A non-fluorinated, weakly solvating electrolyte for efficient sodium–sulfurized polyacrylonitrile batteries

Abstract

Room-temperature sodium–sulfur (RT/Na–S) batteries are gaining increasing attention due to their high energy density (1274 W h kg⁻¹) and the abundance of sodium and sulfur. However, traditional carbonate- and glyme-based electrolytes exhibit significant drawbacks, including capacity loss from nucleophilic reactions and polysulfide solubility, respectively. In this study, we propose a novel weakly solvating electrolyte (WSE) utilizing tetrahydropyran (THP) as an eco-friendly solvent. The WSE, composed of 2 M sodium bisfluorosulfonyl imide in THP, forms an anion-rich solvation sheath that enhances the stability of both the solid-electrolyte and cathode-electrolyte interfaces. This design promotes uniform sodium deposition and reversible sulfur redox reactions. The WSE achieves a remarkable Na plating/stripping coulombic efficiency of 99.1% over 1000 cycles at 1.0 mA cm⁻². Coupled with sulfurized polyacrylonitrile (SPAN) cathodes, the Na‖SPAN full cells deliver a reversible capacity of 365 mA h g⁻¹ at 1 A g⁻¹ (based on the mass of SPAN), sustaining 90% of their initial capacity over 500 cycles. This WSE represents a promising avenue for developing high-performance and environmentally friendly electrolytes for long-life Na–S batteries.

---

Junxiong Wu

Junxiong Wu received his B.S. from Fuzhou University and M.E. from Tsinghua University in 2013 and 2016, respectively. He earned his PhD in Mechanical Engineering from the Hong Kong University of Science and Technology in 2020 and joined the College of Environmental and Resource Sciences and College of Carbon Neutral Modern Industry at Fujian Normal University in 2021. His research focuses on new materials and advanced characterizations for alkali metal-sulfur batteries.

1. Introduction

Room-temperature sodium–sulfur (RT/Na–S) batteries have attracted significant attention due to the abundance, cost-effectiveness, and high theoretical capacities of sodium (1166 mA h g⁻¹) and sulfur (1675 mA h g⁻¹), which collectively contribute to their impressive energy density of 1274 W h kg⁻¹.^1–9 However, a variety of challenges, including volume variation of the cathode during charge–discharge processes, unstable electrolyte–electrode interfaces, and high self-discharge caused by the polysulfide shuttle effect, have hindered the practical implementation of RT/Na–S batteries. In response to these challenges, various strategies have been proposed, such as electrode design, interface modification, binder optimization, and electrolyte engineering.^10–13 Among these, electrolyte engineering has drawn significant interest, as many of the key issues faced by Na–S batteries stem from poor compatibility between the electrolyte and the electrode.^14–17

Two primary types of electrolytes have been explored in RT/Na–S batteries: carbonate-based and glyme-based electrolytes. Unfortunately, traditional carbonate electrolytes undergo nucleophilic reactions with sodium polysulfides (NaPSs), intermediates formed during sulfur reduction, leading to severe capacity loss.^18–20 Additionally, the high reactivity of Na metal with the carbonate electrolyte results in diminished Na plating/stripping coulombic efficiencies (CEs) and hazardous Na dendrite growth. In contrast, glyme-based electrolytes offer better reduction stability against Na metal anodes.^21–24 However, the high solubility of NaPSs in glyme-based ethers results in pronounced NaPS shuttling, depleting active sulfur species at the cathode and causing significant Na corrosion.^25,26 This ultimately leads to rapid capacity attenuation and shortened battery lifespan. Therefore, it is crucial to develop advanced electrolyte systems to achieve stable cycling of RT/Na–S batteries.

In recent years, several novel electrolytes, including high-concentration electrolytes (HCEs),²⁷ localized high-concentration electrolytes (LHCEs),^28,29 and fluorinated electrolytes,³⁰ have been developed to enhance the interfacial stability between the electrolyte and the electrodes in RT/Na–S batteries. These electrolytes exhibit unique solvation structures distinct from those of conventional dilute electrolytes and have demonstrated good stability for both alkali metal anodes and sulfur cathodes. However, despite their promising performance, these strategies face significant challenges in practical implementation. For instance, the high cost and extensive utilization of Na salts, along with increased solution viscosity and density, compromise the use of HCEs in commercial batteries. LHCEs, formed by diluting HCEs with anti-solvents (typically fluorinated ethers), reduce electrolyte viscosity and improve electrode wettability. Nevertheless, the high cost and density of fluorinated ethers negatively impact overall energy density, while the inclusion of fluorine poses environmental concerns (Fig. S1†). Similarly, replacing conventional solvents with fluorinated alternatives significantly raises both battery costs and environmental impact.^27,31 Therefore, there is considerable interest in identifying suitable solvents that exhibit excellent compatibility with Na metal while facilitating reversible sulfur redox reactions by mitigating NaPS dissolution.

