Integrated Thick Flexible Iodine Electrode with Polyiodide Immobilization Hydrogelation Interphase for Practical Ah-Level Aqueous Zinc-Iodine Pouch Cells

Abstract

Aqueous zinc-iodine batteries are promising candidates for grid-scale energy storage, but their practical application is hindered by low areal-capacity and shuttling of polar polyiodides on positive-electrode side. Herein, a high-areal-capacity iodine electrode (up to 14 mAh cm⁻²) with mechanical flexibility is developed through an integrated thick electrode processing technology, which synergistically combines enhanced ion transport with a polyiodide-immobilization hydrogelation interphase. A viscoelastic polyamide thermoplastic elastomer network is in situ formed throughout electrode bulk via a rapid phase inversion method, acting as structural skeleton to preserve transient slurry-level uniformity, concurrently ensuring interconnected ion transport network and mechanical robustness even under high thickness (>2 mm). Moreover, the polyamide interphase, rich in functional amide groups, demonstrates strong binding affinity for polyiodides, effectively immobilizing them within the hydrogelation layer and facilitating their reutilization during redox cycling. Consequently, the high-loading flexible electrode (70 mg cm⁻²) delivers an average areal-capacity of 11.5 mAh cm⁻² at 0.62 A g⁻¹ with over 1600 cycles. Notably, the integrated electrode can be directly used to assemble Ah-level single-layer pouch cells (~1.6 Ah), demonstrating stable operation over 1200 cycles and validating practical applications in powering wearable electronics and high-power devices, thereby paving the way toward high-energy-density zinc-iodine batteries.

Aqueous zinc-iodine batteries face low areal capacity and polyiodide shuttling. Here, authors develop an integrated thick iodine electrode with a polyamide binder that enables flexible, robust electrodes with polyiodide immobilization.

Introduction

The rapid exploitation of renewable energy has substantially increased global demand for efficient and safe energy storage systems^1,2. However, the widespread adoption of lithium-ion batteries is constrained by the flammability of organic electrolytes and their relatively high cost^3–5. In contrast, aqueous zinc metal batteries have garnered growing attention as promising alternatives for grid-scale storage due to the intrinsic safety and environmental compatibility of aqueous electrolytes, the low redox potential of Zn (−0.76 V vs. SHE), and the high abundance of zinc^6–9. Among them, aqueous zinc-iodine batteries (AZIBs), operating through a reversible conversion reaction between I₂ and I⁻, are particularly attractive due to the favorable redox potential (0.536 V vs. SHE), high theoretical capacity (211 mAh g⁻¹), and the natural abundance of iodine (55 μg L⁻¹ in ocean)^10,11. Nevertheless, the practical deployment of iodine-positive electrodes is significantly impeded by limited areal capacity, as well as the spontaneous formation of soluble polyiodide intermediates (e.g., I₃⁻, I₅⁻) during cycling, which cause irreversible capacity loss and Zn electrode corrosion (Fig. 1a)^12,13.

Fig. 1. Design of TPAE-based integrated thick electrode.

a Schematic diagram of the polyiodide dissolution issues encountered in dry-process PTFE@I₂AC electrodes, compared with the polyiodide immobilization in TPAE@I₂AC electrodes. b Schematic diagram of the phase inversion, along with photographs of the TPAE@I₂AC electrode embedded with Ti mesh current collectors.

From a practical application perspective, achieving a high areal capacity of iodine positive electrode (e.g., > 10 mAh cm⁻²) is a direct route to improving cell-level gravimetric energy density, as it reduces the mass fraction of inactive components (e.g., current collectors, separators, and casing) and provides greater design flexibility in optimizing the N/P ratio^14,15. However, fabricating such high-loading iodine electrodes via conventional slurry-casting is challenging, as binder migration during drying often results in microstructural heterogeneity, cracking, and poor adhesion^16–20. Recently, dry-processing techniques have been explored to overcome these mechanical limitations^21–23. For instance, Qiao et al. pioneered a PTFE (polytetrafluoroethylene)-based dry-processing method to prepare a ZnI₂ positive electrode with an areal capacity of 7.9 mAh cm⁻²²¹. Notably, as electrode loading further increases (over 50 mg cm⁻²), the limited pore-filling electrolyte becomes insufficient to sustain uniform reaction activity, and the system increasingly relies on ion supply from the bulk electrolyte^24,25. However, the extended transport paths, combined with the hydrophobic PTFE matrix, render ion migration from the bulk electrolyte into the electrode interior rate-limiting at high rates, fundamentally restricting iodine-conversion kinetics and rate performance^26,27.

During electrochemical cycling, soluble polyiodide intermediates (e.g., I₃⁻, I₅⁻), which are more polar and bind more weakly to nonpolar activated carbon (AC), will generate spontaneously due to the coexistence of I₂ and I⁻, and are easily extracted into the electrolyte by the polar aqueous interphase^13,28–30. Under high-iodine-loading conditions, where ionic transport within the bulk electrode becomes a rate-limiting step, redox reactions and active iodine species tend to localize near the electrode surface, further aggravating polyiodide shuttling^31–33. Meanwhile, chemically inert binders such as PTFE or CMC (sodium carboxymethylcellulose) lack the ability to chemically immobilize polyiodides; once shuttled, these polyiodides are difficult to reclaim into the redox cycle, leading to continuous capacity fade and Zn corrosion^34,35. Conventional approaches have introduced various functional modifications (e.g., Fe, Co, Ni, N sites) onto AC surfaces to chemically adsorb polyiodides and mitigate shuttling in low-loading systems (typically < 2 mAh cm⁻²)^13,35–40. Although these strategies improve cycle life, they often involve complex synthesis steps and remain ineffective against interfacial diffusion. Moreover, their cost-effectiveness and efficacy in practical thick electrodes are still unclear^21,41. Therefore, the central challenge for a reliable iodine positive electrode is to concurrently enable efficient ion transport throughout a thick electrode while chemically confining polyiodides within a mechanically robust architecture.

