Amide Monomer‐Mediated Solubilization Strategy Enables Nonflammable Deep Eutectic Gel Polymer Electrolytes for High‐Temperature‐Stable Lithium Metal Batteries

ABSTRACT

Deep eutectic gel polymer electrolytes (DEGPEs), combining intrinsic non‐flammability with outstanding thermal stability, are attractive candidates for next‐generation lithium metal batteries (LMBs). However, their practical deployment in high‐energy‐density LMBs has been fundamentally constrained by poor interfacial stability with the lithium metal anode and limited tolerance toward the high‐voltage cathode. We report a new amide‐monomer‐mediated DEGPE that achieves comprehensive performance via a LiNO₃ solubilization strategy. The N‐methylacrylamide (NME) units in the poly(N‐methylacrylamide) (PNME) framework enhance LiNO₃ solubility through hydrogen bonding and Li⁺ coordination, forming a stable inorganic‐rich interphase. Concurrently, it immobilizes free N‐methyltrifluoroacetamide (NMTFA) via hydrogen bonds, suppressing transition‐metal dissolution and preventing electrolyte leakage. The amide monomer‐mediated DEGPE‐based NCM811||Li cells achieve 80.1% capacity retention after 500 cycles with an average Coulombic efficiency of 99.67%, a performance that surpasses state‐of‐the‐art (deep eutectic electrolyte) DEE‐based systems. More impressively, LCO||Li cells retain 89.6% capacity after 300 cycles even at an elevated temperature of 80°C, far exceeding the thermal stability limits of conventional electrolytes and underscoring its remarkable interfacial stability under extreme operational conditions. This work establishes a molecularly engineered solvation and interfacial regulation strategy for DEGPEs, providing both fundamental insight and a practical pathway toward safe, high‐energy, and high‐temperature‐tolerant LMBs.

The PNME framework promotes LiNO₃ dissolution via dual Li⁺/NO₃ ⁻ coordination, immobilizes NMTFA to enable high‐voltage NCM811 compatibility, and yields a high‐voltage‐tolerant, flame‐retardant PNME‐1.5 electrolyte with excellent interfacial stability.

1. Introduction

Lithium metal batteries (LMBs), owing to their exceptionally high theoretical energy density, are widely regarded as a leading candidate for next‐generation rechargeable energy‐storage systems and have attracted intense interest from both academia and industry [1, 2, 3, 4, 5]. However, their practical deployment remains severely constrained by safety hazards, most notably fire and explosion, arising from the use of conventional carbonate‐ or ether‐based electrolytes [6, 7, 8, 9]. Developing electrolyte chemistries that can reconcile high safety with high electrochemical performance, therefore, represents a central challenge for LMB technologies. Among emerging alternatives, deep eutectic electrolytes (DEEs) have gained increasing attention because of their intrinsic non‐flammability and outstanding thermal stability. When combined with polymer solid‐state electrolytes to form deep eutectic gel polymer electrolytes (DEGPEs), these systems promise a good combination of high ionic conductivity, favorable electrode wettability, and enhanced safety [10, 11].

Despite these advantages, the application of DEGPE in high‐energy‐density LMBs has been fundamentally limited by poor interfacial compatibility with lithium metal anodes and insufficient oxidative stability against high‐voltage cathodes. These shortcomings trigger severe parasitic reactions, rapid capacity decay, and premature cell failure, preventing DEGPEs from meeting the stringent demands of advanced LMB chemistries [12, 13, 14, 15, 16]. Although NMTFA （N‐methyltrifluoroacetamide） in DEEs can act as fluorinated hydrogen bond donors (HBDs) and help stabilize electrodes via LiF‐rich interfaces [17, 18, 19, 20], their strong electron‐withdrawing character weakens coordination with lithium salts, leading to reduced salt solubility and compromised ionic conductivity [21, 22]. Moreover, continued interfacial decomposition of fluorinated HBDs ultimately undermines long‐term stability [23, 24]. Meanwhile, lithium nitrate (LiNO₃) is well known for its dual interfacial functionality: its reductive decomposition yields Li⁺‐conductive, dendrite‐suppressing SEI layers at lithium metal anodes [25], while its oxidative participation stabilizes cathode‐electrolyte interphases, mitigating capacity fading and improving rate performance [26]. Unfortunately, LiNO₃ exhibits extremely poor solubility in fluorinated DEE due to strong Li⁺‐NO₃ ⁻ interactions and weakened coordination of fluorinated HBDs [22]. Consequently, designing DEGPEs that simultaneously stabilize lithium metal, tolerate high‐voltage cathodes, and maintain high ionic conductivity without sacrificing safety remains a long‐standing and unresolved challenge.

Here, we report a class of DEGPE that overcomes these intrinsic limitations through a molecularly engineered solvation and interfacial regulation strategy. Central to this design is an amide‐monomer‐mediated LiNO₃ solubilization mechanism, in which cooperative hydrogen bonding between nitrate anions and N‐methylacrylamide (NME) units within the poly(N‐methylacrylamide) (PNME) framework, together with strong carbonyl coordination to Li⁺, markedly enhances LiNO₃ solubility in deep eutectic gel solid‐state electrolytes [27, 28]. This enables the in situ construction of a uniform, inorganic‐rich interphase that stabilizes lithium deposition and suppresses parasitic interfacial reactions. Concurrently, hydrogen‐bond interactions between the NME units and NMTFA molecules immobilize NMTFA within the polymer matrix [29], thereby suppressing strong C = O‐driven coordination with transition metal ions and effectively mitigating transition‐metal (TM) dissolution. These synergistic mechanisms endow the DEGPE with simultaneous high‐voltage stability, lithium‐metal compatibility, high ionic conductivity, and intrinsic flame retardancy, a combination that has remained a central challenge for DEE systems.

PNME demonstrates unique comprehensive advantages in constructing high‐performance gel polymer electrolytes. Compared to other amide‐based polymers (e.g., PDMAA), which lack an N‐H group, PNME's abundant N‐H groups can directly anchor LiNO₃ through hydrogen bonding, promoting its controlled decomposition at the electrode interface to form a uniform inorganic‐rich interphase, thereby avoiding side reactions in the bulk electrolyte [30]. Relative to polyether‐based materials such as PEO, the amide structure of PNME confers higher oxidation stability, enabling compatibility with high‐voltage cathodes, while its intrinsic non‐flammability significantly improves battery safety [31, 32]. In contrast to polyesters (e.g., PCL), PNME exhibits stronger intrinsic compatibility with LiNO₃, enabling effective anchoring and regulation of its decomposition, whereas polyesters themselves hardly dissolve or stabilize LiNO₃ [33].

This work innovatively polymerizes the NME to construct the polymer framework PNME. This framework not only provides mechanical support but also features NME units that act as “multifunctional molecular anchors.” These anchors dissolve LiNO₃ via hydrogen bonding and simultaneously reconstruct the Li⁺ solvation sheath and immobilize free NMTFA molecules through coordination and hydrogen bonding. This represents a shift from a “passive reliance on solvent formulation” to a paradigm of “active regulation of solution chemistry and interfacial processes by the polymer framework.”

The amide monomer‐mediated DEGPE exhibits a wide electrochemical stability window exceeding 4.8 V and a high ionic conductivity of 1.28 × 10⁻³ S cm⁻¹. Combustion tests confirm its non‐flammability, with the electrolyte retaining its structural integrity even after direct flame exposure. The performance metrics achieved under demanding conditions underscore the electrolyte's robustness. While a 500‐cycle retention of 80.1% for Li||LiNi_0.8Co_0.1Mn_0.1O₂ (NCM811) is notable, the true test lies in the high‐temperature performance. It is remarkable that Li||LiCoO₂ (LCO) cells operated at 2C retain 89.6% capacity after 300 cycles at 80°C, a condition that typically accelerates electrolyte decomposition and interfacial degradation. This result demonstrates exceptional thermal and interfacial stability rarely achieved in prior DEE systems.

