Energetic All‐Polymer Fiber Batteries Enabled by Interface‐Interlocked Water‐In‐Network Electrolytes for Wearable Electronics

ABSTRACT

All‐polymer batteries configured in a fibrous form by integrating polymeric electrodes with aqueous quasi‐solid polymer electrolytes (AQPEs) represent a promising solution for wearable electronics with safety and sustainability. However, their practical development is constrained by the narrow electrochemical stability window (ESW) of AQPEs and the structural instability of electrode–electrolyte interfaces under deformation. Here, we present a “water‐in‐network” (WIN) electrolyte that precisely engineers water activity and dynamics by modulating the crosslinking density of the polymer network. We unanticipatedly discover a distinct “water confinement” effect wherein water activity exhibits a non‐monotonic dependence on crosslinking density, originating from the structural evolution of the network. Benefiting from this mechanism, the ESW significantly expands to 3.4 V, endowing an all‐polymer sodium‐ion fiber battery with an energy density of 92.4 Wh kg⁻¹. Crucially, in situ interfacial polymerization following electrolyte pre‐infiltration forms a mechanically interlocked electrode–electrolyte interface, effectively suppressing water‐induced delamination and preserving 78% capacity after 12 000 bending cycles.When seamlessly woven into a shirt, the fiber batteries can power the fabric‐based chemical sensor, enabling real‐time, on‐body health monitoring across various activities.

A “water‐in‐network” strategy that simultaneously addresses narrow electrochemical stability windows and interfacial instability. By modulating crosslinking density to engineer water confinement and employing in situ polymerization to construct mechanically interlocked interfaces that suppress water‐induced delamination, this approach yields an all‐polymer sodium‐ion fiber battery featuring high energy density and exceptional deformability for wearable applications.

1. Introduction

With the rapid development of flexible and wearable electronic devices, energy storage systems must deliver not only high energy density and mechanical resilience but also inherent breathability and operational safety [1, 2, 3]. In contrast to rigid lithium‐ion batteries, devices based on a fibrous architecture present a promising path forward, offering superior wearing comfort and seamless textile integration [4, 5, 6, 7, 8, 9]. Yet, the full potential of such fiber‐based devices remains constrained by the prevalent reliance on liquid electrolytes. While liquid electrolytes offer high ionic conductivity, their complex injection and sealing processes, susceptibility to leakage, and volatilization risks severely hinder their practical implementation in flexible energy storage devices [10, 11].

All‐polymer batteries, composed of polymeric electrodes and aqueous quasi‐solid polymer electrolytes (AQPEs), offer a pragmatic solution by merging the high ionic conductivity of aqueous electrolytes with the mechanical robustness of solid matrices, thereby enabling safer and more reliable wearable energy platforms [12, 13, 14]. However, the application of AQPEs remains largely limited to a few inorganic electrode systems, primarily because AQPEs typically inherit the narrow electrochemical stability window (ESW) of water, which restricts the stable operation of polymer electrodes [15, 16, 17]. We recently demonstrated that in highly concentrated polymer aqueous solutions, the dense hydrogen‐bonded network formed between polymer chains and water molecules can significantly suppress hydrogen evolution, thereby expanding the ESW [18]. Yet, extending this concept to quasi‐solid‐state systems remains challenging. The transition from a liquid state to a crosslinked polymer network fundamentally alters the water environment due to steric confinement and restricted polymer chain dynamics [19, 20]. Moreover, unlike liquids that naturally wet electrode surfaces, AQPEs struggle to maintain a stable and highly conformal electrode–electrolyte interface under long‐term cycling and mechanical deformation due to the presence of interfacial water between the polymer electrolyte and the electrode [21, 22]. Therefore, realizing high‐performance fiber batteries demands a preparation strategy that simultaneously modulates water activity and engineers a robust electrode–electrolyte interface.