Herein, we propose an integrated thick-electrode strategy that simultaneously addresses mechanical and electrochemical challenges. A polyamide thermoplastic elastomer (TPAE) binder is introduced as both a flexible structural framework and a polyiodide-confining hydrogelation interphase (Fig. 1b). From the electrode structure view, the rapid and uniform solidification via phase inversion method (no drying step) realizes the transient slurry-level electrode uniformity reservation, thereby ensuring mechanical integrity and creating an interconnected ion transport network throughout the thick electrode, enhancing Zn²⁺ transport and iodine conversion kinetics. This also allows for direct in situ embedding of the current collector, ensuring uniform contact and high electronic conductivity across large areas. From the interfacial specie interaction perspective, abundant amide groups in the TPAE matrix exhibit strong binding affinity for polyiodide anions, effectively immobilizing them within the viscoelastic hydrogelation interphase. Moreover, the conformal interface between the TPAE hydrogel and AC host shortens polyiodide diffusion pathways and promotes their reutilization in the redox process. Consequently, the flexible, high-loading iodine positive electrode (70 mg cm⁻²) achieves a high areal capacity of 14 mAh cm⁻² and good rate performance (9 mAh cm⁻² at 1.06 A g⁻¹). Importantly, the integrated electrode can be directly used to assemble Ah-level single-layer pouch cells, eliminating the need for complex multilayer electrode stacking, providing a practical cell-level capacity of 1.6 Ah and stable cycling over 1200 cycles, demonstrating significant potential for scalable AZIB applications.

Results

Enhanced mechanical performance and fast ion-transport capability in thick electrodes

The integrated thick TPAE@I₂AC electrode was fabricated through an aqueous phase inversion process. This method leverages the reversible hydrogen bonding of polar amide groups within the TPAE^42–44. In concentrated LiTFSI aqueous solution (10 m), Li⁺ cations coordinate with the electron-rich carbonyl groups, while electron-deficient -NH groups engage in weak hydrogen bonding with the TFSI⁻ anions. These associated Li⁺-TFSI^- ion pairs effectively disrupt the intrinsic hydrogen-bonding network between polyamide (PA) chains, leading to the dissolution of TPAE^42,43. When the slurry containing AC@I₂, conductive agents, and TPAE binder is immersed in pure water, LiTFSI immediately diffuses into the aqueous phase, allowing the reformation of strong hydrogen bonds between PA chains. This contributes to rapid and homogeneous phase inversion, transforming the homogenous slurry into a robust solid-state electrode. As presented in Fig. 2a, b, the resulting TPAE@I₂AC electrode maintains good structural integrity without delamination or cracking at a millimeter-scale thickness. Furthermore, TOF-SIMS analysis (Fig. 2i and Supplementary Fig. 1) confirms a highly uniform distribution of characteristic ionic fragments (C₆⁻ from activated carbon, CNO⁻ and CN⁻ from TPAE, and I⁻ from iodine) and an intact morphology (Supplementary Fig. 2), demonstrating the homogeneous chemical distribution achieved by the phase inversion.

Fig. 2. Enhanced mechanical performance and fast ion-transport capability in thick electrodes.

a, b SEM images of TPAE@I₂AC electrode. (c) Micro-CT images of TPAE@I₂AC electrode and (d) pore distribution. e, f SEM images of PTFE@I₂AC electrode. g Micro-CT images of PTFE@I₂AC electrode and h pore distribution. i 3D distribution of the characteristic ionic fragments in the TPAE@I₂AC electrode acquired by TOF-SIMS (time-of-flight secondary ion mass spectrometry). j The contact angle tests between aqueous ZnSO₄ solution and PTFE@I₂AC electrode or TPAE@I₂AC electrode. k The tensile testing results of various electrodes. l The CV curves of Zn | |TPAE@I₂AC full coin cell at different scanning rates, and m corresponding fitting results between oxidation peak current and scan rates. n The ion-diffusion testing results for various electrodes. o The rate performance of full coin cells with various positive electrodes (using 100 μm Zn foil as negative electrode) and p corresponding voltage hysteresis analysis (calculated from the voltage difference at the mid-point capacity). Based on the theoretical specific capacity of I₂, 1 C corresponds to a specific current of 0.21 A g⁻¹.

According to previous reports on TPAE dissolution, effective disruption of the interchain hydrogen bonds in TPAE requires sufficiently concentrated dissolution salt to form efficient ion pairing (Supplementary Fig. 3), with the cation possessing adequate Lewis acidity and the anion having strong affinity for the amide N-H groups, a sufficiently large ionic volume, and chaotropic character^42,43,45. For instance, Zn(TFSI)₂ can also be used to dissolve TPAE and prepare TPAE-based thick electrodes (Supplementary Fig. 4 and Supplementary Fig. 5). However, divalent Zn²⁺ tends to remain trapped within the PA matrix during the hydrogen bond reformation stage, hindering the complete removal of Zn(TFSI)₂ (Supplementary Tables 1 and 2). The locally residual TFSI^-, owing to its typical chaotropic nature and low charge density (Supplementary Fig. 6), increases the solubility of I₂ in the electrolyte (Supplementary Fig. 7) and accelerates polyiodide diffusion (Supplementary Fig. 8), thereby reducing the Coulombic efficiency of the cell (Supplementary Figs. 5, 9, and 10)⁴⁶. In contrast, LiTFSI can be effectively removed from the electrode via washing after phase inversion (Supplementary Figs. 11, 12, Supplementary Tables 3, 4) due to the monovalent nature of Li⁺, preventing its participation in electrochemical processes during battery operation.

From a structural perspective, the transient slurry-level electrode uniformity reservation of phase-inversion process enables a continuous internal architecture with homogeneously distributed porosity in TPAE@I₂AC, as confirmed by micro-CT (micro-computed tomography) analysis (Fig. 2c, d). Although the dry-processed PTFE@I₂AC electrode is capable of forming an integrated structure via fibrillated PTFE binders (Supplementary Fig. 13), it exhibits a denser and less porous morphology (Fig. 2e, f). As presented in Fig. 2g, h, the obvious active material agglomeration and heterogeneous, elongated pore defects can be observed within the PTFE@I₂AC electrode, featuring that significantly hinders uniform ion transport in thick electrodes. These structural defects originate from the mechanical shearing and calendaring inherent to the dry-processing route, which preclude fine control over pore architecture. Notably, excessive fibrillation of PTFE must also be avoided to preserve electrode functionality, imposing a further constraint on structural design²². As a result, the TPAE@I₂AC electrode demonstrates good mechanical robustness, and the tensile testing shows that the TPAE@I₂AC electrode can withstand a tensile strain of up to 34% (Fig. 2k). This stems from a hydrogen-bonded network formed by flexible polyamide segments, which endows the freestanding electrode with the ability to withstand significant stretching and twisting without failure (Supplementary Fig. 14). In comparison, conventional wet-processed CMC-based electrodes are prone to severe cracking and delamination caused by binder migration during drying (Supplementary Figs. 15 and 16), rendering them unsuitable for high-loading applications (Supplementary Fig. 17).

