Chelating additive enabled dual-action hydrogel polymer electrolyte: suppressing dendrite formation and crosstalk in aqueous rechargeable zinc metal batteries

Abstract

Aqueous rechargeable zinc metal batteries (ARZMBs) have garnered significant attention as a sustainable energy storage solution offering high capacity, cost-effectiveness, and ecofriendliness. However, challenges such as uncontrolled zinc dendrite formation and inter-electrode crosstalk hinder their electrochemical stability and long-term cycling performance. In this work, we developed a dual-function gel electrolyte incorporating a chelating additive within a polymer hydrogel matrix to address these limitations. The chelating additive regulates the Zn²⁺ ion flux, suppressing the dendritic growth and facilitating uniform Zn plating/stripping, thereby enhancing the anode reversibility. Simultaneously, the polymer hydrogel electrolyte provides a mechanically robust framework with high ionic conductivity, mitigating Mn dissolution from the MnO₂-based cathode and suppressing the cathode–anode crosstalk, even at practical cathode loadings. The synergistic effects of the chelating agent and polymeric network significantly enhance the electrochemical performance, as evidenced by improved cycling stability, superior rate capability, and increased coulombic efficiency compared to conventional aqueous electrolytes. The computational insights from DFT identify strong Zn²⁺–polymer coordination and restricted Mn²⁺ dynamics, corroborating the experimentally observed stabilization. This study emphasizes the crucial role of electrolyte engineering in tailoring the performance of high-performance ARZMBs and offers a strategic approach for stabilizing the Zn metal interfaces through the design of functionalized electrolytes.

---

Introduction

Batteries that are safe, economical, and environmentally sustainable hold the promise of revolutionizing energy storage for a more sustainable future. In the pursuit of advanced energy storage technologies, aqueous rechargeable zinc–metal batteries (ARZMBs) have emerged as a promising contender, offering a unique combination of safety, environmental friendliness, and cost efficiency.^1–3 However, the practical deployment of ARZMBs is fraught with interconnected challenges that severely hinder their long-term performance and reliability. At the anode, zinc (Zn) deposition during battery cycling is expected to occur uniformly across the anode surface, ensuring stable performance. However, this ideal scenario is rarely achieved in practice due to the heterogeneous nature of the anode surface. Factors such as parasitic reactions, impurities, and uneven surface energy create localized regions of varying resistance and current density.⁴ Eventually, this leads to the growth of non-uniform needle-like structures, called dendrites, that extend toward the separator and counter-electrode, posing significant challenges to the battery's operation.^5,6 Although aqueous electrolytes are designed to enhance the performance of ARZMBs, they can also contribute to battery failure due to their inherent instability. The electrochemical decomposition of the electrolyte can lead to the formation of undesirable byproducts, side reactions, such as hydrogen evolution and parasitic corrosion of the zinc anode, which accelerate the capacity fading and reduce the cycling stability.^7,8

In addition to affecting the Zn anode and the stability of the solid–electrolyte interphase (SEI) layer, the aqueous electrolyte also poses significant challenges for the cathode. Manganese (Mn)-based cathodes, particularly manganese dioxide (MnO₂), are highly favored in ARZMBs due to their exceptional theoretical capacity, high operating potential, and affordability. However, there are major obstacles to its actual implementation. A primary issue is the gradual dissolution of Mn into the electrolyte on encountering water. The dissolution not only reduces the amount of the active cathode material available for energy storage but also triggers undesirable consequences such as side reactions, contamination of the electrolyte, and integrity of the electrode structure over time. Once dissolved, the Mn²⁺ ions diffuse through the electrolyte and migrate toward the Zn anode due to the concentration gradient and electric field during battery operation. This migration is problematic because the Mn²⁺ ions do not remain inert; instead, they participate in the side reactions and localized deposition of the manganese species (e.g., Mn oxides or hydroxides), which disrupts the uniformity of the Zn surface. This process, called cathode–anode cross-talk, causes serious degradation in the plating–stripping process of the Zn-batteries.^9–12