2. Results and Discussion

2.1. The Reaction Mechanism and Characteristics of PNME‐1.5

LiNO₃ exhibits intrinsically poor solubility in the fluorinated DEE (F‐DEE) composed of LiTFSI and NMTFA, while lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) readily forms a homogeneous eutectic phase. This contrast arises from the electron‐deficient carbonyl carbon in NMTFA, induced by the strong electron‐withdrawing effect of the C‐F bonds, together with the high polarity of the NO₃ ⁻ anion, which binds strongly to Li⁺ and suppresses salt dissociation. By contrast, the lower polarity and charge delocalization of TFSI⁻ facilitate Li⁺ dissociation and its coordination with the carbonyl oxygen of the NMTFA. In addition, N─H···O hydrogen bonding between NMTFA and LiTFSI promotes their spontaneous liquefaction into a stable F‐DEE (Figure 1a) [34].

FIGURE 1.

Mechanistic illustration of (a) the interface‐unstable F‐DEE composed of NMTFA and LiTFSI is unsuitable for high‐voltage cathode materials. Versus (b) interface‐stable, high‐voltage‐tolerant, and high‐ion‐conductivity DEGPE.

Despite this favorable eutectic formation, the resulting F‐DEE exhibits severe interfacial instability in high‐energy lithium metal batteries. When paired with high‐voltage cathodes such as NCM811, the unstable cathode‐electrolyte interphase (CEI) fails to suppress TM dissolution. The dissolved high‐valence TM ions (for example, Ni²⁺ and Mn⁴⁺) subsequently catalyze electrolyte decomposition and parasitic reactions, leading to the formation of a non‐uniform, organic‐rich, and mechanically fragile CEI. In parallel, the F‐DEE is intrinsically incompatible with the lithium metal anode and undergoes secondary reactions with metallic lithium, producing a porous and heterogeneous solid–electrolyte interphase (SEI). Migration of dissolved TM ions to the anode further disrupts the SEI, accelerating lithium dendrite growth. These coupled interfacial degradation processes drive rapid capacity fading and premature cell failure.

This work demonstrates that NME units within the PNME framework actively promote LiNO₃ dissolution through a dual coordination mechanism, simultaneously chelating Li⁺ via the highly polar carbonyl (C═O) groups and forming N─H···O hydrogen bonds with NO₃ ⁻ anions [27]. Upon electrochemical cycling, the decomposition products of LiNO₃, including Li_xN and LiN_xO_y, contribute to the formation of dense, inorganic‐rich solid–electrolyte and cathode–electrolyte interphases (SEI and CEI) at the electrode interfaces [35]. These robust interphases effectively suppress lithium dendrite growth and minimize interfacial side reactions. Meanwhile, the PNME framework immobilizes the fluorinated amide NMTFA through an extended hydrogen‐bonding network, thereby inhibiting its coordination with dissolved TM ions and ensuring compatibility with high‐voltage NCM811 cathodes. Coupled with the intrinsic high‐temperature stability and non‐flammability of NMTFA, this integrated molecular design enables the fabrication of a high‐voltage‐tolerant, flame‐retardant, and high‐performance DEGPE (PNME‐1.5) with exceptional interfacial stability, as illustrated in Figure 1b. For synthesis details, refer to the “Electrolyte preparation” section in the Supporting Information.

To elucidate the mechanism underlying LiNO₃ dissolution and the formation of stable SEI and CEI layers, density functional theory (DFT) calculations were performed. Analyses of the electrostatic potential (ESP) distributions and frontier orbital energies provide key mechanistic insights. As shown in Figure 2a, the higher electron density localized around the carbonyl (C═O) group of NME results in a stronger interaction with Li⁺ (binding energy, BE = −2.51 eV) compared with NMTFA (BE = −2.08 eV), rationalizing the enhanced LiNO₃ solubilization within the PNME framework [27]. In addition, the highest occupied molecular orbital (HOMO) energy of the PNME–NMTFA complex (−6.59 eV) is markedly higher than that of pristine NMTFA (−8.07 eV), indicating an increased oxidation propensity. This electronic modulation enables PNME‐NMTFA to undergo preferential oxidative decomposition at the cathode surface, promoting the formation of an inorganic, LiF‐rich CEI and thereby enhancing interfacial stability [36]. Conversely, the low‐lying lowest unoccupied molecular orbital (LUMO) of LiNO₃ favors its preferential reduction at the lithium metal anode, leading to the formation of Li_xN‐rich SEI layers [37, 38]. Meanwhile, the higher HOMO energy level of LiNO₃ (−7.70 eV) compared to LiTFSI (‐8.67 eV) leads to its prior oxidative decomposition, thereby promoting a stable CEI containing Li_xN and LiN_xO_y. Such SEI chemistry facilitates rapid Li⁺ transport while effectively suppressing lithium dendrite growth (Figure 2b) [25, 26].

FIGURE 2.

(a) ESP and binding energy with Li⁺ for NME and NMTFA. (b) The HOMO‐LUMO values of the molecules in DEGPE. (c) ¹H NMR analysis of NME and NME‐LiNO₃. (d) ¹H NMR spectra of F‐DEE, PNME‐0 and PNME‐1.5. (e) FT‐IR spectroscopy of F‐DEE, PNME‐0 and PNME‐1.5. (f) Raman spectroscopy of NME, F‐DEE, PNME‐0 and PNME‐1.5. (g) ⁷Li NMR spectroscopy of F‐DEE, PNME‐0 and PNME‐1.5. (h) Raman spectroscopy fitting for free and coordinated TFSI⁻. (i, j) MD simulations and RDF analysis for NMTFA, TFSI⁻, NO₃ ⁻ and NME in F‐DEE, and PNME‐1.5. VT‐IR spectral analysis of F‐DEE (k, l) and PNME‐1.5 (m, n).

The formation mechanism of the PNME‐1.5 electrolyte was further elucidated using nuclear magnetic resonance (NMR), Fourier‐transform infrared (FT‐IR), and Raman spectroscopy, providing complementary insights into polymer‐mediated Li⁺ solvation and hydrogen‐bond regulation. The ¹H NMR spectra of NME exhibited a pronounced upfield shift in the amide proton signals upon addition of LiNO₃ (Figure 2c) [27, 28]. The chemical shift of the N‐H proton in NME moves from 2.22 to 2.14 ppm, with Δδ = −0.08 ppm, indicating a strong interaction between NME and the NO₃ ⁻ anion. This shift is attributed to the preferential coordination of Li⁺ to the C═O group of NME, which acts as a potent ligand. The Li⁺ coordination polarizes the adjacent N‐H bond, increasing electron density around the amide proton and enhancing its shielding, thus leading to the observed upfield shift. This interaction effectively “fixes” NO₃ ⁻ anions within the amide network of NME. Such fixation promotes the reductive decomposition of NO₃ ⁻ at electrode interface, thereby facilitating the formation of a nitrogen‐rich solid electrolyte interphase (SEI/CEI), potentially comprising species such as Li_xN [39]. This spectroscopic signature reflects the formation of N─H···O hydrogen bonds and confirms the active participation of NME units in stabilizing nitrate species. Consistently, LiNO₃ remains undissolved in F‐DEE composed solely of LiTFSI and NMTFA, whereas it dissolves completely in the presence of NME, yielding a transparent and homogeneous solution (Figure S1a,b). These observations provide direct experimental evidence that the PNME framework fundamentally alters LiNO₃ solvation thermodynamics, enabling salt dissolution that is otherwise prohibited in F‐DEE.