It should be noted that the inherent chemical inertness and strong hydrophobicity of PTFE result in poor electrolyte wettability, as demonstrated by a high contact angle of 116° between the PTFE@I₂AC electrode and aqueous ZnSO₄ electrolyte (Fig. 2j). In contrast, the TPAE@I₂AC electrode exhibits strong hydrophilicity, with a much lower contact angle of only 32°. In fact, this enhanced wettability, combined with its uniformly interconnected pore structure, is highly conducive to efficient ion transport within the electrode bulk. To quantitatively compare ion-transport properties, the ion-diffusion experiments are conducted using an H-cell configuration (Supplementary Fig. 18). As presented in Fig. 2n, the diffusion rate of Zn²⁺ passing through the TPAE@I₂AC electrode from the 2 M ZnSO₄ solution to the pure water phase reaches approximately 0.1398 mmol h⁻¹ cm⁻², whereas the PTFE@I₂AC electrode shows a significantly lower rate of only ~0.00885 mmol h⁻¹ cm⁻². This corresponds to a 16-fold enhancement in ion transport kinetics for the TPAE@I₂AC electrode, a key factor that underpins its better high‑rate performance.

Furthermore, rate-dependent cyclic voltammetry (CV) measurements are performed to quantitatively assess the iodine conversion kinetics in high-mass-loading electrodes. As illustrated in Fig. 2l and Supplementary Fig. 19, the current responses of both the TPAE@I₂AC and PTFE@I₂AC electrodes are comparable at a low scan rate (0.1 mV s⁻¹). However, as the scan rate increases, the peak current of the PTFE@I₂AC electrode increases significantly more slowly than that of the TPAE@I₂AC, indicating the constrained iodine conversion kinetics in PTFE@I₂AC. Fitting the power-law relationship between peak current (i_p) and scan rate (v), expressed as log(iₚ) = log(a) + b × log(v), yields a b-value of 0.551 for the oxidation process of the PTFE@I₂AC electrode (Fig. 2m), which is consistent with the typical diffusion-limited behavior. In contrast, the TPAE@I₂AC electrode shows a higher b-value of ~0.744, suggesting a mixed control by both ion diffusion and surface-mediated processes (iodine redox on the AC surface). Complementary in situ electrochemical impedance spectroscopy (EIS) reveals a pronounced increase in charge‑transfer resistance (R_ct) for the Zn | |PTFE@I₂AC full cell at the end of discharge (0.8 V, Supplementary Fig. 20a), which is attributed to the significant Zn²⁺ accumulation at the positive electrode surface caused by restricted ion migration in the PTFE-based electrode. In contrast, the impedance of Zn | |TPAE@I₂AC maintains stable throughout the entire cycling (Supplementary Fig. 20b), confirming that the pore structure and better wettability of TPAE electrode effectively alleviate ion-transport limitations, reducing the diffusion-dominated kinetic constraints in thick electrodes.

According to the rate performance tests (Fig. 2o and Supplementary Fig. 21), both the PTFE@I₂AC and TPAE@I₂AC electrodes deliver comparable specific capacities (~185 mAh g⁻¹ and ~201 mAh g⁻¹, respectively) at a low rate of 0.5 C (0.106 A g⁻¹). However, as the specific current increases, the PTFE@I₂AC electrode undergoes significant capacity decay, retaining only about ~106 mAh g⁻¹ at 3 C (0.63 A g⁻¹) and ~70 mAh g⁻¹ at 5 C (1.06 A g⁻¹). In contrast, the Zn | |TPAE@I₂AC full cell shows lower voltage hysteresis (Fig. 2p), and still maintains high specific capacities of approximately ~171 mAh g⁻¹ at 3 C (0.63 A g⁻¹) and ~138 mAh g⁻¹ at 5 C (1.06 A g⁻¹), corresponding to areal capacities of ~11.9 mAh cm⁻² and 9.6 mAh cm⁻² with a high-mass-loading of 69.5 mg cm⁻². As presented in the dQ/dV analysis (Supplementary Fig. 22), with increasing current rates, the PTFE@I₂AC electrode displays broader peak spacing and reduced peak intensity compared to the TPAE@I₂AC, indicating slower reaction kinetics and increased polarization. Even with the ZnI₂ as the active material, the Zn | |ZnI₂@AC full cell still shows unsatisfactory rate performance at high-mass-loading conditions (Supplementary Fig. 23). Notably, under a higher loading of 80.5 mg cm⁻², the TPAE@I₂AC electrode maintains specific capacities of 155.8 mAh g⁻¹ and 117.2 mAh g⁻¹ at 3 C (0.63 A g⁻¹) and 5 C (1.06 A g⁻¹), respectively (Supplementary Fig. 24). Even at an high iodine loading exceeding 100 mg cm⁻², the electrode continues to cycle stably, delivering an areal capacity over 20 mAh cm⁻² (Supplementary Fig. 25). These results underscore the TPAE matrix’s critical role in preserving slurry-level electrode uniformity, interconnected ion transport network, and ensuring good wettability for achieving fast reaction kinetics in thick electrodes.

Investigation of polyiodide immobilization by TPAE

Given the limited solubility of iodine in water (only 0.33 g L⁻¹)⁴⁷, elemental iodine (I₂) is effectively confined to the hydrophobic, nonpolar surface of activated carbon (AC) through van der Waals interactions and physical entrapment within its porous structure⁴⁸. However, in practical zinc‑iodine systems where I₂ and I⁻ coexist, soluble polyiodide intermediates (e.g., I₃⁻, I₅⁻) are spontaneously generated^29,30. These negatively charged polyiodides exhibit higher polarity and consequently show much weaker affinity for the nonpolar AC surface. Driven by the polar, water‑rich interphase, they readily desorb and migrate into the electrolyte, thereby triggering the shuttle effect (Supplementary Figs. 26 and 27).

Fortunately, the polar amide groups in TPAE enable specific polyiodide confinement through the formation of hydrogen bonds between the electron-deficient -NH groups and polyiodide anions, analogous to the mechanism by which TPAE dissolution is mediated by associated Li⁺-TFSI⁻ ion pairs. This chemical affinity is demonstrated in a visual adsorption experiment (Fig. 3a). When a Ti@TPAE substrate (TPAE hydrogel film fabricated on a Ti foil via phase inversion) is immersed in a solution containing polyiodides (prepared by adding I₂ to ZnI₂), the color of the solution fades almost completely within 300 minutes. Correspondingly, in situ UV-Vis spectroscopy shows that the characteristic I₃⁻ absorption intensity decreases rapidly upon introduction of the Ti@TPAE film, nearly vanishing within 80 minutes (Fig. 3b). In contrast, control samples of Ti@PTFE and Ti@CMC show no significant change in the I₃⁻ signal (Fig. 3c and Supplementary Fig. 28), confirming their chemical inertness towards polyiodides. Moreover, the post-adsorption stability of TPAE (denoted TPAE@I₃⁻) was further evaluated: after polyiodide saturation, the initially white TPAE film turned black; When immersed in pure ZnSO₄ solution, no colored species leached into the solution over 240 minutes (Fig. 3d), and in situ UV-vis spectra showed no new polyiodide-related peaks (Supplementary Fig. 29), confirming strong immobilization.

