Ingyun
Shin†
a,
Yeong-Seok
Oh
a,
Seung-Woo
Seo
a,
Junyoung
Heo
bc,
Jeong-Jin
Yang†
d,
Moongook
Jeong
*d,
Jun-Woo
Park
*bc and
Seongki
Ahn
*a
aDepartment of Chemical Engineering, Research Center of Chemical Technology, Hankyong National University, 27, Jungangro, Anseong-si, Gyeonggi-do, Republic of Korea. E-mail: skahn@hknu.ac.kr
bNext Generation Battery Research Center, Korea Electrotechnology Research Institute (KERI), Changwon-si, Gyeongsangnam-do, Republic of Korea. E-mail: parkjw@keri.re.kr
cDepartment of Electro-Functionality Materials Engineering, University of Science and Technology (UST), Daejeon, Republic of Korea
dResearch Organization for Nano and Life Innovation, Waseda University, 3-4-1, Okubo, Shinjuku-ku, Tokyo, Japan. E-mail: mg-jeong@aoni.waseda.jp
First published on 24th November 2025
Quasi-solid-state electrolytes (QSSEs) have attracted significant attention as a promising solution to safety and shuttle-effect issues in lithium-sulfur batteries (LSBs) owing to their high lithium-ion transference numbers (tLi+), which suppress lithium dendrite formation and enhance safety and electrochemical stability. In this study, a binder-free (BF) cathode with a QSSE is fabricated via in situ polymerization of 1,3-dioxolane into poly(1,3-dioxolane) (PDOL). The QSSE serves simultaneously as an electrolyte and a binder. Despite being BF, the electrode exhibits stable electrochemical performance and mechanical strength, even under deformation. The polyvinylidene fluoride (PVDF) cathode shows a lower initial capacity of 885.8 mAh g−1 because the PVDF binder impedes capillary absorption, preventing deep electrolyte infiltration and generating voids that hinder charge transport and reduce coulombic efficiency. The BF cathode achieves 1059.3 mAh g−1 at 0.2C owing to infiltration of polymerized PDOL into the porous structure, enhancing interfacial integration and wettability. This study is the first to employ PDOL as a bifunctional binder–solid polymer electrolyte in LSBs, exploiting its strong adhesion and high lithium-ion conductivity. The BF@QSSE pouch cell is exceptionally flexible and safe under cycling and mechanical abuse, demonstrating the potential of combining BF cathodes with in situ-formed PDOL to fabricate flexible LSBs.
In this study, a quasi-solid-state LSB was developed via the in situ polymerization of DOL to form bifunctional PDOL. By exploiting the strong adhesion of the in situ-polymerized PDOL, binder-free (BF) cathodes were fabricated and evaluated without the use of insulating polyvinylidene fluoride (PVDF) binders, enabling improved electrode flexibility and enhanced electrochemical performance. The BF cathode utilizing bifunctional PDOL achieved an initial capacity of 1059.3 mAh g−1 and 700.4 mAh g−1 after 100 cycles at 0.2C. In comparison, the PVDF-based cathode delivered 885.8 mAh g−1 initially and 370.9 mAh g−1 after 100 cycles. These results demonstrate that the BF cathode combined with QSSE provides enhanced cycling stability. Additionally, the solid form of the QSSE not only facilitates efficient ion transport in solution but also effectively suppresses the polysulfide shuttle effect, mitigating lithium anode corrosion.18 This approach offers a promising design pathway for BF configurations in LSBs, particularly when integrated with QSSEs, enabling both mechanical cohesion and enhanced interfacial stability. Notably, our investigation revealed that the use of a PVDF binder obstructed the electrolyte infiltration pathways within the electrode, in turn suppressing capillary absorption. This blockage led to the formation of internal voids, which were identified as critical factors contributing to the reduced electrochemical performance of PVDF cathodes. By contrast, the porous structure of the BF cathode allows the polymerized PDOL to infiltrate deeply, fill internal gaps, and improve interfacial integration and wettability, consequently enhancing the electrochemical performance. Furthermore, in cycling and abuse tests, the pouch cell exhibited exceptional flexibility and reliable safety without any internal short circuits, highlighting its strong potential for practical applications.
