Ultrathin Polymer Electrolyte With Fast Ion Transport and Stable Interface for Practical Solid‐state Lithium Metal Batteries

Abstract

Ultrathin solid‐polymer‐electrolytes (SPEs) are the most promising alternative substituting for the conventional liquid electrolyte to enable high‐energy‐density, safe lithium‐metal‐batteries (LMBs). Nevertheless, developing ultrathin SPEs with both high ionic conductivity, and strong Li dendrite retardant is still a significant challenge. Here a scalable fabrication of high‐performance ultrathin (≈7.8 µm) polycarbonate‐based electrolyte (UPCE) is proposed via electrolyte structural engineering, phase separation‐derived poly(vinylidene fluoride‐co‐hexafluoropropylene) (PVH) porous scaffold, without use of additional liquid additives. The rational electrolyte structural modulation with 1‐fluoro‐4‐(1‐methylethenyl)benzene (FMB) enables a weakened Li⁺‐polymer interaction due to weak Li⁺ solvation with fluorine, benzene ring, facilitates the formation of LiF‐rich solid‐electrolyte‐interphase on Li metal surface. As a result, the designed UPCE delivers a high ionic conductivity of 4.8 × 10⁻⁴ S cm⁻¹, an ultrahigh critical current density of 11.5 mA cm⁻² at 25 °C. The solid‐state Li symmetric cell attains unprecedented ultralong cycling over 6000 h at 0.5 mA cm⁻². Furthermore, the Li|LiCoO₂ cell cycles stably over 1500 cycles at a high operating voltage of 4.5 V, and the pouch cell can achieve a high energy density of 495 Wh kg⁻¹ excluding the packaging. This work offers a new pathway inspiring efforts to commercialize ultrathin SPEs for high‐energy solid‐state LMBs.

High‐performance ultrathin (≈7.8 µm) polycarbonate‐based electrolyte (UPCE) is fabricated, without the use of additional liquid additives. The designed UPCE delivers a high ionic conductivity (4.8 × 10⁻⁴ S cm⁻¹) and an ultrahigh critical current density (11.5 mA cm⁻²) at 25 °C. The 4.5 V solid‐state Li|LiCoO₂ cell demonstrates an ultralong lifespan cycling stability over 1500 cycles at 1 C.

1. Introduction

Li metal batteries have been recognized as the most promising next‐generation advanced energy storage due to their high energy density.^{[ 1 ]} However, the serious Li dendrite growth and inferior high‐voltage stability of liquid electrolytes severely fettered the practical implementation of Li metal batteries. Developing solid‐state Li metal batteries (SSLMBs) using solid electrolytes (SEs) substituting for liquid electrolytes is regarded as an effective and promising strategy for achieving safe and high‐energy‐density energy storage.^{[ 2 ]} However, the currently reported SSLMBs often exhibit unsatisfied properties toward practical application, including limited cycle life and poor rate capability due to the low ionic conductivity of SEs, excessive electrolyte thickness and uncontrollable Li dendrite growth especially at relatively high current densities. To realize high‐energy‐density (exceeding 500 Wh kg⁻¹) and practical SSLMBs,^{[ 3 ]} it is essential to developing high‐performance ultrathin SEs (below 15 µm) with superior ionic conductivity and robust Li dendrite suppression capability.

Although inorganic solid electrolytes (ISEs) possess relatively high ionic conductivity and superior mechanical strength, their inherent brittleness, complex preparation process and huge challenge for scaling up have seriously fettered their practical utilization. Additionally, maintaining a relatively large thickness is typically required for ISEs due to their high brittleness, but this inevitably results in a substantial compromise of both battery rate performance and energy density.^{[ 4 ]} In comparison, solid polymer electrolytes (SPEs) demonstrate prominent superiorities of high flexibility and easy processability. However, SPEs also encounter severe challenges inherent to their polymeric nature, such as low ionic conductivity at room temperature. Researchers have attempted to boost polymer segmental motion to enhance ion transport, via decreasing the crystallinity of polymer matrices and increasing the amorphous phase ratio.^[2c,5] However, a deeper understanding of factors influencing rapid ion diffusion in SPEs remain unclear. For example, the strong interaction between Li⁺ and the polymer chain is a key factor that hinders the fast Li⁺ diffusion.^{[ 6 ]} Recently, a low‐polarity linear carbonate segment has been reported to modify polyvinyl ethylene carbonate matrix to enable a weak Li⁺ solvation environment and promote the fast Li⁺ diffusion.^{[ 7 ]} However, the Li⁺ solvation of the modification segment remains relatively strong due to the existence of strong Li⁺ solvation ester group, and innovative strategies for electrolyte structural design of SPEs to optimize the solvation properties are still lacking. Notably, fluorination of solvent is a well‐known effective way to weaken Li⁺ solvation in liquid electrolytes.^{[ 8 ]} Regarding SPEs, past research primarily focuses on improving the oxidation resistance of electrolyte or modulating the interphase layer on Li metal surface using fluorine‐containing segments.^{[ 9 ]} However, the impact of fluorination on Li⁺ transport in SPEs has been scarcely reported.