Further insights into polymer‐solvent interactions were obtained from the ¹H NMR spectra of the F‐DEE‐based gel electrolytes. The N‐H proton signals in PNME‐0 (F‐DEE+NME) and PNME‐1.5 (F‐DEE+NME+LiNO₃) exhibit systematic upfield shifts relative to the F‐DEE, indicative of hydrogen‐bond formation between PNME chains and NMTFA molecules (Figure 2d). This interaction was independently corroborated by FT‐IR spectroscopy, where the N‐H stretching vibration progressively shifts to higher wavenumbers from F‐DEE to PNME‐0 and then to PNME‐1.5 (Figure 2e), reflecting the evolution and strengthening of hydrogen‐bond networks within the gel matrix [29, 40]. In the PNME system, the hydrogen bond acceptors for the N─H groups are limited to the available carbonyl oxygens (C═O), forcing the N‐H groups to compete for these sites. Upon addition of LiNO₃, Li⁺ acts as a strong Lewis acid and preferentially coordinates with the carbonyl oxygen of the amide (Li⁺···O═C). This coordination deprives the C═O of its ability to act as a hydrogen bond acceptor, thereby physically disrupting the original N─H···O═C hydrogen bonding network. As a result, the N─H groups originally bound to C═O are “released.” [41].

At the same time, a large number of NO₃ ⁻ anions are introduced into the system. Although a single N─H···O(NO₃ ⁻) hydrogen bond is weaker than N─H···O═C, and the deshielding effect on the N─H proton is reduced, leading to a blue shift in the N─H stretching vibration, the number of available NO₃ ⁻ sites far exceeds that of the original C═O sites. According to Le Chatelier's principle and the law of mass action, the excess NO₃ ⁻ drives the equilibrium toward the formation of N─H···O(NO₃ ⁻) interactions, achieving a statistical thermodynamic compensation where “numerical advantage makes up for the lower individual strength.” Therefore, it is not that a weak hydrogen bond actively replaces a strong one; rather, the strong coordination of Li⁺ first breaks down the original hydrogen bonding network, forcing the released N‐H groups to turn to the most abundant new acceptor (NO₃ ⁻) in the electrolyte environment, thus restructuring the entire interaction landscape of the electrolyte. This process effectively dissociates LiNO₃ and confines NMTFA within the gel, creating a tailored chemical microenvironment. This microenvironment enables the in situ formation of a beneficial nitrogen‐rich interphase on the electrode surface, which enhances electrolyte stability (preventing leakage) and suppresses carbonyl‐mediated transition metal ion dissolution, as evidenced by inductively coupled plasma (ICP), X‐ray photoelectron spectroscopy (XPS) analysis and the reduced TM deposition observed on the cycled lithium metal anode (Section 2.3) [42].

Evidence for the formation of a continuous PNME network is provided by Raman spectroscopy, where the characteristic C═C stretching band at 1613 cm⁻¹ disappears completely after in situ polymerization in both PNME‐0 and PNME‐1.5 (Figure 2f), confirming the full consumption of vinyl groups [43]. This molecular‐level transformation is accompanied by a pronounced macroscopic phase transition. Optical images show the conversion from a free‐flowing liquid to a non‐flowing deep‐eutectic gel after polymerization, with the electrolyte firmly adhering to the vial wall even upon inversion (Figure S2a,b,c), unambiguously demonstrating successful gelation and network formation. Concurrently, the shift of the N‐H band at 1542 cm⁻¹ to higher wavenumbers and the disappearance of the C═O band at 1658 cm⁻¹ further support the hydrogen bonding formation [29].

Further insight into the Li⁺ coordination environment was obtained from ⁷Li NMR spectroscopy (Figure 2g). Upon introducing NME into the F‐DEE, the Li⁺ resonance associated with LiTFSI exhibits a pronounced high‐field shift (PNME‐0). Upon the addition of LiNO₃ (PNME‐1.5), a slight downfield shift is observed, indicating an increase in mobile Li⁺ species enabled by the NME‐assisted dissolution of LiNO₃ [28].

To further resolve the Li⁺ solvation structures, Raman spectroscopy was employed (Figure 2h). The characteristic S‐N‐S stretching modes of TFSI⁻ at ∼723 and ∼733 cm⁻¹, corresponding to free and Li⁺‐coordinated TFSI⁻, respectively, confirm the coexistence of solvent‐separated ion pairs (SSIPs) and contact ion aggregates (AGGs) [43]. The proportions of ion‐pair species in different electrolytes were quantified by peak‐fitting analysis of the characteristic TFSI⁻ peaks in the Raman spectra: in F‐DEE, the proportions of SSIPs and CIPs were 29% and 71%, respectively; in PNME‐0, the corresponding proportions were 34% and 66%; while in PNME‐1.5, SSIPs and CIPs accounted for 35% and 45%, respectively, accompanied by the appearance of 20% AGGs (anion aggregates). The increase in the proportion of SSIPs from F‐DEE to PNME‐0 stems from the competitive displacement of some NMTFA coordination sites originally occupied by TFSI⁻ by the NME units in the PNME framework, which possess a higher binding energy with NMTFA (−2.51 eV), thereby pushing part of TFSI⁻ out of the first solvation shell. Upon introducing LiNO₃ to form PNME‐1.5, NO₃ ⁻, owing to its unique electronic structure and synergistic hydrogen‐bonding interaction with PNME, draws TFSI⁻ back into the Li⁺ coordination layer, thereby constructing an anion‐rich AGG structure. The observed AGG signal (20%) in the Raman spectra precisely reflects this type of anion‐aggregate structure formed with the participation of TFSI⁻ [44, 45, 46]. Comparative analysis indicates that F‐DEE and PNME‐0 are enriched in solvent‐separated ion pairs (SSIPs), where an NMTFA‐dominated solvation environment favors organic reduction pathways, yielding mechanically weak, organic‐rich interphases. In contrast, PNME‐1.5 exhibits a higher population of ionic aggregates (AGGs), which promotes the preferential reduction of inner‐shell TFSI⁻ and NO₃ ⁻ species, thereby generating a dense, inorganic‐rich SEI/CEI dominated by LiF and Li_xN components. These robust SEI/CEI layers provide enhanced mechanical strength, fast Li⁺ transport, and superior chemical stability, thereby underpinning the simultaneously improved lithium‐metal compatibility and high‐voltage cathode tolerance of the DEGPE.

