The study employs the following chemicals: Anhydrous sodium carbonate (Na₂CO₃, AR, Sinopharm Chemical Reagent Co. Ltd.), tannic acid (TA, C₇₆H₅₂O₄₆, 98%, Shanghai Macklin Biochemical Co. Ltd.), nickel-iron-manganese ternary hydroxide ((NiFeMn)_1/3(OH)₂, battery grade, Zhejiang Huayou Cobalt Co. Ltd.), polyvinylidene fluoride (PVDF, AR, Sinopharm Chemical Reagent Co. Ltd.), N-methyl-2-pyrrolidone (NMP, C₅H₉NO, AR, Shanghai Aladdin Biochemical Technology Co. Ltd.), anhydrous ethanol (AR, Sinopharm Chemical Reagent Co. Ltd.), VC (C₃H₂O₃, 99%, Shanghai Aladdin Biochemical Technology Co. Ltd.), conductive carbon black (C, battery grade, Timcal Graphite & Carbon), aluminum foil (battery grade, Shenzhen Tianchenghe), coin cell case (CR2032, Shenzhen Kejing Star), and glass fiber separator (GF/D, Whatman). All agents were used without further purification. The electrolyte consisted of sodium perchlorate (NaClO₄), a PC/EMC solvent mixture, and the FEC additive in a volume ratio of 50:48:2; it was purchased from Duoduo Chemical Technology Co. Ltd. and denoted as PE.

The O3-type Na(NiFeMn)_1/3O₂@CTA-4 (NFM@CTA-4) was synthesized following our previously reported method^[26]. The pristine O3-type NFM was prepared via high-temperature solid-state reaction. (NiFeMn)_1/3(OH)₂ and Na₂CO₃ were mixed in a molar ratio of 2:1.05. After grinding, the mixture was preheated at 500 °C for 3 h and then calcined at 900 °C for 12 h to obtain NFM. The as-obtained NFM was then ultrasonically dispersed in ethanol, followed by the addition of TA (4 wt%). The mixture was stirred at 80 °C until complete dissolution. The resulting solution was evaporated and dried under vacuum at 60 °C for 12 h. Finally, the product was calcined at 700 °C for 3 h under a nitrogen atmosphere (Kejing, OTF-1200X-II, China) to finish the carbon coating, yielding NFM@CTA-4. The electrodes were prepared by weighing the active material, conductive carbon black, and PVDF at a mass ratio of 8:1:1, using NMP as the solvent. Then, the suspension was cast onto aluminum foil using a doctor blade with a fixed gap of 100 µm, followed by vacuum drying at 80 °C for 12 h. The mass loading of NFM@CTA-4 on cathodes was around 2 mg cm^-2.

Electrolyte preparation was carried out in an argon glovebox (MIKROUNA, Super, China, H₂O < 0.1 ppm, O₂ < 0.1 ppm). The PE electrolyte (10 mL) was mixed with VC additive at volume ratios of 99:1, 98:2, and 96:4, respectively. The mixtures were thoroughly shaken and allowed to stand until complete dissolution, yielding electrolytes denoted as PE1VC, PE2VC, and PE4VC, respectively.

Scanning electron microscopy (SEM, ZEISS Sigma 360, GER) coupled with energy dispersive spectroscopy (EDS, OXFORD ULTIM MAX 100, UK) was used to characterize the microstructure of the samples. X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha⁺, USA) was used to characterize the composition of CEI films. The CR2032 coin cells were assembled with a working electrode, sodium metal foil, glass fiber separator, and 100 μL electrolytes (PE, PE1VC, PE2VC, and PE4VC) in an argon glovebox (H₂O < 0.1 ppm, O₂ < 0.1 ppm). Two half-cells were assembled for each electrolyte, and the cell exhibiting the representative performance within each group was selected for subsequent evaluation. Galvanostatic charge-discharge curves (GCD), rate performance, and cycling stability were tested on a battery testing system (Neware, CT-4008Q, China). High-temperature testing was performed in a programmable temperature-controlled oven (Ruijia, DHG-9030A, China), calibrated to maintain temperature fluctuations within ±1 °C. Cyclic voltammetry (CV) and electrochemical impedance spectra (EIS) were tested by an electrochemical workstation (Chenhua, CHI690F, China). CV was performed over a voltage window of 2~4 V (vs. Na⁺/Na) at scan rates ranging from 0.1 to 1 mV s^-1. EIS was recorded over a frequency range of 100 kHz to 0.01 Hz with an AC amplitude of 0.5 mV, and the obtained spectra were fitted by ZView software using a nonlinear least-squares method.