On the other hand, another major challenge in developing ultrathin SPEs is the high risk of lithium dendrite penetration. Inorganic‐enriched solid electrolyte interphase (SEI) layer, particularly LiF‐rich one, could effectively suppress Li dendrite growth due to the lithiophobic nature of the LiF‐rich SEI and the Li migration along the Li/SEI interface.^{[ 10 ]} Notably, fluoroethylene carbonate (FEC) liquid additive has often been added in SPEs to promote the formation of LiF‐rich SEI layer,^{[ 11 ]} however, this does not fundamentally solve the critical issues associated with SPEs and the safety issues associated with liquid electrolyte still persist. Therefore, developing high‐performance SPEs solely through electrolyte structural engineering, without relying on any extra liquid additives, to enhance Li⁺ diffusion within the electrolyte and optimize the SEI layer is crucial for advancing the practical application of ultrathin SPEs.

In this work, ultrathin polycarbonate‐based electrolytes (UPCE) with fast Li⁺ transport and stable interface are rationally designed using electrolyte structural engineering and phase separation‐derived poly(vinylidene fluoride‐co‐hexafluoropropylene) (PVH) scaffold, without the need for additional liquid additives. We propose the introduction of 1‐fluoro‐4‐(1‐methylethenyl)benzene (FMB) segment to weaken the Li⁺‐polymer interaction to enable the fast Li⁺ diffusion in polycarbonate electrolyte and promote the formation of LiF‐rich SEI layer. In sharp contrast to fluorinated SPEs designed to enhance oxidation resistance or modify the interphase layer, our work aims to tailor the Li⁺‐polymer interaction to enable fast ion transport. A weakly solvated segment containing both fluorine and benzene ring is introduced to construct highly ionically conductive polycarbonate SPEs via in situ polymerization. Both the fluorine and benzene ring have the weak interaction with Li⁺, and hence the Li⁺‐polymer interaction environment of designed polycarbonate SPEs is weakened. As a consequence, the achieved UPCE delivers a high ionic conductivity of 4.8 × 10⁻⁴ S cm⁻¹ (25 °C), high Li⁺ transfer number (0.68), as well as strong Li dendrite suppression capability evidenced by the ultrahigh critical current density (CCD) of 11.5 mA cm⁻². Moreover, the UPCE also delivers excellent voltage stability up to 5.3 V. The solid‐state Li symmetric cell demonstrates an unprecedent ultralong lifespan cycling durability over 6000 h with dendrite‐free and uniform Li deposition at 0.5 mA cm⁻², 25 °C. The Li|LiCoO₂ (LCO) cell exhibits an ultralong and stable cycling over 1500 cycles at 1 C, 25 °C and an aggressive operating voltage of 4.5 V. Furthermore, the Li|LCO pouch cell with an ultrahigh areal‐capacity (≈4.5 mAh cm⁻²) and an ultrathin Li metal (20 µm) still manifests greatly enhanced cycling stability, achieving a high energy density of 495 Wh kg⁻¹, even under the harsh cycling condition of an extremely low negative‐to‐positive‐capacity (N/P) ratio of only 0.89.

2. Results and Discussion

2.1. Electrolyte Design and Structural Characterization

High performance UPCE with weakened Li⁺‐polymer interaction was prepared via the in situ copolymerization of vinyl ethylene carbonate (VEC) and fluorobenzene segments, using PVH porous membrane. Figure 1a shows the proposed electrolyte design principle of polycarbonate electrolyte. The Li⁺ to fluoride (from FMB) shows a much higher binding energy of 106.67 kcal mol⁻¹ than that Li⁺ coordinated with VEC (83.65 kcal mol⁻¹), revealing a much weaker Li⁺ solvation. Additionally, Li⁺ coordinated to benzene ring (from FMB) also exhibits a much weaker interaction with a higher binding energy of 101.31 kcal/mol. Due to the weak Li⁺ interaction with both fluorine and benzene ring of fluorobenzene segments, the Li⁺‐polymer interaction environment in UPCE could be significantly weakened compared to the strong Li⁺ interaction to carbonyl groups in polyvinyl ethylene carbonate (Figure 1b). A facile and scalable phase separation method was employed to prepare the PVH porous membrane. Driven by the rapid mutual diffusion between the solvent and non‐solvent, a highly porous microstructure within the polymer matrix was achieved.^{[ 12 ]} In stark contrast to the time‐consuming and expensive electrospinning,^[3d,13] the phase separation method is cost‐effective and large‐scalable. UPCE with a thickness of only ∼7.8 µm is constructed by the in situ copolymerization of VEC and FMB segments with trimethylolpropane trimethacrylate (TMPTMA) as the crosslinker (Figure 1c), where PVH scaffold provides high mechanical strength while inert for ion conduction. Notably, in situ polymerization can facilitate excellent electrode/electrolyte interphase wettability and enable the electrolyte preparation and cell assembly simultaneously by eliminating the complicated polymer dissolution and electrolyte drying processes. Hence, in situ polymerization shows an exceptional compatibility with existing battery manufacturing techniques (Figure S1, Supporting Information).

Figure 1.

Electrolyte design with a consideration of weak Li⁺ interaction for high‐performance UPCE. a) Proposed design principle for UPCE electrolyte with the basic monomers, showing the binding energy between Li⁺ and fluorine or benzene ring in different segments. VEC segment shows a strong Li⁺ interaction in contrast to the much weaker interaction of FMB segment. b) Schematic illustrations of Li⁺ interaction with different polymer chains. Left: strong Li⁺‐polymer interaction due to the rich carbonate group in polymer chain; right: weak Li⁺‐polymer interaction owing to the introduction of fluorobenzene segment. c) Schematic illustrations of UPCE preparation via in situ copolymerization, utilizing ultrathin PVH scaffold prepared by the phase separation method.