To elucidate the atomic‑level reconstruction mechanism of the Li⁺ solvation shell by PNME, we performed molecular dynamics (MD) simulations combined with radial distribution function (RDF) analysis and representative structural snapshots. The MD snapshots and corresponding RDFs for F‐DEE (Figure S3a; Figure 2i), PNME‐0 (Figure S3b,d), and PNME‐1.5 (Figure S3c; Figure 2j) are shown, where the solid and dashed curves denote the Li⁺ radial distribution functions and coordination numbers (CNs), respectively. In the pristine F‐DEE, Li⁺ coordination is dominated by oxygen atoms from NMTFA, reflected by a pronounced Li‐O(NMTFA) peak arising from the strong polarity of the carbonyl group. In F‑DEE, Li⁺ is primarily coordinated by NMTFA (coordination number 2.91) and TFSI⁻ (2.02). Upon introduction of PNME (PNME‑0), the carbonyl oxygen of NME enters the first solvation shell, contributing 0.43 coordination sites, while the coordination number of NMTFA decreases to 2.51, indicating that the PNME framework partially replaces NMTFA and coordinates with Li⁺. Upon introducing LiNO₃ (PNME‐1.5), NO₃ ⁻ contributes 0.75 coordination sites, ultimately forming an anion‑rich solvation sheath dominated by anions (TFSI⁻, NO₃ ⁻) and NME. This structural evolution is fully consistent with the calculated binding energies of Li⁺ with NME and NMTFA (Li⁺‐NME > Li⁺‐NMTFA), quantitatively revealing the fundamental reason for the PNME‑driven reconstruction of the solvation structure from both energetic and structural perspectives [47]. Notably, the steric hindrance from the aggregation of NME molecules into NME3 motifs drives a shift in the Li⁺ solvation environment from a free‐solvent‐dominated to a polymer‐regulated regime by reducing the coordination number [48]. This reconfigured solvation chemistry promotes the preferential reduction of anionic species at the electrode interfaces, facilitating the in situ formation of dense, inorganic‐rich SEI/CEI layers dominated by LiF and Li_xN, thereby effectively suppressing lithium dendrite growth and interfacial side reactions.

To elucidate the molecular mechanism underlying the enhanced stability, variable‐temperature infrared (VT‐IR) spectroscopy was performed on the F‐DEE, PNME‐0, and PNME‐1.5 electrolytes. In the synchronous spectra (Figure 2n; Figure S3f), within the range of 1300–1600 cm⁻¹, both PNME‐0 and PNME‐1.5 samples exhibited distinct positive cross‐peaks between ∼1350 cm⁻¹ (assigned to the C‐F‐related vibration or amide II band of the NMTFA solvent) and ∼1550 cm⁻¹ (assigned to the N‐H stretching or amide III band of the PNME backbone amide groups) during heating. This indicates a synchronous variation in the vibrational intensities of these functional groups with increasing temperature, directly confirming the existence of a stable hydrogen‐bond network between PNME and NMTFA. Such a cross‐peak was not observed in the spectrum of F‐DEE (Figure 2l) [49].

Furthermore, in the asynchronous spectra (Figure 2m; Figure S3e), no significant new peaks or peak shifts were observed for PNME‐0 or PNME‐1.5 during heating, indicating that the hydrogen bonds between PNME and NMTFA do not undergo significant rupture or dissociation. This is attributed to the fundamental alteration of the hydrogen‐bonding environment induced by the introduction of PNME polymer chains, which promotes the formation of N‐H···O hydrogen bonds among LiTFSI, NMTFA, and the PNME backbone, thereby suppressing thermally induced dissociation. In contrast, for the F‐DEE sample without a polymer network, distinct cross‐peaks at 1333 cm⁻¹ and 1542 cm⁻¹ were clearly observed in the asynchronous spectrum over the same temperature range (Figure 2k). These peaks arise from the gradual breakdown of the complexation between ‐SO₂ (from TFSI⁻) and ‐NH (from NMTFA) as temperature increases, indicating that the hydrogen‐bond network among solvent molecules is disrupted at elevated temperatures, leading to an increase in free solvent molecules [50, 51].

“Suppression thermally induced dissociation” primarily refers to the stabilization of the polymer–solvent network through a strong hydrogen‐bond network between the PNME backbone and NMTFA solvent molecules, which physically anchors the solvent molecules and prevents their escape from the Li⁺ solvation sheath at high temperatures. This mechanism does not directly inhibit the thermal decomposition of lithium salts (e.g., LiTFSI) nor alleviate the degradation of LiNO₃ (the main role of LiNO₃ is to be preferentially reduced at the interface to form an inorganic‐rich SEI/CEI). Instead, by stabilizing the entire electrolyte network, it indirectly mitigates interface deterioration and side reactions caused by solvent loss. The hydrogen‐bond enhancement mediated by PNME stabilizes the deep eutectic network, resulting in a structurally more robust and thermally stable electrolyte system [39].

Thermal stability is a critical determinant of battery safety and constitutes a defining advantage of the PNME‐1.5 electrolyte developed in this work. The thermal robustness of PNME‐1.5 was systematically evaluated by thermogravimetric analysis (TGA) in comparison with a commercial liquid electrolyte (LE: 1 M LiPF₆ in EC/EMC), the pristine F‐DEE, LiNO₃‐free polymer gel (PNME‐0) and the PNME‐1.5 (Figure S4). Notably, PNME‐1.5 exhibits the highest thermal stability, with an onset decomposition temperature (temperature for 90% mass retention) of 124.3°C, exceeding those of PNME‐0 (109.9°C), F‐DEE (92.6°C), and LE (66.2°C) [27]. This pronounced improvement underscores the enhanced thermal resilience imparted by NME‐moity‐mediated solvation and hydrogen‐bond stabilization, highlighting the intrinsic safety advantage of the designed DEGPE.

The superior fire resistance of PNME‐1.5 was further demonstrated by direct flame tests. Upon exposure to an open flame, the commercial LE ignited instantaneously, sustained combustion after flame removal. In stark contrast, PNME‐1.5 remained non‐flammable and retained its structural integrity even under prolonged flame exposure (Figure S5a,b; Videos S1 and S2). This striking difference unequivocally confirms the excellent flame‐retardant characteristics of PNME‐1.5, a critical prerequisite for safe high‐energy‐density LMB operation.

In addition to thermal safety, the electrochemical stability window (ESW), which delineates the maximum safe operating voltage of the electrolyte, was assessed by linear sweep voltammetry (LSV). The LSV was performed by scanning from the open‐circuit voltage toward both higher and lower potentials, using an activated C||SS cell to better simulate real battery conditions. As shown in Figure S6a, the LSV results indicate that PNME‐1.5 exhibits a wide electrochemical stability window, with a cathodic cut‐off potential reaching 5.3 V (vs. Li⁺/Li). To further rigorously assess the oxidative stability of the electrolytes, potentiostatic polarization (floating) tests were carried out. As shown in Figure S6b, in the Li|F‐DEE|NCM811 cell, continuous side reactions occur at the cathode interface once the voltage exceeds 4.6 V, leading to pronounced fluctuations in leakage current. When the voltage surpasses 4.7 V, the pre‑formed SEI begins to decompose, and the current increases significantly for both F‐DEE and PNME‐0 systems. In contrast, the Li|PNME‐1.5|NCM811 cell does not show an upward trend in leakage current until 4.8 V. Combining the LSV and potentiostatic polarization results, the PNME‐1.5 electrolyte maintains good electrochemical stability within 4.5 V (vs. Li⁺/Li) and meets the cathode‑side stability requirements for high‑voltage lithium‑metal batteries (e.g., NCM811, LCO, etc.).

The critical current density (CCD), a stringent metric for assessing electrolyte compatibility with lithium metal, was evaluated by stepwise increasing the current density under a fixed plating/stripping duration (Figure S6c). The Li|PNME‐1.5|Li symmetric cell exhibits good stability, sustaining reversible cycling up to a high CCD of 3.0 mA cm⁻². By contrast, cells employing the pristine F‐DEE suffer abrupt voltage fluctuations at a lower current density of 2.0 mA cm⁻², indicative of premature interfacial failure and unstable lithium deposition. Although PNME‐0 achieves a CCD comparable to that of PNME‐1.5, it operates under higher polarization, reflecting inferior interfacial kinetics. The better CCD performance of PNME‐1.5 originates from its favorable electrode interface compatibility and ion‐transport properties, including a high ionic conductivity of 1.28 × 10⁻³ S cm⁻¹ at 40°C (Figure S6d). These characteristics collectively signify a stable and kinetically robust Li/electrolyte interface, capable of homogenizing Li⁺ flux and suppressing dendritic growth under high‐current operation. Furthermore, PNME‐1.5 delivers an enhanced Li⁺ transference number (t_Li*  =  0.55), surpassing those of PNME‐0 (0.47) and F‐DEE (0.39) (Figure S6e,f,g). The elevated t_Li+ promotes uniform Li⁺ distribution during plating/stripping, suppresses anion accumulation and concentration polarization, and thereby further stabilizes the lithium metal interface and improves overall electrochemical performance.