Here, we propose a weakly solvating electrolyte (WSE) for RT/Na–S batteries, using tetrahydropyran (THP) as an eco-friendly solvent (Table S1†). The developed WSE electrolyte, consisting of 2 M sodium bisfluorosulfonyl imide (NaFSI) dissolved in THP (denoted as THP hereafter), forms an anion–rich solvation sheath, which governs the formation of robust and protective solid-electrolyte interphase (SEI) and cathode-electrolyte interphase (CEI) layers on the Na metal and SPAN surfaces, respectively. These tailored interfaces promote uniform Na deposition and dissolution, as well as reversible “quasi-solid-phase” sulfur redox reactions, as verified by both theoretical calculations and experimental characterizations. As a result, the optimized THP electrolyte demonstrates an impressive average CE of approximately 99.1% over 1000 cycles at 1.0 mA cm⁻² and 1.0 mA h cm⁻². When paired with sulfurized polyacrylonitrile (SPAN) cathode, the Na‖SPAN full cell with the THP electrolyte delivers a high reversible capacity of 365 mA h g⁻¹ (912.5 mA h g⁻¹ based on the mass of sulfur) at 1 A g⁻¹, with 90% of its initial capacity retained over 500 cycles, demonstrating superior cyclic performance. This WSE presents a promising pathway for designing cost-effective and environmentally friendly electrolytes for long-life RT/Na–S batteries.

2. Results and discussion

2.1. Electrolyte characterization

The physicochemical properties of the solvent significantly influence the solvation structure of electrolytes, hence dictating the electrochemical performance of RT/Na–S batteries. As shown in Fig. 1a, THP features a cyclic and symmetrical structure that decreases the electron density of the oxygen (O) atom in the C–O–C bond, leading to diminished solvation power and lower polysulfide solubility.³² The electrostatic potential (ESP) maps of DME and THP reveal that oxygen atoms exhibit the most negative charge densities in both solvents, confirming that Na⁺ is more likely to bind to oxygen atoms. Notably, the minimum ESP of THP was calculated to be −153.32 kJ mol⁻¹, which is less negative than that of DME (−174.43 kJ mol⁻¹). Notably, a more negative value indicates a stronger interaction, hence implying a weaker interaction between THP and Na⁺ ions compared to DME (Fig. 1b).^22,32 A recent study has shown that the binding energy between Na⁺ and the solvent molecule can serve as a descriptor to evaluate the solvating power of solvents, thereby differentiating the solvents with strongly and weakly solvating power, respectively, as shown in Fig. S2.†³³Fig. 1c demonstrates that the binding energy of Na⁺–THP is much more negative than that of Na⁺–DME, indicating the comparatively weaker binding strength between Na⁺ and THP, consistent with the ESP result. Notably, even when accounting for solvation and counterion effects, the binding strength between Na⁺ and THP remains weaker than that between Na⁺ and DME (Table S2†). Importantly, the weak solvating power of THP is expected to augment the involvement of more FSI⁻ anions into the primary solvation sheath in the electrolyte, hence favoring the formation of anion-derived electrolyte–electrolyte interfaces.

Fig. 1 Electrolyte characterization. (a) Schematic diagrams of the molecular structures of THP and DME. (b) ESP of DME and THP. (c) Binding energies of DME–Na⁺ and THP–Na⁺. (d) Solvation structure and (e) RDF of DME electrolyte. (f) Solvation structure and (g) RDF of THP electrolyte. (h) Raman spectra and (i) ²³Na and ¹⁹F NMR spectra of DME and THP electrolytes. (j) LSV curves using stainless steel electrodes to evaluate the electrochemical oxidation stability of different electrolytes. (k) Ionic conductivities and Na⁺ transference numbers of different electrolytes. (l) Tafel plots of different electrolytes.

In light of its low solvating ability, the solubility of different sodium salts in THP was further investigated. As shown in Fig. S3,† only NaFSI and sodium bis(trifluoromethanesulfonyl)imide (NaTFSI) can be completely dissolved in THP at a salt concentration of 1 M and room temperature (26 °C). Interestingly, as the temperature decreases to 0 and −20 °C, NaTFSI gradually precipitates, resulting in a cloudy solution, while the solution containing NaFSI remains clear, indicating that NaTFSI has poorer solubility in THP compared to NaFSI. Furthermore, the solubility limit of NaFSI in THP was measured to be approximately 5–6 M, and the 2 M THP solution exhibits the highest ionic conductivity of 3.41 mS cm⁻¹ (Fig. S4 and S5†). Consequently, the salt concentration in all electrolytes of interest was fixed at 2 M for further investigation unless otherwise specified. Notably, although most weakly solvating solvents developed for Li metal batteries can effectively dissolve LiFSI salt, the solubility of NaFSI in these solvents is extremely low (Fig. S6†). Therefore, novel weakly-solvating electrolyte systems for Na metal-based batteries cannot be directly obtained by simply replacing the salt; rather, a more sophisticated selection of components is required, which remains a significant challenge.