Herein, we report a “water‐in‐network” (WIN) electrolyte to regulate water activity and dynamics via modulation of the polymer network structure, enabling an energetic aqueous all‐polymer sodium‐ion fiber battery (ASFB) with polyaniline (PANI) symmetric electrodes. While examining the effect of crosslinking density on electrolyte properties, we discovered a water confinement effect within the polymer network, where electrolyte performance exhibits a non‐monotonic dependence on the crosslinking density of the polymer matrix. Although “water confinement” has attracted considerable attention in low‐dimensional materials and porous frameworks [23, 24, 25], this effect in polymer networks has remained largely unexplored. The optimized WIN electrolyte exhibits an expanded ESW of 3.4 V, far exceeding that of conventional aqueous quasi‐solid‐state systems [26, 27, 28, 29, 30, 31] (Table S1). Simultaneously, to tackle interfacial failure, the in situ polymerization constructs a mechanically interlocked interface and facilitates the formation of a compositionally graded solid–electrolyte interphase (SEI), effectively stabilizing the polymer electrode against water‐induced failure. Consequently, the ASFB exhibits a high energy density of 92.4 Wh kg⁻¹, exceptional mechanical durability with 78% retention after 12 000 bending cycles due to the mechanically interlocked interface, and intrinsic scalability for textile integration. This approach establishes a versatile platform for engineering the thermodynamics and dynamics of confined water, creating scalable, sustainable, and high‐performance batteries that advance the frontier of safe, sustainable, and monolithically integrated wearable power sources.

2. Results

2.1. Hierarchical Regulation for All‐Polymer Battery

To achieve a sustainable and safe energy storage solution for wearable electronics, we designed an ASFB with a “WIN” electrolyte that combines high energy characteristics with mechanical flexibility through multiscale regulation spanning molecular interactions to device‐level integration (Figure 1). At the macroscopic level (Figure 1a), the ASFB serves as a weavable power supply that can be directly integrated into a smart textile. This smart textile exhibits high breathability (2358 mm s⁻¹) and excellent flexibility, enabling a self‐powered health management platform that continuously drives wearable fiber electronics. At the device level (Figure 1b), infiltration‐curing of the electrolyte onto an aligned, porous PANI/carbon nanotube (CNT) fiber forms a robust electrode–electrolyte interface that withstands large mechanical deformation. The resulting ASFB delivers a specific capacity of 115.5 mAh g⁻¹ and an energy density of 92.4 Wh kg⁻¹ at 1 C, outperforming reported sodium‐ion fiber batteries (Figure 1c) [1, 32, 33, 34, 35]. At the molecular level, the “WIN” electrolytes were obtained through UV‐initiated photopolymerization of a precursor solution composed of poly(ethylene glycol) methyl ether methacrylate (PEGMA) as the main monomer and bisphenol‐A ethoxylate dimethacrylate (BEMA) as the crosslinker, together with NaTFSI as the electrolyte salt, water as the solvent, and fluoroethylene carbonate (FEC) as an additive (Figure 1d). Tuning the crosslinking density of polymer network modulates water activity via confinement effects, effectively suppressing parasitic reactions and expanding the ESW.

FIGURE 1.

Design concept and interfacial engineering of ASFB with the “WIN” electrolytes. (a), Schematic illustration of the structure and application of the ASFBs. PANI as electrodes are first deposited on the CNT fiber current collectors to form fiber electrodes. The “WIN” electrolyte is then infiltrated and cured, followed by encapsulation to create a weavable power unit. The resulted ASFBs can be further knitted into a breathable and flexible textile power source. (b), Comparison between a dense, conformal electrode–electrolyte interface formed by infiltration‐cured (right) and a defective interphase resulting from surface‐cured (left) strategy. (c), Energy density and specific capacity of the ASFB with the “WIN” electrolytes compared with reported previous aqueous sodium‐ion fiber batteries. (d), Schematic of the “WIN” electrolyte polymer network evolution with increasing crosslinking density.

2.2. “Water‐In‐Network” Electrolyte Design

Previous studies on polymer‐aqueous electrolytes (PAEs) have predominantly focused on parameters such as salt and polymer concentrations, and more recently, molecular weight and end‐group [30, 36, 37, 38]. However, the specific influence of crosslinking density on ESW in polymer electrolytes has remained largely unexplored. Here, we systematically tuned the BEMA crosslinker content (0, 1.6, 3.2, and 4.8 wt.%) of PAEs to probe the structural role of the polymer network and evaluated the corresponding ESW using a 2 m NaTFSI polymer electrolyte containing 6 wt.% H₂O (Figure S1 and Figure 2a). All samples exhibit solid‐like behavior, which is verified by rheological measurements with the storage modulus (G′) substantially exceeding the loss modulus (G″) (Figure 2b). Compared with the PAEs without BEMA, the introduction of 1.6 wt.% BEMA greatly expanded the ESW from 2.2 to 3.1 V, and further increasing the content to 3.2 wt.% expanded the ESW to 3.4 V. However, raising the BEMA content to 4.8 wt.% unexpectedly reduced the ESW to 3.0 V (Figure 2a and Figure S2). The “WIN” electrolyte exhibited an ESW exceeding 3.0 V, and the same crosslinking‐density‐dependent evolution was observed even in the absence of FEC, a standard SEI‐forming additive (Figure S3). Evaluating neat polymer networks with varying crosslinking densities in control experiments (Figure S4a) reveals that the polymer matrix itself has a negligible intrinsic effect on the ESW.