The LE was prepared by dissolving LiTFSI (1.0 M) in a mixture of DOL and DME (1
:
1 v/v), followed by the addition of LiNO3 (1 wt%, Thermo Fisher Scientific).
To prepare the Li2S6 solution, S and Li2S were dissolved in DOL at a mass ratio of 5
:
1 and stirred at 50 °C for 24 h. All procedures were conducted in an argon-filled glovebox.
:
3 to prepare the S/KB composite material. Subsequently, the mixture was melt-diffusion treated at 155 °C for 12 h to ensure uniform infiltration of sulfur into the carbon matrix. For the BF and PVDF cathodes, the S/KB composite and conductive carbon (Super-P) were mixed at weight ratios of 80
:
10 and 80
:
10
:
10 (S/KB
:
Super-P
:
PVDF), respectively. The PVDF binder was added to the PVDF cathode formulation. In both cases, 1-methyl-2-pyrrolidinone (Sigma-Aldrich) was used as the solvent to form a uniform slurry. The coin cells were assembled using a BF@QSSE configuration, in which the BF cathode was first placed into the cell. The QSSE precursor was then dropped onto the PE separator, followed by placement of the Li foil. After assembly, the cells were left to stand at room temperature for 24 h, during which DOL polymerized in situ, forming a QSSE. For comparison, PVDF@QSSE cells with PVDF cathodes were assembled and gelled using the same procedure. As an additional control, BF@LE cells filled with the LE were also assembled. For the pouch-cell assembly, a cell with a total area of 33 cm2 was assembled using the same design as that of the coin cell.
![]() | (1) |
![]() | (2) |
![]() | (3) |
Charge–discharge tests were carried out from 1.8 to 2.8 V (SERIES 4000, Maccor). The lithium plating/stripping behavior was examined using Li/Li symmetric cells (VSP-300, BioLogic); cells assembled with QSSE or LE were tested at current densities of 0.1, 0.2, 0.5, 1.0, and 1.5 mA cm−2.
To assess the morphological changes induced by QSSE injection, scanning electron microscopy (SEM) was used to analyze three cathode configurations: BF cathode (BF), PVDF cathode (PVDF), and QSSE on a BF cathode (BF@QSSE). As shown in Fig. 2a and b, the BF cathode exhibits noticeable interparticle gaps and surface cracks, indicating weak cohesion and poor mechanical integrity. By contrast, the PVDF-based cathode (Fig. 2c and d) shows reduced cracking and improved structural stability. Notably, the BF@QSSE cathode (Fig. 2e and f) presents a smooth and uniform surface, indicating homogeneous integration and structural integrity. To evaluate whether the PDOL-based electrolyte can function as a binder in the cathode, three mechanical tests (adhesion strength, tape peeling, and folding) were performed. As shown in Fig. S2, both PDOL and PVDF supported a 100 g weight without detachment, indicating a comparable adhesion strength. Given PVDF's established role as a binder, the similar performance of PDOL indicates that it can also ensure structural integrity of the cathode. To further examine the mechanical stability, a peel test was conducted using Scotch tape. As shown in Fig. S3, the PVDF-based and QSSE-coated cathodes exhibited moderate adhesion with partial detachment, whereas the BF cathode completely delaminated, exposing the Al foil. These results underscore the importance of binders in maintaining cathode integrity. Additionally, a folding test was conducted to assess mechanical robustness (Fig. S4). The QSSE-coated and PVDF-based cathodes maintain their structural integrity after folding, whereas the BF cathode exhibits delamination and surface cracking. Thus, the QSSE acts effectively as a binder, improving particle cohesion and interfacial adhesion, which is expected to enhance cycling stability and flexibility.