Over the years, extensive research has been conducted on designing advanced electrolytes to address the aforementioned challenges. Strategies such as incorporating additives to regulate the Zn-ion flux or act as Mn²⁺-trapping agents, utilizing separators with anti-dendrite properties, coating the Zn anodes or Mn-based cathodes with protective interfacial layers, and employing engineered substrates to facilitate homogeneous Zn nucleation have all been explored to address the persistent challenges in aqueous Zn–MnO₂ batteries.^13–19 The bottlenecks, particularly the formation of the zinc dendrites on the anodes, electrolyte decomposition, and the dissolution of the Mn²⁺ ion from the cathodes, are not merely isolated problems but interconnected phenomena that undermine the battery stability, efficiency, and lifespan. Addressing these challenges requires an electrolyte that does more than act as an ion transport medium; ideally, it must actively intervene in and control the underlying failure mechanisms.

In this context, we present a multifunctional hydrogel electrolyte engineered to address these issues comprehensively. Polyvinyl alcohol (PVA) serves as the primary polymer matrix in the hydrogel due to its exceptional film-forming ability, hydrophilicity, and mechanical stability. Its hydroxyl (–OH) groups enable the formation of a stable hydrogel network through hydrogen bonding and crosslinking, ensuring structural integrity during battery cycling. The flexibility and adaptability of PVA make it an ideal backbone for incorporating functional additives without compromising the electrolyte's mechanical robustness.^20,21 Polyacrylic acid (PAA) is incorporated into the hydrogel to introduce high-density carboxyl (–COOH) groups, which enhance the ionic conductivity, water uptake properties, and mechanical stability of the interpenetrated polymer network. Additionally, the zincophilic carboxyl groups anchor the Zn²⁺ ions transiently, facilitating a controlled and uniform transport pathway.^22,23 This mechanism reduces the localized Zn-ion flux and minimizes the dendrite formation. The hydroxyl groups in PVA and the carboxyl groups of PAA both have a strong affinity for the water molecules, forming hydrogen bonds that immobilize water within the hydrogel matrix. This reduces the free water content available for direct interaction with the Zn anode surface. By limiting the free water near the Zn–electrolyte interface, the polymer backbone helps suppress the water-induced side reactions. To augment the functional scope of the polymer interface, a chelating additive, ethylenediaminetetraacetic acid (EDTA), is introduced into the PVA–PAA matrix, imparting the system with additional capability for the transition metal ion sequestration. The structure of EDTA, with its four carboxylate groups and two amine groups, allows it to form stable coordination complexes with a wide range of divalent and trivalent metal ions. The ability of EDTA to coordinate with various transition metal ions is employed to sequester the metal ions that would otherwise contribute to undesirable side reactions.^24,25 The additive, selected based on its multidentate coordination functionality (e.g., –COO⁻ and –N donor atoms), exhibits competitive complexation with Zn²⁺, thereby immobilizing the dissolved Mn²⁺ species within the polymer scaffold, preventing their transport to the anode. The chelating-polymer synergy not only mitigates the redox shuttle effects but also enhances the structural retention of the cathode and prolongs the electrochemical lifespan of the system. The EDTA additive selectively transports the Zn²⁺ by forming stable chelation complexes that disrupt the native solvation structure of [Zn(H₂O)₆]²⁺, enabling uniform ion distribution and reducing the water-related side reactions. This coordination suppresses the zinc dendrites by promoting homogeneous nucleation during electrodeposition.

Through a combination of experimental characterizations, electrochemical performance evaluation, and computational analysis, the present work demonstrates the transformative potential of the dual-functional polymer electrolyte with an extended cycle life of more than 400 cycles compared to the extremely low plating–stripping stability of the Zn electrode in the presence of an aqueous electrolyte (∼150 cycles). By tackling both the cathode and anode degradation pathways simultaneously, this approach marks a significant departure from the traditional electrolyte designs, paving the way for more durable and efficient ARZMBs. By leveraging the zincophilic coordination, chelation-driven ion selectivity, and solvation environment engineering, the proposed hydrogel polymer electrolyte system delivers simultaneous improvements in the dendrite suppression, cathode–anode cross-talk mitigation, and corrosion resistance. These advances not only elevate the electrochemical performance and safety of the Zn batteries but also offer a scalable and modular approach to interphase engineering for a broader class of multivalent metal battery chemistries.