In this work, phase separation was employed to fabricate a highly porous membrane, with water serving as the non‐solvent. The PVH slurry rapidly solidifies into a membrane once contacted with water and could quickly detach from the substrate (Figure 2a). Due to the facile preparation process, the PVH membrane can be easily scaled up. The PVH membrane demonstrates rich and uniform micron‐sized pores with a thickness of only ∼7.8 µm (Figure 2b,c). After the in situ copolymerization, the achieved P(VEC‐FMB‐TMPTMA) UPCE exhibits a smooth surface while with almost no thickness variation (Figure 2d,e). The inset digital photograph reveals the homogeneous morphology of PVH membrane with excellent flexibility, and the membrane becomes transparent after being fully infused with the electrolyte. The optical images of P(VEC‐FMB‐TMPTMA) show that the transparent, colorless liquid precursor turns into light yellow solid after the copolymerization reaction (Figure S2, Supporting Information). The possible molecular configuration of P(VEC‐FMB‐TMPTMA) was shown as Figure S3 (Supporting Information) in the supporting information. Fourier transform infrared (FTIR) spectroscopy has been performed to characterize the molecular structure of the P(VEC‐FMB‐TMPTMA). The absence of the characteristic peak at 1638 cm⁻¹, which corresponds to the stretching vibrations of C═C, indicates the complete copolymerization of the monomers (Figure 2f). And the simultaneous disappearance of peak at 3100 cm⁻¹ attributed to the stretching vibration of C‐H bonds in the ─C═C─H group, could further evidence the complete copolymerization of the C═C bonds (Figure S4, Supporting Information). Additionally, the ¹³C nuclear magnetic resonance (NMR) spectra also show the absence of the C═C signal (Figure S5, Supporting Information). The combined FTIR and NMR results reveal that the monomers were highly copolymerized. In the XRD spectra, the P(VEC‐FMB‐TMPTMA) demonstrates weaker characteristic peaks compared to those of P(VEC‐TMPTMA), revealing the reduced crystallinity of electrolyte with the introduction of FMB segment (Figure 2g).

Figure 2.

Morphology and structural characterization of UPCE. a) The schematic illustration of preparation of PVH scaffold. b,d) The top and c,e) side‐view SEM images of b,c) PVH scaffold and d,e) P(VEC‐FMB‐TMPTMA) UPCE. f) The FTIR spectra of different electrolytes, showing the absence of the characteristic peak at 1638 cm⁻¹ corresponding to the stretching vibrations of C═C, indicates the highly copolymerized of the monomers. g) The XRD patterns of P(VEC‐FMB‐TMPTMA), showing weaker characteristic peaks compared to those of P(VEC‐TMPTMA).

The influence of FMB segment on the ionic conductivity of P(VEC‐FMB‐TMPTMA) was systematically investigated. With no FMB, the achieved P(VEC‐TMPTMA) shows a low ionic conductivity of only 1.47 × 10⁻⁵ S cm⁻¹ at 25 °C (Figures 3a and S6, Supporting Information). A progressive increase in the ionic conductivity is observed with the introduction of FMB and increasing its ratio. A high ionic conductivity of 4.8 × 10⁻⁴ S cm⁻¹ was achieved when with 4% FMB, which is more than 32 times higher than that of P(VEC‐TMPTMA). Further increasing the FMB content to 6%, however, results in a decrease in the ionic conductivity. Thus, P(VEC‐FMB‐TMPTMA) with 4% FMB is selected for further investigation. Differential scanning calorimetry (DSC) results reveal that P(VEC‐FMB‐TMPTMA) shows a lower glass transition temperature (−60.5 °C) compared to P(VEC‐TMPTMA) (−51.4 °C), complementing the XRD results confirming the reduced crystallinity of the polymer matrix due to the incorporation of FMB segment (Figure S7, Supporting Information). A deterioration in the polymerization efficiency was detected when the FMB content exceeds 4% (Figure S8, Supporting Information). The liquid precursor with 8% FMB cannot polymerize effectively even with an extended reaction time. The presence of free FMB molecules has a negative effect on the overall ionic conductivity due to its poor lithium salt dissolvability capability itself (Figure S9, Supporting Information). The temperature‐dependent ionic conductivity of electrolyte shows the calculated Li⁺ diffusion energy barrier for P(VEC‐ FMB‐TMPTMA) is only 0.11 eV based on the Arrhenius formula, in sharp contrast to the much higher value of 0.28 eV for P(VEC‐TMPTMA). This reveals a significantly enhanced Li⁺ diffusion owing to the incorporation of FMB segment (Figure 3b). In addition, a significantly enhanced Li⁺ transfer number of 0.68 is also observed in P(VEC‐FMB‐TMPTMA), in sharp contrast to that of only 0.39 for the P(VEC‐TMPTMA) (Figure 3c,d). The high Li⁺ transfer number reveals the reinforced immobilization of TFSI⁻ anions and facilitated high Li⁺ flux in P(VEC‐FMB‐TMPTMA), which could also help suppress dendritic Li metal growth through the reduction of concentration gradient.^[10b,14] The linear sweep voltammetry (LSV) curve reveals that P(VEC‐FMB‐TMPTMA) exhibits a wide ESW of up to 5.3 V (Figure 3e). The significantly improved oxidation resistance of P(VEC‐FMB‐TMPTMA) ensures its compatibility with high‐voltage cathodes. Moreover, the cyclic voltammetry curves of cell based on P(VEC‐FMB‐TMPTMA) show much better reversibility compared to those of cells with P(VEC‐TMPTMA), revealing the greatly enhanced interfacial compatibility of P(VEC‐FMB‐TMPTMA) with Li metal (Figure S10, Supporting Information).