Fig. 3. Investigation of polyiodide immobilization by TPAE.

a Visualized monitoring of ZnI_n solution after introducing Ti@TPAE. b, c In situ UV-vis spectroscopy of ZnI_n solution after introducing (b) Ti@TPAE and (c) Ti@PTFE. d Visualized monitoring of ZnSO₄ solution after introducing TPAE@I₃^- film. e The snapshots of dynamic interactions between the PA chain and Zn²⁺ or I₃⁻. f The electrostatic potential isosurface plots showing the interactions between the amide monomer and Zn²⁺ or I₃⁻. g, h The potential energy of the amide group in the PA chain interacting with (g) I₃⁻ and (h) Zn²⁺. (i) The optimized structure models of amide monomer, long-chain PA, and double-chain PA interacting with I₃⁻.

Furthermore, theoretical calculations are performed to elucidate the adsorption mechanism of polyiodides on TPAE. As presented in Fig. 3f, ESP analysis reveals that within the amide group, the -NH moiety acts as a local positive charge center, while the carbonyl (C=O) group functions as a negative charge center, which facilitates the affinity with I₃⁻ and Zn²⁺, respectively. Statistical trajectory analysis from molecular dynamics (MD) simulations (Fig. 3e) confirms that both Zn²⁺ and I₃⁻ ions preferentially migrate toward the amide groups of the PA chains. Additionally, the calculated potential energy between I₃⁻ and the amide group (Fig. 3g) reaches a minimum value (−0.261 eV) at a distance of 3.75 Å, and remains negative as the distance increases. Similarly, Zn²⁺ exhibits a minimum interaction energy of −1.183 eV at a distance of 1.9 Å (Fig. 3h). The polar nature of the amide groups enables them to exhibit potential affinity for both zinc ions and polyiodides. This phenomenon is consistent with the observation that the poorly solvated Zn²⁺ ions in concentrated Zn(TFSI)₂ can effectively coordinate with the carbonyl groups, thereby disrupting the hydrogen bonds and achieving TPAE dissolution.

It is important to note that the binding of a single amide group to polyiodide alone does not appear sufficient to account for the robust immobilization observed experimentally. The extended DFT (density functional theory) calculations for I₃⁻ binding to more complex structures show that the binding energy increases significantly from −0.964 eV for a long-chain PA to −2.371 eV for a double-chain configuration (Fig. 3i), highlighting the critical role of the entangled PA network in providing enhanced spatial confinement. Furthermore, the electron‑rich carbonyl groups of unpaired amide groups also show affinity for Zn²⁺ cations (Supplementary Fig. 30), and the concurrent adsorption of Zn²⁺ serves to balance the charge introduced by polyiodide adsorption, thereby enhancing both the capacity and stability of TPAE for immobilizing polyiodides. In summary, the powerful confinement of I₃⁻ arises from a synergistic combination of electrostatic interactions with the amide groups, charge compensation by Zn²⁺, and the spatial restriction provided by the entangled molecular architecture of the PA chains.

Enhanced polyiodide conversion efficiency via interphasial confinement

Although soluble polyiodide intermediates are spontaneously generated at the electrode-electrolyte interphase and tend to diffuse outward into the bulk electrolyte, they can still participate in electrochemical redox reactions by being reduced to I⁻ or oxidized back to I₂. However, their conversion is kinetically limited due to their indispensable extra diffusion process through the liquid phase under the influence of electric fields and concentration gradients. This diffusion-induced kinetic bottleneck markedly lowers the reduction efficiency of I₃⁻. To get deeper insight into this behavior, galvanostatic charge-discharge experiments are conducted using a platinum wire as the working electrode in a home-made electrochemical cell, with real-time monitoring of polyiodide dynamics via in situ optical microscopy. As shown in Fig. 4a and Supplementary Movie 1, during the charging process, non-uniform black iodine deposits form on the Pt wire surface, while distinct purple regions appear in the surrounding electrolyte, confirming the generation and outward diffusion of polyiodides. Upon discharging, the iodine deposits rapidly dissolve, disappearing within approximately 40 seconds, coinciding with a sharp voltage drop (Fig. 4c). However, a residual purple zone remains around the wire, indicating incomplete reutilization of polyiodides and a significant loss in discharge capacity. In contrast, when the Pt wire is coated with a TPAE gel interphase via phase inversion (TPAE@Pt), the oxidized iodine species are spatially confined within the interphase during charging (Fig. 4b and Supplementary Movie 2), with no visible diffusion of polyiodides into the electrolyte. During discharging, the confined iodine species on the Pt wire disappear uniformly, and the discharge time is notably extended to 150 seconds (Fig. 4c). These results highlight the good capability of the TPAE interphase in immobilizing polyiodides and enhancing their redox reversibility.

Fig. 4. Enhanced polyiodide conversion efficiency via interphasial confinement.

a, b In situ optical micrographs of (a) bare Pt and (b) TPAE@Pt in 3 M ZnI₂ solution during charging/discharging process in the electrochemical cell at 0.5 mA cm⁻², and c corresponding galvanostatic charge/discharge (GCD) profile. d The atomic ratio of Zn²⁺ under different binding conditions with PA chains (insets are the binding states of Zn²⁺ with H₂O and the PA chain). e, f Top views of structural snapshots from MD simulations for TPAE phase (e) before and (f) after equilibration, and g, h corresponding density distribution of I₃⁻ in the TPAE/ZnSO₄ entire interphase (g) before and (h) after equilibration. i, j The in situ Raman spectra of ZnSO₄ electrolyte during cycling for (i) Zn | |PTFE@I₂AC and (j) Zn | |TPAE@I₂AC electrochemical cells at 2 C (0.42 A g⁻¹) and corresponding GCD profiles. k, l The in situ UV spectra of ZnSO₄ electrolyte for (k) Zn | |PTFE@I₂AC and (l) Zn | |TPAE@I₂AC electrochemical cells during cycling at 1 C (0.21 A g⁻¹) and corresponding GCD profiles.