FIGURE 2.

Water‐confinement mechanism in “WIN” electrolytes. a), The ESW of “WIN” electrolytes with varying BEMA contents (0, 1.6, 3.2, and 4.8 wt.%). (b and c), Rheological analysis (b) and FTIR spectra (c) of “WIN” electrolytes with varying BEMA contents (0, 1.6, 3.2, and 4.8 wt.%). (d and e), Raman spectra (d) and deconvolution of the O─H stretching Raman (e) of the “WIN” electrolyte containing 3.2 wt.% BEMA. Deconvolution of the O─H stretching Raman envelope into asymmetric H‐bonded (AS), symmetric H‐bonded (S), and non‐H‐bonded (N) components. (f), Fractional contributions of AS, S, and N components as a function of BEMA content. (g), LF‐NMR relaxation profiles with varying BEMA contents (0, 1.6, 3.2, and 4.8 wt.%) revealing water dynamics.

A combination of rheological and morphological analyses corroborates this structure‐induced water modulation. Rheological measurements reveal a monotonic increase in G′ with higher BEMA content, indicating enhanced crosslinking density and improved structural robustness [39, 40, 41] (Figure 2b). Scanning electron microscopy (SEM) further confirms progressive network densification as the BEMA content increases, revealing a clear transition from loosely packed domains to much denser structures (Figure S5). Given that polymer architectures evolve from loose linear chains to denser crosslinked networks, this structural transition is expected to influence water activity. At the optimal crosslinking density (3.2 wt.%), the network effectively disrupts the hydrogen‐bond network of water and lowers its activity. In contrast, excessive crosslinking (4.8 wt.%) may restrict polymer swelling [42, 43], driving water molecules to reaggregate into bulk‐like clusters that deteriorate the ESW. To elucidate the structural origin of this aqueous reaggregation, small‐angle x‐ray scattering (SAXS) and x‐ray diffraction (XRD) analyses were conducted (Figure S6). XRD shows no detectable crystalline peaks for all samples, excluding crystallization as the origin of the non‐monotonic ESW trend, whereas SAXS reveals markedly enhanced scattering intensity for the 4.8 wt.% BEMA sample at q > 0.02 Å⁻¹, indicative of increased nanoscale heterogeneity, possibly associated with local phase separation [44]. This structural heterogeneity suggests that excessive crosslinking may alter the local distribution and confinement environment of water, thereby influencing the ESW. Importantly, while the structural state of water dictates the ESW, proper water confinement is equally essential for ion mobility. As shown in Figure S4b, the neat crosslinked polymer electrolytes exhibit consistently low ionic conductivity (0.9 to 1.25 mS cm⁻¹) at different crosslinking densities. However, the introduction of water in the optimal “WIN” electrolyte significantly boosts the conductivity to 2.9 mS cm⁻¹. This contrast explicitly confirms that confined water is indispensable for sustaining rapid ion transport. Thus, an optimal crosslinking window exists that maximizes the water‐confinement effect while maintains solid‐like mechanics and adequate conductivity.

2.3. Mechanistic Insight Into Water Confinement

To understand the origin of the ESW variation with crosslinking density, we first carried out a series of spectroscopic analyses. Fourier transform infrared (FTIR) spectroscopy was used to track the evolution of the hydrogen‐bonding structure of water at different crosslinking densities. In the absence of BEMA, the O─H stretching bands at ∼3300 and ∼3600 cm⁻¹ correspond to bulk water and free water, respectively [45] (Figure 2c). When the BEMA content was increased to 3.2 wt.%, both bands shifted to higher wavenumbers (∼3350 and ∼3650 cm⁻¹), indicating weakened hydrogen bonding in bulk water as the water molecules became increasingly confined within the denser polymer network. At 4.8 wt.% BEMA, however, both bands shifted back to lower wavenumbers, which can be attributed to the reduced swelling capability of the highly crosslinked network. Limited swelling restricts water uptake and forces a portion of the water molecules out of the polymer matrix, leading them to reaggregate into bulk‐like clusters where stronger hydrogen bonding dominates [46, 47]. Deconvolution of the Raman O─H stretching envelope (3200–3700 cm^{− 1}) further supports this picture. In the optimal “WIN” electrolyte containing 3.2 wt.% BEMA, the populations of strongly symmetric H‐bonded (DDAA), partially coordinated (DDA and DA), and non‐H‐bonded water species were fundamentally reorganized (Figure 2d,e). The fraction of non‐H‐bonded water reached its maximum, while strongly H‐bonded water markedly decreased, along with a pronounced free‐water peak at 3701 cm⁻¹ (Figure S7). This can be ascribed to an appropriate crosslinking density, where strong polymer–water interactions disrupt the continuous hydrogen‐bond network of bulk water and substantially suppress water activity [48, 49]. Quantitative peak fitting showed that the fraction of non‐H‐bonded water increased sharply with enhanced crosslinking density, reaching a maximum at 3.2 wt.% BEMA (Figure 2f), while the strongly H‐bonded fraction continuously decreased. This redistribution of water from strongly H‐bonded to non‐H‐bonded states lowers water activity and consequently broadens the ESW.