The Gaussian 16 package was used to optimize the geometric structures of PC, EMC, FEC, and VC molecules. The values of the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) energies were calculated using DFT with the B3LYP/6-311+G (d) functional/basis set. The electrostatic potential (ESP) maps were obtained by further calculating on Gaussian checkpoint files.

LTMOs for SIBs commonly suffer from severe interfacial side reactions. Carbon coating modification and electrolyte engineering are effective strategies to suppress these detrimental reactions. In our previous work, a carbon-coated NFM@CTA-4 cathode was synthesized, delivering 108.6 mAh g^-1 at 1 C, with a capacity retention of 73.5% after 300 cycles^[26]. In this study, we utilize the previously reported high-performance NFM@CTA-4 as the cathode and propose a synergistic electrolyte engineering strategy to further enhance the cycling stability of LTMO-based SIBs.

To predict the binding sites of Na⁺ with electrolyte solvents/additives at the molecular level, the ESP distributions of PC, EMC, FEC, and VC were simulated. Positively charged Na⁺ preferentially bind to the negatively charged sites of solvents/additives. As shown in Figure 1A, all four molecules exhibit the most negative ESP values around the C=O groups, indicating that Na⁺ ions primarily coordinate with solvents/additives via these groups. Subsequently, the binding energies of Na⁺ with the electrolyte components were calculated. The binding energy between a solvent and Na⁺ reflects the strength of their coordination. Higher coordination strength facilitates the dissociation of NaClO₄ salt and the formation of solvated structures; however, it can conversely hinder desolvation at the electrode-electrolyte interface. As shown in Figure 1B, the binding energy between PC and Na⁺ is -139.9 kJ mol^-1, indicating stronger coordination with Na⁺. However, this also results in a higher desolvation energy barrier, which is unfavorable for rate capability. In contrast, the binding energy between EMC and Na⁺ is -110.9 kJ mol^-1, indicating weaker coordination with Na⁺. As a result, the solvation structure loosens, reducing the Na⁺ desolvation energy. Therefore, the combination of PC and EMC can enhance the rate capability by modulating the overall coordination strength. Meanwhile, the binding energies of VC (-115.8 kJ mol^-1) and FEC (-120.6 kJ mol^-1) fall between those of PC and EMC, striking an optimal balance that facilitates both salt dissociation and interfacial desolvation.

To assess the rationale for using FEC and VC as film-forming electrolyte additives, the HOMO and LUMO energy levels of the electrolyte components were calculated, and the results are shown in Figure 1C. These frontier orbitals govern the redox reactivity of electrolyte components, thereby determining the formation of SEI and CEI films^[27]. The SEI and CEI are critical to the capacity, rate capability, and cycling stability of sodium anodes and LTMOs, respectively, and thus dictate the performance of half-cells. FEC, with a low LUMO energy level (-0.037 eV), preferentially undergoes reduction to form a NaF-rich SEI that effectively passivates the sodium anode. Meanwhile, among the four components, VC exhibits both the lowest LUMO energy level (-0.120 eV) and the highest HOMO energy level (-7.211 eV). Therefore, it can be preferentially oxidized and reduced during electrochemical reactions, promoting the formation of both SEI and CEI films that protect the anodes and cathodes from corrosion caused by electrolyte decomposition byproducts.