2.2. Compatibility of PNME‐1.5 With the Lithium Anode

The cycling stability of F‐DEE, PNME‐0, and PNME‐1.5 electrolytes were systematically evaluated in Li||Li symmetric cells by galvanostatic lithium plating/stripping (Figure 3a). At a current density of 0.2 mA cm⁻², the F‐DEE‐based cell exhibits premature failure after 400 h, while the PNME‐0‐based cell extends the cycling lifetime to 800 h. In contrast, the Li|PNME‐1.5|Li cell delivers exceptional long‐term stability, sustaining continuous cycling for over 2500 h with a low and stable overpotential of only 73  mV. This performance advantage is preserved under more demanding conditions. When the current density is increased to 0.5 mA cm⁻², the Li|PNME‐1.5|Li cell remains stable for 1500  h, whereas cells employing F‐DEE and PNME‐0 fail rapidly after 450 h and 800 h, respectively (Figure S7). These results demonstrate the superior tolerance of PNME‐1.5 toward long‐term lithium cycling and sustained interfacial stress. Post‐mortem morphological analysis of the lithium anodes provides direct insight into the origin of this enhanced stability. Scanning electron microscopy (SEM) (Figure 3b; Figure S8b) and in situ dendrite monitoring (Figure 3d; Figure S8a) reveal that lithium deposited in F‐DEE and PNME‐0 electrolytes forms loose, needle‐like, and mossy dendritic structures. Cross‐sectional SEM further confirms that these dendrites propagate deep into the bulk lithium electrode (Figure S8c,d), creating mechanically fragile regions that readily penetrate the separator and trigger internal short circuits. By contrast, lithium anodes cycled in the PNME‐1.5 electrolyte display a dense, smooth, and highly uniform morphology (Figures 3c,e; Figure S8e), with no observable dendritic features. This compact lithium deposition effectively suppresses parasitic interfacial reactions and mechanical instability at the Li/electrolyte interface, thereby underpinning the extraordinary cycling durability of the PNME‐1.5 system.

FIGURE 3.

(a) Li||Li symmetric cells via galvanostatic Li plating/stripping tests at 0.2 mA cm⁻². Post‐cycling analysis of the lithium anodes via SEM (b,c) and in situ dendrite monitoring for F‐DEE and PNME‐1.5 (d,e). (f,g,h) EIS was performed on NCM811|F‐DEE|Li, NCM811|PNME‐0|Li, and NCM811|PNME‐1.5|Li cells during cycling. (i,j,k) TOF‐SIMS analysis of the SEI on lithium anodes in NCM811|F‐DEE|Li, NCM811|PNME‐0|Li, and NCM811|PNME‐1.5|Li cells, respectively.

To further evaluate interfacial stability under realistic operating conditions, electrochemical impedance spectroscopy (EIS) was conducted on NCM811||Li cells employing F‐DEE, PNME‐0, and PNME‐1.5 electrolytes during extended cycling (Figures 3f,g,h). Cells based on F‐DEE and PNME‐0 exhibit a pronounced and progressive increase in interfacial impedance with cycling, manifested by the continuous expansion of the semicircle in the Nyquist plots. This behavior is indicative of persistent SEI/CEI fracture, repeated electrolyte decomposition, and the accumulation of resistive interfacial byproducts, consistent with their rapid electrochemical degradation. In contrast, the NCM811|PNME‐1.5|Li cell displays an opposite impedance evolution. With continued cycling, the interfacial resistance gradually decreases, as evidenced by the progressive shrinking and eventual fading of the semicircle in Figure 3h. This atypical impedance decay reflects the in situ formation of a stable, ionically conductive, and mechanically robust SEI/CEI, which effectively suppresses parasitic interfacial reactions and facilitates sustained charge transport across the electrode‐electrolyte interfaces. To quantitatively elucidate the role of LiNO₃ in regulating interfacial charge‐transfer kinetics, exchange current densities were extracted from fitted Tafel plots (Figure S9). The PNME‐1.5 electrolyte exhibits a markedly higher exchange current density than PNME‐0, signifying substantially accelerated Li⁺ charge‐transfer kinetics at the SEI interface. This enhancement directly corroborates the formation of an inorganic‐rich, Li_xN/LiF‐dominated interphase enabled by LiNO₃ solubilization, which lowers interfacial energy barriers and underpins the superior electrochemical stability of PNME‐1.5 under prolonged cycling.

To elucidate the role of LiNO₃ in regulating the SEI chemistry of PNME‐1.5, X‐ray photoelectron spectroscopy (XPS) was performed on lithium metal anodes after cycling in F‐DEE, PNME‐0, and PNME‐1.5 electrolytes. The C 1s spectra (Figure S10a,e,i) are broadly similar across all samples, featuring a dominant C‐F component at 292.8 eV arising from the decomposition of LiTFSI‐derived species. Simultaneously, the C─O and Li₂CO₃ signals observed in Figures S10b,f,j are mainly attributed to oxygen‐containing functional groups on the electrode material surface and the electrochemical reduction of the electrolyte at the interface. In contrast, pronounced differences emerge in the nitrogen chemistry. The N 1s spectra of the F‐DEE‐ and PNME‐0‐cycled anodes (Figure S10c,g) display a single N‐SO_x signal at 399.6 eV, attributable to the decomposition of TFSI⁻ and NMTFA. The cycled Li anode in the PNME‐1.5 (Figure S10k) exhibits an additional, well‐resolved peak at 398.0 eV, assigned to Li_xN species generated from LiNO₃ reduction [52], alongside the N‐SO_x component at 399.9 eV. The emergence of this Li_xN‐rich phase constitutes direct spectroscopic evidence that LiNO₃ actively participates in SEI formation when rendered soluble by the PNME framework. Li_xN species are well known for their high Li⁺ ionic conductivity and mechanical robustness, enabling rapid and homogeneous Li⁺ transport across the SEI. Their presence promotes uniform lithium deposition, mitigates local current‐density amplification, and effectively suppresses dendritic growth, consistent with the markedly improved plating/stripping stability observed for PNME‐1.5.

The F 1s spectras (Figure S10d,h,l) further reveal a characteristic LiF signal at 684.8 eV for all electrolytes [53]. As an electronically insulating yet ionically conductive component, LiF contributes to passivating the lithium surface, suppressing continuous electrolyte reduction, and internal leakage current. Notably, in PNME‐1.5, the synergistic coexistence of LiF and Li_xN gives rise to a compact, inorganic‐rich SEI that combines chemical stability with fast ion transport, underpinning the exceptional interfacial durability of this electrolyte system [54].