Figure 3.

The electrochemical performance of UPCE. a) The ionic conductivity of P(VEC‐FMB‐TMPTMA) with different FMB ratios. b) Arrhenius plots showing temperature‐dependent ionic conductivity. Li⁺ transference number measurements of Li symmetric cells under a voltage of 10 mV, and the Nyquist profiles (the inset) of Li symmetric cells before and after polarization with c) P(VEC‐FMB‐TMPTMA) and d) P(VEC‐TMPTMA). e) The LSV curves for assessing the oxidation stability in a Li|UPCE|stainless steel configuration at a scan rate of 1 mV s⁻¹. f) The Li⁷ solid‐state NMR spectra of different electrolytes, showing a noticeable downfield shift of P(VEC‐FMB‐TMPTMA) and a more active Li⁺ coordination environment. g) FTIR spectroscopy shows peaks at 1810 and 1765 cm⁻¹ attributed to free carbonyl and Li⁺‐solvated carbonyl of cyclic polycarbonates. The binding energies of h) VEC and i) FMB segments to TFSI⁻ anions.

To gain a deeper understanding for the enhanced ionic conductivity of P(VEC‐FMB‐TMPTMA), solid‐state NMR spectra and molecular dynamics (MD) theoretical simulations have been conducted. Compared to P(VEC‐TMPTMA), a noticeable downfield shift of P(VEC‐FMB‐TMPTMA) can be observed in the ⁷Li solid‐state NMR spectra (Figure 3f), indicating a more active Li⁺ coordination environment due to the weakened Li⁺‐polymer interaction. Hence, fast Li⁺ diffusion in P(VEC‐FMB‐TMPTMA) are facilitated. The MD simulation results show that Li⁺ can be solvated well in P(VEC‐FMB‐TMPTMA) (Figure S11, Supporting Information). The calculated lithium ion diffusion coefficient of P(VEC‐FMB‐TMPTMA) is 9 × 10⁻⁹ cm² s⁻¹, much higher than of 1.6 × 10⁻¹⁰ cm² s⁻¹ for P(VEC‐TMPTMA), revealing the enhanced lithium ion transfer kinetics for the P(VEC‐FMB‐TMPTMA). Moreover, for the FTIR spectroscopy, the peak at 1810 and 1765 cm⁻¹ can be attributed to free carbonyl and Li⁺‐solvated carbonyl of cyclic polycarbonates, respectively (Figure 3g).^[7a] And the small peak at 1719 cm⁻¹ can be attributed to the weak Li⁺ coordination with the carbonyl of TMPTMA. With the introduction of FMB segment, the peak at 1719 cm⁻¹ becomes much stronger, which suggests that the Li⁺ coordination environment was altered and the overall solvation ability of the polymer chain is greatly reduced owing to the introduction of FMB segment. The binding energies of P(VEC‐FMB‐TMPTMA) and P(VEC‐TMPTMA) to TFSI⁻ anions have also been calculated. As shown in Figure 3h,i, the binding energy of P(VEC‐FMB‐TMPTMA) with TFSI⁻ anion is ‐19.88 kca mol⁻¹, much lower than that of P(VEC‐TMPTMA) with the TFSI⁻ anion (‐16.79 kca/mol), revealing that P(VEC‐FMB‐TMPTMA) can facilitate the effect anchoring of TFSI⁻ anions and thereby afford a high Li⁺ flux.