The ionic conductivities of the LE and QSSE were measured over the temperature range of 30–60 °C, with values of 7.05 × 10−3 and 2.07 × 10−3 S cm−1 at 30 °C, respectively. Although the QSSE exhibited lower conductivity than the LE, as shown in Fig. 3a, it still demonstrated a remarkably high ionic conductivity at room temperature, which is considered excellent for solid-state electrolytes. As shown in Fig. 3b and S5, the electrochemical stability windows of the QSSE and LE were evaluated using linear sweep voltammetry (LSV). The onset of a distinct oxidation current is defined as the upper voltage limit of the electrochemical stability window. As illustrated in the magnified view in Fig. S5, compared with the LE, which exhibited oxidative decomposition at 3.9 V, the QSSE remained stable up to 4.15 V, indicating improved oxidative resistance. This enhancement is attributed to PDOL, as the ring-opening polymerization initiated by Lewis acid catalysts improves the oxidative stability of DOL by reducing the number of free solvent molecules that are otherwise easily oxidized under high-voltage conditions.35,36 A higher lithium-ion transference number (tLi+) can effectively mitigate concentration polarization during the charge–discharge process of lithium–sulfur batteries, thereby promoting uniform Li+ deposition and suppressing the formation of Li dendrites.37 As shown in Fig. 3c and d, tLi+ of the QSSE-based cell (0.74) was higher than that of the LE-based cell (0.34), demonstrating improved Li+ transport. This enhancement may be attributed to the favorable interaction between the PDOL-based quasi-solid matrix and the Lewis acid initiator (SiCl4), which likely coordinates with the TFSI− anions and restricts their mobility, thereby increasing the effective tLi+.38 Additional measurements were conducted for various electrolyte compositions (Fig. S6, tLi+ values of 0.62 for the 1 M, 20 mM system; 0.19 for the 2 M, 10 mM system; and 0.40 for the 2 M, 20 mM system). Based on these results, the 1 M, 10 mM precursor formulation, exhibited the best performance, was selected for all subsequent QSSE preparations in this study.
Fig. 4a shows the cyclic voltammetry (CV) curves of the LSB with BF@QSSE measured at a sweep rate of 0.1 mV s−1. The CV curves of the BF@QSSE cell in the initial three cycles exhibited high repeatability, indicating good electrochemical reversibility. Subsequent CV measurements were carried out at scan rates ranging from 0.1 to 0.4 mV s−1 to evaluate the kinetic features of the redox reactions (Fig. 4b). Two distinct cathodic peaks were observed during the discharge process. The first cathodic peak (C1) at around 2.3 V corresponds to the initial reduction of elemental sulfur (S8) into soluble long-chain polysulfides (Li2S8 and Li2S6) through S–S bond cleavage and multi-electron transfer. The resulting Li2Sx species (x ≥ 6) dissolve into the electrolyte, forming active intermediates that enable further redox reactions. The second cathodic peak (C2) near 2.0 V is attributed to the subsequent reduction of these soluble polysulfides into short-chain and insoluble species (Li2S2/Li2S), which nucleate and deposit on the cathode surface, completing the conversion of sulfur to Li2S.39 Two distinct anodic peaks (A1 and A2) were observed during the charging process at approximately 2.3 V and 2.4 V, respectively. The first anodic peak (A1) near 2.3 V corresponds to the oxidation of insoluble discharge products(Li2S2/Li2S) into soluble higher-order polysulfides such as Li2S4, representing the reverse of the low-voltage reduction process. During this step, the solid Li–S species are reoxidized and dissolved into the electrolyte, regenerating electrochemically active intermediates. The second anodic peak (A2) at around 2.4 V is attributed to the further oxidation of these polysulfides into long-chain species (Li2S6,/Li2S8) and elemental sulfur (S8).40 As shown in Fig. 4c, the linear relationship between the anodic and cathodic peak currents and the square root of the scan rate indicates that the redox processes were governed by a diffusion-controlled mechanism.41
![]() | (3) |
To quantitatively evaluate the effect of BF@QSSE on Li+ diffusion kinetics, diffusion coefficients were derived from the CV curves obtained at varying scan rates using the Randles–Sevcik equation, as described in eqn (3). The coin cell constructed with BF@QSSE exhibited Li+ diffusion coefficients of 5.63 × 10−10, 9.38 × 10−10, 3.45 × 10−10, and 6.35 × 10−10 cm2 s−1 for the A1, A2, C1, and C2 peaks, respectively, at a scan rate of 0.1 mV s−1 (Fig. 4d). A full overview is provided in Table S1, which summarizes the diffusion coefficients corresponding to A1, A2, C1, and C2 at scan rates of 0.1, 0.2, 0.3, and 0.4 mV s−1, showing comparatively favorable values.