Time‐of‐flight secondary ion mass spectrometry (TOF‐SIMS) was further employed to probe the chemical architecture and spatial evolution of the SEI. As shown in Figure 3k, the lithium anode cycled in the PNME‐1.5 electrolyte yields a distinctly different SEI composition compared with that in F‐DEE (Figure 3i) and PNME‐0 (Figure 3j). In addition to abundant C₂HO⁻ and LiF fragments, the PNME‐1.5 spectrum exhibits a pronounced LiN⁻ signal, whereas only C₂HO⁻ and LiF are detected for F‐DEE and PNME‐0, confirming that LiNO₃ actively participates in SEI formation exclusively when solubilized by the PNME framework. 3D ion‐mapping analysis reveals that, in PNME‐1.5, LiF is distributed more deeply and uniformly throughout the SEI thickness, a feature known to facilitate rapid Li⁺ transport and enhance the mechanical integrity of the interphase. Simultaneously, the strong LiN⁻ signal accompanied by an almost negligible NO₃ ⁻ signal indicates that LiNO₃ is nearly completely converted into Li_xN species at the lithium interface, consistent with preferential anion reduction driven by the anion‐rich solvation environment.

Li₂CO₃ signals are observed in all three electrolytes (Figure S11), possibly originating from the reductive decomposition of NMTFA and PNME. However, their spatial distributions differ markedly. In the F‐DEE‐derived SEI, Li₂CO₃ is highly concentrated near the surface, forming a brittle layer that is prone to cracking during repeated cycling [55]. Such mechanical instability leads to continuous SEI rupture and regeneration, accelerating the consumption of active lithium and electrolyte and thereby promoting rapid capacity decay. In contrast, in PNME‐1.5, Li₂CO₃ is uniformly distributed as a moderately concentrated layer near the surface. Together with LiF and Li_xN, it collaboratively forms a compact, mechanically resilient inorganic framework that stabilizes the SEI during long‐term cycling. Notably, different depth profiles of C₂HO⁻, LiF, LiN⁻, and NO₃ ⁻ further reveal that the SEI formed in PNME‐1.5 adopts a hybrid architecture, comprising an organic matrix interpenetrated by uniformly distributed inorganic species. This composite SEI structure simultaneously facilitates fast Li⁺ transport and homogeneous Li⁺ flux at the interface, thereby enabling uniform lithium deposition and effectively suppressing dendrite formation.

2.3. Compatibility of the PNME‐1.5 With NCM811

To assess the applicability of DEGPEs in high‐voltage NCM811 systems, it is crucial to elucidate the morphological and structural evolution of the cathode under high‐voltage conditions. Accordingly, NCM811||Li half‐cells were assembled to systematically investigate the structural stability of NCM811 in F‐DEE, PNME‐0, and PNME‐1.5 electrolytes. When cycled at room temperature at 1C (200 mA g⁻¹, 2.7–4.3 V), the NCM811|F‐DEE|Li and NCM811|PNME‐0|Li cells exhibited rapid capacity decay, suffering nearly complete capacity loss within 200 cycles. In contrast, the NCM811|PNME‐1.5|Li cell delivered markedly improved cycling stability, maintaining stable operation for over 500 cycles. The cell exhibited an initial discharge specific capacity of 204 mAh g⁻¹ and retained 80.1% of its capacity after prolonged cycling (Figure 4a; Figure S12). These results demonstrate the superior compatibility of the PNME‐1.5 electrolyte with high‐voltage NCM811 cathodes and underscore the critical role of DEGPEs in stabilizing cathode structure under demanding electrochemical conditions.

FIGURE 4.

(a) NCM811||Li half‐cells were assembled in F‐DEE, PNME‐0, and PNME‐1.5 electrolytes tested at room temperature and 1C (200 mA g⁻¹, 2.7–4.3 V). (b) The cycled Li metal anodes in NCM811|F‐DEE|Li, NCM811|PNME‐0|Li, and NCM811|PNME‐1.5|Li cells were examined by ICP spectroscopy. (c,d,e) Depth‐profiling XPS was used to analyze the composition gradients of the CEI of F in NCM811|F‐DEE|Li and NCM811|PNME‐1.5|Li cells. (f,g,h) Depth‐profiling XPS was used to analyze the composition gradients of the CEI of N in NCM811|F‐DEE|Li and NCM811|PNME‐1.5|Li cells. (i,m) Morphological inspection of the cycled NCM811 cathode in the F‐DEE and PNME‐1.5 electrolytes. (j,n) TEM was employed to examine the CEI evolution on cycled NCM811 cathodes in cells using F‐DEE and PNME‐1.5 electrolytes. EELS analysis in the charged state from the surface to a depth of ≈20 nm on NCM811 cathodes of O‐K edge (k,o) and Ni‐L edge (l,p) in cells using F‐DEE and PNME‐1.5 electrolytes. (q,r) In situ XRD tests in the charged state for cells cycled in F‐DEE and PNME‐1.5.

To elucidate the origin of the observed performance differences, the Li metal anodes were examined after prolonged cycling. ICP spectroscopy revealed substantial accumulation of Ni and Mn species on the Li surface in cells employing F‐DEE and PNME‐0 electrolytes, whereas only trace amounts were detected for PNME‐1.5 (Figure 4b). This indicates severe TM dissolution and cross‐talk in F‐DEE and PNME‐0, which is effectively suppressed in PNME‐1.5. Consistent with these compositional differences, pronounced morphological contrasts were observed. The Li anode cycled in PNME‐1.5 exhibited a smooth, dendrite‐free surface, indicative of a uniform and robust SEI. In contrast, Li surfaces from F‐DEE and PNME‐0 were completely covered by black, insulating byproducts arising from severe parasitic reactions, which accelerate impedance growth and premature cell failure (Figure S13).

To further investigate the CEI chemistry, depth‐profiling XPS was employed to analyze composition gradients within the CEI. In the F‐DEE system (Figures 4c,f), the LiF content decreases progressively from the surface toward the interior of the NCM811 secondary particles (atomic content decreasing from 8.2 wt% to 5.3 wt%), while in the PNME‐1.5 sample (Figures 4d,g), it exhibits a gradual increase inward (atomic content increasing from 5.1 wt% to 12.3 wt%). This opposing LiF distribution trend (Figure 4e) reveals that the PNME polymer framework transforms the formation of LiF from a passive decomposition process confined to the particle surface into an orchestrated interfacial construction process that extends deep into the secondary particles. The resulting unique gradient LiF‐rich interphase provides holistic reinforcement to the NCM811 secondary particles, from the exterior to the interior.

In contrast, the N atom signal in PNME‐1.5 is significantly higher than in F‐DEE and exhibits a monotonic decrease from the surface toward the interior (atomic content decreasing from 7.1 wt% to 4.5 wt%) (Figure 4h), which is consistent with the preferential surface decomposition of LiNO₃ and the formation of a Li_xN/LiN_xO_y‐rich CEI. Further morphological examination of the cycled NCM811 cathodes revealed pronounced innerparticle cracking of the secondary spheres in the F‐DEE electrolyte (Figure 4i), which can be attributed to severe high‐voltage interfacial degradation. Such structural damage leads to irreversible mechanical collapse and rapid capacity decay, in contrast to the preserved cathode integrity observed in PNME‐1.5. After cycling in PNME‐0, the NCM811 secondary spheres exhibited severe deformation, characterized by loose interparticle contact and pronounced gaps between primary grains (Figure S14a). In contrast, the cathode cycled in PNME‐1.5 maintains a smooth surface, with tightly bonded primary particles and no visible cracks (Figure 4m), indicating both mechanical and chemical stability at the interface. This significant difference demonstrates that relying solely on the polymer network is insufficient for cathode protection; the addition of LiNO₃ is essential for the in situ formation of a robust, nitrogen‐rich CEI layer, which effectively mitigates severe structural degradation. XPS N 1s analysis of the cycled cathode surface reveals almost no detectable nitrogen signal in PNME‐0, with only a weak Li‐C‐N peak (binding energy = 399.8 eV, Figure S15g) originating from the polymer backbone. In contrast, the sample with LiNO₃ addition (PNME‐1.5) exhibits distinct features in the N 1s spectrum: at 399.5 and 397.5 eV, corresponding to nitrogen‑containing inorganic species such as LiN_xO_y and Li₃N, respectively (Figure 4g). These results indicate that the introduction of LiNO₃ promotes the formation of a CEI rich in inorganic nitrides on the cathode surface. In PNME‐0, the absence of such a nitrogen‑rich CEI leaves the cathode particles directly exposed to the electrolyte during cycling, leading to continuous side reactions and structural degradation of the active material. The enlarged gaps and loosened contact between primary particles observed in the SEM images are a direct consequence of this lack of physical and chemical protection. In PNME‐1.5, the nitrogen‑rich CEI effectively buffers volume variations and suppresses side reactions, thereby preserving the integrity and tight contact of the cathode particles [56].

