Li–S batteries have emerged as promising energy-storage technologies for next-generation electric vehicles and large-scale energy-storage systems owing to their high theoretical energy density (2600 Wh kg^–1), elemental sulfur abundance, and affordability. Compared with conventional Li-ion batteries (LIBs), Li–S batteries theoretically provide more than twice the energy density and have distinct advantages in terms of cost efficiency and environmental sustainability [1–3].

Sulfur has excellent economic and environmental advantages. As sulfur is abundant, inexpensive, and non-toxic, it is a promising alternative to the heavy metal-based cathode materials commonly used in current LIBs [4,5]. However, several technical challenges inhibit the practical application of Li–S batteries. The primary issues include dissolution of intermediate lithium polysulfides (Li₂S₈, Li₂S₆, Li₂S₄, etc.) into the electrolyte during cycling, resulting in a shuttle effect and decreased coulombic efficiency; structural instability caused by volumetric expansion during repeated charge–discharge cycles; and inherently low electrical conductivity, which restricts charge transfer. Extensive research has been devoted to overcoming these issues to accelerate Li–S battery commercialization [6,7]. Among these challenges, the shuttle effect caused by lithium polysulfide dissolution and subsequent diffusion toward the lithium metal anode is particularly detrimental because it leads to the loss of active sulfur species from the cathode, lowers coulombic efficiency, and shortens battery lifespan [8]. A recent comprehensive review [9] reaffirmed that the shuttle effect remains the major bottleneck in the practical implementation of Li-S batteries, emphasizing the need for multifaceted design strategies such as constructing in-anode capture structures, optimizing compositions, catalyst design [10–12], and forming interfacial protective layers to curb it.

Compared with sulfur cathode active materials, lithium metal anodes, separators, and electrolytes, polymeric binders have been overlooked when it comes to their ability to improve the performance of Li–S batteries. For example, polyvinylidene fluoride (PVDF), a binder widely used for electrode fabrication, exhibits several major limitations, including swelling and partial dissolution in organic electrolytes, poor mechanical flexibility, and weak interactions with sulfur-based active materials. These drawbacks result in structural degradation of the electrode and loss of the active material, consequently lowering Li–S battery performance. Therefore, the development of optimized polymeric binder systems is essential to enhance the performance of Li–S batteries [13].

To overcome the limitations of PVDF, recent studies have actively explored the design of novel binders incorporating functional groups. In particular, self-healable polyelectrolyte network binders have been reported to autonomously restore their structures upon mechanical damage and effectively immobilize polysulfides through dynamic hydrogen bonding and ionic interactions [14]. Moreover, bio-derived polymer binders have been shown to enhance interfacial stability by adsorbing polysulfides via a dual anchoring mechanism involving amino (–NH₂) and carboxyl (–COOH) groups [15]. Additionally, functional binders combined with deep eutectic solvents [16] and oxygen-rich polymer networks capable of regulating Li₂S nucleation and growth uniformity [17] have also been reported, demonstrating that binders are evolving beyond their conventional adhesive roles into active mediators that regulate ion transport and reaction kinetics.

In line with this trend, composite binder systems composed of PEO and PVP have also attracted considerable attention for their ability to enhance Li-S battery performance [18]. Specifically, PEO promotes lithium-ion mobility, but can induce excessive lithium polysulfide dissolution owing to its high affinity for the electrolyte. Conversely, PVP suppresses the shuttle effect through chemical interactions with lithium polysulfides [19]. Previous studies have confirmed that PVP directly interacts with lithium polysulfides, forming stable complexes that reduce active material loss and significantly improve capacity retention [18,20,21]. However, excessive PVP content can lead to the formation of insoluble complexes, negatively affecting electrode reactivity. Therefore, careful optimization of the PEO/PVP ratio is essential for not only suppressing the shuttle effect but also precisely controlling the chemical interactions with lithium polysulfides.

However, the binder systems based on PEO and PVP (4/1 weight ratio) exhibited limited long-term cycling stability [22]. To address this issue, this study introduced carboxymethylcellulose (CMC) as an additional binding component. The high elastic modulus of CMC contributes to the enhanced dimensional stability of the sulfur cathodes. Additionally, the negatively charged carboxyl groups in CMC can mitigate the excessive affinity of PEO for electrolytes by engaging in moderate chemical interactions with the lithium polysulfides.