2.2. Li metal Interfacial Compatibility Investigation of P(VEC‐FMB‐TMPTMA) UPCE

To reveal the interfacial compatibility of P(VEC‐FMB‐TMPTMA) UPCE with Li metal, Tafel plots of Li symmetric cells have been first investigated. A significantly enhanced exchange current density of 3.16 × 10⁻² mA cm⁻² was detected on Li symmetric cells with P(VEC‐FMB‐TMPTMA), which is ≈11 times higher than that with P(VEC‐TMPTMA) (Figure S12, Supporting Information), revealing an enhanced lithium ion interfacial transport kinetics at Li metal/P(VEC‐FMB‐TMPTMA) interface. In addition, to demonstrate the viability of P(VEC‐FMB‐TMPTMA), Li symmetric cells were assembled and tested at current densities ranging from 0.5–2 mA cm⁻². Figure 4a shows the galvanostatic cycling performance of Li|P(VEC‐FMB‐TMPTMA)|Li, exhibiting an ultralong and stable cycling durability over 6000 h with a small voltage polarization of only ≈62 mV at 0.5 mA cm⁻² with a Li plating/stripping capacity of 0.1 mAh cm⁻². By comparison, the cell with P(VEC‐TMPTMA) deteriorates quickly only after 200 h with a much larger polarization voltage of ≈168 mV, primarily due to the serious interfacial degradation and inferior Li dendrite suppression. The much lower polarization voltage of Li symmetric cells can also reveal the greatly enhanced interfacial charge transfer kinetics between Li metal and P(VEC‐FMB‐TMPTMA) UPCE. The evolution of electrochemical impedance spectroscopy (EIS) reveals an extremely stable interfacial resistance upon cycling for the cell featuring P(VEC‐FMB‐TMPTMA), indicating a significantly stabilized electrolyte/electrode interphase (Figure 4b). In contrast, an evident increase in interfacial resistance of cells with P(VEC‐TMPTMA) was detected during the cycling (Figure 4c). Impressively, the Li symmetric cell with P(VEC‐FMB‐TMPTMA) maintains an ultra‐long lifespan stable operation over 6000 cycles (1200 h) even at a relatively high current density of 1 mA cm⁻² (Figure S13, Supporting Information). An outstanding cycling stability over 1500 h at 0.5 mA cm⁻², 0.5 mAh cm⁻² (Figure S14, Supporting Information) and 400 h at 1 mA cm⁻², 1 mAh cm⁻² can also be attained (Figure 4d). In addition, the Li symmetric cell with P(VEC‐FMB‐TMPTMA) still demonstrates superior cycling over 200 h even at a high current density of 2 mA cm⁻² (Figure S15, Supporting Information). CCD is a crucial parameter assessing the Li dendrite suppression capability of SEs. Rapid short‐circuiting was identified for the cell with the P(VEC‐TMPTMA) electrolyte, characterized by a low CCD value of only 1.2 mA cm⁻² (Figure S16, Supporting Information). Remarkably, P(VEC‐FMB‐TMPTMA) UPCE exhibits an exceptionally CCD as high as 11.5 mA cm⁻², which is ≈10 times greater than that of P(VEC‐TMPTMA) (Figure 4e; Figure S17, Supporting Information). The CCD comparison with recently reported works reveals that the P(VEC‐FMB‐TMPTMA) demonstrates a significantly higher CCD compared to other reported SEs, highlighting its considerable potential for practical applications at high current densities and large areal capacities (Figure 4f).^[3d,15] The excellent Li dendrite retardant capability of UPCE can be attributed to the in situ constructed LiF‐rich SEI on Li metal surface. The lithiophobic LiF could effectively facilitate the lateral Li deposition and suppressed the dendritic Li metal growth.^[10b,10c] Moreover, a comparison of cycling performance of Li symmetric cells also reveals that the cells with P(VEC‐FMB‐TMPTMA) exhibit an extraordinarily ultralong‐term cycling stability, far surpassing other reported works.^{[10b,11a,15g,16]} (Figure 4g; Table S1, Supporting Information)

Figure 4.

The electrochemical cycling performance of UPCE. The galvanostatic cycling performance of solid‐state Li symmetric cells at a) 0.5 mA cm⁻² with a capacity of 0.1 mAh cm⁻². The electrochemical impedance spectroscopy evolution of Li symmetric cells coupled with b) P(VEC‐FMB‐TMPTMA) and c) P(VEC‐TMPTMA) at 0.5 mA cm⁻² with a capacity of 0.1 mAh cm⁻². d) The cycling performance of Li symmetric cells at 1 mA cm⁻² with a capacity of 1 mAh cm⁻². e) The critical current density of P(VEC‐FMB‐TMPTMA). f) The critical current density comparison with reported works. g) Comparison of the cycling performance of Li symmetric cells with previously reported works.

2.3. Performance of Half Cells with P(VEC‐FMB‐TMPTMA) UPCE

The feasibility of P(VEC‐FMB‐TMPTMA) electrolyte was further tested in SSLMBs coupled with different cathodes. The solid‐state Li|LFP cell demonstrates excellent rate performance with high specific capacities of 154.8, 153.6, 150.1, 144.1 and 130.7 mAh g⁻¹ at 0.2, 0.5, 1, 2, and 5 C (Figure 5a,b). A high specific capacity of 154.6 mAh g⁻¹ still can be attained when the current density returns to 0.2 C. In addition, the Li|LFP cell also exhibits an ultra‐long lifespan cycling durability over 1800 cycles at 1 C, manifesting an ultralow decay rate of 0.01% per cycle and an extraordinarily high average Coulombic efficiency of 99.93% (Figure 5c; S18, Supporting Information). The excellent cycling stability can be attributed to the ultra‐stable electrode/electrolyte interphase as revealed by the corresponding EIS curves (Figure S19, Supporting Information). As a contrast, the Li|LFP cell with liquid electrolyte delivers short cycle life due to serious interfacial side reactions and notorious Li dendrite growth. Additionally, the P(VEC‐FMB‐TMPTMA) UPCE also demonstrates excellent environmental adaptability under a wide temperature range of −10–80 °C. The Li|P(VEC‐FMB‐TMPTMA)|LFP cell can stably cycle over 300 cycles with almost no capacity degradation at a low temperature of −10 °C (Figure 5d). And a superior cycling stability over 500 cycles can also be achieved even at a high temperature of 80 °C (Figure S20, Supporting Information). The wide ESW of P(VEC‐FMB‐TMPTMA) ensures its superior interfacial compatibility with high‐voltage cathodes. The solid‐state Li|P(VEC‐FMB‐TMPTMA)|LCO cell tested at an aggressive cutoff voltage of 4.5 V delivers a highly reversible specific capacity of 172.5 and 166.1 mAh g⁻¹ at 0.2 and 0.5 C. Impressively, superior capacities of 152.3, 142.1 and 118.4 mAh g⁻¹ still can be achieved at 1, 2 and 5 C, respectively (Figure S21, Supporting Information). During long‐term cycling at 1 C, the Li|P(VEC‐FMB‐TMPTMA)|LCO cell achieves exceptional ultralong lifespan cycling stability over 1500 cycles with a high capacity retention of 91.6% (1300^th) and 84.8% (1500^th) (Figure 5e). Remarkably, the solid‐state Li|P(VEC‐FMB‐TMPTMA)|LCO cell in our work outperforms those previously reported works (Table S2, Supporting Information). The excellent long‐term cycling performance of Li|P(VEC‐FMB‐TMPTMA)|LCO can be attributed to the wide electrochemical stability window of P(VEC‐FMB‐TMPTMA) and its high compatibility with high‐voltage LCO cathode. Moreover, the Li|P(VEC‐FMB‐TMPTMA)|LCO pouch cell with a high areal‐capacity of ≈1.71 mAh cm⁻² still demonstrates excellent cycling stability over 120 cycles (Figure S22, Supporting Information). Furthermore, the Li|LCO pouch cell with ultra‐high areal capacity (∼4.5 mAh cm⁻²) cathode and ultrathin Li metal (20 µm) still demonstrates excellent cycling, showing a remarkably high energy density of 495 Wh kg⁻¹ at 25 °C, even at the harsh cycling condition of an ultra‐low negative‐to‐positive‐capacity (N/P) ratio of 0.89 (Figure 5f; Figure S23 and Table S3, Supporting Information). The room‐temperature energy density of our work surpasses those utilizing LFP, LCO or LiNi_xCo_yMn_zO₂ cathodes (Figure 5g; Table S4, Supporting Information).^{[3a,3b,16h,17]} Remarkably, the 6.3 Ah Li|LiNi_0.8Co_0.1Mn_0.1O₂ (NCM811) pouch cell can still operate effectively at 25 °C even with a low‐porosity Celgard separator, demonstrating a high energy density of 338 Wh kg⁻¹ based on the total mass of the cell including the package (Figure S24 and Table S5, Supporting Information). It is worth mentioning that the energy density of the whole cell can be further enhanced by utilizing PVH host and lean electrolyte. Additionally, the pouch cell still can operate well under harsh cycling conditions such as folding, piercing, and even multiple cutting, ensuring exceptional safety in various practical conditions (Figure S25, Supporting Information).

