Design and integration of glass-fiber-cloth networks in PEO–LLZTO composite: a multifunctional approach for electrolyte engineering

Abstract

Targeting next-generation batteries, a multifunctional composite polymer electrolyte (CPE) was developed by integrating glass-fiber-cloth (GFC) with PEO–Li_6.4La₃Zr_1.4Ta_0.6O₁₂ (LLZTO). LLZTO improves thermal stability and ionic conductivity. GFC framework further reinforces mechanical strength and facilitates Li-ion dissociation, creating a Li-ion gradient. This synergistic combination presents a promising approach for strategic electrolyte design.

---

Lithium metal batteries (LMBs) are considered strong candidates for next-generation rechargeable batteries due to their high theoretical specific capacity (3860 mA h g⁻¹), low density (0.59 g cm⁻³), and extremely low redox potential (−3.040 V vs. the standard hydrogen electrode).¹ Nevertheless, the growth of lithium dendrites² during cycling presents significant safety concerns, such as the risk of thermal runaway and explosions. Additionally, liquid electrolyte exacerbates safety issues due to the properties of flammability and leakage issues. Solid-state electrolytes (SSEs) have emerged as a promising solution to overcome these challenges, owing to their non-flammability and low reactivity with lithium metal surfaces.³ Among the various SSEs, solid polymer electrolytes (SPEs) demonstrate benefits for commercializing LMBs, including enhanced electrode contact, ease of fabrication, and cost-effectiveness.

Polyethylene oxide (PEO)-based solid polymer electrolytes (SPEs) are the earliest and the most extensively studied in quasi-solid-state electrolyte systems. SPEs offer a large variety of advantages, including good flexibility, cost-effectiveness, and excellent processability. Nonetheless, low ionic conductivity of PEO-based SPEs at ambient temperature and their tendency to crystallize hinder the possibility of practical application. To overcome these limitations, extensive research has explored incorporating fillers into polymer matrices to form composite polymer electrolytes (CPEs).^4,5 This approach leverages filler properties to enhance both mechanical properties and ionic conductivity. Cubic-phase Li₇La₃Zr₂O₁₂ (LLZO) exhibits superior ionic conductivity and distinguished mechanical properties. Consequently, incorporating LLZO-based filler into a PEO-based polymer electrolyte reduces the polymer matrix's crystallinity,⁶ enhances its mechanical strength,⁷ improves ionic conductivity,⁸ and enhances interfacial compatibility.⁹ Furthermore, this Li-ion conducting filler, when distributed architecturally within the PEO matrix, can significantly affect the Li diffusion gradient, enabling advancements in ionic transport engineering.¹⁰ Inspired by Li-ion-conducting filler-reinforced electrolytes, we further advanced the development of innovative hybrid CPEs by demonstrating a fiber-network-reinforced PEO/garnet Li_6.4La₃Zr_1.4Ta_0.6O₁₂ (LLZTO) composite electrolyte. This composite electrolyte was prepared by incorporating a glass-fiber-cloth (GFC) network into a PEO/LLZTO electrolyte matrix. Due to its remarkable mechanical durability and thermal retention capabilities, GFC serves as a robust scaffold, enhancing the structural integrity of the CPE.​¹¹​ Furthermore, GFC network within the PEO matrix promotes lithium salt dissociation and provides an extra pathway for Li-ion diffusion.

In this work, we present a novel approach to enhance the performance of PEO-based solid polymer electrolytes through the synergistic combination of Ta-doped LLZO (LLZTO) particles and GFC network reinforcement. The incorporation of these dual components addresses multiple challenges simultaneously: the LLZTO particles improve ionic conductivity and reduce crystallinity, while the GFC network enhances mechanical stability and facilitates Li-ion transport by promoting LiTFSI dissociation. Through systematic investigation of the composite's structural, electrochemical, and mechanical properties, we demonstrate that this hybrid design effectively suppresses lithium dendrite growth and enables stable cycling performance.

