Dual redox mediators accelerate the electrochemical kinetics of lithium-sulfur batteries

Lithium-sulfur (Li-S) batteries are attractive for next-generation energy storage due to sulfur’s high theoretical energy density, environmental benignity, and low cost. However, their practical implementation is hindered by sluggish electrochemical kinetics involving multiple polysulfide intermediates and by the shuttling of soluble species, which degrades capacity and cycling life. Prior approaches have focused on confining sulfur in conductive scaffolds (porous carbons, graphene, CNTs) and introducing barriers to mitigate shuttle, but high-energy and high-power Li-S cells remain challenging because the identities and reaction pathways of the active sulfur species are not fully understood. The study aims to elucidate the geometric and electronic structures of sulfur species via first-principles calculations, construct an electronic energy diagram to reveal reaction pathways and the origin of sluggish kinetics, and propose a strategy that couples fast electrochemical reactions with spontaneous chemical reactions using redox-mediating pseudocapacitive oxides to accelerate both discharge and charge processes in sulfur cathodes.

• Conventional strategies: Confinement of sulfur and polysulfides within conductive scaffolds (porous carbon, graphene, carbon nanotubes) to improve conductivity and reduce shuttle; physical/chemical barriers and interlayers to suppress polysulfide migration. These mitigate but do not fully resolve sluggish kinetics or enable high power in practical thick electrodes with lean electrolyte.
• Kinetic enhancers explored: Various polar hosts and catalysts including oxides, sulfides, nitrides, and carbides have been studied to enhance redox kinetics of sulfur species. Nonetheless, a clear, mechanistically grounded design linking electronic structure alignment of mediators with active sulfur intermediates has been lacking.
• Knowledge gaps: Difficulty in pinpointing critical active species and their reaction pathways; limited understanding of molecular origins of slow electrochemical steps; lack of strategies that independently optimize discharge and charge pathways given their differing redox requirements.

Computational:

• DFT calculations to determine geometric/electronic structures of sulfur species and mediators.
• Sulfur species modeled: cyclo-S8; solvated Li2Sn chains (n=4–8) with explicit 0–3 DOL molecules in the first Li+ solvation shell and implicit solvent via a polarizable continuum model; radical species LiSm (m=2–5) with various DOL solvation; crystalline Li2S2 structure adopted from global optimization literature.
• Functionals/basis: B3LYP/6-311++G(d,p) for isolated molecular species (Gaussian09) to compute HOMO/LUMO; SCAN and HSE06 for periodic oxides and Li2Sx (VASP), plane-wave cutoff 400 eV, dense k-mesh ensuring <0.01 eV/unit cell energy convergence.
• Band alignment: Absolute band positions (VBM/CBM or HOMO/LUMO) aligned to vacuum using semicore orbital references (Nb 4s, Mn 3s, Li 1s for Li2Sx) and thick (>35 Å) slabs to minimize Madelung potential differences; band gap centers (BGCs) used as robust redox proxies.
• Oxide mediators: Orthorhombic Nb2O5 and birnessite MnO2 structures optimized pre-/post-lithiation (LiNb2O5, Li0.5MnO2). DOS and BGCs computed to assess redox alignment and conductivity changes upon lithiation.

Materials synthesis:

• RGO via Hummers oxidation of graphite followed by ascorbic acid reduction.
• RGO–Nb2O5: Ethanol-based NbCl5 precursor with oleylamine and trace water; 75 °C solvothermal, washing, freeze-drying, then 600 °C Ar anneal.
• RGO–MnO2: Room-temperature redox deposition using KMnO4/MnSO4 onto RGO in water; 12 h stir, wash, freeze-dry.
• S-composites (S–RGO, S–Nb2O5, S–MnO2, S–Nb2O5–MnO2 (1:1 oxides) by liquid infiltration at 159 °C for 4 h; sulfur:host mass ratio 4:1.