Fig. 5a–c show the galvanostatic charge–discharge profiles of the BF@QSSE, BF cathode with LE (BF@LE), and PVDF cathode with QSSE (PVDF@QSSE) cells recorded at the 1st, 25th, 50th, 75th, and 100th cycles at a current rate of 0.2C (1C = 1675 mAh g−1). During the discharge process, two distinct voltage plateaus are observed, which correspond to the typical stepwise redox reactions of lithium–sulfur batteries. The first plateau at around 2.3–2.2 V arises from the electrochemical reduction of solid sulfur (S8) to soluble long-chain polysulfides (Li2S8), while the subsequent plateau at 2.1–1.9 V is associated with the further reduction of these intermediates into short-chain and insoluble lithium sulfides (Li2S2/Li2S). This multistep conversion involves solid–liquid and liquid–solid phase transitions, consistent with the CV results. Among the three configurations, the BF@QSSE cell exhibited the highest initial discharge capacity of 1059.3 mAh g−1, outperforming the BF@LE (917.2 mAh g−1) and PVDF@QSSE (885.8 mAh g−1) cells, confirming its superior sulfur utilization and more efficient redox kinetics. As shown in Fig. 5a, the BF@QSSE cell displays two well-defined discharge plateaus at ≈2.4 and 2.1 V, characteristic of a catholyte-mediated redox mechanism. The first charge curve is the key feature to focus on. The overpotential observed during the initial charge process reflects the energy barrier associated with Li2S oxidation (Fig. S7). Notably, the lower overpotential of the QSSE-based cell indicates superior reaction kinetics during this conversion step.42,43 Moreover, QSSE-based cells displayed a lower polarization potential than LE-based cells (Fig. S8), as determined by the voltage gap, ΔE (measured at 50% discharge capacity); specifically, the BF@QSSE cell showed a lower ΔE of 150 mV than the BF@LE cell (183 mV). This improvement is attributed to the higher Li+ transference number of the QSSE, which enhances Li+ transfer dynamics and reduces concentration polarization during cycling.
However, from the 75th cycle onward, the plateaus in the BF@QSSE cell gradually faded and shortened, likely owing to the increased solution resistance stemming from the elevated viscosity caused by high concentrations of dissolved polysulfides.44–46 Nevertheless, the BF@QSSE cell, benefiting from the high tLi+ of the QSSE and its reversible redox behavior, maintained superior performance, delivering an initial discharge capacity of 1059.3 mAh g−1 and retaining 700.4 mAh g−1 after 100 cycles (66.12% retention, Fig. 5d). The BF@LE cell showed rapid capacity fading, likely owing to inadequate suppression of the polysulfide shuttle and the formation of unstable SEI and Li dendrites.18,47 Interestingly, although PVDF was expected to provide a stable and highly reversible electrochemical performance, the PVDF@QSSE electrode exhibited poor reversibility and a low capacity retention of only 41.88%. This unexpected result indicates that the use of PVDF binders must be carefully reconsidered when combined with in situ-polymerized solid electrolytes.
Fig. 5e compares the rate capabilities of Li–S cells under varying C-rates. As the current density increases from 0.1C to 2.0C, the BF@QSSE cell delivers discharge capacities of 1072, 851.9, 767.4, 712.6, 624.9, and 522.8 mAh g−1, respectively. These values are consistently higher than those of the BF@LE cell (979.8, 476.7, 406.5, 360.5, 297.1, and 248.4 mAh g−1) and the PVDF@QSSE cell (797.4, 626.5, 551.9, 489.2, and 320.9 mAh g−1). When the current rate is returned to 0.1C, the discharge capacities of the BF@QSSE, BF@LE, and PVDF@QSSE cells recover to 686.8, 455.7, and 529.6 mAh g−1, respectively. Notably, the BF@QSSE cell exhibited excellent capacity recovery, indicating excellent structural resilience and rate reversibility, in agreement with the CV test results and confirming its highly reversible redox behavior. By contrast, the rapid capacity loss of the BF@LE cell at 0.1C is likely attributed not only to reduced structural stability caused by the absence of a binder but also to additional capacity fading induced by LiPS dissolution. Furthermore, a comparison of BF@QSSE and PVDF@QSSE across all current densities reveals a pronounced capacity gap, which is likely due to enhanced electrolyte infiltration within the BF cathode, thereby increasing the efficiency of charge transport.