The Na⁺ solvation structures in DME and THP-based electrolytes were further analyzed using molecular dynamics (MD) simulations. The snapshot in Fig. 1d reveals that the presence of FSI⁻ anions in the primary solvation sheath in the DME electrolyte is negligible, with almost all Na⁺ ions coordinated by DME molecules. Moreover, the radical distribution function (RDF) and coordination number (CN) analysis (Fig. 1e) show that the coordination number of Na⁺–O_DME is 5.8, while that of Na⁺–O_FSI⁻ is 0.3, suggesting a solvent-dominant solvation structure in the DME electrolyte.³⁴ Additionally, Fig. S7† illustrates the four most dominant solvation structures in the DME electrolyte, with the Na⁺(DME)₃ structure accounting for 84.2%. In contrast, as shown in Fig. 1f and g, a significant number of FSI⁻ anions are involved in the primary Na⁺ solvation sheath in the THP electrolyte, with an increased CN of Na⁺–O_FSI⁻ (2.1) and a low CN of Na⁺–O_THP (2.5). Furthermore, Fig. S8† shows that the three most prevalent solvation structures in the THP electrolyte all contain at least one FSI⁻ anion. The stark difference between the DME and THP electrolytes demonstrates that the weaker coordination capability of THP allows anions to enter the primary solvation sheath, thereby modulating the electrolyte properties.^17,28 These findings suggest that FSI⁻ anions are more likely to be involved in the primary Na⁺ solvation sheath in the THP electrolyte, impacting both the desolvation process and SEI formation, as will be discussed later.

Raman spectroscopy was employed to investigate the coordination of Na⁺–FSI⁻ in different electrolytes (Fig. S9†). The peaks in the range of 700–760 cm⁻¹ correspond to the stretching/vibration of the S–N–S bond in FSI⁻ anions.³⁵ Compared to the DME electrolyte, the characteristic peak of FSI⁻ in the THP electrolyte showed a blue shift to approximately 740 cm⁻¹, indicating the presence of fewer free FSI⁻ anions.³⁶ Notably, even in 1 M THP electrolyte, a Raman shift of 737 cm⁻¹ was observed, suggesting stronger cation–anion (i.e., Na⁺–FSI⁻) coordination when a weakly solvating solvent is used. Fig. 1h further decouples the coordination environment of FSI⁻ into solvent-separated ion pairs (SSIP, 722 cm⁻¹), contact ion pairs (CIPs, 731 cm⁻¹), and aggregates (AGGs, 742 cm⁻¹).^32,34 In the THP electrolyte, FSI⁻-anions predominantly exist in the form of CIPs and AGGs. Nuclear magnetic resonance (NMR) spectroscopy, using coaxial NMR tubes, was performed to elucidate the coordination environment in the DME and THP electrolytes. As shown in Fig. 1i, the stronger Na⁺–FSI⁻ ion pairing increases electron density around Na⁺, resulting in an upfield in the ²³Na NMR chemical shift from DME to THP. Meanwhile, a downfield shift is observed in the ¹⁹F NMR spectrum of FSI⁻ in the THP electrolyte compared to the DME counterpart. This shift indicates decreased electron density in FSI⁻ anions due to enhanced Na⁺–FSI⁻ pairing in the THP electrolyte, consistent with the ²³Na NMR spectra. Overall, the MD simulations, Raman, and NMR results collectively demonstrate that FSI⁻ anions actively interact with Na⁺ and enter the primary solvation sheath in the THP electrolyte, forming an anion-dominated solvation structure characterized by CIPs and AGGs. Notably, the detailed solvation structure of electrolytes can be further investigated using small-angle X-ray scattering, which offers a non-destructive method to study the molecular clusters, anion–cation pairs, aggregates, solvation sheath, and domain sizes of liquids.^37,38

The electrochemical stability of the 2 M NaFSI DME (denoted as DME hereafter) and THP electrolytes was evaluated using linear sweep voltammetry (LSV). As shown in Fig. 1j, the DME electrolyte decomposes first, with an onset potential of 3.8 V vs. Na/Na⁺, while the THP electrolyte exhibits enhanced oxidation resistance, reaching 4.1 V vs. Na/Na⁺, respectively. Next, the ionic conductivities and Na⁺ transference numbers (t_Na⁺) of two electrolytes were measured, as shown in Fig. 1k and S10.† Although the THP electrolyte had a lower conductivity of approximately 3.4 mS cm⁻¹ compared to 12.3 mS cm⁻¹ of the DME electrolyte, a significantly higher t_Na⁺ of 0.62 was achieved in the THP electrolyte. It is important to note that a high t_Na⁺ is advantageous for mitigating concentration polarization and promoting uniform Na deposition. Moreover, the kinetics of Na plating/stripping on the electrode surface was evaluated by measuring the Tafel plots of symmetrical Na‖Na cells. As shown in Fig. 1l, the exchange current density with the THP electrolyte was 20.6 μA cm⁻², which significantly exceeds that with the DME electrolyte (I₀ = 12.6 μA cm⁻²). The higher exchange current density allows faster electrodeposition kinetics, thus benefitting uniform and reversible Na plating/stripping.^39,40