The purpose of this research was to identify the optimal polymeric binder ratio of PEO to PVP and CMC to precisely control lithium polysulfide interactions, effectively suppress shuttle effects, and enhance capacity retention and long-term cycling stability in high-capacity Li–S batteries.

The polymeric binder ratio was optimized before fabricating the sulfur electrode slurries. To determine the optimal CMC content in the binder formulation, the dispersion stabilities of the binder solutions were systematically evaluated by varying the CMC concentration while maintaining a fixed PEO/PVP weight ratio of 5/5. After optimizing the CMC content, the weight ratio of PEO to PVP was adjusted to optimize the binder formulation.

An abnormally high instability index of 0.846 was observed at a CMC concentration of 0.2 wt.%, whereas the index sharply decreased to 0.226 when the CMC concentration was increased to 0.4 wt.% (Fig. 1a,b). During centrifugation for 100,000 s, the instability indices of the polymeric solutions at both 0.2 wt% and 0.4 wt.% CMC remained comparable. This indicates that the initially high instability index of 0.2 wt.% CMC was not due to material sedimentation but microphase separation induced by the blending of PEO and PVP [18,23]. The reduced instability index observed for 0.4 wt.% CMC indicates that CMC effectively suppresses microphase separation between PEO and PVP [24]. At higher concentrations (0.6 and 0.8 wt.%), a slight increase in the instability index was noted, which can be attributed to the higher viscosity of the polymer solution and the resulting local aggregation phenomena.

These results indicate that a PEO/PVP polymer solution containing 0.4 wt.% CMC optimally balances the sulfur electrode slurry’s physical stability, viscosity, and flow characteristics. This weight ratio is expected to be advantageous in terms of compatibility, coating uniformity, and structural integrity after drying during electrode fabrication. Consequently, the 0.4 wt.% CMC formulation was identified as the optimal condition for simultaneously achieving dispersion stability and slurry processability, and it was employed as the reference ratio in subsequent binder formulation evaluations and electrochemical performance tests in this study.

For the sulfur electrode slurry containing 0.4 wt.% CMC, a uniform coating was observed on the aluminum current collector surface (Fig. 2a). By contrast, the CMC-free slurry exhibited poor coating uniformity (Fig. 2b). Peel tests were conducted to evaluate the structural integrity of the fabricated sulfur electrodes. The sulfur electrodes containing CMC exhibited minimal material transfer to the adhesive tape (Fig. 2c, d), indicating strong adhesion between the electrode components and current collector.

By contrast, the CMC-free electrodes almost completely delaminated the active material, which was fully transferred to the tape during the test (Fig. 2e, f). These results demonstrate that the CMC-containing polymer binder improved the dispersion stability of the sulfur electrode slurry and significantly enhanced interfacial adhesion within the sulfur electrode. This could be attributed to the microphase separation between PEO and PVP, as previously discussed, which hindered the homogeneous distribution of the binder within the electrode, consequently weakening the binding of the electrode constituents.

Fig. 3 shows the voltage profiles of the first cycle at a rate of C/10 for the sulfur electrodes with varying binder ratios. Among the tested formulations, the PEO/PVP/CMC binder ratio of 5/5/0.4 yielded the highest initial discharge capacity of 1439.9 mAh g^–1, corresponding to 85.9% of the theoretical sulfur capacity (1673 mAh g^–1) [25]. By comparison, PEO/PVP/CMC binder ratios of 2/8/0.4 and 8/2/0.4 exhibited significantly lower initial capacities of 971.1 mAh g^–1 (57.9%) and 700.4 mAh g^–1 (41.8%), respectively. These findings underscore the significant impact of binder ratio on sulfur utilization and the uniformity of the electrochemical reactions within the sulfur cathode.