The long-term cycling performances at 1C and 2C were also evaluated (Fig. 5f). At 1C, the BF@QSSE cell delivered an initial discharge capacity of 1048.7 mAh g−1 and retained 418.8 mAh g−1 after 300 cycles, outperforming BF@LE (initial: 873.3 mAh g−1, 300th: 254.5 mAh g−1) and PVDF@QSSE (initial: 859.5 mAh g−1, 300th: 225.3 mAh g−1). At 2C, the BF@QSSE cell still exhibited stable cycling, retaining 352.4 mAh g−1 after 400 cycles, whereas BF@LE and PVDF@QSSE showed much lower capacities of 159.43 and 199.41 mAh g−1, respectively. The superior performance of BF@QSSE was attributed to the enhanced electrolyte wettability and Li+ transport enabled by the porous BF structure, which effectively accommodated volume changes during cycling.48 Although the BF@LE cell exhibited a comparable initial capacity (542.2 mAh g−1) to the BF@QSSE cell (610 mAh g−1) at 2C, its capacity declined more severely after 400 cycles. This degradation is likely attributable to the formation of lithium dendrites, as inferred from the previously observed higher overpotential and polarization potential of the BF@LE cell. By contrast, the PVDF@QSSE cell showed a significantly lower capacity at high current densities, likely owing to the use of the PVDF binder, which can block active pores and hinder ion transport. This pore blockage increases the interfacial resistance and impairs electrode–electrolyte interactions, ultimately limiting the electrochemical performance under high current density conditions.49
Electrochemical impedance spectroscopy (EIS) was performed to evaluate the kinetics and resistance behavior of the LSB containing BF@QSSE, BF@LE, and PVDF@QSSE before and after 1C cycling (Fig. S9, Table S2). The high-frequency intercept, high- and medium-frequency semicircles, and low-frequency slope correspond to bulk resistance (Rb), SEI resistance (RSEI), charge transfer resistance (RCT), and Warburg impedance (Wo), respectively.50–52 First, a comparison between BF@QSSE and BF@LE, both employing the same BF cathode, indicates that the initially higher RSEI and RCT observed for BF@LE could be attributed to the reduced mechanical stability owing to the absence of a binder. Notably, BF@QSSE exhibited a stable RSEI, whereas the RSEI of BF@LE significantly changed from 100.93 Ω to 34.14 Ω after 300 cycles. This stability can be attributed to the improved interfacial contact between the QSSE and lithium anode, achieved via in situ polymerization, which facilitates uniform lithium deposition and efficient charge transfer.53 Consequently, a robust SEI film was formed during cycling, contributing to shuttle suppression and lithium anode protection.54,55
Similarly, when comparing BF@QSSE and PVDF@QSSE, both employing the same QSSE, the BF@QSSE cell exhibited an increase in Rb from 2.48 Ω to 9.12 Ω and in RSEI from 18.17 Ω to 21.4 Ω after prolonged cycling. These increases align well with the earlier observation in Fig. 5a, supporting our explanation that the fading and shortening of the discharge plateaus originate from the viscosity-induced rise in solution and interfacial resistances caused by the accumulation of dissolved polysulfide. Nevertheless, the BF@QSSE cell still demonstrated a markedly lower RCT than the PVDF@QSSE counterpart even after extended cycling. The elevated RCT in the PVDF-based cathode is attributed to the insulating nature of PVDF, which forms resistive polymer layers and, owing to its strong dipole moment (2.1 D), induces Li+ dipole interactions that hinder both electron and ion transport within the cathode.48,56 Therefore, despite the gradual rise in Rb and RSEI, the BF@QSSE configuration maintained superior interfacial kinetics and overall electrochemical stability compared with PVDF@QSSE.