2.2. Na plating/stripping performance

To assess Na plating/stripping CEs in different electrolytes, Na‖carbon-coated aluminum (Al/C) half cells were assembled. As shown in Fig. S11,† Na‖Al/C cells utilizing the THP electrolyte exhibited stable plating/stripping performance with high CEs at 0.5 mA cm⁻²/1 mA h cm⁻², outperforming that of the cells with DME and EC/PC electrolytes. Additionally, the half cell with the THP electrolyte sustained low polarization voltages over 500 cycles (Fig. S12†). At a higher areal current of 1 mA cm⁻², the Na‖Al/C cell using the THP electrolyte exhibited an impressive initial CE of 90.1%, surpassing those using DME and EC/PC (2 M NaFSI EC/PC, v/v = 1/1) electrolytes (Fig. S13†). Notably, a superior and stable average CE of 99.1% over 1000 cycles was achieved in the THP electrolyte at 1 mA cm⁻²/1 mA h cm⁻² (Fig. 2a and b). In comparison, the half cells using EC/PC and DME electrolytes exhibited comparatively lower CEs of 47.04% and 96.43% over 35 and 150 cycles, respectively (Fig. S13a and b†). The exceptional Na plating/stripping behavior in the THP electrolyte can be attributed to its excellent reductive stability and comparatively weaker solvation affinity with Na⁺, which permits anions into the Na⁺ primary solvation sheath and generates robust SEI.

Fig. 2 Na deposition behavior. (a) Sodium plating/stripping CEs in different electrolytes at 1.0 mA cm⁻² with a deposition capacity of 1 mA h cm⁻². (b) Voltage profiles of Na‖Al/C cell using THP electrolyte at 0.5 mA cm⁻² for a plating/stripping capacity of 1 mA h cm⁻² over different cycles. Voltage profiles of Na‖Na symmetric cells at current densities of (c) 0.5 mA cm⁻² and (d) 1 mA cm⁻² in different electrolytes.

To further evaluate the cycling stability of Na plating/stripping, symmetric Na‖Na cells using different electrolytes were assembled and tested. The symmetric cells with the THP electrolyte exhibited outstanding stability over 1100 h, maintaining a stable and small polarization voltage of 42 mV at a current density of 0.5 mA cm⁻² for a capacity of 0.5 mA h cm⁻² (Fig. 2c and S14†). In comparison, the EC/PC electrolyte demonstrated poor compatibility with Na metal, characterized by large polarization voltages and a short lifespan of approximately 200 h. Similarly, the symmetric cell with the DME electrolyte experienced electrical short circuits after only 400 h. When the current density was increased to 1.0 mA cm⁻², Na‖Al/C cell operating in EC/PC and DME electrolytes exhibited reduced cycle lifetimes of less than 150 h (Fig. 2d), whereas the cell employing the THP electrolyte remained stable for over 600 h, underscoring the excellent compatibility between Na metal and the THP electrolyte.

2.3. Characterization of SEI and CEI

Scanning electron microscopy (SEM) was employed to analyze the morphologies of Na deposits in different electrolytes. The deposition was conducted at a current density of 0.5 mA cm⁻² with a deposition capacity of 2 mA h cm⁻². As shown in Fig. 3a, the EC/PC electrolyte led to porous and non-uniform Na deposition with an average thickness of approximately 103 μm. This morphology, characterized by a large surface area, exacerbates side reactions between the electrolyte and the deposited Na, leading to low plating/stripping CEs observed in the EC/PC electrolyte. In contrast, the Na electrodeposits in the DME electrolyte were denser with fewer pores, and the deposition thickness was measured to be approximately 26.7 μm (Fig. 3b), reflecting better compatibility between DME and Na metal. However, the presence of cracks still negatively impacts the reversibility of Na deposition/dissolution. Strikingly, the THP electrolyte facilitated uniform and dense Na deposition with a thickness of only 22.3 μm and a smooth surface (Fig. 3c). Additionally, optical images (Fig. S15†) further confirm the improved Na deposition morphology in the THP electrolyte. Such desirable Na deposits effectively mitigate parasitic reactions, hence contributing to high Na plating/stripping CEs and prolonged cell lifespan.

Fig. 3 Top-view and cross-sectional SEM images of 2 mA h cm⁻² Na deposition on Al/C in different electrolytes: (a) EC/PC, (b) DME, and (c) THP. Characterizations of SEI. (d) XPS spectra of SEI components formed on Na metal after cycling in DME and THP electrolytes. (e) R_a obtained from AFM results. (f) 3D AFM images of surface modulus distribution for Na metal cycled in DME and THP electrolytes. (g) Schematic illustration of robust SEI formed in THP electrolyte to promote uniform Na deposition.