In addition, MD simulations reveal that the vast majority of polyiodides in the ZnSO₄ solution phase spontaneously migrate into the TPAE phase (Fig. 4e–h, Supplementary Fig. 31). Interestingly, the adsorption behavior of the amide groups toward polyiodide anions and Zn²⁺ ions in the ZnSO₄ electrolyte used for battery operation exhibits a significant difference. For instance, time-resolved atomic distribution analysis over a 5 ns post-equilibration period reveals that I₃⁻ remains stably bound to amide groups throughout the simulation (Supplementary Fig. 32), while most Zn²⁺ ions exhibit only weak or transient interactions (Fig. 4d). Furthermore, in situ UV-vis measurements confirm that the TPAE membrane exhibits no active adsorption of I⁻ in ZnI₂ solution, indicating that it does not actively adsorb Zn²⁺ ions either (Supplementary Fig. 33). In contrast to the dissolution phenomenon of TPAE in concentrated Zn(TFSI)₂ and corresponding DFT calculation results, which show a strong affinity between Zn²⁺ and the carbonyl groups, the weak adsorption of Zn²⁺ under operating ZnSO₄ conditions is attributed to the significant differences in ion concentration, solvation structure of Zn²⁺, and the type of supporting anion between the two environments⁴⁵. In dilute ZnSO₄ solution, Zn²⁺ exists in a fully solvated form with six water molecules, with its charge density being substantially reduced. Moreover, the kosmotropic SO₄²⁻ anion cannot effectively break the as-reformed hydrogen bonds between PA chains. As a result, TPAE matrix exhibits considerable stability during battery operation, as evidenced by the absence of Zn²⁺ residue in the TPAE@I₂AC structure after soaking in ZnSO₄ electrolyte (Supplementary Tables 5 and 6). Furthermore, the moderate, dynamic interaction between Zn²⁺ and the TPAE network allows for rapid association and dissociation. In fact, Zn²⁺ is present at a high concentration as the charge carrier in the supporting electrolyte, whereas polyiodides exist only as trace reaction intermediates. The comparable ionic conductivity between the TPAE@ZnSO₄ interphase and bulk ZnSO₄ electrolyte (Supplementary Figs. 34 and 35) further confirms that Zn²⁺ transport within the TPAE interphase follows a hydrogel‑like mechanism, with migration occurring predominantly through hydrated regions inside the polymer matrix, which ensures efficient ion transport and sustains high‑rate performance without significant kinetic limitation.

Thicker electrodes inherently promote greater generation of polyiodides, an issue that is exacerbated when restricted ion transport slows the overall reaction kinetics. This limitation causes the electrochemical reactions to localize predominantly at the electrode surface, thereby accelerating the shuttle effect. To corroborate this behavior, in situ Raman and UV-vis spectroscopy are employed to monitor the redox behavior of iodine species during cycling (Supplementary Figs. 36 and 37). As presented in Fig. 4i, the in situ Raman spectra show clear polyiodides’ signals during the mid-to-late charging stage in the Zn | |PTFE@I₂AC cell, coinciding with an inefficient electrochemical process, indicating that the non‑polar activated carbon and chemically inert PTFE binder are insufficient to immobilize the generated polar polyiodides. Similarly, a detectable level of polyiodides appears in the electrolyte of the Zn | |CMC@I₂AC cell by mid-charging stage, though the extent of shuttling is relatively lower, likely due to the lower electrode loading and a more confined electrolyte environment (Supplementary Fig. 38). In contrast, the release of polyiodides is strongly suppressed in the Zn | |TPAE@I₂AC cell throughout cycling (Fig. 4j). Even under conditions of excess electrolyte (as monitored by in situ UV-vis), the TPAE-based cell shows minimal polyiodide escape, and the conversion process remains more reversible (Fig. 4l). For the inert PTFE or CMC-based systems, however, a dramatic increase in I₃⁻ concentration is observed during charging process, with significant amounts persisting in the electrolyte after charging is complete (Fig. 4k and Supplementary Fig. 39). Therefore, the TPAE gel framework serves as an effective polyiodide-immobilizing interphase. This confinement shortens the polyiodide diffusion path, accelerates redox reactions, and improves the reversibility of the iodine positive electrode.

Application in practical high-areal-capacity full cells

Furthermore, long-term cycling tests are conducted to evaluate the stability and practical applicability of the different electrode systems. Due to the slow iodine-conversion kinetics stemming from limited ion transport, the Zn | |PTFE@I₂AC full cell delivers average discharge areal capacities of only 7.6 mAh cm⁻² at 3 C and 4.4 mAh cm⁻² at 5 C, despite its high mass loading of 78 mg cm⁻². Notably, this cell suffers from rapid failure after only ~650 cycles at 3 C and ~200 cycles at 5 C (Fig. 5a and Supplementary Fig. 40), manifested as a sharp voltage plunge during charging, a characteristic sign of an internal short circuit. After replacing the separator, zinc negative electrode, and electrolyte, the failed cell resumes normal operation (Supplementary Fig. 41), confirming the negative electrode as the failure origin. While the PTFE@I₂AC positive electrode itself remains stable, substantial basic zinc sulfate by-products can be detected on the negative electrode side (Supplementary Fig. 42). In addition, visible black dendritic penetration sites can be observed on the positive-electrode-facing side of separator, alongside numerous residual dead zinc fragments on the negative-electrode-facing side (Supplementary Fig. 43). Severe pitting corrosion can be detected on the cycled zinc negative electrode from failed Zn | |PTFE@I₂AC cell (Supplementary Fig. 44a). Given that it delivers lower areal capacity compared to that of Zn | |TPAE@I₂AC full cell, such accelerated failure can be attributed to that the active polyiodides migrate to and corrode the Zn negative electrodes, intensifying dendrite growth. In addition, conventional slurry-based CMC@I₂AC electrode is limited to a maximum loading of only 25.6 mg cm⁻² due to cracking and delamination issues, resulting in a much lower areal capacity of 4.4 mAh cm⁻² at 0.5 C (Supplementary Fig. 45). Even at this reduced loading, the inherent poor morphological homogeneity and weak polyiodide confinement of the wet-processed electrode can still lead to eventual short-circuit failure (Supplementary Figs. 46 and 47).

Fig. 5. Application in practical high-areal-capacity full cells.

a The cyclic performance of coin full cells at 3 C (0.63 A g⁻¹) with various positive electrodes (insert is the corresponding GCD profiles). b The self-discharge tests of Zn | |TPAE@I₂AC full coin cell. c The cyclic performance of Ah-level Zn | |TPAE@I₂AC pouch cell at 2 C (0.42 A g⁻¹, insert is the photo of the pouch cell and GCD profile). d Photos of batch-assembled Ah-level Zn | |TPAE@I₂AC pouch cells. e The performance comparison of Ah-level Zn | |TPAE@I₂AC pouch cell with other reported AZIBs pouch cells (The source of the literature data is listed in Supplementary Table 8). f The bending test of Zn | |TPAE@I₂AC pouch cell. g The photos of flexibility demonstration for TPAE@I₂AC electrode and the wearable applications. h The photos of a series-connected Zn | |TPAE@I₂AC pouch cell power the take-off of a drone.