Contrary to our initial expectation that a quasi-solid-state battery with a PVDF-based cathode would exhibit the most stable electrochemical characteristics, this battery was found to be poorly reversible. As mentioned earlier, this indicates that the use of PVDF binders may be detrimental when combined with in situ-polymerized quasi-solid-state electrolytes. To validate this assumption, cross-sectional analyses of the electrodes were conducted using FE-SEM, as shown in Fig. 6.
Fig. 6a and b display cross-sectional SEM images of the BF@QSSE and PVDF@QSSE cathodes, respectively, and Fig. 6a-i,ii and b-i,ii show the corresponding energy-dispersive X-ray spectroscopy (EDS) mapping images. In BF@QSSE, the polymerized PDOL effectively infiltrates the porous structure, penetrating deep into the cracks and reaching the Al foil, highlighting its superior wettability and interfacial integration. The porous structure of the BF cathode facilitated electrolyte penetration and active material utilization. By contrast, in a PVDF cathode, the PVDF binder is expected to occupy a significant fraction of the pore volume of the carbon host, thereby reducing the accessible surface area and limiting electrolyte infiltration. This blockage hinders effective charge transport and prevents deep electrolyte penetration, leading to the formation of voids within the cathode. The sulfur confined in these voids remained electrochemically inactive, essentially becoming dead space, which explains the reduced capacity and inferior cycling performance of the PVDF@QSSE cell.57–60 This behavior is closely related to the capillary effect.61 To quantitatively substantiate the hypothesis that the PVDF cathode restricts electrolyte infiltration by limiting the accessible surface area, Brunauer–Emmett–Teller (BET) analysis was performed for both the BF and PVDF cathodes. As shown in Fig. S10a and b, the BF cathode exhibited a higher specific surface area of 56.6 m2 g−1, compared with 45.7 m2 g−1 for the PVDF cathode. In addition, the pore size distribution results (Fig. S10c) revealed that the BF cathode possessed a larger average pore width (75.9 nm) than the PVDF cathode (40.8 nm), indicating a more open porous framework. These differences in surface area and pore size quantitatively confirm that the PVDF binder significantly reduces the accessible pore volume of the carbon host, thereby impeding electrolyte infiltration. The restricted penetration likely leads to the formation of voids within the cathode. To clarify the correlation between electrode and electrolyte uptake, the contact angle was measured (Fig. S11). The BF cathode (17°) exhibited a smaller contact angle than the PVDF cathode (28°), indicating better electrolyte wettability. Additionally, the electrolyte absorption was evaluated by immersing the electrodes (1 × 4 cm) into the electrolyte for controlled times of 5, 10, and 15 s; the weight change is shown in Fig. S12. The PVDF-based electrode rapidly absorbed the electrolyte within the first 5 s, but barely increased afterward, whereas the BF electrode continued to absorb the electrolyte over time. These results indicate that in the PVDF electrodes, the DOL solution could not penetrate deeply during in situ polymerization, leading to the formation of internal voids and degraded electrode performance (Fig. 6c).
Li-polysulfide (0.1 M Li2S6) diffusion tests were conducted using H-type cells equipped with either a QSSE-coated polyethylene (PE) membrane or a bare PE membrane, as shown in Fig. 7a. In the H-cell with the QSSE-coated PE, no visible color change occurred on the right side, even after 4 h, indicating effective suppression of Li-polysulfide permeation. By contrast, the H-cell with bare PE showed gradual yellowing of the solution on the right side within 4 h owing to the rapid diffusion of Li2S6 from the left chamber through the membrane. This result indicates that the QSSE-coated membrane can effectively mitigate the shuttle effect in LSB by blocking the migration of soluble polysulfides. After the H-type cell experiment, the variation in the concentration of polysulfides in the electrolyte on the right side of the cell was analyzed using UV-vis spectroscopy. As shown in Fig. 7b, the polysulfide solutions exhibit broad absorption peaks ranging from 250 to 600 nm. The QSSE-coated PE significantly reduced the Li2S6 absorbance intensity compared with that of the bare PE membrane, indicating that only a trace amount of Li2S6 permeated the QSSE-coated PE. This result confirms that the BF@QSSE cell suppressed the shuttle effect more effectively than the BF@LE cell.