We hypothesize that the relationship between non‐H‐bonded water and crosslinking density originates from changes in water dynamics under network induced water confinement. Furthermore, the water molecular mobilities can be evaluated by low‐field nuclear magnetic resonance (LF‐NMR) through transverse spin‐spin relaxation time (T₂ ) distributions. T₂ values of 0.1‐3, 3–20, and 150–2200 ms typically correspond to bound water, immobilized water, and bulk water, respectively [50]. Notably, as shown in Figure 2g and Figure S8, at the optimal density of 3.2 wt.% BEMA, the bulk‐water peak nearly disappears, and its area fraction reaches a minimum (3%) according to peak area fitting (Figure S9), which indicates that most bulk water has transitioned into immobilized states, thereby effectively minimizing free water activity. Conversely, the bulk water ratio rebounds at 4.8 wt.% BEMA, confirming that excessively tight networks expel water, which explains the narrowed ESW at higher crosslinking densities. Finally, Density functional theory (DFT) calculations and molecular modeling provide the theoretical mechanism for this widened ESW. Compared to H‐bonded H₂O clusters, non‐H‐bonded H₂O molecules exhibit a lowered HOMO and an elevated LUMO. This widened energy gap renders them thermodynamically less susceptible to oxidation and reduction (Figure S10). As shown in Figure S11, the introduction of the polymer reduces the average H‐bond number of water from 4.48 to 2.92, effectively disrupting the H‐bond network and broadening the density of states (DOS) bandgap from 3.66 to 4.23 eV. These results demonstrate that the transition to non‐H‐bonded states substantially lowers water activity, culminating in a wide ESW of 3.4 V.

Crucially, this water confinement enhances thermodynamic stability without compromising macroscopic ionic mobility. Raman spectral fitting, using the neat crosslinked polymer electrolyte as a reference, confirms that ether groups and water jointly participate in Na⁺ solvation in the WIN electrolyte, with ether groups contributing 72% (Figure S12). All these features show that the ESW is governed by confinement effects, where moderate crosslinking density minimizes water activity by restructuring the H‐bond network, while excessive crosslinking density limits swelling and induces bulk‐like water clustering that narrows the ESW.

2.4. A High‐Performance ASFB With Stable Interface

In fiber battery applications, interfacial instability between AQPEs and fiber electrodes under deformation remains a major bottleneck for device durability. We employed aligned CNT fibers with a hierarchical porous structure as 1D current collectors (Figure S13). This hierarchical porosity enables ready infiltration of the low‐viscosity electrolyte precursor (67.4 mPa·s, Figure S14). To minimize interfacial defects, we implemented an infiltration‐curing strategy, where the fiber electrodes were first immersed in the precursor solution of electrolyte, and subsequently polymerized, thereby densifying the electrode–electrolyte interphase. Benefiting from this architecture, the assembled ASFB delivered a high capacity of 115.5 mAh g⁻¹ and an energy density of 92.4 Wh kg⁻¹ at 1 C, surpassing most reported aqueous sodium fiber batteries [1, 32, 33, 34, 35] (Figure 1c and Table S2). In addition, the infiltration‐cured ASFB exhibited prolonged cycling stability, maintaining 86 mAh g⁻¹ (74% of initial capacity) after 330 cycles with an average Coulombic efficiency (CE) of 96% (Figure 3a,b). For comparison, we prepared a control device using a surface‐cured approach, where the electrolyte was polymerized immediately after injection. This control device suffered from rapid capacity fading, retaining only 42% of its initial capacity after 100 cycles with a low average CE of 79% (Figure 3a and Figure S15). Moreover, the surface‐cured device showed poor mechanical resilience (72% after 10 cycles at 180° bending, Figure S16), underscoring the critical role of infiltration‐cured interphase formation in ensuring both electrochemical and mechanical durability.