Benefitting from the good homogeneity, fast iodine-conversion kinetics, and polyiodide immobilization within the TPAE matrix, the Zn | |TPAE@I₂AC full cell demonstrates better cycling stability, maintaining an average areal capacity of 11.5 mAh cm⁻² for over 1600 cycles at 3 C (Fig. 5a). At a higher specific current of 5 C (Supplementary Fig. 40), the full cell still remains stable for over 1600 cycles with an average areal capacity of 9.6 mAh cm⁻². As presented in Supplementary Fig. 44b, the Zn negative electrode from cycled Zn | |TPAE@I₂AC exhibits a smoother surface with significantly fewer by-products, indicating the fewer polyiodide-corrosion. Even when coupled with a thin 50 μm-thick zinc negative electrode, the Zn | |TPAE@I₂AC full cell can still deliver an areal capacity of exceeding 10 mAh cm⁻² for over 300 cycles at 3 C (Supplementary Fig. 48), while operating at a reduced N/P ratio of 2.45. Post-cycling characterization confirms the structural and chemical integrity of the TPAE@I₂AC electrode. As presented in Supplementary Fig. 49, the cycled TPAE@I₂AC positive electrode retains uniform surface morphology without significant morphological changes. Meanwhile, XRD (X-ray diffraction) and XPS (X-ray photoelectron spectroscopy) analyses show no detectable impurity phases or significant shifts in spectral features, verifying the chemical stability of the electrode during cycling (Supplementary Figs. 50 and 51). Furthermore, linear sweep voltammetry confirms that TPAE does not undergo electrochemical decomposition within the operational voltage window (Supplementary Fig. 52), meeting the stability requirements for a polymeric binder.

The influence of polyiodide shuttling can be directly reflected in the self-discharge tests. As presented in Fig. 5b, the Zn | |TPAE@I₂AC full cell exhibits high capacity retention of 97.43%, 96.46%, 95.70%, 94.91%, and 92.46% after resting for 12 h, 24 h, 36 h, 48 h, and 64 h, respectively. This high retention reflects the strong ability of the TPAE framework to suppress polyiodide dissolution. In contrast, the Zn | |PTFE@I₂AC full cell shows markedly lower retention, only 87.59% after 12 h and 78.43% after 24 h (Supplementary Fig. 53). Moreover, an internal short circuit happens after 24 h of rest and subsequent discharge, which is attributed to the migrated polyiodides during storage.

More importantly, the integrated thick electrode processing technology enables pre-embedding of current collectors directly within the slurry, facilitating the formation of monolithic, structurally integrated large-size electrodes with high areal capacities suitable for pouch cells. As presented in Supplementary Fig. 54, the 5 cm × 5 cm Zn | |TPAE@I₂AC pouch cell with an iodine loading of 43 mg cm⁻² demonstrates stable cycling for over 4000 cycles (5000 h) at 2 C, maintaining an average areal capacity of 5.6 mAh cm⁻². Under stringent conditions with an N/P ratio of ~1.53 (using an electrodeposited zinc negative electrode) and a limited E/C ratio of 15.3 mL Ah⁻¹, the pouch cell can still deliver a capacity of 0.8 Ah and cycled stably for over 200 cycles (Supplementary Figs. 55, 56, and Supplementary Table 7). When further scaled up to 10 cm × 10 cm sized and a higher I₂ loading of 90 mg cm⁻², the pouch cell with single-layer positive electrode achieves a high cell capacity of 1.6 Ah, and maintains stable cycling for over 1200 cycles (Fig. 5c), delivering competitive performance to some recent reports on AZIB pouch cells (Fig. 5e). Moreover, the aqueous phase solution containing LiTFSI after phase inversion can be concentrated via heating and reused for TPAE dissolution in subsequent batches (Supplementary Fig. 57), facilitating cost-effective and scalable Ah-level pouch cell production (Fig. 5d). These results underscore the potential of the integrated thick electrode processing technology in constructing mechanically robust, high-loading electrodes suited for practical large-scale energy storage.

In addition, the mechanical flexibility of the TPAE@I₂AC electrode also imparts mechanical resilience to pouch cells (Fig. 5f and g). In bending tests, the cell retains ~90% of its capacity after repeated bending and winding (Fig. 5f), making it suitable for wearable electronics. For instance, the flexible Zn | |TPAE@I₂AC pouch cell can power LEDs in various form factors (Fig. 5g). Moreover, the good rate capability of TPAE@I₂AC enables the pouch cells to deliver high-power discharge. As presented in Fig. 5h, the Ah-level TPAE@I₂AC cells can successfully power the take-off of a high-power drone by connecting five cells in series, with an instantaneous output exceeding 6 W, confirming the practical potential of TPAE@I₂AC electrode. More importantly, the TPAE@I₂AC with a reduced TPAE content (10 wt%) does not significantly compromise electrochemical performance (Supplementary Fig. 58). Accordingly, the TPAE content can be strategically tuned to meet specific application requirements, for example, using higher TPAE loading in mechanically demanding or flexible configurations, and lower content when maximizing energy density is the primary objective. This compositional tunability highlights the broad adaptability of the TPAE-based electrode design. Looking forward, the continued development of high-energy-density zinc batteries will also benefit from strategies like positive-electrode-free configurations and designing high-voltage or multi‑electron‑transfer positive electrode architectures, to further push the energy density limits^49,50.

Discussion

In conclusion, this work addresses two fundamental challenges in high-performance iodine positive electrodes: poor mechanical robustness and dramatically deteriorated ion transport kinetics under high-loading conditions, as well as inherent polyiodide dissolution in aqueous interphase. Moving beyond slurry-based or dry-processed electrode fabrication methods, an integrated thick-electrode processing technology is employed to fabricate high-loading iodine-positive electrodes that enable in situ current-collector embedding. Leveraging the hydrogen‑bonded bulk network formed by flexible polyamide segments and the transient, slurry-level electrode uniformity, this technology reconciles tunable pore architecture with robust mechanical integrity in thick electrodes. Beyond serving as a structural framework, the TPAE with polyiodide-absorption ability can function as a polyiodide-immobilization gelation interphase, effectively compensating for the limited confinement capability of activated carbon for polar polyiodide species that can spontaneously dissolve in an electrolyte environment. Furthermore, the tunable loading capacity and good flexibility of the TPAE@I₂AC electrode enable zinc-iodine batteries to achieve Ah-level capacities and provide potential applications in flexible devices, significantly advancing the practical deployment of aqueous zinc-iodine batteries.