Li–S batteries typically exhibit two distinct discharge plateaus, at approximately 2.3 V and 2.1 V [26]. The higher-voltage plateau (~2.3 V) corresponds to the transformation of elemental sulfur (S₈) into soluble highorder polysulfides (Li₂S_n, n = 4–8), involving the reduction of S⁰ to S^–0.5. This step accounts for ~25% of the theoretical discharge capacity [25,26]. The lower-voltage plateau (~2.1 V) is associated with the further reduction of lithium polysulfides to insoluble lower-order products, such as Li₂S₂ and Li₂S, corresponding to the conversion from S^–0.5 to S^2–. This stage accounted for 75% of the theoretical discharge capacity. Realizing the theoretical capacity of Li–S batteries critically depends on mitigating the shuttle effect caused by the crossover of soluble polysulfides.

Although all three electrodes exhibited two well-defined voltage plateaus during discharge, the PEO/PVP/ CMC formulation with a ratio of 5/5/0.4 showed the smallest overpotential. Interestingly, although the absolute discharge capacities corresponding to each plateau varied with the binder ratio, the relative contributions of each plateau to the total discharge capacity remained nearly identical. The high-voltage plateau (~2.3 V) regions of the samples with PEO/PVP/CMC binder ratios of 5/5/0.4, 2/8/0.4, and 8/2/0.4 contributed 35.6% (510.0 mAh g^–1, total discharge capacity is 1439.9 mAh g^–1), 35.95% (349.08 mAh g^–1, total discharge capacity is 971.1mAh g^–1), and 35.09% (241.8 mAh g^–1, total discharge capacity is 700.4 mAh g^–1), respectively; similarly, the low-voltage plateaus (~2.1 V) account for 64.6% (929.8 mAh g^–1, total discharge capacity is 1439.9 mAh g^–1), 64.05% (622.0 mAh g^–1, total discharge capacity is 971.1 mAh g^–1), and 64.91% (458.6 mAh g^–1, total discharge capacity is 700.4 mAh g^–1), of the total voltage, respectively.

These results indicate that, although the amount of active material involved in the redox reactions varies with the binder ratio, the underlying electrochemical reaction mechanism remains consistent across different sulfur electrodes.

Cyclic voltammetry (CV) measurements were conducted on sulfur electrodes prepared with varying binder ratios to gain deeper insight into the electrochemical behavior occurring within the electrodes (Fig. 4a–d). In contrast to the similar voltage profiles observed during the pre-cycling tests, the CV results revealed significant differences depending on the polymer binder formulation.

For the electrode with a PEO/PVP/CMC binder ratio of 5/5/0.4, distinct reduction peaks were observed at approximately 2.3 V and 2.1 V during discharge, corresponding well with the plateau regions identified in the pre-cycling data (Fig. 3). During the charging process, clear oxidation peaks appeared near 2.3 V and 2.4 V, aligning with previously observed plateau voltage. By contrast, the electrodes with binder ratios of 2/ 8/0.4 and 8/2/0.4 exhibited broad, poorly defined current peaks over a wide voltage range.

The distinct voltage plateaus observed during the pre-cycling measurements indicate well-defined redox transitions. By contrast, the broad current peaks observed in the CV profiles can be primarily attributed to fundamental differences in the measurement techniques. While galvanostatic cycling maintains the voltage at specific levels during the redox processes, CV involves a continuous sweep of the applied potential, preventing the system from equilibrating at specific reaction voltages. Consequently, when the redox kinetics are relatively slow, the corresponding current response is distributed over a broader voltage range, resulting in broadened peaks. Furthermore, in sulfur electrodes, multistep conversion reactions involving various polysulfide intermediates may overlap in potential, further contributing to the smearing of the current response.

Considering these observations, a PEO/PVP/CMC binder ratio of 5/5/0.4 can be inferred to enhance the electrochemical reaction kinetics in the sulfur electrode. Moreover, more sulfur active materials appear to participate in redox reactions than in electrodes incorporating other polymer binder formulations.

Rate capability evaluations of the sulfur electrodes revealed that the cell featuring the PEO/PVP/CMC binder in a 5/5/0.4 ratio performed the best (Fig. 5a). Specifically, the electrode with a 5/5/0.4 binder ratio retained 55.5% (811.376 mAh g^–1) of its initial discharge capacity at the 21^st cycle at a 2C rate, whereas the binders incorporated in 8/2/0.4 and 2/8/0.4 ratios retained only 35.3% (508.663 mAh g^–1) and 1.86% (604.409 mAh g^–1), respectively. This result is consistent with the superior electrochemical reactivity observed in the CV measurements (Fig. 4).