Figure 5.

The cycling performance of full cells with UPCE. a) The voltage‐capacity profile and b) the rate performance of solid‐state Li|LFP cells. c) The long‐term cycling performance of solid‐state Li|LFP cell at 1 C. d) The cycling performance of solid‐state Li|LFP cell at −10 °C. e) The long‐term cycling performance of solid‐state Li|LCO cell (1 C = 180 mAh g⁻¹) and f) the cycling performance of 4.5 V solid‐state Li|LCO pouch cell with a high areal capacity (≈4.5 mAh cm⁻²) at 0.2 C. g) The energy density comparison of SSLMBs with reported works (the energy density is obtained at RT unless otherwise specified).

2.4. Electrode/Electrolyte Interface Investigation with P(VEC‐FMB‐TMPTMA) UPCE

To reveal the excellent electrochemcial performance of P(VEC‐FMB‐TMPTMA), the Li deposition morphology was first investigated (Figure 6a). Serious Li dendrite growth is detected on the cycled Li metal surface when using P(VEC‐TMPTMA) after 50 cycles at a current density of 0.5 mA cm⁻² with a capacity of 0.1 mAh cm⁻². By comparison, the Li deposition exhibits a dense and homgenesous morphology when utilizing the P(VEC‐FMB‐TMPTMA) electrolyte, revealing the uniform Li⁺ flux and thus effectively regulated Li deposition behavior. In addition, the HOMO and LUMO energy levels of the P(VEC‐FMB‐TMPTMA) were also calculated to elucidate the interfacial chemistry on Li metal side. The results clearly show that the incorporation of FMB segment reduces the LUMO energy level of the electrolyte (Figure 6b), and FMB segment will be preferentially reduced by Li metal, facilitating the formation a LiF‐enriched SEI layer on Li metal surface. To further validate our hypothesis, the SEI layer composition was also examined by the XPS spectra. The SEI layer presents fewer C signals and more F signals, revealing organic‐less and LiF‐enriched SEI layer formation on Li metal surface when employing the P(VEC‐FMB‐TMPTMA) UPCE (Figures S26,S27, Supporting Information). For the F 1s spectra, a much higher content of LiF can be clearly detected compared with the counterpart with the P(VEC‐TMPTMA) (Figure 6c). Additionally, the notable much lower CO₃ ²⁻ content detected in the SEI layer when using P(VEC‐FMB‐TMPTMA) can further evidence the significantly reduced side reactions of electrolytes with Li metal (Figure S28, Supporting Information). The Ar⁺ depth profiling result shows a high proportion of LiF distribution throughout the entire SEI layer (Figure 6d). In contrast, a relatively high content of CO₃ ²⁻ can also be observed in the inner SEI layer when employing P(VEC‐TMPTMA), revealing its serious side reactions with Li metal. The significantly increased LiF content in the SEI layer with P(VEC‐FMB‐TMPTMA) could be attributed to the preferential decomposition of fluorinated polymer segments.^{[ 18 ]} TEM characterization of the SEI layer on Li metal surface has also been conducted. As demonstrated, typical crystal plane of Li metal can be detected, showing the interplanar crystal spacing of 2.4 Å designated to the (110) crystal plane of Li metal. Moreover, obvious interplanar crystal spacing of 2.0 Å was also detected corresponding to the (002) crystal plane of LiF. Furthermore, electron energy‐loss spectroscopy (EELS) was also performed on the regions marked by a dotted rectangle in the high‐angle annular dark‐field scanning transmission electron microscopy (HAADF‐STEM) image of deposited Li metal (Figure S30a, Supporting Information). The spectrum results in Figure S30b (Supporting Information) show obvious signals of LiF, which can firmly verify the formation of LiF‐enriched SEI layer on Li metal surface. Time‐of‐flight secondary ion mass spectrometry (TOF‐SIMS) results further elucidate the depth‐resolved composition and structure of the SEI layer. The LiF₂ ⁻ and LiO⁻ ion fragments can be ascribed to LiF and Li₂O, respectively. And the C₂HO⁻ ion fragment represents the organic species resulting from the polymer segments decomposition (Figure S31, Supporting Information). The TOF‐SIMS results can further verify the LiF‐enriched SEI formation when employing P(VEC‐FMB‐TMPTMA) (Figure 6e). Both the experimental and theoretical calculation results reveal the excellent interfacial compatibility of P(VEC‐FMB‐TMPTMA) with Li metal, which can guarantee the extraordinary long lifespan cycling stability of solid‐state Li metal batteries. Moreover, to reveal the excellent long‐term cycling performance of P(VEC‐FMB‐TMPTMA) coupled with a high‐voltage cathode, the cathode electrolyte interface (CEI) layer on cycled LCO surface was also analyzed by the transmission electron microscopy (TEM) technique. A thick and inhomogeneous CEI layer was detected on cycled LCO surface, when using P(VEC‐TMPTMA) (Figure 6f), revealing its poor electrochemical stability and the serious side reactions with LCO cathode. In comparison, a thin (≈4.3 nm) and uniform amorphous CEI layer on cycled LCO surface was observed when using the P(VEC‐FMB‐TMPTMA) UPCE (Figure 6g). The thin and stable CEI layer is beneficial to suppressing the further polymer degradation. And the high‐resolution XPS spectra of cycled LCO cathode can further evidence the formation of a F‐rich CEI layer on cycled LCO cathode surface when employing the P(VEC‐FMB‐TMPTMA) UPCE (Figure S32, Supporting Information), which can ensure the excellent ultralong‐term cycling stability of high‐voltage Li|LCO cells.^{[ 19 ]}