The interfacial stability between the electrolyte and lithium metal anode is critical for the reliable operation of lithium-based batteries.62 To investigate the Li+ deposition behavior, the cycling performances of Li symmetric cells were evaluated at current densities of 0.1, 0.2, 0.5, 1.0, and 1.5 mA cm−2, as shown in Fig. 8a. Symmetric Li cells were assembled using QSSE, whereas the control cells were treated with LE. The QSSE-based cell exhibited lower voltage polarization across all current densities, indicating superior compatibility and interfacial stability with lithium metal compared with the LE-based cell. This can be attributed to the faster ion transport and enhanced interfacial stability of the QSSE, even at high current densities, which ultimately led to superior electrochemical performance. The slight asymmetry of the overpotential around 0 V originates from the different kinetics of Li plating and stripping. During plating, nucleation and growth of new Li deposits induce a higher interfacial resistance, whereas stripping occurs from pre-formed Li, leading to a lower overpotential.63 SEM was used to analyze the Li plating/stripping behavior by examining the surface morphology of the Li metal after the Li plating/stripping test of the Li/Li symmetric cells. As shown in Fig. 8b and c, the Li surface of the QSSE-based symmetric cell exhibits a smooth, silver-white morphology, indicating uniform Li plating/stripping and suppressed dendrite formation. By contrast, the LE-based cell exhibited a rough, dark-gray surface with byproducts and a powdery texture, implying uncontrolled deposition and parasitic reactions. These SEM and optical observations (inset images) confirmed that the QSSE promotes interfacial stability by mitigating dendrite growth and inactive Li accumulation during cycling.39,64
Atomic force microscopy (AFM) was employed to further investigate the Li plating/stripping behavior by analyzing the surface morphology and roughness of the Li metal electrodes after cycling. As shown in Fig. 8d, the QSSE-based cell exhibited a significantly smoother and denser lithium surface with a reduced surface roughness (Sq) of 227.7 nm. This uniform topography indicates more stable Li plating/stripping behavior and effective suppression of dendrite growth. By contrast, the LE-based cells (Fig. 8e) displayed a rough and irregular surface characterized by coarse lithium particles and branch-shaped dendritic structures, resulting in a hilly and uneven topography with an Sq of 596 nm. This morphology indicates uncontrolled Li deposition, which can accelerate electrolyte consumption and promote undesirable side reactions owing to the increased surface area.65,66 These findings highlight the beneficial role of the QSSE in stabilizing the Li interface and promoting homogeneous Li+ transport during cycling.
The electrochemical performance of the flexible battery under repeated deformation is shown in Fig. 9a. An initial high discharge capacity of 952.8 mAh g−1 was maintained. The excellent mechanical flexibility and stable cycling performance of the Li–S cells were primarily attributed to the highly conductive and free-standing BF sulfur cathodes, along with the QSSE, which provided enhanced electrochemical performance and mechanical integrity. Moreover, a well-packaged pouch battery can steadily drive a red light-emitting diode (LED) under various deformation states (e.g., bending and folding) in a fully charged state (Fig. 9b). Even after cutting, as shown in Fig. S13, the pouch cell continued to light an LED lamp without any flame, smoke, or liquid leakage, clearly demonstrating the excellent safety and flexibility of the BF@QSSE-based cell.
The authors confirm that the data supporting the findings of this study are available within the article and its supplementary information (SI). The supplementary information contains LSV and EIS data, BET analysis, and additional supporting figures and tables. See DOI: https://doi.org/10.1039/d5ta08211k.
Footnote |
| † These authors contributed equally to this work. |
| This journal is © The Royal Society of Chemistry 2026 |