Since the composition of the interphase dramatically impacts Na deposition behaviors and the electrochemical performance, a wide range of characterizations have been conducted to evaluate the interfaces formed in different electrolytes. Fig. 3d presents X-ray photoelectron spectroscopy (XPS) measurements of the SEI films. The peaks observed at 284.8 eV, 286.9 eV, and 289.2 eV in the C 1s spectra correspond to C–C/C–H, C–O, and O–C–O, respectively, which arise from solvent decomposition.³⁴ While the SEI compositions formed in both the DME and THP electrolytes are similar (Table S3†), the DME system exhibits a higher carbon content, suggesting that THP undergoes less decomposition than DME. In the O 1s and F 1 s spectra, peaks at 530.9 eV, 687.9 eV, and 683.8 eV are attributed to S–O, S–F, and Na–F species, respectively, which are derived from FSI⁻ anions.^35,41 Notably, in the DME electrolyte, the intensities of S–O and S–F are higher, while Na–F is lower. In contrast, the S–F does not appear in the THP electrolyte. These results indicate that the SEI formed in the DME electrolyte cannot effectively prevent the decomposition of FSI⁻. Overall, the SEI formed in the THP electrolyte contains a great proportion of inorganic components, such as NaF and Na₂S, which are beneficial for forming a high-quality SEI layer. Additionally, a peak corresponding to Na₂O is evident in the O 1s spectrum of the SEI formed in the THP electrolyte but is absent in the DME electrolyte. The presence of Na₂O-rich SEI may enhance Na⁺-ion transport kinetics due to its superior ionic conductivity compared to NaF.

To further demonstrate that the SEI formed in the THP electrolyte is flatter and more robust, atomic force microscopy (AFM) was employed for characterizing the Na metal anode after cycling 10 times in a Na‖SPAN cell. The surface roughness of the Na metal cycled in DME and THP electrolytes, characterized by the arithmetic squared difference of contours (R_a), is shown in Fig. 3e. The THP-derived SEI exhibited a R_a of 29.2 nm, which is significantly lower than that of the DME-derived SEI (R_a = 72.1 nm), indicating that the SEI formed in the THP electrolyte is smoother than that formed in the DME electrolyte. As illustrated in Fig. 3f, Young's modulus of the SEI derived from the DME electrolyte was as low as approximately 0.21 GPa, whereas the value for the THP electrolyte increased to approximately 1.23 GPa. This significant improvement in Young's modulus of the anion-derived SEI in the THP electrolyte facilitates dense and flat Na deposition, resulting in remarkable CEs. Therefore, when a strongly solvated solvent is employed, the solvation structure is dominated by the solvent, leading to the formation of a vulnerable SEI on Na metal surfaces and low reversibility of Na deposition (see Fig. 3g). In contrast, an anion-rich solvation structure synergistically constructs SEIs from both solvents and anions, resulting in a robust SEI that supports reversible Na plating/stripping.

Since the failure of RT/Na–S batteries is primarily attributed to the dissolution of NaPSs,^30,42 the remarkable inhibition of NaPS shuttling by the CEI layer formed in THP electrolyte was validated by in situ ultraviolet (UV) spectroscopy. A homemade cell setup used to collect the in situ UV spectra is shown in Fig. S16.† As illustrated in Fig. 4a, the characteristic peak of NaPSs located at 320 nm appears and the intensity increases over time in the DME electrolyte.⁴³ Conversely, no characteristic peaks of NaPSs were detected in the THP electrolyte (Fig. 4b). This observation indicates that SPAN experiences a solid–solid conversion in the THP electrolyte without the formation of soluble NaPSs. Moreover, cyclic voltammetry curves also confirm the absence of NaPS dissolution in the THP electrolyte (Fig. S17a and b†). To investigate the stabilizing mechanism of the THP electrolyte on the SPAN electrode, we conducted an exchange experiment by reassembling the SPAN cathode pre-cycled in the THP electrolyte. As shown in Fig. S18,† the reassembled Na‖SPAN cell exhibited improved cyclic performance relative to the cell with the DME electrolyte at 0.2C, but demonstrated inferior performance compared to the cell utilizing the THP electrolyte. This result indicates that the significantly improved cycling stability of the Na‖SPAN cell can be attributed to the synergy between the robust CEI and the low solubility of polysulfides in THP.

Fig. 4 In situ UV-vis spectra during the first discharge of Na‖SPAN cells using (a) DME and (b) THP electrolytes. TEM images of SPAN cathodes cycled in (c) DME and (d, e) THP electrolytes. (f) F 1s XPS spectra of CEI components formed on SPAN. (g) 3D AFM images of surface modulus distribution for cycled SPAN. (h) Schematic illustration of the roles of THP electrolyte in stabilizing SPAN cathode.