FIGURE 3.

Electrochemical properties and interphasial chemistry of ASFBs. (a), Cycling performance and CE of ASFBs fabricated by infiltration‐ and surface‐cured during charge‐discharge cycles. (b), Galvanostatic charge‐discharge curves of ASFBs at first, 10th, 50th, 100th, and 200th cycles. (c and d), Cross‐sectional SEM images (c) and corresponding EIS (d) of infiltration‐cured and surface‐cured ASFBs. Scale bar: 30 µm. e to g), XPS results of the SEI films on PANI anodes cycled in “WIN” electrolyte: F1s spectra at pristine, 10th, and 30th cycles (e), quantified atomic ratios of the elements in SEI (f), and Na1s spectra after 0, 1, and 3 min of Ar⁺ sputtering (g). (h) TOF‐SIMS 3D renders showing the depth‐dependent spatial distribution of various ions, including NaF⁻ and NaO⁻ within the SEI. (i), Rate performance at current densities from 1 C to 5 C.

Cross‐sectional SEM imaging further revealed that the infiltration‐cured “WIN” electrolyte forms intimate contact with the electrode surface and fully infiltrates the porous architecture. In contrast, the surface‐cured sample exhibits pronounced interfacial voids and defects that likely impede Na⁺ transport [51]. 3D visual evidence of this mechanical interlocking is provided by non‐destructive x‐ray micro‐computed tomography (micro‐CT) in Figure S17, Movies S1 and S2. It is seen that the “WIN” electrolyte seamlessly penetrates the fiber electrode, resulting in a mutually embedded 3D architecture. Such structural interpenetration provides the foundation for accommodating dynamic bending stresses and maintaining macroscopic interfacial stability. Consistently, Nyquist plots from electrochemical impedance spectroscopy (EIS) showed that the interfacial resistance of the infiltration‐cured device was significantly lower than that of the surface‐cured counterpart, indicating the formation of a more stable and compatible electrode–electrolyte interface (Figure 3d). The distribution of relaxation times (DRT) of the infiltration‐cured and surface‐cured devices further supports this conclusion (Figure S18). Two characteristic relaxation processes are displayed, the S1 peak (10⁻⁴ to 10⁰ s) corresponds to SEI formation and contact resistance, whereas the S2 peak (10⁰ to 10¹ s) reflects diffusion‐dominated charge‐transfer processes [52, 53]. For the infiltration‐cured electrolyte, the S1 peak intensity varied only minimally with relaxation time, suggesting a more stable SEI layer. Meanwhile, the S2 region exhibited shorter relaxation times and lower resistance, indicative of faster charge transfer. These results demonstrate that infiltration‐curing not only enhances physical interfacial conformity but also improves electrochemical kinetics, thereby enabling the superior performance of the ASFB.

2.5. ASFB With Dual‐Ion Doping Mechanism and Interphasic Chemistry

To elucidate the charge storage behavior of our ASFB, we analyzed the scan‐rate dependence of the cyclic voltammetry response (Figure S19). The relationship between current and scan rates follows the following equation:

$$i=a{v}^b$$ (1)

where i is peak current, v is potential scan rate and a and b are fitting parameters [54]. The parameter b serves as an indicator of the electrochemical behavior of the system. Specifically, the b‐value approaching 0.5 indicates a diffusion‐controlled process characteristic of battery‐type charge storage, whereas the b‐value close to 1 suggests a surface‐controlled process associated with capacitor‐like behavior. The b value of this all‐polymer device is 0.74, implying a mixed charge storage mechanism involving both diffusion and capacitive contributions.

To further quantify the electrochemical mechanism and delineate the relative contributions of capacitive and diffusion‐controlled processes across scan rates, we introduced another empirical relation:

$$I\left(v\right)=k_{1}v+k_2{v}^{0.5}$$ (2)

where I (v) is peak current, v is scan rate, k₁ and k₂ are proportionality constants representing surface‐controlled and diffusion‐controlled process, respectively [13]. When the scan rate was below 5 mV s⁻¹, the diffusion‐controlled contribution consistently dominated (Figure S19c), indicating that the device exhibits pronounced battery‐type behavior rather than capacitive characteristics, highlighting its potential for developing high‐energy polymer‐based aqueous fiber batteries. Operando FTIR synchronized with galvanostatic cycling confirmed reversible dual‐ion doping in PANI, whereby TFSI⁻ and Na⁺ were reversibly incorporated at the cathode and anode, respectively, in agreement with our group previous works [18]. Notably, this mechanism was demonstrated in a symmetric PANI fiber architecture using a “WIN” electrolyte (Figures S20 and S21).