To investigate interfacial charge transfer, EIS measurements were conducted on Na||Na symmetric cells. From the EIS spectra in Figure 1D and the inset, the electrolyte resistances (R_e) were determined from the high-frequency intercept on the Z' axis. Notably, compared with PE2VC and PE4VC, PE and PE1VC exhibit lower R_e values, indicating lower bulk electrolyte resistance. The Na||Na symmetric cells using PE2VC and PE4VC electrolytes exhibit higher R_e than those using PE and PE1VC electrolytes. The elevated R_e observed for PE2VC and PE4VC is attributed to the electrically insulating nature of VC. To further elucidate the influence of the VC additive on Na⁺ transport, the Na⁺ transference numbers of PE and PE1VC were subsequently calculated via the chronoamperometric curves [Figure 1E], as well as the EIS before and after the chronoamperometric test [Figure 1F and G]. The Na⁺ transference number was calculated using:

(1) $$ \begin{equation} \begin{aligned} t_{N a^{+}}=\frac{I_{S S}\left(\Delta V-I_{0} R_{0}\right)}{I_{0}\left(\Delta V-I_{S S} R_{S S}\right)} \end{aligned} \end{equation} $$

where ΔV is the applied polarization voltage (20 mV), I₀ and I_SS are the initial and steady-state currents, respectively, and R₀ and R_SS are the initial and steady-state resistances of the passivation films on the Na electrode, respectively^[28]. The resistance values were determined from the diameters of the semicircles in Figure 1F and G. The Na⁺ transference numbers were calculated to be 0.63 for PE and 0.90 for PE1VC. Furthermore, a high Na⁺ transference number can effectively suppress concentration polarization, thereby enhancing rate capability. These results indicate that PE1VC strikes an optimal balance between bulk ionic conductivity and the Na⁺ transference number.

To gain further insight into electrolyte properties, electrochemical measurements were conducted on Na||Na symmetric cells with PE and PE1VC electrolytes at a current density of 2 mA cm^-2 [Supplementary Figures 1 and 2]. The voltage-time profiles show that the symmetric cell with the PE electrolyte has a lower overpotential during the initial 30 h. However, after 30 h of cycling, the overpotential of the PE-based cell continues to increase, whereas that of the PE1VC-based cell gradually decreases and eventually stabilizes. This result confirms that the preferential decomposition of the VC additive facilitates the formation of a stable SEI film that effectively protects the Na anode while enabling efficient Na⁺ transport. In contrast, the PE electrolyte undergoes continuous decomposition, resulting in excessive SEI thickening that ultimately impedes Na⁺ transport.

To clarify the influence of the VC additive on the electrochemical Na⁺ storage performance, GCD tests were conducted at various current rates using half-cells assembled with four distinct electrolytes: PE, PE1VC, PE2VC, and PE4VC. As shown in Figure 2A-D, the charge profiles of all four samples at 1 C display a dominant plateau at around 3.3 V, corresponding to the Ni²⁺/Ni³⁺ oxidation. The discharge profile consists primarily of a plateau near 2.5 V, which is associated with a two-phase transition reaction, and a sloping region near 3.0 to 4.0 V attributable to a solid-solution reaction. A sudden voltage surge was observed at the beginning of charging at 5 C. This surge is attributed to the superposition of instantaneous ohmic and activation polarizations induced by the abrupt transition from low-rate to high-rate operation. Specifically, the current step induces an immediate voltage jump governed by ohmic resistance, while the electrode interfaces require additional overpotential to drive the suddenly accelerated electrochemical reaction kinetics at high current^[29].