These findings indicate that a balanced ratio of PEO to PVP (i.e., 5/5/0.4) provides a more favorable electrode environment for high-rate operation. The improved ionic and electronic transport and better mechanical integrity provided by the optimized polymer network likely contributed to the enhanced utilization of the sulfur active material and the stable electrochemical performance. By contrast, unbalanced binder ratios (i.e., either PVP- or PEO-rich) may suffer from poor ion transport or insufficient electrode cohesion, leading to rapid capacity fading at high current densities.

Interestingly, although the PEO/PVP/CMC binder with a 2/8/0.4 ratio exhibited relatively high discharge capacities at low C-rates (3C/10 and 2C/5), its performance declined significantly at high rates. At 1C and above, the binder formulated with a 2/8/0.4 ratio had noticeably lower capacities than that with a 8/2/0.4 ratio. This trend became particularly pronounced at 2C, where the discharge capacity of the PEO/PVP/CMC binder (2/8/0.4) sharply dropped to below 50 mAh g^–1.

This sharp capacity fading of the PEO/PVP/CMC binder (2/8/0.4) at high rates may be attributed to its high PVP content. PVP is known to enhance the dispersion of active materials and improve electrode wettability owing to its surfactant behavior [18,19]. Furthermore, PVP interacts strongly with lithium polysulfides, resulting in their retention at the cathode [18]. However, its weaker mechanical binding capability compared with that of PEO may lead to insufficient electrode cohesion during fast cycling. At low C-rates (e.g., 3C/10, 2C/5), the binding capability of the polymeric binders was less pronounced, allowing the electrode to retain relatively high capacities. By contrast, at high rates ( 1C), the electrodes' lack of structural integrity becomes critical, resulting in increased polarization and a drastic drop in discharge capacity, as observed at 2C. By contrast, PEO-rich binders, such as that in a 8/2/0.4 ratio, provide improved mechanical robustness and reaction kinetics, leading to relatively stable behavior at high rates.

The cycling performances of sulfur electrodes with varying polymeric binder ratios were evaluated. Consistent with the rate capability results, the PEO/PVP/CMC binder with a 5/5/0.4 ratio exhibited the best cycling stability among the tested ratios (Fig. 5b). After 300 cycles, the 5/5/0.4 binder retained 65.03% (628.1 mAh g^–1) of its initial capacity, whereas the 2/8/0.4 and 8/2/0.4 binders maintained around 48.73% (263.1 mAh g^–1) and 55.96% (282.9 mAh g^–1), respectively. Notably, the 5/5/0.4 binder initially delivered a higher discharge capacity and experienced a rapid capacity drop within the first 25 cycles. However, it stabilized and maintained a relatively consistent capacity over the remaining cycles. By contrast, the electrodes with the 2/8/0.4 and 8/2/0.4 binders demonstrated more gradual and continuous capacity fading over the entire test period. This initial decline was commonly observed across all binder ratios and can be attributed to the electrode-stabilization process, which typically occurs during the early cycling stages. This process likely involves the formation and reorganization of solid-phase reaction products and the gradual establishment of stable electrode/electrolyte interfaces, which are characteristic of sulfur cathodes.

In the PEO/PVP/CMC binder with an 8/2/0.4 ratio, a higher proportion of PEO enhanced Li-ion transport and accelerated the reaction kinetics. However, this also promotes the excessive dissolution of lithium polysulfides into the electrolyte, intensifying the shuttle effect and ultimately reducing long-term cycling stability. By contrast, the PEO/PVP/CMC binder with a 2/8/0.4 ratio, which contains a higher content of PVP, effectively suppressed the shuttle effect owing to the strong interaction between PVP and lithium polysulfides. Nevertheless, the increased PVP content also resulted in higher internal resistance within the electrode, slowing the electrochemical reactions. This kinetic limitation is believed to contribute to the degradation in cycling performance over prolonged operation [18]. The binder with balanced PEO/PVP ratio, that is, a 5/5/0.4, can mitigate the shuttling problem of the sulfur electrodes while maintaining proper electrochemical kinetics.