Electrode/cell preparation and testing:

• Cathodes: Slurry of carbon/sulfur composite:carbon fiber:sodium alginate = 8:1:1, cast on carbon-coated Al foil; dried at 70 °C under vacuum. Thick electrodes used with areal sulfur loading ~7 mg cm−2 and controlled electrolyte-to-sulfur ratio (E/S=7), unless varied.
• Electrolyte: 0.5 M LiTFSI + 2 wt% LiNO3 in DOL/DME (1:1 v/v). Separator: Celgard 2500. Anode: Li metal. Voltage window: 1.7–2.8 V.
• Catholyte for kinetic tests: 0.5 M LiTFSI with Li2S6 (e.g., 50–100 mM) in DOL/DME. Li2Sx solutions prepared by reacting S with Li2S at 130 °C for 24 h.
• Electrochemical measurements: Cyclic voltammetry (Bio-Logic VMP3, three-electrode with Li counter/reference) at 0.1–0.5 mV s−1; linear voltammetry/Tafel analysis (Solartron) to extract exchange current i0; galvanostatic cycling at 0.05–0.1C and rate tests at 0.5–3 mA cm−2 (single cycles at each rate to avoid Li-anode degradation).

Characterization:

• XRD (Cu Kα), TGA (5 °C min−1 to 700 °C in air), SEM/TEM, XPS (AXIS Ultra DLD) with Ar+ depth profiling; spectra fitted with Gaussian–Lorentzian peaks and Shirley background; C 1s at 284.5 eV as reference.

Analytical approaches:

• Tafel plots to estimate exchange currents and infer standard rate constants k0 trends.
• Randles–Sevcik analysis of CV peak currents (Ip vs v0.5) to compare apparent Li+ diffusion coefficients D(Li+) among electrodes for specific redox steps.
• XPS to confirm spontaneous redox between mediators and polysulfides (changes in Nb 3d, Mn 2p, S 2p states).

• Electronic structure of sulfur intermediates:
  • S8 is highly insulating (band gap ~4.59 eV; conductivity ~1×10−15 S cm−1).
  • Closed-shell solvated Li2Sn–4DOL (n=4–8) have large band gaps of 2.91–3.76 eV, increasing with shorter chains; LUMOs rise as n decreases, indicating increasing reduction difficulty.
  • Radical LiSm–2DOL (m=2–5) exhibit narrower band gaps (1.72–2.07 eV) and lower LUMO/BGC energies (BGC −4.12 to −4.18 eV) than Li2Sn–4DOL (BGC −3.9 to −3.05 eV), implying radicals are preferentially and more easily reduced and are key active intermediates during discharge.
  • Li2S2 band gap computed as 2.29 eV with SCAN (1.8 eV PBE; 3.04 eV HSE06).
• Mediator selection and alignment:
  • Pseudocapacitive Nb2O5/LixNb2O5 active at 1.2–2.0 V vs Li/Li+ functions as electron–ion source during discharge; MnO2/LiyMnO2 active at 2.4–3.6 V functions as electron–ion drain during charge.
  • Lithiation of mediators lowers band energies and partially fills conduction bands (LiNb2O5, Li0.5MnO2), conferring metallic character and high conductivity; minimal structural distortion supports fast kinetics.
• Kinetic enhancement (electroanalytical evidence):
  • In Li2S6 catholyte, Nb2O5/RGO shows much higher cathodic exchange current i0 vs RGO (2.0 mA vs 0.41 mA). During anodic scans, i0: RGO 1.0 mA, Nb2O5/RGO 0.85 mA, MnO2/RGO 1.82 mA (MnO2 82% faster than RGO), validating Nb2O5 accelerates reduction and MnO2 accelerates oxidation of sulfur species.
  • CV Randles–Sevcik analysis: both oxides promote S8→Li2Sn; only Nb2O5 assists Li2Sn→Li2S2/Li2S on discharge; MnO2 especially facilitates oxidation back to S8 on charge.
• Spontaneous chemical mediation (XPS evidence):
  • LiNb2O5 mixed with Li2S6: Nb4+ fraction decreases from 48.8% to 30% (oxidation of LiNb2O5), and Li2S6 is reduced to Li2S2 (161.7 eV) and Li2S (160.0 eV); thiosulfate surface species (~166.8 eV) observed.
  • MnO2 mixed with Li2S6: Mn4+ reduced to Mn3+ (~642 eV) and Mn2+ (~640.2 eV); sulfur oxidized to S8 (~163.3 eV); polythionate species (~167.9 eV) formed.
  • Depth-profile XPS confirms Nb2O5 → LiNb2O5 formation within electrode even when held at 2.4 V, supporting kinetically driven local lithiation during discharge and mediated redox cycling.
• Cell-level performance in thick, lean-electrolyte cathodes (areal S ~7 mg cm−2, E/S=7):
  • At 0.1C, after 50 cycles: S–Nb2O5–MnO2 retains 767.2 mAh g−1 vs S–RGO 329.9 mAh g−1; S–Nb2O5 307.6 mAh g−1; S–MnO2 359.8 mAh g−1; Coulombic efficiency significantly improved with dual mediators.
  • Self-discharge loss after 24 h rest at 2.1 V (5th cycle): S–RGO 13.88%, S–Nb2O5 12.53%, S–MnO2 7.7%, S–Nb2O5–MnO2 5.93%.
  • Rate capability (single-cycle per rate to avoid Li anode degradation): S–Nb2O5–MnO2 maintains high capacities from 0.5 to 3 mA cm−2. S–RGO suffers severe loss at 2 mA cm−2 (available capacity ~206.4 mAh g−1), indicating failure to utilize the second plateau without mediators.
  • Benefit increases at lower E/S (e.g., E/S=5): S–Nb2O5–MnO2 preserves voltage profile and capacity better than S–RGO, yielding higher cell-level energy density under lean electrolyte.
• Mechanistic insight: The dual mediators act as fast electron–Li+ reservoirs enabling coupled fast electrochemical steps with spontaneous chemical reactions between mediators and soluble sulfur radicals, thereby bypassing intrinsically slow direct electrochemical pathways at carbon interfaces.