The long‐term reversibility of this dual‐ion mechanism relies critically on a stable SEI on the PANI anode. A conformal layer composed of nanoscale particles on the electrode surface was observed from SEM images, consistent with the formation of an inorganic SEI [55], upon extended cycling, this layer without causing evident active material pulverization or macroscopic interfacial delamination (Figure S22). The chemical transformation of the SEI evolution was further tracked by x‐ray photoelectron spectroscopy (XPS). In the initial cycles, the F1s spectrum was dominated by ─CF_x (687 eV), whereas by the 30th cycle, a distinct Na─F peak (683.5 eV) emerged as the primary component. Concurrently, the Na1s spectrum evolved from Na─O interactions to a stable inorganic Na─F environment (Figure 3e and Figure S23), while C1s spectra indicated the progressive accumulation of carbonate species (Na₂CO₃) and sodium alkyl carbonate (R–OCO₂Na) species within the SEI [56]. To resolve the depth‐dependent architecture of this interphase, we performed XPS depth profiling with Ar⁺. As shown in Figure 3f, the O1s signal decreased with etching time, while F1s and Na1s signals increased and then plateaued. The C─F signal disappeared within 1 min of etching, C─O emerged, and C═O persisted throughout (Figure 3f,g and Figure S24). These spectra supported a compositionally graded SEI. More direct evidence for this differentiated spatial architecture was provided by time‐of‐flight secondary‐ion mass spectrometry (TOF‐SIMS) (Figure 3h and Figure S25). Visual 3D renders vividly revealed that inorganic fragments (F⁻, NaF⁻) were predominantly located in the outer layer of the SEI [57, 58]. In contrast, organic fragments (CO₂ ⁻) were enriched at the inner interface adjacent to the PANI electrode, consistent with the XPS depth profiles. Benefiting from the unique interphase chemistry induced by the “WIN” electrolyte, this compositionally graded interphase functions as a highly effective chemo‐mechanical buffer (Figure S26). It suppresses parasitic water activity while accommodating the volume fluctuations of fiber electrodes, fundamentally preserving interfacial structural integrity.

Crucially, the formation of this robust SEI fundamentally redirects the degradation pathway of the aqueous battery. While conventional aqueous systems typically fail rapidly due to structural collapse, the “WIN”‐induced SEI successfully preserves electrode integrity. The DRT analysis of interfacial resistance (Figure S27) reveals a progressive increase in charge‐transfer resistance (R_ct ) with cycling. As a result, the long‐term capacity decay (Figure 3a) is primarily associated with progressive sluggishness charge‐transfer kinetics. Robust interface engineering yields exceptional performance, allowing the ASFB to retain 58 mAh g⁻¹ even at 5 C (Figure 3i). Notably, when the current density returned to 1 C, a high capacity of 98 mAh g⁻¹ (86% of initial capacity) was retained. Although aqueous batteries typically suffer severe performance degradation near 0°C due to water freezing, the “WIN”‐based ASFB maintained a discharge capacity of 98 mAh g⁻¹ at 0.2 C even under harsh conditions at −10°C (86% of the initial capacity, Figure S28), demonstrating remarkable low‐temperature operational stability. This all‐weather tolerance originates from polymer‐confinement‐induced suppression of ice nucleation and stabilizes the non‐ice hydrogen‐bonding networks, as confirmed by differential scanning calorimetry and temperature‐dependent Raman analyses (Figure S29) [59, 60].

This general electrolyte design strategy can also be readily applied to conventional inorganic intercalation systems. By pairing a Na_0.44MnO₂ cathode with a NaTi₂(PO₄)₃ anode, the fiber battery maintains a discharge capacity of 58 mAh g⁻¹ at 1 C with a 72% capacity retention over 100 cycles (Figure S30). Compared with reported aqueous Na‐ion fiber batteries [32], this inorganic‐based battery delivers a higher discharge capacity with comparable cycling stability and an average Coulombic efficiency of 93%, confirming the broad compatibility of the WIN electrolyte to aqueous fiber batteries.

2.6. Mass Production for ASFB

The compatibility of the infiltration‐curing strategy with continuous manufacturing enabled the production of hundred‐meter‐scale ASFBs, facilitating seamless textile integration without compromising mechanical and electrochemical integrity (Figure 4a). During fabrication, two parallel CNT fibers were drawn under controlled tension and sequentially coated with electrode slurries, producing continuous fiber electrodes exceeding 100 m in length (Figure S31). Subsequently, the “WIN” electrolyte was applied using the same coating method, and then UV‐cured. The resulting anode and cathode fibers were twisted together, with the “WIN” electrolyte functioning simultaneously as an ion conductor and a separator. To ensure a continuous electrolyte layer and reliable ionic conduction, an additional cycle of electrolyte coating and UV curing was performed on the twisted assembly, followed by protective encapsulation.