Although the electrodes with 2/8/0.4 and 8/2/0.4 binder ratios exhibited coulombic efficiencies close to 100% over 300 cycles, the 5/5/0.4 binder showed a gradual decline in coulombic efficiency after 50 cycles, reaching ~83.49% at the end of testing. As this ratio demonstrated the best rate capability and capacity retention, this decrease was unexpected.

Coulombic efficiency, defined as the ratio of the discharge capacity to the charge capacity in each cycle, reflects the reversibility of the electrochemical reactions. A decrease in the coulombic efficiency typically indicates that some lithium ions inserted during charging are not fully recovered during discharge. In conventional Li–S batteries, such a decrease is often attributed to polysulfide shuttling and parasitic side reactions with the electrolyte, which lead to the irreversible loss of the active material and electrolyte components. A sustained low coulombic efficiency is usually accompanied by a gradual decline in the discharge capacity as the cell becomes depleted of usable sulfur species and lithium ions.

However, unusual behavior was observed for the PEO/PVP/CMC binder with a 5/5/0.4 ratio. Despite the gradual decrease in coulombic efficiency after 50 cycles, the discharge capacity remained relatively stable over prolonged cycling (Fig. 5b). This indicates that the mechanism responsible for the reduced coulombic efficiency may differ from the conventional degradation pathways typically observed in Li–S batteries.

These results indicate that the coulombic efficiency does not necessarily provide a direct measure of overall electrochemical stability. Typically, a decrease in coulombic efficiency is associated with irreversible reactions that consume lithium ions, electrolytes, or active materials, ultimately leading to performance degradation. However, such a correlation was not observed in this study, suggesting an anomalous electrochemical behavior. Similar phenomena have been reported in several previous studies [27,28], although their underlying mechanisms remain unclear. In this work, we interpret this behavior as originating from the presence of delayed reactivation species.

In Li–S batteries, lithium ions migrate from the lithium metal anode to the sulfur cathode during discharge, where they react with sulfur to form lithium polysulfides. These lithium polysulfides decomposed into lithium ions and sulfur during the charging process. This reversible process underpins the fundamental redox mechanism of Li–S cells.

Despite prolonged cycling, the discharge capacity of the electrode employing the PEO/PVP/CMC binder with a 5/5/0.4 ratio remains relatively stable, indicating that the amount of active material involved in the discharge reaction did not significantly diminish. However, the concurrent decrease in the coulombic efficiency indicates that the amount of lithium ions recovered during charging exceeded the amount consumed during discharging, producing an imbalance. Therefore, lithium ions participate in charging processes that do not contribute to reversible discharge, which may be associated with transient trapping and delayed reactivation. Identifying the presence and role of such “unaccounted for” lithium ions—electrochemically active during charging but not during discharging—could be key to understanding the mechanism behind the anomalous decrease in coulombic efficiency despite stable discharge capacity.

This phenomenon can be explained by the unique synergistic effects of the PEO/PVP/CMC binder system. PVP has a strong affinity for lithium polysulfides, which helps immobilize them within the cathode region. The CMC contributes to the mechanical stability of the electrode by mitigating structural degradation, whereas the PEO facilitates ionic transport through its segmental motion. Together, these functionalities may prevent the loss of polysulfides to the lithium metal anode and ensure their retention in the cathode composite.

Over prolonged cycling, changes in electrode morphology and internal spatial rearrangement owing to volumetric expansion/contraction may temporarily render some polysulfides electrochemically inactive (Fig. 6). However, these trapped species, tethered by a polymeric binder, may gradually reenter the electrochemical reaction pathway during discharge owing to the significant volumetric expansion of sulfur (~80%) upon lithiation [2,25,29]. If this occurs, more lithium ions may be generated during charging (owing to the reactivation of previously inaccessible Li polysulfides). At the same time, fewer lithium ions participated in the subsequent discharge, leading to an apparent decrease in coulombic efficiency despite a steady capacity output.