Methods

Preparation of materials

Preparation of I₂AC: Typically, 15 g of activated carbon powder (99.9 wt%, CANRD), 2.1 g of Super P (TIMCAL), and 2.1 g of VGCF-H (vapor grown carbon fiber, RESONAC, ~150 nm in diameter and ~6 μm of fiber length) were pre-mixed uniformly using a high-speed blender (CHIGO) for 1 min to obtain a homogeneous carbon mixture. Subsequently, 19.2 g of elemental iodine (99.8%, MACKLIN) was ground into fine powder and mixed with the carbon mixture at a mass ratio of 1:1. The mixture was divided into multiple 50 mL serum bottles and heated at 80 °C for 2 hours to allow uniform sublimation and deposition of iodine onto the activated carbon surface, yielding I₂AC powder. Preparation of TPAE solution: 86.1 g LiTFSI (99%, LEYAN) powder was dissolved in 30 mL of ultrapure water (>18 MΩ), thus obtaining approximately 50 mL of 10 m LiTFSI solution. Then, 9.6 g of TPAE particles (MACKLIN) were added and stirred at 75 °C for 12 h until complete dissolution, resulting in a homogeneous TPAE solution. Preparation of TPAE@I₂AC electrode: 38.4 g of I₂AC powder was dispersed into 50 mL TPAE solution in a 250 mL beaker with mechanically stirring for 2 h to form a uniform TPAE@I₂AC slurry. Then the TPAE@I₂AC slurry was cast into a smooth polytetrafluoroethylene (PTFE) mold and immersed in a petri dish containing 10 times its volume of ultrapure water to conduct the phase inversion process, forming a self-supporting TPAE@I₂AC electrode film. The aqueous phase from phase inversion was evaporated to recover and concentrate LiTFSI for reuse. The film was immersed and rinsed extensively with ultrapure water three times to remove residual LiTFSI. In this work, the mass-loading of iodine in TPAE@I₂AC was controlled as 55-75 mg cm⁻². Similarly, TPAE@I₂AC (fabricated with Zn(TFSI)₂) was obtained by replacing the 86.1 g of LiTFSI with 93.5 g of Zn(TFSI)₂ (99.8%, MACKLIN, to prepare a 5 m Zn(TFSI)₂ solution) and following the identical preparation steps. For assembly of pouch cells, a commercial 100-mesh titanium grid was embedded into the slurry before phase inversion to fabricate an integrated current collector-electrode composite for a subsequent pouch cell. The pure TPAE films were prepared by casting the TPAE solution directly onto a glass plate, followed by phase inversion.

Preparation of CMC@I₂AC electrodes: 4 g of I₂AC powder was dispersed into 50 g of 1 wt% CMC (GUANG ZHOU SONGBAI) solution under magnetic stirring for 3 h, followed by adding 1.25 g of 40 wt% SBR emulsion (GUANG ZHOU SONGBAI) and further stirring for 0.5 h. The resulting slurry was coated onto a 20 μm-thick Ti foil (CANRD) by an automatic coating machine (AFA-Ⅲ, MODERNER), followed by drying at 35 °C for 12 h to form the CMC@I₂AC electrode. Preparation of PTFE@I₂AC electrodes: A mixture of 14 g activated carbon, 2 g vapor-grown carbon fiber, 2 g Super P carbon black, and 1.08 g polytetrafluoroethylene (PTFE, F-104, DAIKIN) powder was dry-mixed in a high-speed homogenizer for 1 min to ensure uniform dispersion of solid components while simultaneously fibrillating the PTFE binder. The resulting powder was divided equally into nine 50 mL serum bottles, each supplemented with 2 g pulverized iodine. The bottles were heated at 80 °C for 2 hours to promote iodine sublimation and its homogeneous deposition onto the high-surface-area AC, yielding a well-distributed I₂AC composite. The final powder was lightly moistened with a minimal amount of isopropanol (>99.5%, MACKLIN, ~250 μL per gram of material) to improve plasticity and then calendered using a dual-roller system (MRX-JS200, MINGRUIXIANG) into a free-standing electrode film with a uniform thickness of approximately 1.4 mm, corresponding to an iodine loading of 75 ~ 80 mg cm⁻². The film was then heated at 35 °C for 12 h to remove residual isopropanol.

Electrochemical measurements

For coin cell assembly (CR2032 type, 304 stainless steel case), the self-supporting TPAE@I₂AC, PTFE@I₂AC, or CMC@I₂AC electrodes were punched into Φ12 mm disks by a manual die cutter (MSK-T10, SHENZHEN KEJING) and assembled with Zn foil negative electrodes (Φ16 mm, 100 μm of thickness) and glass fiber separators (GF-D, Whatman, Φ19 mm, 675 μm of thickness and 2.7 μm of average pore size). A stainless steel spacer (Φ15.8 mm, 0.5 mm thick) and a spring (304 stainless steel) were sequentially stacked in the cell, in which 150 μL of 2 M ZnSO₄ electrolyte was added to wet the separator. Coin cells were tested at a constant temperature of 25 °C. The Ah-level pouch cell was assembled using one piece of TPAE@I₂AC electrode (10 cm × 10 cm, two‑side coated) as the positive electrode, which was sandwiched between two glass fiber separators (11 cm × 11 cm) and Zn foil negative electrodes (10.5 cm × 10.5 cm, 100 μm of thickness), and sealed with aluminum plastic film. The Zn foil negative electrodes were polished with 3000‑grit sandpaper to remove the surface oxide layer. 50 μL cm⁻²_separator of electrolyte was added to wet separator. Prior to battery assembly, the positive electrodes were pre-wetted with electrolyte at a dosage of approximately 2.5 μL mg⁻¹_iodine. An external pressure of ~0.2 MPa was applied during cycling. The pouch cells were tested at 25 °C ± 3 °C without formation or gas‑release steps.

All of the cells were assembled in an air environment and tested with the Neware BTS test system. The specific capacity of positive electrode was calculated based on the loading mass of I₂, while the specific current densities were calculated based on the theoretical specific capacity of the iodine cathode (211 mAh g⁻¹), with 1 C corresponding to 0.21 A g⁻¹. All of the Zn-iodine cells were discharged to 0.8 V and charged to 1.6 V. The E/C and N/P ratios were calculated based on an actual specific capacity of ~200 mAh g⁻¹ for the iodine-positive electrode. The dQ/dV analysis was performed using data exported from the Neware battery test system and processed with its built-in BTSDA software. The rate-dependent cyclic voltammetry measurements and in situ EIS results were obtained on Metrohm AUTOLAB PGSTAT302N. The in situ EIS tests were carried out by applying a 10 mV AC signal after holding at the set voltage for 10 min during a galvanostatic discharge-charge-discharge cycle (0.106 A g⁻¹). The EIS spectra were recorded in the frequency range from 10⁻¹Hz to 10⁵Hz with 10 data points per decade.