As shown in Fig. 1(a)​, the fiber network reinforced solid polymer electrolytes were prepared via a solution casting method poly(ethylene oxide) (PEO, M_w = 600 000, ACROS organic) was mixed with LiTFSI at an ethylene oxide [EO] to Li⁺ molar ratio of 10 : 1 and dissolved in solvent under magnetic stirring to form a homogeneous solution. To improve the performance of the polymer electrolyte, 15 wt% of Li_6.4La₃Zr_1.4Ta_0.6O₁₂ (LLZTO, diameter: 5–8 μm, China Glaze) ceramic filler was incorporated into the polymer matrix. Afterward, a GFC framework (thickness: 30 μm, PENG SHENG Co., Ltd) was impregnated with the PEO–LiTFSI–LLZTO electrolyte to form the composite membrane, referred to as PEO–LiTFSI–LLZTO@GFC. The structural details of the GFC framework are shown in Fig. 1(b). The GFC consists of braided fiber bundles forming a reticular network. This interconnected structure provides mechanical reinforcement and facilitates the integration of the polymer matrix.

Fig. 1 (a) Schematic illustration for preparation of CPEs and LFP‖Li cells. (b) SEM images of glass-fiber-cloth, and (c) cross-section image of PEO–LiTFSI–LLZTO@GFC electrolyte. (d) Temperature-dependent morphological evolution (25–220 °C) of CPEs.

Fig. 1(c) presents the cross-sectional view of the polymer membrane. The image illustrates that the GFC framework is embedded centrally within the PEO–LiTFSI–LLZTO polymer electrolytes. The central positioning of the GFC enhances the structural integrity of the membrane while ensuring sufficient interaction between the polymer matrix and the fiber network. Fig. 1(d) reveals the temperature-dependent morphological evolution of three electrolyte systems (PEO–LiTFSI, PEO–LiTFSI@GFC, and PEO–LiTFSI–LLZTO@GFC) from 25 °C to 220 °C. At 60 °C, all samples become more transparent as the temperature approaches the theoretical melting point of PEO–LiTFSI matrix. At 100 °C, while the transparency transition continues, PEO–LiTFSI@GFC and PEO–LiTFSI–LLZTO@GFC maintain their structural integrity. At 160 °C, the pristine PEO–LiTFSI shows edge deformation due to melting. As the temperature reaches 220 °C, the pristine electrolyte completely transitions into a liquid state, while both PEO–LiTFSI@GFC and PEO–LiTFSI–LLZTO@GFC composites preserve their dimensional stability.

This enhanced thermal stability can be attributed to the GFC framework, which acts as a robust scaffold within the CPEs. Beyond thermal resistance, the GFC-containing composites also demonstrate evident mechanical strength, as evidenced by bending and twisting tests (Fig. S1, ESI†). This mechanical robustness is particularly crucial for battery applications, as it helps prevent lithium dendrite penetration, thereby improving cycle life and ensuring battery safety.

To further understand the structural characteristics of these CPEs, X-ray diffraction (XRD) analysis was conducted, as shown in Fig. 2(a). The pristine PEO–LiTFSI electrolyte exhibits strong and sharp diffraction peaks at 19° and 23.15° at room temperature, indicating its highly crystalline nature. While the incorporation of GFC and together with LLZTO both maintain these characteristic peaks, a significant reduction in T_m can be clearly observed from the DSC analysis (Fig. S2(a), ESI†), indicating an effective crystallinity suppression.

Fig. 2 Structural and spectroscopic analyses of CPEs: (a) XRD patterns, (b) FTIR spectra, (c) Raman spectra, and (d) schematic of Li-ion transport mechanisms.

Previous studies have shown that Li-ion conduction primarily occurs in the amorphous phase of PEO-based electrolytes.¹² The LLZTO-induced crystallinity suppression therefore suggests an enhanced ionic transport capability in these composite electrolytes. Moreover, thermogravimetric analysis reveals that the LLZTO-modified CPEs exhibit superior thermal stability with a major decomposition occurring at ∼400 °C, significantly higher than that of pristine PEO–LiTFSI which begins continuous weight loss from ∼170 °C (see Fig. S2(b), ESI†). This microstructural advantage, combined with the previously demonstrated thermal and mechanical stability, indicates that the LLZTO-modified CPEs could offer improved overall electrochemical performance.