As shown in Figure 2E, the cell with the PE1VC electrolyte delivers an initial discharge specific capacity of 120.2 mAh g^-1 at 1 C (1 C = 150 mA g^-1), whereas those with PE, PE2VC, and PE4VC electrolytes deliver 115.8, 120.1, and 116.8 mAh g^-1, respectively. Moreover, PE1VC exhibits superior rate capability at high rates, delivering the highest discharge specific capacities of 89.8 mAh g^-1 at 5 C and 68.9 mAh g^-1 at 10 C. This behavior can be attributed to the rapid formation of a stable SEI film with a high Na⁺ transference number, which ensures fast Na⁺ transport kinetics and excellent stability (discussed above). In addition, VC facilitates the formation of a thermally stable and mechanically robust CEI film, which protects the cathode from corrosion (discussed later).

As shown in Figure 2F, the Coulombic efficiencies of all four samples are nearly 100%. After 300 cycles at 1 C, the retained capacities of the PE, PE1VC, PE2VC, and PE4VC cells are 74.9, 84.8, 83.5, and 79.8 mAh g^-1, respectively, indicating that PE1VC provides significantly improved cycling stability. To further assess durability, all four samples were subjected to 600 GCD cycles at 10 C. As shown in Figure 2G, after 600 cycles, PE1VC exhibits the highest reversible capacity (62.7 mAh g^-1) and the best capacity retention (78.4%), outperforming PE (39.4 mAh g^-1, 56.5%), PE2VC (59.7 mAh g^-1, 77.7%), and PE4VC (43.5 mAh g^-1, 71.5%). Notably, all VC-containing electrolytes exhibit superior capacity retention compared with PE.

To evaluate electrochemical performance at elevated temperatures, rate capability and cycling stability were evaluated at 90 °C. As shown in Figure 2H, PE1VC delivers high specific capacities of 104.48 mAh g^-1 at 5 C and 94.4 mAh g^-1 at 10 C. The enhanced capacities at 90 °C, compared with those at 25 °C, result from reduced electrolyte viscosity, improved sodium salt dissociation, and accelerated electrode kinetics. Although the rate capability is enhanced, the elevated temperature simultaneously accelerates side reactions and structural degradation of the electrode material, thereby compromising cycling stability. After 300 cycles at 90 °C, the discharge specific capacity of PE1VC drops to 54.8 mAh g^-1, corresponding to a capacity retention of 47.24% [Figure 2I]. In contrast, PE exhibits much poorer cycling stability under the same conditions, retaining only 14.5 mAh g^-1 (18.68% retention) [Supplementary Figure 3]. Additionally, the button cells with PE and PE1VC electrolytes exhibited no pressure-induced rupture during or after cycling at 90 °C and retained their original structural integrity.

The boiling point of VC (162 °C) lies between those of PC (241.7 °C) and EMC (109 °C). Owing to its relatively high boiling point, VC remains thermally stable at typical operating temperatures (≤ 90 °C), thereby ensuring sustained electrochemical performance^[30]. Furthermore, among the electrolyte components, VC exhibits the lowest LUMO energy level and the highest HOMO energy level, indicating its thermodynamic tendency to undergo preferential reduction at the anode and preferential oxidation at the cathode. Thus, the VC additive contributes to the rapid formation of SEI films on the Na foils and CEI films on the cathodes. Specifically, during charging, it facilitates cleavage of the C=C bond and the subsequent ring-opening reactions of VC, leading to the formation of poly-VC and other oligomeric/polymeric species that constitute the CEI. These species coalesce into a protective film on the cathode surface, thereby effectively shielding the NFM@CTA-4 active material from corrosive byproducts generated during electrolyte decomposition^[31-33]. Besides, CEI and SEI films formed in the presence of VC usually exhibit enhanced thermal stability^[34,35].