The evolution of the CEI was further investigated by transmission electron microscopy (TEM). For F‐DEE, the interfacial layer of the cathode particle is intrinsically fragile and insufficient to withstand aggressive high‐voltage reactions. As a result, extensive electrolyte corrosion and TM dissolution occurred, leading to the formation of a rough, highly non‐uniform CEI layer with an average thickness of 16.8 nm on the NCM811 particle surface (Figure 4j). Electron Energy‐Loss Spectroscopy (EELS) analysis of the charged state indicates that the region from the surface to a depth of ≈20 nm corresponds to the CEI layer, which contains lattice defects related to oxygen pressure (O‐K edge) and electrochemically inert nickel (Ni‐L edge) (Figure 4k,l). Although PNME‐0 forms a gel electrolyte, its interfacial chemistry lacks effective passivating components. Consequently, a relatively thick (≈13.0 nm) and non‐uniform CEI layer was still observed (Figure S14b), offering only limited protection against high‐voltage interfacial degradation. In contrast, PNME‐1.5 enabled the formation of a stable, compact, and inorganic‐rich CEI through the controlled interfacial decomposition of LiNO₃. The cycled NCM811 particles were uniformly coated with a thin CEI layer of only 5.1 nm, which effectively resisted electrolyte attack and suppressed TM dissolution (Figure 4n). EELS provided further evidence of a distinct contrast under high‐potential charging: while an electrochemically inert CEI layer (≈5 nm) forms on the surface, the bulk region (within the 20 nm depth) is characterized by a pronounced decrease in oxygen defects and a concomitant shift of nickel toward a higher oxidation state (Figure 4o,p).

In situ X‐ray diffraction (XRD) measurements provide direct evidence for the enhanced structural stability of NCM811 in PNME‐1.5. Upon charging, the (003) reflection exhibits pronounced shifts of 1.07° and 0.65° for cells cycled in F‐DEE and PNME‐0, respectively (Figure 4q; Figure S16), indicative of substantial lattice distortion associated with severe interfacial side reactions. In contrast, the NCM811 cathode cycled in PNME‐1.5 shows a smaller peak shift of only 0.61° (Figure 4r), reflecting reduced lattice parameter variation and improved structural reversibility during high‐voltage cycling. To further elucidate the chemical origin of this stability, XPS was conducted on the cycled NCM811 cathode (Figure S15). In the O 1s spectra of F‐DEE‐based cells, prominent signals corresponding to M─O coordination bonds were detected after cycling, revealing the formation of a chemically unstable CEI layer. The emergence of M─O bonds indicates strong coordination between dissolved TM and oxygen originating from electrolyte decomposition products, which accelerates TM dissolution and contributes to rapid performance degradation upon prolonged cycling. By contrast, the cathode cycled in PNME‐1.5 exhibits a fundamentally different interfacial chemistry. Strong LiN_xO_y and LiF signals are observed in the N 1s and F 1s spectra, respectively, confirming the formation of a LiNO₃‐derived, inorganic‐rich CEI. These robust interfacial species effectively suppress oxidative electrolyte decomposition at high potentials and reduce charge‐transfer resistance at the cathode interface [57]. Notably, the M‐O bond signal is completely absent in PNME‐1.5, providing evidence for the effective inhibition of TM dissolution.

Collectively, a direct quantitative correlation between the solvation structure and the interfacial chemical properties was established. As the proportion of AGGs in the electrolyte increased (from F‑DEE to PNME‑1.5), the content of inorganic components (LiF, Li_xN) in the CEI derived from anion‑prioritized reduction showed a monotonic increasing trend (Figure 4e,h), while the CEI thickness significantly decreased from 16.8 nm (F‑DEE) to 5.1 nm (PNME‑1.5) (Figure 4j,n). Furthermore, the AGG's proportion was negatively correlated with the amount of TM deposition on the lithium anode surface (Figure 4b). These correlations directly confirm the core mechanism: an anion‑rich solvation structure drives anion‑prioritized reduction, thereby inducing the formation of a thin, dense, and inorganic‑enriched stable interface [58, 59, 60]. This interphase not only mitigates parasitic interfacial reactions but also preserves lattice integrity and redox reversibility of NCM811, thereby effectively suppressing capacity fading and enabling outstanding long‐term cycling stability under high‐voltage operation.

2.4. Electrochemical Performance of the Cells Based on PNME‐1.5

To evaluate the high‐temperature applicability of the DEGPE, full cells with a high cathode loading of 19 mg cm⁻² were assembled using the NCM811|PNME‐1.5|Li configuration. When tested at 60°C and 0.1 C, the cell delivered a high initial discharge specific capacity of 222 mAh g⁻¹ and retained 81.5% of its capacity after 80 cycles, demonstrating favorable cycling stability under elevated‐temperature operation. By contrast, the NCM811|F‐DEE|Li cell exhibited rapid performance degradation under identical conditions: pronounced capacity decay commenced as early as the 10th cycle, and nearly complete capacity loss was observed within 60 cycles (Figure 5a). The results demonstrate that the PNME‐1.5 gel polymer electrolyte effectively stabilizes the electrode‐electrolyte interface at high temperatures by leveraging the high polarity of NME to enhance LiNO₃ dissolution and facilitate the formation of a uniform, heat‑resistant interphase at the electrode interfaces. Concurrently, the PNME framework suppresses thermally induced dissociation, as evidenced by the robust hydrogen‑bond network formed between PNME and the NMTFA solvent, which physically anchors the solvent molecules and prevents their escape from the Li⁺ solvation sheath under elevated temperatures [61]. This enables durable high‐capacity operation even under high cathode areal loadings and elevated temperature.

FIGURE 5.

