Tungsten Nanoparticles Accelerate Polysulfides Conversion: A Viable Route toward Stable Room‐Temperature Sodium–Sulfur Batteries

Abstract Room‐temperature sodium–sulfur (RT Na–S) batteries are arousing great interest in recent years. Their practical applications, however, are hindered by several intrinsic problems, such as the sluggish kinetic, shuttle effect, and the incomplete conversion of sodium polysulfides (NaPSs). Here a sulfur host material that is based on tungsten nanoparticles embedded in nitrogen‐doped graphene is reported. The incorporation of tungsten nanoparticles significantly accelerates the polysulfides conversion (especially the reduction of Na2S4 to Na2S, which contributes to 75% of the full capacity) and completely suppresses the shuttle effect, en route to a fully reversible reaction of NaPSs. With a host weight ratio of only 9.1% (about 3–6 times lower than that in recent reports), the cathode shows unprecedented electrochemical performances even at high sulfur mass loadings. The experimental findings, which are corroborated by the first‐principles calculations, highlight the so far unexplored role of tungsten nanoparticles in sulfur hosts, thus pointing to a viable route toward stable Na–S batteries at room temperatures.

pure PC/EC, after dried, the electrodes were covered in the XPS holder and transferred to the XPS chamber. Thermogravimetric analysis (Mettler Toledo TGA/DSC 3+) was conducted under the N 2 atmosphere by heating from RT to 750 ℃ at 5 ℃ min -1 . Ultraviolet/visible absorbance spectroscopy was performed in the spectral range of 200-800 nm using a Cary 5000 UV-vis variable wavelength spectrophotometer to evaluate the sodium NaPSs absorption capability of W@N-G and NG composite (the Na 2 S 6 solution was synthesized by mixing Na 2 S and sulfur in a stoichiometric ratio of 2:6 in DME (dimethoxyethane)).
The W Nanoparticles Size Statistical Analysis: The W nanoparticles size distribution was calculated by the Photoshop soft, briefly, enlarge the STEM image 100 times (Figure 2c), select 50 W nanoparticles (random) and measure the largest length of each nanoparticle, and finally calculate the size distribution.
Electrochemical Measurements: The working electrodes for the Na-S cells were fabricated by mixing the as-synthesized composites (W@N-G/S or NG/S), carbon black, and polyvinylidene difluoride (PVDF) with a weight ratio of 8: 1: 1 in N-methyl-2-pyrrolidone (NMP) to form a slurry, then uniformly pasted on the aluminum foil followed by drying under vacuum oven at 60 °C overnight. The Na-S cells were assembled with metallic sodium as the anode and the W@N-G/S (or NG/S) as the cathode by the CR2032 coin cell, glass fiber (Whatman GF/F) as the separator, and 1 M NaClO 4 in 1:1 (volume ratio) ethylene carbonate/propylene carbonate (PC/EC) with 3 wt. % fluoroethylene carbonate (FEC) additive as the electrolyte, and 5 μL mg -1 and 10 μL mg -1 electrolyte was used in low and high (> 3 mg cm -2 ) sulfur loading electrodes, respectively. The assembly of all the Na-S cells was carried out in the Ar-filled glovebox (MBraun) with water and oxygen 5 concentrations less than 0.1 ppm. The electrochemical properties of the W@N-G/S and NG/S cathodes were investigated by the LAND CT 2001A charge/discharge system with a cut-off voltage range from 0.8 to 2.6 V (vs. Na/Na + ). The cyclic voltammetry (CV, at a scan rate of 0.1 mV s -1 ) and electrochemical impedance spectroscopy (EIS) of the cells were conducted using Metrohm Auto-lab.
Computational Methods: Based on density functional theory, the first-principles calculations were performed using LCAO calculator as implemented in QuantumATK package to investigate the equilibrium configurations and adsorption energies of NaPSs (Na 2 S n (n=1, 2, 4, 6, 8)) and S 8 on N-doped graphene film. [1] The valence electrons and core interactions were described with PseudoDojo pseudopotentials. A generalized gradient approximation (GGA) proposed by Perdew, Burke, and Ernzerhof (PBE) was used to treat the exchange-correlation functional. [2] A density mesh cut-off of 120 Ha was used to ensure reliable accuracy. The van der Waals (vdW) correction was also considered by using a Grimme DFT-D3 dispersion term. [3] A 6 × 7 × 1 supercell of graphene was utilized to adsorb NaPSs. A vacuum layer of at least 20 Å perpendicular to the graphene film was applied to avoid the interaction between neighboring images. The first Brillouin zone was sampled using a 2 × 2 × 1 and 6 × 5 × 1 Monkhorst-Pack k-point scheme for structural optimization and adsorption energy calculations. All the structures were fully relaxed until the residual Hellmann-Feynman force on each atom is smaller than 0.01 eVÅ -1 . The total energy convergence criterion was 1 × 10 -6 eV. The adsorption energy was defined as: 240 (5 C) [4] CNF-L@Co/S 55% 745 (0.5 C) 538 (after 150 cycles) 442.7 (1.5 C) [5] S/TiN-TiO 2 @MCCFs 43.1% 1150 (0. 349 (3 C) [11] W@N-G   14 Figure S7. The electronic properties of W@N-G with and without the adsorption of NaPSs.
15 Figure S8. Photos showing the glass fiber separators after three cycles.
16 Figure S9. SEM of (b) N-G/S and (b) W@N-G/S cathodes after three cycles. Figure S10. First three charge/discharge curves of (a) N-G/S and (b) W@N-G/S cathodes. Figure S11. Ex-situ XPS spectra of N-G/S cathode after discharged to 0.8 V (after three cycles).