To elucidate the molecular interactions within the CPEs, we conducted comprehensive spectroscopic analyses. Fourier Transform Infrared (FTIR) Spectroscopy revealed characteristic C–O–C stretching vibration triplets centred at 1056, 1090, and 1134 cm⁻¹ (Fig. 2(b)).¹³ Notably, the C–O–C stretching vibration peak at 1090 cm⁻¹ in pure PEO–LiTFSI shifted to 1104 cm⁻¹ upon incorporation of LLZTO and GFC, indicating enhanced cation complexation in the GFC-containing samples.¹⁴ Further investigation of LiTFSI dissociation behavior using Raman spectroscopy (Fig. 2(c)) revealed asymmetric peaks, suggesting two distinct components. Peak deconvolution analysis, performed using Gaussian fitting after linear background subtraction, identified two characteristic peaks at 744 and 748 cm⁻¹, corresponding to free TFSI⁻ anions and undissociated Li-TFSI ion pairs, respectively.¹⁵ The quantitative analysis showed progressively increasing free TFSI⁻ percentages: 65.84% for PEO–LiTFSI, 71.76% for PEO–LiTFSI@GFC, and 75.90% for PEO–LiTFSI–LLZTO@GFC. This enhancement in salt dissociation can be attributed to the abundant Si–O and O–H polar groups on the GFC surface, which facilitate TFSI interactions and promote Li-ion dissociation and transport.¹⁶

The spectroscopic evidence clearly demonstrates that the GFC membrane embedded in the polymer electrolyte enhances lithium salt dissociation. To understand how this enhanced dissociation contributes to Li-ion transport, we propose a mechanism based on the structural similarity between GFC and PDMS (polydimethylsiloxane) solid electrolyte, as illustrated in Fig. 2(d). Since GFC and PDMS share similar Si–O–Si backbone structures¹⁷ with abundant oxygen atoms, and given that Li-ion transport in polymer electrolytes occur through complexation with oxygen atoms, we anticipate that GFC can provide additional pathways for Li-ion conduction. In this system, Li-ions can migrate through the LLZTO, conventional PEO matrix via complexation with ether oxygen atoms assisted by polymer chain segmental motion, and through the oxygen-rich sites along the GFC framework. The enhanced transport behavior, combining the inherent polymer–salt complexation and GFC-promoted LiTFSI dissociation, effectively increases the overall ionic conductivity of the composite polymer electrolyte.

To validate our proposed ionic transport enhancement mechanism and evaluate the practical performance of our composite electrolytes, we conducted comprehensive electrochemical characterizations. The ionic conductivity, a critical parameter for energy storage applications, was measured using electrochemical impedance spectroscopy (EIS) at various temperatures, as shown in Fig. 3(a) and Fig. S3 (ESI†). At 50 °C, the ionic conductivity increased from 2.00 × 10⁻⁴ S cm⁻¹ for PEO–LiTFSI to 2.34 × 10⁻⁴ S cm⁻¹ for PEO–LiTFSI@GFC, and further improved to 4.90 × 10⁻⁴ S cm⁻¹ for PEO–LiTFSI–LLZTO@GFC. The corresponding activation energies (E_a), derived from temperature-dependent conductivity measurements, decreased from 0.278 eV for PEO–LiTFSI to 0.152 eV for PEO–LiTFSI@GFC, and 0.181 eV for PEO–LiTFSI–LLZTO@GFC. E_a is calculated at 27–50 °C because PEO–based electrolytes begin to transition into a molten amorphous phase near 60 °C, altering their Li-ion conductivity properties. The improvements in both ionic conductivity and reduced activation energy align well with our spectroscopic observations of enhanced salt dissociation and validate our proposed mechanism of additional transport pathways through the GFC framework, with LLZTO particles further facilitating ion migration.

Fig. 3 (a) Arrhenius plots of log σ versus 1000/T of PEO–LiTFSI, PEO–LiTFSI@GFC and PEO–LiTFSI–LLZTO@GFC electrolytes. (b) The schematic illustration of Li-ion concentration profile of PEO–LiTFSI and PEO–LiTFSI–LLZTO@GFC electrolytes. (c) The voltage profiles of Li plating/stripping in Li|CPEs|Li symmetric cells at 50 °C (0.05 mA cm⁻²): (c) PEO–LiTFSI, (d) PEO–LiTFSI@GFC, and (e) PEO–LiTFSI–LLZTO@GFC electrolytes.