(a) The cycle performance of the coin cells (NCM811|PNME‐1.5|Li, NCM811|PNME‐0|Li, NCM811|F‐DEE|Li) with a high cathode loading of 19 mg cm⁻² at 0.1 C and 60°C. (b) The rate capability of the coin cells (NCM811|PNME‐1.5|Li, NCM811|PNME‐0|Li, NCM811|F‐DEE|Li). (c) The cycle performance of the coin cells (NCM811|PNME‐1.5|Li, NCM811|PNME‐0|Li, NCM811|F‐DEE|Li) after 200 cycles at 5C. (d) The cycle performance of a pouch cell consisting of a high‐areal‐capacity NCM811 cathode (12 mg cm⁻²) with PNME‐1.5 and an ultra‐thin Li anode (50 µm). (e) NCM811||Li half‐cells were assembled in F‐DEE, PNME‐0, and PNME‐1.5 electrolytes tested at room temperature and 0.5 C (220 mA g⁻¹, 2.7–4.5 V). (f) The cycle performance of the full‐cells (NCM811|PNME‐1.5|Li, NCM811|PNME‐0|Li, NCM811|F‐DEE|Li) with a high cathode loading of 12 mg cm⁻² at 0.1 C and 25°C (220 mA g⁻¹, 2.7–4.5 V). (g) The cycle performance of the coin cells on LCO|PNME‐1.5|Li at 80 °C and 2 C (2.8–4.3 V). (h) The cycle performance of a pouch cell consisting of a high‐areal‐capacity LCO cathode (15 mg cm⁻²) and an ultra‐thin Li anode (50 µm). (i,j,k) The nail penetration test monitored by infrared thermography evaluates the safety performance of pouch cells on NCM811|F‐DEE|Li, NCM811|PNME‑0|Li, and NCM811|PNME‑1.5|Li. (l,m) The gas generation from the heating‐induced swelling test on NCM811|F‐DEE|Li and NCM811|PNME‑1.5|Li. (n) Folding and cutting tests of pouch cells with the NCM811|PNME‑1.5|Li configuration. (o) The multi‐parameter radar chart comparison of the amide‐based DEGPEs in this work with literature data.

The rate capability, cycling durability, and safety characteristics of the NCM811|PNME‐1.5|Li system were systematically evaluated. The NCM811|PNME‐1.5|Li coin cells exhibit outstanding rate performance, delivering high discharge specific capacities of 210, 207, 204, 202, 177, 153, 133, and 94 mAh g⁻¹ at current densities of 0.1, 0.2, 0.5, 1, 2, 3, 5, and 10 C, respectively (Figure 5b). The system exhibits high capacity retention even at extreme rates, particularly with 96.4% retention after 200 cycles at 5C, which underscores the rapid Li⁺ transport kinetics and robust interfacial stability achieved by the PNME‐1.5 electrolyte (Figure 5c).

To further validate its practical applicability, pouch cells were assembled using a high‐areal‐loading NCM811 cathode paired with an ultrathin Li metal anode (50 µm). When employing the PNME‐1.5 electrolyte, the NCM811||Li pouch cell delivered a high initial discharge specific capacity of 211 mAh g⁻¹ at 0.1 C (2.7–4.3 V), while maintaining 80.6% capacity retention and a high average Coulombic efficiency of 99.38% over 60 cycles (Figures 5d). To further verify the applicability of the electrolyte at higher voltages, cycling tests of NCM811||Li full cells and coin cells at a cutoff voltage of 4.5 V were performed. The experimental results indicate that even at 4.5 V, the PNME‐1.5 cell can still cycle stably. Specifically, the coin cell delivers a discharge capacity of 196.4 mAh g⁻¹ at 0.5 C with a capacity retention of about 80.6% after 300 cycles (Figure 5e), while the full cell achieves a discharge capacity of 235.6 mAh g⁻¹ at 0.1  C, retains 92.8% of its initial capacity after 50 cycles, and maintains stable performance throughout cycling (Figure 5f). These results confirm the outstanding interfacial stability and tolerance of the electrolyte system under a high voltage of 4.5 V.

The generality of the PNME‐1.5 electrolyte was further demonstrated in a high‐voltage LCO cathode system. At an elevated temperature of 80°C and a high rate of 2C (2.8–4.3 V), the LCO|PNME‐1.5|Li cell delivered an initial specific capacity of 130.0 mAh g⁻¹ and retained 89.6% of its capacity after 300 cycles (Figure 5g). Notably, in a more practically relevant pouch‐cell configuration with a high‐loading LCO (15 mg cm⁻²) cathode and an ultrathin Li anode (50 µm), a capacity retention of 87.0% was achieved after 80 cycles (Figure 5h).

To comprehensively assess the safety performance of the pouch cells, nail‐penetration tests monitored by infrared thermography and heating‐induced swelling tests were conducted. As shown in Figure 5i,j,k, the maximum hotspot temperatures recorded during nail penetration for the NCM811|F‐DEE|Li, NCM811|PNME‐0|Li, and NCM811|PNME‐1.5|Li pouch cells are 39.9°C, 37.4°C, and 35.6°C, respectively. The NCM811|PNME‐1.5|Li pouch cell shows the lowest short‐circuit temperature, indicating progressively enhanced resistance to thermal propagation and internal short‐circuit‐induced heat accumulation. Consistent with these observations, heating‐induced swelling tests (Figure 5l,m) revealed a pronounced suppression of gas generation in pouch cells employing the PNME‐1.5 electrolyte, reflecting mitigated parasitic reactions and improved thermal robustness at elevated temperatures. The concurrent reductions in temperature rise and gas evolution collectively demonstrate that PNME‐1.5 possesses the strongest capability to suppress thermal runaway among the tested electrolytes.

Beyond thermal stability, PNME‐1.5 also exhibits good mechanical resilience. The assembled pouch cell continued to power an LED light even after being cut and folded, without triggering short‐circuit failure (Figure 5n). This extreme abuse tolerance highlights the intrinsic safety advantages imparted by the gel polymer electrolyte architecture. In summary, the amide‐based DEGPE developed in this work delivers superior interfacial stability and intrinsic safety, as comprehensively benchmarked by a multi‐parameter radar chart comparison against state‐of‐the‐art literature systems (Figure 5o; Table S2). These advantages originate from the dual functionality of the amide monomer, which simultaneously enables efficient LiNO₃ solubilization and immobilizes reactive NMTFA within the polymer network. This synergistic design directly translates into outstanding cycling stability, high‐voltage compatibility, and markedly enhanced safety in practical lithium‐metal batteries.

3. Conclusion

In this study, we develop an amide‐monomer‐mediated LiNO₃ solubilization strategy to construct a DEGPE that simultaneously achieves high‐voltage stability, robust lithium‐metal compatibility, high ionic conductivity, and intrinsic flame retardancy. Central to this design is a molecularly programmed, cooperative hydrogen‐bonding interaction between the NME units in the PNME framework and nitrate anions, together with strong carbonyl coordination to Li⁺, which markedly enhances LiNO₃ solubility and enables its controlled interfacial decomposition to form uniform, inorganic‐rich, and mechanically robust SEI/CEI layers. These inorganic interphases effectively regulate Li⁺ transport and suppress parasitic reactions at both electrodes. Concurrently, PNME‐NMTFA hydrogen bonding immobilizes reactive free NMTFA molecules within the polymer network, inhibiting TM dissolution at high voltages and stabilizing the electrolyte against leakage. This dual‐function molecular design establishes a synergistic regulation of solvation chemistry and interfacial evolution, overcoming long‐standing trade‐offs in deep‐eutectic electrolyte systems. Consequently, this design enables DEEs to simultaneously deliver long‐term cycling stability, exceptional high‐temperature tolerance, and intrinsic safety, a combination that is essential for the practical realization of high‐energy‐density batteries. This work establishes a molecular engineering paradigm for electrolytes, linking solvation and interphase chemistry. It enables a shift from passive solvent formulation to active regulation via a polymer framework, offering key insights and a practical route to high‐performance, safe LMBs.

Conflicts of Interest

The authors declare no conflict of interest.

Supporting information

Supporting File 1: adma73748‐sup‐0001‐SuppMat.docx.

Supporting File 2: adma73748‐sup‐0002‐VideoS1.mp4.

Supporting File 3: adma73748‐sup‐0003‐VideoS2.mp4.

Data Availability Statement

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