The enhanced cycling stability observed at room temperature, at high rates, and at 90 °C is attributed to the rapid formation of stable SEI films on Na foils and CEI films on NFM@CTA-4 cathodes, enabled by the VC additive. On the one hand, VC exhibits the lowest LUMO energy level among the mixed electrolytes, so it can be preferentially reduced, thereby facilitating the formation of a robust, ionically conductive SEI film. On the other hand, VC possesses a higher HOMO energy level than the others (FEC, PC, EMC), facilitating its preferential oxidation at the cathode interface and driving the in situ formation of a CEI film that combines chemical inertness, thermal resilience, and mechanical integrity with fast Na⁺ transport kinetics. The CEI film can protect cathodes from electrolyte-induced corrosion, mitigate side reactions, and suppress cracking and pulverization, as will be discussed later.

To investigate electrode redox reactions, CV measurements were conducted on NFM@CTA-4 cathodes with PE, PE1VC, PE2VC, and PE4VC electrolytes. The CV profiles of the first three cycles are shown in Figure 3A and B (PE and PE1VC) and Supplementary Figures 4 and 5 (PE2VC and PE4VC). The Ni³⁺/Ni²⁺ redox couple is considered the primary contributor to capacity in NFM-type cathode materials. The redox peaks observed at 2.8/3.3 V (Peak C/A) are predominantly assigned to the Ni²⁺/Ni³⁺ couple, whereas the peaks at around 3.55/3.6 V (Peak D/E) are widely ascribed to the Fe³⁺/Fe⁴⁺ redox couple^[31,36]. The reduction peak at 2.5 V (Peak B) is tentatively attributed to the partial reduction of Mn⁴⁺ to Mn³⁺ during the initial charge/discharge cycles^[37].

To further elucidate the electrode reaction kinetics in the presence of the VC additive, CV curves were recorded at scan rates ranging from 0.1~1 mV s^-1. As shown in Figure 3C and D (PE and PE1VC) and Supplementary Figures 6 and 7 (PE2VC and PE4VC), the peak current increases monotonically with increasing scan rate. Notably, the CV profiles of the bare PE electrolyte exhibit pronounced peak distortions, whereas those of the VC-containing electrolytes show minimal scan-rate dependence, which is attributable primarily to the formation of a robust CEI film induced by the VC additive. The reaction kinetics can be quantitatively assessed through the power-law formula given in:

(2) $$ \begin{equation} \begin{aligned} i=a v^{b} \end{aligned} \end{equation} $$

(3) $$ \begin{equation} \begin{aligned} \log i=b \log v+\log a \end{aligned} \end{equation} $$

Where i is the peak current and v is the scan rate. The b-value was extracted from the slope of the log i versus log v plot via least-squares linear regression. The value of b provides insight into the charge-storage mechanism: b ≈ 0.5 indicates diffusion-controlled behavior, while b ≈ 1 suggests adsorption-controlled behavior^[38]. Based on the peak currents in Figure 3C and D, the b-values of the four peaks are calculated to be 0.98, 0.71, 0.59, and 0.96 for PE, and 0.97, 0.85, 0.67, and 0.96 for PE1VC, respectively [Figure 3E and F]. Most b-values are close to 1, indicating that adsorption-controlled charge-transfer processes dominate the electrode reactions. Notably, the PE1VC electrolyte exhibits relatively higher b-values (averaging 0.86) than PE, PE2VC, and PE4VC [Supplementary Figures 8 and 9], suggesting faster Na⁺ transfer under high-rate and high-capacity conditions at high current. This observation is consistent with the superior rate stability of PE1VC. This trend also indicates that PE and PE1VC electrolytes promote faster redox reaction kinetics. The relatively sluggish reaction kinetics observed for PE2VC and PE4VC arise primarily from the electrically insulating nature of VC. An additional contributing factor is their higher charge transfer resistance (R_ct) compared to those of PE and PE1VC, as quantified by EIS and discussed later.