Figure 6.

Electrode/electrolyte interface investigation. a) The SEM images of different cycled Li metal after 50 cycles at 0.5 mA cm⁻², 0.1 mAh cm⁻². b) The HOMO and LUMO energy levels and the corresponding electron density distribution. c) The F 1s XPS spectra and d) the XPS depth profiling analysis of cycled Li metal after 50 cycles at 0.5 mA cm⁻², 0.1 mAh cm⁻² and the relative proportions of the components of the SEI layer. e) The schematic illustrations of cycled Li metal anodes, showing the cycled Li metal with P(VEC‐FMB‐TMPTMA) exhibits dense and uniform Li deposition in contrast to the serious Li dendrite growth when using P(VEC‐TMPTMA). TEM images of cycled LCO cathodes after 100 cycles at 1 C with f) P(VEC‐TMPTMA) and g) P(VEC‐FMB‐TMPTMA).

3. Conclusion

We have designed fast ion transport and scalable UPCE, using the electrolyte structural engineering and phase separation‐derived PVH scaffold. The designed P(VEC‐FMB‐TMPTMA) facilitates fast Li⁺ transport due to the weakened Li⁺‐polymer interaction and ensures a LiF‐rich SEI to stabilize Li metal. As a result, the solid‐state Li symmetric cells demonstrate high‐rate performance and record‐breaking ultralong lifespan cycling stability (6000 h). The solid‐state Li|LCO cell with a high cutoff voltage of 4.5 V also exhibits an ultralong‐term cycling durability over 1500 cycles. Moreover, the Li|LCO pouch cell with a high‐areal‐capacity (4.5 mAh cm⁻²) and ultrathin Li metal (20 µm) still shows excellent cycling even at the harsh cycling condition of an ultralow N/P ratio of 0.89, manifesting an exceptional high energy density of 495 Wh kg⁻¹. This work provides a highly efficient and large‐scalable strategy for fabricating high performance SSLMBs, which could expedite their practical usage and commercialization.

4. Experimental Section

Preparation of PVH and UPCE

PVH was dissolved into N, N‐dimethylformamide with a concentration of 1 g ml⁻¹ and stirred vigorously at 25 °C to get a homogeneous slurry. Then, the slurry is evenly coated on the glass substrate with a doctor blade. The porous membrane was achieved by promptly dipping the glass substate into water as the non‐solvent. The density of the PVH host is ≈0.95 mg cm⁻² with a porosity of ≈70.9%. The porous PVH membrane was carefully lifted using a PTFE substrate and then washed thoroughly with anhydrous ethanol and dried overnight at 100 °C. Bis(trifluoromethane) sulfonamide lithium (LiTFSI, 99.9%), vinylethylene carbonate (VEC, 99%), FMB, trimethylolpropane trimethacrylate (TMPTMA) were purchased from Sigma‐Aldrich. The preparation of SPE was performed in the Ar‐filled glovebox with the concentration of oxygen and water both below 1 ppm. Liquid precursors of VEC, FMB, and TMPTMA with a molar ratio of 100: 4: 1 were mixed together to get a homogeneous solution under vigorous stirring. 2, 2′‐Azobis(2‐methylpropionitrile) (AIBN) was added as the initiator. Then a certain amount of LiTFSI was added with a mass ratio of 2:1 (VEC: LiTFSI) to form a uniform solution. The solid‐state electrolyte was synthesized by the in situ copolymerization of the liquid precursors at 80 °C for 8 h. For the preparation of P(VEC‐FMB‐TMPTMA), the PVH host with varied thickness was immersed into the liquid precursor on the PTFE membrane. After being fully filled, the excess liquid solution was scraped off. The PVH porous membrane, saturated with the precursor solution, is transferred into the battery. After polymerization at 80 °C, P(VEC‐FMB‐TMPTMA) with different thicknesses can be obtained.

