically performed at various charge/discharge rates within a potential window of 1.5–3.0 V versus Li+/Li. The electrochemical impedance spectroscopy data were recorded using a Zennium 40088 electrochemical workstation by applying a sine wave with an amplitude of 10 mV over a frequency range of 100 kHz to 10 mHz.

Additional information

How to cite this article: Li, G. et al. Three-dimensional porous carbon composites containing high sulfur nanoparticle content for high-performance lithium–sulfur batteries. Nat. Commun. 7:10601 doi: 10.1038/ncomms10601 (2016).