The study links the electronic structures of sulfur intermediates with their electrochemical activity, identifying solvated radical species as the most readily reduced intermediates during discharge. By aligning mediator band gap centers relative to sulfur radicals, Nb2O5/LiNb2O5 and MnO2/LiyMnO2 were rationally chosen to selectively accelerate discharge and charge, respectively. The mediators provide rapid, localized electron–Li+ buffering and engage in spontaneous redox with soluble radicals, stabilizing transiently reduced/oxidized sulfur species and maintaining reaction continuity even under mass-transport-limited conditions of thick, lean-electrolyte cathodes. Electroanalytical and spectroscopic evidence corroborates the mediated pathways: higher exchange currents reflect accelerated kinetics; CV indicates step-selective enhancement; and XPS confirms spontaneous bidirectional redox between mediators and polysulfides. At the device level, the dual-mediator strategy substantially improves cycling stability, Coulombic efficiency, rate performance, and reduces self-discharge in realistic thick electrodes, directly addressing the core challenges of sluggish kinetics and polysulfide management. The framework of coupling fast electrochemical reactions with spontaneous chemical mediation suggests a generalizable path to overcoming slow electrochemical steps in other conversion/alloying electrodes.

• Constructed an electronic energy diagram of key sulfur species, revealing that solvated sulfur radicals have lower redox energies and narrower band gaps than closed-shell Li2Sn species, making them primary active intermediates during discharge.
• Designed and validated a dual-mediator concept using pseudocapacitive Nb2O5 (electron–ion source for discharge) and MnO2 (electron–ion drain for charge) with band alignment tailored to sulfur radicals.
• Demonstrated spontaneous mediator–polysulfide redox that, coupled with fast electrochemical mediator cycling, circumvents slow direct electrochemical pathways, markedly accelerating sulfur redox kinetics.
• Achieved strong performance gains in thick, lean-electrolyte Li–S cathodes: higher retained capacities (e.g., 767.2 mAh g−1 at 0.1C after 50 cycles), improved Coulombic efficiency, reduced self-discharge, and superior high-rate capability. Future directions include extending the mediator design principle to other electrode chemistries with sluggish kinetics (e.g., silicon, phosphorus) and optimizing mediator composition, loading, and architecture for practical pouch cells under ultralean electrolyte and high sulfur loading.

• Computational simplifications: Molecular dynamics for explicit solvation was not employed due to cost; instead, ensembles with varying numbers of explicit DOL molecules plus continuum solvent were used. Band gaps show functional dependence, though band gap centers are less sensitive.
• Electrochemical testing constraints: High-rate tests were limited to single cycles at each current density to avoid lithium metal anode degradation, so long-term stability under sustained high power was not assessed.
• Electrode dependence: Benefits are pronounced in thick, lean-electrolyte electrodes; in thin electrodes with abundant transport pathways, improvements can be marginal, indicating context-dependent efficacy.
• Thermodynamic vs kinetic operation: Nb2O5 redox lies below sulfur cathode voltages; mediator lithiation is proposed to occur locally under high-rate kinetic conditions—validated by ex situ XPS—but in situ/operando confirmation and quantification under varied rates would strengthen the claim.