FIGURE 4.

Continuous fabrication, scalable production and mechanical robustness of ASFBs. (a), Schematic of the set‐up for continuous ASFBs fabrication. (b), Mechanical stability of fiber electrodes, with resistance retention rates of more than 90% after 2500 cycles of twisting, bending and pressing. (c), Cycling stability of ASFBs under different bending angles (0, 45, 90, 135, 180°). (d), Electrochemical performance of the ASFBs as a function of repeated bending cycles at a fixed 90°. (e,f), Photographs of the ASFB textile, demonstrating the potential for mass production. Scale bar: 10 cm.

Owing to the inherent compliance of the polymer electrodes and “WIN” electrolyte, the ASFB exhibited excellent mechanical robustness. The “WIN” electrolyte allowed the fibers to withstand large bending and fully recover without fracture (Figure S32). The fiber electrodes showed minimal resistance variation under twisting, bending and pressing (ΔR/R₀ < 10%, Figure 4b). Consequently, the ASFBs exhibit exceptional flexibility and durability, maintaining stable charge‐discharge characteristics during 0°–180° bending (Figure 4c) and retaining 78% of their initial capacity after 12 000 bending cycles (Figure 4d). Even under knotting or repeated severe twisting, the device preserved structural integrity (Figure S33). These findings indicate that an all‐polymer, compliant architecture effectively mitigates mechanically induced performance decay, enabling the ASFBs to be coiled and seamlessly woven into textiles for practical applications (Figure 4e,f and Figure S34).

To demonstrate real‐world applicability, we integrated the ASFBs into a smart shirt designed for personalized health monitoring during activities such as running, cycling, and elliptical training. The garment integrates fiber sweat sensors in the axillary region, powered by an ASFB woven along the lateral seam (Figure 5a,b). Specifically, a 300‐cm fiber battery was woven into the fabric and integrated with biosensors and a flexible wireless power‐management module for real‐time data transmission (Figure 5c‐(i) and Figure S35). During physical activity, the system continuously monitored multiplexed biochemical signals in sweat, including Na⁺, K⁺, pH, lactate (Lac), uric acid (UA) and glucose (Glu), and wirelessly transmitted the readouts to a receiver (Figure 5d), supporting early diagnosis and preventive healthcare.

FIGURE 5.

System integration of ASFB textiles for health monitoring. (a), Illustration of a smart health‐management shirt with the ASFB textile integrated to power the fiber sensors. WE1 comprises Na⁺, K⁺, and pH fiber electrodes, whereas WE2 consists of lactate (Lac), uric acid (UA), and glucose (Glu) fiber electrodes. (b), System block diagram of smart shirt and the health monitoring interface. (c), Photograph of (i) the smart shirt integrated with the ASFB textile and (ii) air permeability of a conventional pouch‐type flexible battery and the ASFB‐based textile battery. Scale bar: 5 cm. (d), Real‐time multiplexed monitoring of sweat biomarkers recorded from multifiber biosensors powered by the ASFB textile. Working electrodes (WE), reference electrodes (RE).

Crucially, beyond functional integration, the smart shirt addresses the long‐standing challenge of wearing comfort. Breathability is a critical metric for the usability of textile electronics. Conventional e‐textiles relying on rigid, impermeable pouch cells often create locally occlusive areas, leading to skin redness and discomfort (Figure 5c‐(ii)). In sharp contrast, the ASFB‐based textile maintains high air permeability, ensuring superior thermal and moisture management while fully preserving system functionality. This combination of scalable manufacturing, low‐energy solution processing with UV curing, and intrinsic textile compatibility offers a viable and sustainable pathway for next‐generation wearable energy systems.