The enhanced ionic conductivity can be further rationalized by examining the spatial distribution of Li-ions within the composite structure. Based on our FTIR and Raman spectroscopic analyses, which demonstrated enhanced salt dissociation in GFC-containing samples, we propose that the GFC membrane creates a unique Li-ion concentration profile across the electrolyte. Specifically, as illustrated in Fig. 3(b), the centrally positioned GFC membrane acts as a high-concentration region, leading to the formation of a concentration gradient that decreases radially towards both interfaces of the polymer matrix. This gradient distribution mechanism is analogous to previous studies on LLZO fillers, where gradual gradient distribution has been shown to facilitate consistent and rapid Li-ion transport.¹⁸ In our system, the naturally formed Li-ion concentration gradient, combined with the enhanced salt dissociation near the GFC membrane, synergistically contributes to the observed improvement in ionic transport properties.

To further validate the enhanced ionic transport properties and evaluate the practical applicability of our CPEs, we investigated their compatibility with lithium anodes using lithium symmetric cells (Li|CPEs|Li). The Li plating/stripping profiles of CPEs were measured at a current density of 0.05 mA cm⁻² and 50 °C, as shown in Fig. 3(c–e). Initially, the Li|PEO–LiTFSI|Li cell exhibited a relatively small voltage hysteresis of 60 mV. However, the performance deteriorated significantly after 70 hours of cycling, manifesting random voltage fluctuations and a dramatic increase in voltage hysteresis to over 1000 mV after 100 hours. This sharp voltage increase can be attributed to severe uneven Li deposition, a common failure mechanism in conventional polymer electrolytes.¹⁹ In stark contrast, both the Li|PEO–LiTFSI@GFC|Li and Li|PEO–LiTFSI–LLZTO@GFC|Li cells demonstrated significant stability, maintaining consistent voltage hysteresis of approximately 30 mV and 40 mV, respectively, with no short circuits observed throughout 300 hours of stripping-plating cycles. This enhanced stability can be understood through the mechanical perspective: previous studies have established that lithium dendrite growth can be effectively suppressed when the separator's shear modulus exceeds 1.8 times that of lithium metal (4.9 GPa).²⁰ In our system, the GFC membrane, possessing a high shear modulus of approximately 70 GPa,²¹ provides a robust mechanical framework. When strategically positioned in the middle layer of CPEs, the GFC membrane not only enhances the overall structural rigidity but also effectively prevents lithium dendrite penetration.

Notably, both GFC-containing systems also exhibited an expanded electrochemical stability window, with both PEO–LiTFSI@GFC and PEO–LiTFSI–LLZTO@GFC demonstrating stability exceeding 4.9 V, compared to 4.0 V for the conventional PEO–LiTFSI system (Fig. S4, ESI†). This widened electrochemical window, coupled with the improved mechanical properties, enhanced ionic conductivity, and favourable Li-ion concentration gradient, collectively contributes to the uniform Li deposition and effective suppression of dendrite formation in these composite systems.

Following the promising results from symmetric cell tests, we further evaluated the practical applicability of our CPEs in LiFePO₄ (LFP)|CPEs|Li quasi-solid-state batteries. The cycling performance of these batteries is presented in Fig. 4(a). While all cells showed capacity fading over extended cycling, cells incorporating PEO–LiTFSI@GFC and PEO–LiTFSI–LLZTO@GFC electrolytes demonstrated significantly improved capacity retention compared to those with conventional PEO–LiTFSI electrolyte. Specifically, approaching 50 cycles, at 0.2C and 50 °C, the PEO–LiTFSI–LLZTO@GFC maintained approximately 65% of its initial capacity, while the PEO–LiTFSI@GFC retained about 45%. In contrast, the conventional PEO–LiTFSI system showed severe capacity degradation, retaining only about 10% of its initial capacity. The charge–discharge profiles at 0.2C (Fig. 4(b)) show favorable initial coulombic efficiencies (CE) for all configurations, with the composite polymer electrolytes demonstrating incrementally higher values: PEO–LiTFSI achieved 95.76% (151.38/158.09 mA h g⁻¹), PEO–LiTFSI@GFC reached 96.99% (150.45/155.13 mA h g⁻¹), and PEO–LiTFSI–LLZTO@GFC exhibited the highest at 98.79% (149.91/151.75 mA h g⁻¹).