Furthermore, the properties of the CEI layer formed in DME and THP electrolytes were systematically investigated. Fig. 4c shows that the CEI in the DME electrolyte is uneven and exhibits a greater thickness, spanning from 13.0 to 20.6 nm. In contrast, the CEI on the SPAN cathode in the THP electrolyte is uniform, with a thickness of approximately 5.6 nm (Fig. 4d). This indicates that the weakened ion–dipole interactions in the THP electrolyte contribute to a more stable interface on the SPAN cathode. In addition, the high-resolution TEM image in Fig. 4e reveals the presence of NaF in the CEI layer. XPS was also employed to characterize the CEI component. The two peaks observed at 683.8 eV and 687.9 eV in the F 1s spectra correspond to NaF and S–F species, respectively (Fig. 4f).^41,44 The higher intensity of the Na–F peak in the THP electrolyte suggests the formation of a NaF-rich CEI on the SPAN cathode. The presence of NaF in the CEI is beneficial as it helps to inhibit the diffusion of NaPSs.⁴⁵ Moreover, the surface roughness and Young's modulus of the CEIs formed in the DME and THP were compared. As shown in Fig. 4g and S19† the CEI layer derived from THP electrolyte has a flatter surface (R_a = 8.2) and a higher Young's modulus (∼1.54 GPa). The robust CEI derived from the proposed THP electrolyte can better accommodate the volume changes of SPAN during charging and discharging processes and the weakly solvating power of THP mitigates the dissolution of NaPSs, thus preserving their electrochemical activity, as illustrated in Fig. 4h.

2.4. Performance of Na‖SPAN cells

To further assess the feasibility of THP electrolyte in practical application, Na‖SPAN full cells were evaluated within a voltage window of 0.6–2.8 V vs. Na/Na⁺, as shown in Fig. 5. Sulfurized polyacrylonitrile (SPAN) was selected as the cathode material due to its facile preparation and reduced NaPS solubility. Based on the high Na plating/stripping CE and the suppression of NaPS dissolution, the Na‖SPAN cell utilizing the THP electrolyte displayed exceptional cycling performance, maintaining a reversible capacity of 354 mA h g⁻¹ after 500 cycles at 0.2C, with approximately 84% capacity retention (Fig. 5a). In contrast, cells using the DME and EC/PC electrolytes sustained reversible capacities of only 173 mA h g⁻¹ and 282 mA h g⁻¹ after 100 cycles, respectively. The rapid capacity degradation observed in the DME electrolyte is associated with the notorious polysulfide shuttle effect, which not only leads to the loss of active species but also corrodes the Na metal anode.^36,44 While the SPAN cathode exhibits acceptable cyclic stability in the EC/PC electrolyte during the first 50 cycles, a dramatic capacity loss is noted after 100 cycles, which can be attributed to the low Na plating/stripping CE in the carbonate electrolyte, thus resulting in the depletion of Na reservoir.⁴⁴Fig. 5b compares the cyclic performance of Na‖SPAN cells using THP, DME, and EC/PC electrolytes at 1C. The cell employing the THP electrolyte preserved 330 mA h g⁻¹ over 500 cycles, demonstrating a capacity retention of 90% and an average CE of approximately 100%. Meanwhile, cells in the DME and EC/PC electrolytes exhibited rapid capacity decay, maintaining only 164 and 220 mA h g⁻¹ after 100 cycles, respectively. Furthermore, Na‖SPAN cells with the THP electrolyte delivered excellent rate capability with reversible capacities of 474, 424, and 334 mA h g⁻¹ achieved at 0.1C, 0.5C, and 2C, surpassing those employing the DME and EC/PC electrolytes (Fig. 5c and S20†). In addition, Na‖SPAN cells with the THP electrolyte also exhibited superior high-rate (2C) and low-temperature (−10 °C) performance (Fig. S21 and S22†), underscoring the practical viability of the THP electrolyte. Fig. 5d and Table S4† compare the electrochemical performance of Na‖SPAN batteries using the THP electrolyte with other tailored electrolytes.^{28–30,45–48} It is worth noting that the proposed WSE electrolyte exhibits the best cyclic stability over a prolonged lifespan of 500 cycles.

Fig. 5 Cyclic performance of Na‖SPAN batteries using different electrolytes at (a) 0.2C and (b) 1C. (c) Rate capability of Na‖SPAN batteries using different electrolytes. (d) Comparison of full cell performance using THP electrolyte with other reported electrolytes.

3. Conclusion

In conclusion, we have developed a WSE using THP as an eco-friendly solvent for RT/Na–S batteries. The WSE, composed of 2 M NaFSI in THP, facilitates the formation of stable and protective SEI and CEI layers on the Na metal anode and SPAN cathode, respectively. These tailored interfaces ensure uniform Na deposition and dissolution, while enabling reversible “quasi-solid-phase” sulfur redox reactions. The optimized THP electrolyte exhibited an impressive average CE of 99.1% over 1000 cycles at 1.0 mA cm⁻² for a plating/stripping capacity of 1.0 mA h cm⁻². When coupled with a SPAN cathode, the Na‖SPAN full cells delivered a high reversible capacity of 365 mA h g⁻¹ at 1 A g⁻¹, with an effective capacity of 912.5 mA h g⁻¹ at 2.5 A g⁻¹ based on sulfur mass, and retained 90% of their initial capacity after 500 cycles, demonstrating excellent long-term stability. This study highlights a viable path for designing weakly solvating solvents that overcome key challenges in RT/Na–S batteries.

Data availability

Date will be made available on request.