Characterizations and Electrochemical Measurement

The morphology of the sample was observed by the SEM (FEI Quanta 450). TEM characterization was performed on JEOL JEM‐F200. For the TEM analysis of the SEI layer on Li metal surface, the samples were transferred to TEM column using vacuum‐transfer method without exposure to air. FTIR spectra were collected on a Bruker Vertex 70 FTIR spectrometer. Differential scanning calorimetry (DSC) test was performed in the temperature range of −150 to 30 °C at a heating rate of 10 °C min⁻¹. The ionic conductivity was tested by the AC impedance spectroscopy via the solid electrolyte sandwiched by stainless steel symmetric cells and tested within the frequency of ≈0.1‐1 M Hz via the Solartron potentiostat. The XPS analysis was conducted using the Shimadzu (Kratos) AXIS SUPRA+ with an Al Kα (λ = 0.83 nm, hυ = 1486.7 eV). The C 1s peak at 284.8 eV was used as a reference. The Li⁺ transfer number was determined using Bruce‐Vincent method by the chronoamperometry and AC impedance spectra via a Li symmetric cell with a constant voltage of 10 mV according to the equation:

$$t_{+}=\frac{I_{ss}\left(\mathrm{\Delta }V-I_0{R}_0\right)}{I_0\left(\mathrm{\Delta }V-I_{ss}{R}_{ss}\right)}$$ (1)

where I₀ and I_ss are denoted to the initial and steady state currents, ΔV is the applied voltage. R ₀ and R_ss are the initial and steady state interphase resistance (R_SEI ).

For the LiFePO₄ cathode, LiFePO₄ (MTI Corporation), super P (MTI Corporation), PVDF (MTI Corporation) were mixed at the ratio of 8:1:1 in N‐methyl‐2‐pyrrolidone (NMP) under vigorous stirring to achieve a uniform slurry. For the high voltage LCO cathode, LCO (Guangdong Canrd New Energy Technology Co., Ltd.), super P (MTI Corporation) and PVDF (MTI Corporation) was mixed in the NMP solution with a weight ratio of 8:1:1. The slurry was homogeneously casted on the Al foil with a doctor‐blade and then dried at 80 °C overnight. For the high‐loading LCO cathode preparation, LCO, super P, PVDF was mixed in the NMP solution with a weight ratio of 92: 4: 4. The electrochemical performance of cell was evaluated at the LAND batteries test systems.

Simulation

All structures in this work were optimized using the density functional theory (DFT) method with the B3LYP‐D3BJ^{[ 20 ]} functional via the software package of Gaussian 16 C.01.^{[ 21 ]} During the structure optimization, the valence electrons of H, O, C, F, N, and S were described by the 6–31G(d) basis set for neutral complexes and cations and the 6–31+G(d) basis sets for anions. The valence electrons of Li were described by def2‐SVP basis set. Frequency calculations were performed to confirm that the optimized structures were local minimum at the potential energy surface. To ensure the accuracy, the single point energy (SPE) calculations were performed using the def2‐TZVP basis set for neutral structures and cations, while the def2‐TZVPD basis set was specifically employed for anions. The binding energy (E_b ) was calculated with the following equation:

$$E_b=E_{mol+A}-E_{mol}-E_{A^{\prime }}$$ (2)

where E _{mol + A} , E_mol and E_A are the SPE of VEC/FMB/VEC‐FMB molecule‐A system, VEC/FMB/VEC‐FMB molecule and A ion, respectively.

All molecular dynamics (MD) simulations were performed using the OPLS‐AA force field.^{[ 22 ]} The LIGPARGEN was employed to obtain the OPLS‐AA force field topology.^{[ 23 ]} The box size of 3.0 × 3.0 × 3.0 nm³ were used in all simulation models. The cut‐off distance of 1.2 nm was used for Lennard‐Jones potential. The Coulombic potential was measured using Particle Mesh Ewald (PME) with a cut‐off distance of 1.2 nm and Fourier grid spacing of 0.12. All bonds were constrained with LINCS algorithm. Periodic boundary conditions were applied in all directions. The MD simulations were started by running initial energy minimization, followed by 5000 ps of NVT simulation and 5000 ps of NPT simulation with an integration time step of 0.001 ps. All the simulations systems were finally maintained at 298 K using the Nose‐Hoover thermostat for 45 ns to collect simulation data. A time constant of 1 ps was applied for the temperature coupling. The investigations of mean‐square displacements and diffusion coefficients were analyzed for the last 25 ns of simulation time.

Conflict of Interest

The authors declare no conflict of interest.

Supporting information

Supporting Information

Xia S., Zhang X., Jiang Z., et al. “Ultrathin Polymer Electrolyte With Fast Ion Transport and Stable Interface for Practical Solid‐state Lithium Metal Batteries.” Adv. Mater. 37, no. 38 (2025): 37, 2510376. 10.1002/adma.202510376

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.