3. Conclusion

In summary, we have demonstrated an energetic ASFB enabled by a “WIN” electrolyte design that regulates the thermodynamics and dynamics of confined water. Critically, our investigation uncovers a non‐monotonic dependence of water activity on the polymer network. At intermediate crosslinking levels, network confinement disrupts water clusters and stabilizes weakly hydrogen‐bonded water molecules, thereby broadening the ESW to 3.4 V. In contrast, when the crosslinking density becomes too high, the reduced free volume of the polymer network promotes the reaggregation of water molecules into bulk‐like clusters, leading to a narrowing of the ESW. Moreover, a mechanically interlocked electrode–electrolyte interface is realized via the pre‐infiltration of porous CNT‐based composite electrodes followed by in situ interfacial polymerization. This process simultaneously fosters a compositionally graded SEI that effectively suppresses water‐induced interfacial failure. This multiscale design yields an ASFB delivering a high specific capacity of 115.5 mAh g⁻¹ and an energy density of 92.4 Wh kg⁻¹, while supporting hundred‐meter‐scale continuous fabrication. Moreover, the fiber batteries successfully powered chemical fiber sensors, creating a smart and breathable shirt for real‐time health monitoring. Overall, this work establishes a generalizable protocol for manipulating the thermodynamics and dynamics of confined water via polymer network architecture, offering a versatile pathway for designing energetic, stable, and sustainable energy storage systems suitable for next‐generation wearables.

4. Experimental Section

4.1. Preparation of the “WIN” Electrolyte

The aqueous quasi‐solid polymer electrolyte was prepared via one‐step photopolymerization. The precursor comprised poly(ethylene glycol) methyl ether methacrylate (PEGMA, 28.9 wt.%) as the main monomer and bisphenol‐A ethoxylate dimethacrylate (BEMA) as the crosslinker, along with NaTFSI (2 mol/ kg) as the electrolyte salt, water as the solvent, and FEC as an additive. With 2‐hydroxy‐2‐methylpropiophenone as the photoinitiator, the flowable precursor was rapidly photopolymerised to yield a free‐standing aqueous quasi‐solid polymer electrolyte (Figure S36). All solutions were dried over molecular sieves and stored in the dark prior to use. Dissolved oxygen was removed by bubbling Ar for 10 min, and the final precursor was prepared in an Ar‐filled glovebox. To investigate the compositional effect of crosslinking density and water contents, precursor solutions were prepared with varying BEMA (0, 1.6, 3.2, 4.8 wt.%) or H₂O (0, 3, 6, 9, and 12 wt.%) contents under otherwise identical conditions.

4.2. Preparation of the PANI Fiber Electrode

The CNT fiber electrodes were prepared using the floating chemical vapor deposition method according to the literature [61]. The active materials (PANI, synthesis details in Method S1), conductive agent (KB), and binders (PVDF) were dispersed in NMP and mechanically stirred for 24 h to form a uniform slurry with a solid content of 10%. The slurry was then coated onto CNT fibers with 3 m min⁻¹, followed by twisting at 100 r min⁻¹ for 2 min and vacuum drying at 80°C for 12 h to obtain the composite electrode.

4.3. Fabrication of ASFB

For infiltration‐cured ASFBs, the PANI fiber electrodes were first immersed in the electrolyte precursor solution for 1 h to ensure complete infiltration. Subsequently, the assembly was exposed to UV irradiation (7 W, 365 nm, 15 min) to initiate photopolymerisation, resulting in compact electrode–electrolyte interfaces. The 300 cm‐long anode and cathode fibers were then aligned in parallel, twisted together at 100 r min⁻¹ for 1.5 min using a motorized stage, and finally encapsulated within a heat‐shrinkable tube. All assembly procedures were performed in an Ar‐filled glovebox. For surface‐cured ASFBs, the electrolyte was cured immediately after the electrodes were assembled, omitting the infiltration process.

4.4. Assembly of Multifiber Biosensors

The fabrication of multifiber biosensors followed our previous work [62]. Specifically, silicone fiber substrates were fabricated by curing Ecoflex in polyethylene terephthalate tubes. Ion‐selective (Na⁺, K⁺) and pH fiber electrodes were helically wound around a silicone fiber to form working electrode 1, whereas enzymatic sensors (Lac, UA and Glu) were assembled on a separate fiber to constitute working electrode 2. To prevent short‐circuiting and minimize signal crosstalk, non‐sensing regions were insulated with a hydrophobic silicone layer. The functionalized fibers were then stitched onto a textile substrate alongside a reference electrode for biosignal monitoring. Detailed fabrication procedures are provided in the Supporting Information.

Funding

This work is financially supported by the Shenzhen Medical Research Fund (D2502004), National Natural Science Foundation of China (22575070), Guangdong Basic and Applied Basic Research Foundation (2023A1515011332).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: adma73599‐sup‐0001‐SuppMat.docx.

Supporting Movie 1: adma73599‐sup‐0002‐MovieS1.mp4.

Supporting Movie 2: adma73599‐sup‐0003‐MovieS2.mp4.

Data Availability Statement

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