Author contributions

J. W. and Y. M. designed and performed the experiments and wrote the manuscript. Y. M., D. L., X. C. and J. H. performed the characterization of electrolytes and co-wrote the manuscript. J. P., Z. G., W. X. and J.L. participated in the molecular dynamics simulations and DFT calculations. Yue Chen participated in AFM characterization. J. W., X. L. and Y. C. supervised the project and edited the manuscript. All authors discussed the results and reviewed the manuscript.

Conflicts of interest

The authors declare no competing interests

Acknowledgements

This project was financially supported by the National Natural Science Foundation of China (No. 22209027 and No. 22179022), the Hundred Talents Plan of Fujian Province, the Top Young Talents of Young Eagle Program of Fujian Province, the Youth Innovation Fund of Fujian Province (No. 2022J05046), the Award Program for Fujian Minjiang Scholar Professorship, and the Talent Fund Program of Fujian Normal University. This work is supported by High-performance Computing Public Platform (Shenzhen Campus) of Sun Yat-Sen University.

References

1. R. Carter, L. Oakes, A. Douglas, N. Muralidharan, A. P. Cohn and C. L. Pint, Nano Lett., 2017, 17, 1863–1869.
2. H. Chen, M. Li, C. Li, X. Li, Y. Wu, X. Chen, J. Wu, X. Li and Y. Chen, Chin. Chem. Lett., 2022, 33, 141–152.
3. Y. Lei, H. Liu, Z. Yang, L. Zhao, W. Lai, M. Chen, H. Liu, S. Dou and Y. Wang, Adv. Funct. Mater., 2023, 33, 2212600.
4. Y.-X. Wang, W.-H. Lai, S.-L. Chou, H.-K. Liu and S.-X. Dou, Adv. Mater., 2020, 32, 2070145.
5. J. Wu, Q. Liang, X. Yu, Q. F. Lü, L. Ma, X. Qin, G. Chen and B. Li, Adv. Funct. Mater., 2021, 31, 2011102.
6. Y. Chen, M. Xu, Y. Huang and A. Manthiram, Chem, 2022, 8, 312–318.
7. L. Ma, Y. Lv, J. Wu, Y. Chen and Z. Jin, Adv. Energy Mater., 2021, 11, 2100770.
8. H. Chen, J. Wu, M. Li, J. Zhao, Z. Li, M. Wang, X. Li, C. Li, X. Chen, X. Li, Y.-W. Mai and Y. Chen, eScience, 2025, 5, 100281.
9. J. Huang, Z. Song, J. Wu, Y. Miao, M. Li, D. Lin, K. Zhu, X. Chen, X. Li and Y. Chen, Energy Mater. Devices, 2024, 2, 9370051.
10. W. Tang, M. K. Aslam and M. Xu, J. Colloid Interface Sci., 2022, 606, 22–37.
11. Z. Yan, L. Zhao, Y. Wang, Z. Zhu and S.-L. Chou, Adv. Funct. Mater., 2022, 32, 2205622.
12. Z. C. Yan, Y. R. Liang, J. Xiao, W. H. Lai, W. L. Wang, Q. B. Xia, Y. X. Wang, Q. F. Gu, H. M. Lu, S. L. Chou, Y. Liu, H. K. Liu and S. X. Dou, Adv. Mater., 2020, 32, 1906700.
13. J. Zhao, Z. Yuan, J. Wu, L. Tong, X. Li, M. Wang, M. Li, X. Li, Z. Li, X. Chen and Y. Chen, ACS Nano, 2025, 19, 8266–8276.
14. M. Kohl, F. Borrmann, H. Althues and S. Kaskel, Adv. Energy Mater., 2016, 6, 1502185.
15. B. Tang, P. W. Jaschin, X. Li, S.-H. Bo and Z. Zhou, Mater. Today, 2020, 41, 200–218.
16. Z. Hou, R. Zhou, Y. D. Yao, Z. W. Min, Z. H. Lu, Y. Zhu, J. M. Tarascon and B. Zhang, Angew. Chem., Int. Ed., 2022, 61, e202214796.
17. J. Wu, M. Li, L. Ma, X. Li, X. Chen, J. Long, Y. Wang, X. Li, J. Liu, Z. Guo and Y. Chen, ACS Nano, 2024, 18, 32984–32994.
18. L. Hu, Y. Lu, X. Li, J. Liang, T. Huang, Y. Zhu and Y. Qian, Small, 2017, 13, 1603533.
19. J. T. Kim, X. Hao, C. Wang and X. Sun, Matter, 2023, 6, 316–343.
20. X. Chen, S. Yan, T. Tan, P. Zhou, J. Hou, X. Feng, H. Dong, P. Wang, D. Wang, B. Wang, M. Ouyang and K. Liu, Energy Storage Mater., 2022, 45, 182–190.