Fig. 4 (a) Cycling performance of LFP‖Li cells with different CPEs at 0.2C and 50 °C. (b) Charge–discharge profiles of LFP‖Li cells with various CPE systems. (c) Schematic illustration comparing dead Li formation and dendrite growth mechanisms.

It should be noted that while these results demonstrate the advantages of our composite design, the overall performance metrics still leave room for improvement compared to some literature reports. This performance decay can be attributed to the relatively lower LLZTO filler loading, leading to uneven SEI formation, and the absence of continuous stacking pressure during measurement affecting interfacial contact. Nevertheless, the significant performance enhancement observed with the GFC-based composite structure underscores the effectiveness of our design strategy. Fig. 4(c) illustrates the Li deposition behavior within the CPEs. In conventional PEO electrolyte systems, uneven Li deposition leads to dead Li formation and dendrite growth during cycling. The improved performance of the PEO–LiTFSI–LLZTO@GFC system can be attributed to the establishment of a Li-ion concentration gradient and the structural stability provided by the mechanical framework, which collectively enhance cycling stability in both symmetric cells and LFP‖Li battery configurations.

In summary, we developed a glass-fiber network-reinforced solid polymer electrolyte (PEO–LiTFSI–LLZTO@GFC) by incorporating a GFC network into a PEO-based solid electrolyte. The GFC network enhances mechanical properties and suppresses lithium dendrite formation while providing improved stability under high-temperature conditions. Moreover, it facilitates gradual gradient Li-ion distribution and creates additional pathways for ion transport, leading to enhanced ionic conductivity. The addition of LLZTO reduces PEO crystallinity while providing supplementary Li-ion channels. As demonstrated, the PEO–LiTFSI–LLZTO@GFC electrolyte shows improved electrochemical performance with reduced dead lithium and dendrite formation compared to conventional PEO-based systems, presenting a promising approach for developing high-performance composite solid electrolytes.

The authors gratefully acknowledge financial support from the National Science and Technology Council under project numbers NSTC 113-2636-E-007-006 and NSTC 113-2221-E-007-046.

Data availability

The data supporting this article have been included in the ESI.†

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. W. Xu, et al. , Energy Environ. Sci., 2014, 7, 513–537.
2. T. Yi, et al. , eScience, 2024, 4, 100182.
3. X.-B. Cheng, et al. , Chem, 2019, 5, 74–96.
4. Y. Zheng, et al. , Chem. Soc. Rev., 2020, 49, 8790–8839.
5. K.-Q. He, et al. , Rare Met., 2025, 1–11.
6. H. Zhuang, et al. , J. Alloys Compd., 2021, 860, 157915.
7. H. Huo, et al. , Adv. Energy Mater., 2019, 9, 1804004.
8. J. Zhang, et al. , Nano Energy, 2016, 28, 447–454.
9. Z. Zhang, et al. , Ceram. Int., 2020, 46, 11397–11405.
10. P.-Y. Chen, et al. , MRS Energy and Sustainability, 2022, 9, 360–368.
11. L. Wang, et al. , eScience, 2023, 3, 100090.
12. J. Feng, et al. , Nano Convergence, 2021, 8, 1–12.
13. M. Zhang, et al. , J. Poly. Environ., 2019, 27, 2369–2379.
14. S. Wen, et al. , J. Electroanal. Chem., 1996, 408, 113–118.
15. Z. Zhang, et al. , J. Membr. Sci., 2021, 621, 118940.
16. X. B. Cheng, et al. , Adv. Mater., 2016, 28, 2888–2895.
17. L. Gao, et al. , Nanoscale, 2020, 12, 14279–14289.
18. X. Zhang, et al. , J. Colloid Interface Sci., 2024, 658, 836–845.
19. K.-H. Chen, et al. , J. Mater. Chem. A, 2017, 5, 11671–11681.
20. X. Zhang, et al. , Acc. Chem. Res., 2019, 52, 3223–3232.
21. M. Gunasegeran and E. Sudhagar, Polym. Compos., 2022, 43, 2683–2697.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5cc01390a

---