Based on the CV profiles, the adsorption-controlled contribution (k₁v) and the diffusion-controlled contribution (k₂v^1/2) for PE and PE1VC were quantitatively determined using:

(4) $$ \begin{equation} \begin{aligned} i=k_{1} v+k_{2} v^{1/2} \end{aligned} \end{equation} $$

Where i is the current at a given potential V, v is the scan rate, k₁v is the current component from the adsorption contribution, and k₂v^1/2 is the current component from the diffusion contribution^[39]. Consequently, the adsorption-controlled current component (i_surf = k₁v) was determined at each potential point. The total capacitive contribution (represented by the blue-shaded area in Figure 3G and H) was then quantified by integrating i_surf across the voltage window and normalizing it against the total charge of the measured CV curve. At a scan rate of 0.1 mV s^-1, the adsorption-controlled contributions are 83.84% for PE and 86.25% for PE1VC. As shown in Figure 3I, the adsorption-controlled contributions of PE1VC are 86.3%, 89.5%, 91.0%, 93.0%, and 95.4% as the scan rate rises from 0.1 to 1 mV s^-1, whereas those of PE show a fluctuating trend [Supplementary Figure 10]. The consistently high and progressively increasing adsorption-controlled contribution observed for PE1VC indicates that the VC additive promotes Na⁺ transport kinetics and improves electrochemical reversibility. This is a major reason for the good rate capability and cycling stability.

To further investigate the influence of the VC additive on CEI film formation, EIS was conducted on NFM@CTA-4||Na half-cells assembled with the four electrolytes [Figure 4A-D]. The simulated impedance spectra exhibit excellent agreement with the experimental EIS data, validating the proposed equivalent circuit model [Supplementary Figure 11]. The corresponding extracted impedances under different conditions are listed in Table 1. The electrolyte resistance (R_e) values of PE, PE1VC, PE2VC, and PE4VC remained similar across different tested conditions (the pristine state, after 3 cycles at 1 C, after 300 cycles at 1 C, and after 600 cycles at 10 C). The surface film resistance (R_sf) of the cell with the PE electrolyte increased markedly from 760.3 to 4,884 Ω after 3 cycles at 1 C, and then decreased to 948 Ω after 300 cycles. This is likely attributed to the initial formation and subsequent decomposition of an unstable CEI. Although its R_ct decreased continuously, remarkable capacity decays were observed during charge-discharge cycling, reflecting that the evolving structure facilitates charge transfer yet compromises Na⁺ storage capacity mainly due to irreversible structural reconstructions within the electrode^[40]. In contrast, the incorporation of VC fundamentally alters the interfacial evolution. All VC-containing systems display a progressive increase in R_sf during cycling at 1 C up to 300 cycles, yet the growth rate is highly concentration-dependent. Specifically, for the cell with PE1VC, R_sf increased slightly from 2,312 to 2,950 Ω after 3 cycles at 1 C and remained relatively stable over 300 cycles, while R_ct decreased from 69,817 to 18,229 Ω after 3 cycles and remained nearly unchanged thereafter. The initial decrease is attributed to interfacial activation^[41]. After that, the almost unchanged R_ct indicates that an optimal amount of VC promotes the formation of a stable and highly ionically conductive CEI film on the cathode, thereby resulting in significantly smaller impedance changes during cycling with PE1VC than with PE. By comparison, the cells with PE2VC and PE4VC electrolytes also exhibited an initial decrease in R_ct after 3 cycles but showed a progressive increase after 300 cycles. This demonstrates that excessive VC triggers over-polymerization within the CEI, thereby hindering Na⁺ transport kinetics and underscoring the critical trade-off between interfacial passivation and ionic conductivity. Notably, EIS measurements of the four half-cells after 600 cycles at a high rate of 10 C [Figure 4D] reveal substantially lower R_ct and R_sf values compared to those recorded after 300 cycles at 1 C [Figure 4C and Table 1]. This demonstrates that the NFM@CTA-4 cathode material, paired with the PC/EMC/FEC-based electrolyte, exhibits superior interfacial kinetics and stability under high-rate operation, consistent with its excellent rate capability and markedly enhanced cycling stability at 10 C relative to 1 C [Figure 2E-G].