During the cycling of Li–S batteries, soluble lithium polysulfides generated at the sulfur cathode diffuse toward the lithium metal anode, driven by a concentration gradient. Upon reaching the lithium metal surface, these polysulfides undergo spontaneous chemical reduction, forming insoluble sulfur-containing species, such as Li₂S, within the solid electrolyte interphase (SEI) layers [3,25,30]. This process leads to increased SEI inhomogeneity and promotes the formation of lithium dendrites, thereby compromising interfacial stability. Once the SEI is fractured by volume expansion during repeated cycling, the freshly exposed lithium characterized by its high reactivity, readily reacts with soluble lithium polysulfides, further accelerating the degradation processes. This situation is exacerbated by dendritic lithium deposits, which have a high specific surface area and thus facilitate extensive side reactions. The interaction between lithium polysulfides and dendritic lithium disrupts the structural integrity of the lithium metal anode and leads to electrochemically inactive “dead” lithium, particularly under high-rate conditions [31,32]. Consequently, the presence of soluble lithium polysulfides in the electrolyte significantly shortens the cycling lifespan of the lithium metal anode.

FE-SEM and energy-dispersive X-ray spectroscopy (EDS) analyses were conducted on the cells after 100 cycles to evaluate the effect of the sulfur cathode binder ratio on the interfacial morphology and chemical characteristics of the lithium metal anode.

The surface and cross-sectional FE-SEM images of the lithium metal anodes retrieved from the Li–S cells incorporating different polymeric binder formulations did not show significant morphological differences (Fig. 7a–f). In all cases, the lithium surfaces were covered with rounded lithium dendrites, likely attributed to the use of the functional electrolyte additive LiNO₃, which is known to promote the formation of more uniform and less mossy lithium deposits [33,34].

Nevertheless, considering that the discharge capacity of the Li–S cells using the PEO/PVP/CMC binder with a 5:5:0.4 ratio was significantly higher than those with other ratios, the comparable lithium morphology across all samples implies that the lithium metal anode in the 5/5/0.4 system remained in a relatively favorable condition throughout cycling. This indicates that the enhanced electrochemical performance observed in this system was not achieved at the expense of lithium metal degradation.

Surface and cross-sectional EDS analyses were conducted to investigate the interfacial characteristics of the lithium metal anodes (Fig. 7g–o). Notably, the surface of the lithium metal retrieved from the cell fabricated with the PEO/PVP/CMC binder with a 5/5/0.4 ratio exhibited a lower concentration of fluorine-containing species. These fluorinated compounds are the major components of the SEI layer and typically originate from the decomposition of electrolyte salts, indicating that the lithium surface is likely covered with a more uniform and compact SEI layer. Such an improvement in interfacial stability may be attributed to the reduced contact with soluble lithium polysulfides, resulting from either their lower generation or improved confinement at the cathode side.

This study systematically investigated the effects of the polymeric binder ratio, specifically a ternary system of PEO, PVP, and CMC, on the electrochemical performance and interfacial stability of Li–S batteries. Dispersion stability analysis and peel tests revealed that CMC addition significantly improved slurry processability and electrode adhesion, particularly at an optimized concentration of 0.4 wt.%. Electrochemical characterizations demonstrated that the PEO/PVP/CMC binder composition of 5/5/0.4 delivers the highest initial discharge capacity (1439.9 mAh g^–1) and outstanding rate and cycling performance, attributed to the synergistic balance between ionic transport, structural integrity, and polysulfide confinement.

Notably, despite a gradual decline in the coulombic efficiency after 50 cycles, the binder with the 5/5/0.4 ratio maintained a stable discharge capacity over prolonged cycling. This phenomenon indicates delayed reactivation of polysulfides previously immobilized by the polymer binder network, providing new insights into the electrochemical reversibility of Li–S batteries. Furthermore, FE-SEM and EDS analyses of the lithium metal anodes after cycling revealed a lower concentration of F-containing species and a more compact SEI in the cells incorporating the 5/5/0.4 binder, indicating reduced parasitic reactions and mitigated dendrite formation.

Overall, this study highlights the critical role of binder engineering in regulating polysulfide behavior, stabilizing electrode interfaces, and extending the operational lifespan of Li–S batteries. The optimized PEO/ PVP/CMC ratio offers a promising direction for developing high-energy, long-life Li–S cells, and provides a foundation for future studies on functional binder systems.