Materials Characterizations

Ultraviolet-visible (UV-Vis) spectroscopy was conducted on a Shimadzu UV-3600i Plus spectrophotometer. Scanning electron microscopy (SEM) imaging was carried out with a Zeiss GeminiSEM 500 microscope, using an electron high tension (EHT) of 5 kV. Raman scattering spectra were acquired using a Horiba LabRAM HR Evolution spectrometer with an excitation laser wavelength of 532 nm and a power of 28 mW. Micro-computed tomography (Micro-CT) was performed on a high-resolution SkyScan 1272 system (Bruker). Mechanical tensile tests were conducted using a SYSTESTER TSL−1002 electronic tensile tester. Rectangular specimens with a width of 10 mm were cut from the electrode films and mounted using parallel-plate grips secured with screws. The initial gauge length between grips was 21 mm, and the tensile test was performed at a speed of 30 mm min^-1 at 25 °C with a custom protocol. The ICP-OES results were obtained on SPECTRO SPECTROBLUE FMX36. The optical images were acquired on CX40M metallographic microscope (NINGBO SUNNY INSTRUMENTS). TOF.SIMS 5 (IONTOF) was employed to obtain the TOF-SIMS results. The XPS profile was acquired using Thermo Scientific ESCALAB Xi+ (Al-Kα, 1486.68 eV). X-ray diffractometer (Malvern PANalytical Empyrean diffractometer with CuKα radiation) was used to obtain the XRD patterns. After cycling, the coin cells were disassembled in an ambient atmosphere. The harvested electrodes or separators were gently rinsed with ultrapure water to remove residual electrolyte, and then dried at 35 °C for 12 h before ex situ characterization.

DFT calculations

Quantum chemical calculations were performed using Gaussian 09 E.01 software⁵¹. Geometries of all molecular structures were optimized, and vibrational frequency analyses were conducted at the same theoretical level to confirm the absence of imaginary frequencies, ensuring that the optimized structures correspond to true energy minima. The calculations employed the B3LYP-D3 functional^52,53, which accounts for dispersion interactions and is well-regarded for its accuracy across various elements. The 6-311 + G (d, p)^54–56 basis set was used for all elements. Solvent effects of water (ε = 78.4) were incorporated using the SMD⁵⁷ implicit solvation model. Subsequently, high-precision single-point energy calculations were carried out at the B3LYP-D3/def2-TZVPP^58,59 level of theory. Electrostatic potential (ESP) and restrained electrostatic potential (RESP)⁶⁰ analyses were performed based on the optimized structures using the Multiwfn⁶¹ software package.

Binding energies between the TPAE matrix and other species were calculated using the Vienna Ab initio Simulation Package (VASP)⁶². The projector-augmented-wave (PAW)⁶³ method was used to describe the core electrons, and the exchange-correlation functional was treated within the generalized gradient approximation (GGA) using the Perdew-Burke-Ernzerhof (PBE) functional⁶⁴. A plane-wave basis set with a cutoff energy of 500 eV was adopted. Structural relaxations were performed until the forces on all atoms were less than 0.02 eV Å⁻¹ and the total energy change was below 1 × 10^-5eV. The optimized structures were visualized using VESTA software. The atomic coordinates of the optimized models for DFT calculations are provided in Supplementary Data 1.

Molecular dynamics simulations

Molecular dynamics (MD) simulations were conducted using GROMACS 2018.8⁶⁵ to investigate the interactions between iodine species, zinc ions, and TPAE matrix. The initial structures of small molecules (TPAE segmer, water, ZnSO₄) were optimized at the B3LYP-D3/def2-SVP^52,53,59 level of theory using Gaussian 09 E.01 software. The Amber99SB⁶⁶ force field was employed for the system, with specific parameters for Zn²⁺ ions and the SPC/E⁶⁷ water model. Force field parameters for the sulfate ion and polyiodides were generated using the GAFF⁶⁸ force field via the ACPYPE program⁶⁹, while parameters for TPAE were generated using the Sobtop⁷⁰ software. The Lennard-Jones parameters for zinc atoms were set to σ = 2.44 Å and ε = 3.022 kJ mol⁻¹. RESP atomic charges, scaled by a factor of 0.78 to account for ionic polarization effects, were used to describe electrostatic interactions.

The simulation system, constructed using the PACKMOL software⁷¹, contained a mixture of 2 M ZnSO₄, TPAE, and I₃⁻ ions within a periodic cubic box of 5 × 5 × 10 nm³. The system composition included 75 TPAE monomers, 120 Zn²⁺ ions, 108 SO₄²⁻ ions, 24 I₃⁻ ions, and 2833 water molecules. To constrain the TPAE chain segments during the molecular dynamics simulations, an inert electrode layer was constructed at the bottom of the simulation box. The system was first energy-minimized using the steepest descent algorithm for 10,000 steps. Subsequently, a 20 ns equilibrium simulation was performed in the NPT ensemble at 298.2 K and 1 bar, followed by a 20 ns production simulation in the NVT ensemble at 298.2 K. A time step of 1 fs was used for integration, and a cutoff of 1.0 nm was applied for non-bonded interactions. Molecular dynamics simulations were carried out in the NVT ensemble for 15 ns with an applied electric field of 0.1 V nm⁻¹ to drive ion migration, which was used solely for the comparative analysis of ion migration trends⁷². The initial and final configurations of the MD simulations are also provided in Supplementary Data 1.

Author contributions

M.H.Z., Y.T.X. and C.X. Sun are equal to this work. M.H.Z., Y.T.X. and Y.Y. conceived and designed the investigation. M.H.Z. and Y.T.X. conducted the material and electrochemical tests, analyzed the data, and wrote the manuscript. C.X. Sun conducted the theoretical calculations. H.L., W.H.Y., H.D. J., Z.H.Lv, F.X.M., Q.L.W., J.Y. and X.L. assisted with the interpretation of results. W.H.Y. and X.L. provided valuable advice on data analysis. J.B.Z. and Y.Y. supervised and guided the project and revised the manuscript. All authors discussed the results and commented on the manuscript.

Peer review

Peer review information

Nature Communications thanks Dong-Joo Yoo and other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Funding

Y. Yang discloses support for the research of this work from the National Natural Science Foundation of China (Grant Nos. 22579141, 22379125) and the Fundamental Research Funds for the Central Universities (Grant No. 20720250114).

Data availability

The data supporting this study have been included in the main text and Supplementary Information. Source data are provided with this paper.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41467-026-77387-2.