To elucidate the role of VC in modulating the composition and interfacial stability of the CEI on the NFM@CTA-4 cathode, XPS analysis was performed on the cathode sample in PE1VC cycled for 3 cycles at 1 C. As shown in Figure 4E, the F 1s spectrum exhibits two peaks located at 684.8 eV (Na-F), and 688.1 eV (C-F), which indicate NaF is an important component of the CEI film^[32,42]. In the C 1s spectrum [Figure 4F], the main oxygen containing contributions are observed at 285.4 eV (C-O) and 288.8 eV (C=O), consistent with the polyether-type backbone generated via the ring-opening polymerization of VC^[31]. This is further supported by the O 1s spectrum [Supplementary Figure 12], where the intense peaks at 531.7/533.4 eV correspond to C=O/C-O species. No evident shift or emergence of new features is observed in the Mn 2p spectrum upon cycling compared with that of the pristine NFM@CTA-4 powder (as reported in our previous work^[26] and shown in [Supplementary Figure 13], demonstrating minimal transition-metal dissolution and passivating capability of the CEI film. Collectively, these results demonstrate that VC functions as an effective electrolyte additive, facilitating the in situ formation of a chemically stable CEI.

To directly visualize the impact of the VC additive on the mechanical stability of the CEI film, SEM characterization was performed on the cathodes of the half-cells with PE and PE1VC electrolytes after 300 cycles at 10 C. As shown in Figure 5A-C, the NFM@CTA-4 particles in the PE electrolyte exhibit degraded morphological regularity, with visible surface cracks, indicative of severe structural degradation. In contrast, the cathode cycled in the PE1VC electrolyte [Figure 5D-F] demonstrates substantially improved structural integrity, with virtually no surface cracks and a more uniform, densely packed particle distribution. This observation suggests that the VC additive enables the formation of a mechanically robust CEI film that effectively preserves the structural integrity of the cathode particles. Such a robust CEI film not only promotes homogeneous and stable electrode reactions but also suppresses electrolyte decomposition. Additionally, the EDS [Figure 5G-J] and quantitative elemental analyses [Supplementary Tables 1 and 2] indicate that the Na content in the CEI film of the PE-based electrode is higher than that in the PE1VC-based electrode. This difference is attributed to the more serious accumulation of electrolyte decomposition products on the cathode surface in the baseline PE electrolyte during cycling, which impedes Na⁺ transport and promotes interfacial Na⁺ accumulation^[43]. Moreover, distinct morphological evolutions were observed after 300 cycles. In our previous work, the NFM@CTA-4 cathode exhibited a spherical morphology^[26]. In contrast, the NFM@CTA-4 electrode developed a flower-like microstructure after cycling in the PE-based electrolyte, indicating severe structural degradation. However, the PE1VC-based electrode retained its original spherical morphology, reflecting exceptional structural integrity. The morphological differences between PE and PE1VC are mainly due to the VC additive, which facilitates the formation of a chemically stable CEI film, thereby suppressing morphological degradation under attack by electrolyte decomposition products. As a result, the cathode exhibited outstanding cycling stability at high rates and elevated temperatures.

This study demonstrates that the synergistic electrolyte strategy effectively enhances electrochemical performance in Na-metal half-cells. Admittedly, Na-metal half-cells face substantial challenges in practice, including continuous electrolyte decomposition, dendritic sodium deposition, and anodic overpotential drift^[44]. In contrast, full cells based on relatively stable anodes (e.g., hard carbon) offer significantly improved interfacial stability and suppressed parasitic side reactions^[45], offering greater promise for practical implementation. Thus, future investigations should extend this synergistic electrolyte engineering strategy to full-cell configurations, particularly under industrially representative conditions, such as high areal loadings of active materials.