21. R. Xu, J. F. Ding, X. X. Ma, C. Yan, Y. X. Yao and J. Q. Huang, Adv. Mater., 2021, 33, 2105962.
22. J. Zhang, Q. Li, Y. Zeng, Z. Tang, D. Sun, D. Huang, Y. Tang and H. Wang, ACS Energy Lett., 2023, 8, 1752–1761.
23. M. He, X. Li, N. G. Holmes, R. Li, J. Wang, G. Yin, P. Zuo and X. Sun, ACS Appl. Mater. Interfaces, 2021, 13, 38296–38304.
24. Y. Liao, M. Zhou, L. Yuan, K. Huang, D. Wang, Y. Han, J. Meng, Y. Zhang, Z. Li and Y. Huang, Adv. Energy Mater., 2023, 13, 2301477.
25. J. Zhang, Q. Li, Y. Zeng, Z. Tang, D. Sun, D. Huang, Z. Peng, Y. Tang and H. Wang, Energy Storage Mater., 2022, 51, 660–670.
26. C. Liu, Z. Yang and J. Sun, Energy Storage Mater., 2022, 53, 22–31.
27. J. Lee, Y. Lee, J. Lee, S. M. Lee, J.-H. Choi, H. Kim, M.-S. Kwon, K. Kang, K. T. Lee and N.-S. Choi, ACS Appl. Mater. Interfaces, 2017, 9, 3723–3732.
28. D. Guo, J. Wang, T. Lai, G. Henkelman and A. Manthiram, Adv. Mater., 2023, 35, 2300841.
29. M. H. Pai, T. Lai and A. Manthiram, Adv. Funct. Mater., 2024, 20, 2407450.
30. R. Zou, W. Liu and F. Ran, InfoMat, 2022, 4, e12319.
31. D.-J. Yoo, S. Yang, K. J. Kim and J. W. Choi, Angew. Chem., Int. Ed., 2020, 59, 14869–14876.
32. W. Yao, M.-H. Pai and A. Manthiram, Angew. Chem., Int. Ed., 2025, e202424547.
33. Z. Wang, T. Zheng, S. Wang, X.-G. Zhang, Y. Gu, S. Tang and Y. Fu, J. Am. Chem. Soc., 2025, 147, 5962–5970.
34. W. Yao, M.-H. Pai and A. Manthiram, J. Am. Chem. Soc., 2025, 147, 12061–12074.
35. M. Li, H. Chen, Y. Wang, X. Chen, J. Wu, J. Su, M. Wang, X. Li, C. Li, L. Ma, X. Li and Y. Chen, J. Mater. Chem. A, 2023, 11, 11721–11729.
36. T. Ma, Y. Ni, D. Li, Z. Zha, S. Jin, W. Zhang, L. Jia, Q. Sun, W. Xie, Z. Tao and J. Chen, Angew. Chem., Int. Ed., 2023, 62, e202310761.
37. K. Qian, R. E. Winans and T. Li, Adv. Energy Mater., 2021, 11, 2002821.
38. X. Liu, A. Mariani, T. Diemant, M. E. Di Pietro, X. Dong, P.-H. Su, A. Mele and S. Passerini, ACS Energy Lett., 2024, 9, 3049–3057.
39. A. Schön, E. Kaminska, F. Schelter, E. Ponkkonen, E. Korytiaková, S. Schiffers and T. Carell, Angew. Chem., Int. Ed., 2021, 60, 3850.
40. T. Liebmann, C. Heubner, M. Schneider and A. Michaelis, Mater. Today Energy, 2021, 22, 100845.
41. X. Liu, T. Diemant, A. Mariani, X. Dong, M. E. Di Pietro, A. Mele and S. Passerini, Adv. Mater., 2022, 34, 2207155.
42. Q. Jin, X. Qi, F. Yang, R. Jiang, Y. Xie, L. Qie and Y. Huang, Energy Storage Mater., 2021, 38, 255–261.
43. W. Zhang, B. Song, M. Wang, T. Miao, X.-L. Huang, E. Zhang, X. Zhan, Y. Yang, H. Zhang and K. Lu, Energy Environ. Sci., 2024, 17, 5273–5282.
44. S. Weng, Y. Liu, S. Lu, J. Xu, J. Xue, H. Tu, Z. Wang, L. Liu, Y. Gao, G. Sun, H. Li and X. Wu, Angew. Chem., Int. Ed., 2025, 64, e202421602.
45. J. Wu, J. Liu, Z. Lu, K. Lin, Y.-Q. Lyu, B. Li, F. Ciucci and J.-K. Kim, Energy Storage Mater., 2019, 23, 8–16.
46. L. Wang, N. Ren, W. Jiang, H. Yang, S. Ye, Y. Jiang, G. Ali, L. Song, X. Wu, X. Rui, Y. Yao and Y. Yu, Angew. Chem., Int. Ed., 2024, 63, e202320060.
47. Y. Ren, T. Lai and A. Manthiram, ACS Energy Lett., 2023, 8, 2746–2752.
48. J. He, A. Bhargav, W. Shin and A. Manthiram, J. Am. Chem. Soc., 2021, 143, 20241–20248.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ta00999e

‡ These authors contributed equally to this work: Yuhui Miao, Danjin Lin, Jiapeng Liu, Xiaochuan Chen.

---