Experimental section

Synthesis of PVA : PAA hydrogel polymer electrolyte (HPE)

The PVA : PAA hydrogel polymer electrolyte (HPE) was synthesized by a simple solution casting method. 1 g of PVA was dissolved in 20 mL of deionized (DI) water, and maintained at 90 °C for 1 h under constant stirring to ensure complete dissolution of the PVA. Upon achieving a homogeneous solution, the polymer mixture was allowed to cool to room temperature. To facilitate the polymerization of acrylic acid (AA) and form an interpenetrating network, 5 mg of APS initiator and 4 mg of BIS crosslinking agent were added to the cooled PVA solution. AA was then introduced to the solution in varying weight ratios.

The solution was stirred well and then heated at 80 °C for 10 minutes to initiate the polymerization process. Afterward, the polymerized solution was cast into a Petri dish and allowed to dry in an oven at 70 °C for 24 h. The polymer can be peeled off the Petri dish and immersed in a 0.5 M zinc trifluoromethane sulfonate (Zn(OTf)₂) electrolyte to obtain the PVA–PAA interpenetrating hydrogel polymer electrolyte, denoted as HPEx, where ‘x’ represents the PAA to PVA ratio (thickness ∼ 0.202 mm). For the final hydrogel samples, EDTA as a chelating additive was incorporated at various molar ratios, denoted as HPExCy, where ‘y’ represents the amount of the chelating agent in mM. The final sample HPE1C1, where y = 1 mM, represents the sample with a chelating agent added to the HPE1 formulation.

Result and discussion

Fabrication and evaluation of the hydrogel polymer electrolyte (HPE)

The PVA–PAA hydrogel polymer electrolyte (HPE) loaded with EDTA was successfully synthesized using a simple solution casting method, as detailed in the Experimental section and illustrated in Scheme 1a. The PVA–PAA hydrogel electrolyte's interpenetrating polymer network (IPN) structure was chosen because of its potential to deliver a multitude of significant benefits, including superior ionic conductivity, water uptake and retention, and mechanical strength.^26–28 EDTA acts as a vital additive by interacting with the multiple components in the electrolyte system. EDTA is a chelating agent that forms stable complexes with the metal ions through its four carboxylate groups and two amine groups. Incorporation of EDTA as a functional additive is expected to simultaneously regulate Zn²⁺ deposition and mitigate Mn²⁺ crosstalk through selective chelation.^29–31 The concentration of EDTA is optimized to achieve the highest performance of the hydrogel polymer electrolyte (HPE). The samples containing EDTA are denoted as HPExCy, where ‘x’ indicates the PVA to PAA ratio, and ‘y’ represents the EDTA concentration in mM. Scheme 1b schematically depicts the contribution of the modified HPE and the role of aqueous electrolyte (AE) in deteriorating performance in the full cell configuration.

Scheme 1 (a) The schematic diagram illustrating the preparation of HPE1C1, and (b) the illustration of the challenges associated with the ZMBs equipped with the Zn anode, Mn-based cathodes operating with aqueous electrolytes, and the illustration of the HPE in regulating the ZMB performance.

The ionic conductivity of the synthesized hydrogel polymer electrolytes (HPEs) with varying PVA, PAA, and EDTA concentrations was evaluated by electrochemical impedance spectroscopy (EIS), with the Nyquist plots shown in Fig. S1a and b. The conductivities, calculated using eqn (S1), are summarized in Fig. S1c. A clear trend is observed wherein increasing the PAA : PVA ratio enhances ionic conductivity, owing to the higher density of zincophilic –COOH groups in PAA, as well as the water retention property of PAA, which facilitates Zn²⁺ coordination and migration. The introduction of EDTA as a chelating agent further augments the conductivity by promoting ion dissociation and reducing ion pairing within the hydrogel network. The effect of EDTA concentration is particularly pronounced. At the optimized level (HPE1C1), EDTA enables a balance between chelation and ion mobility, improving the Zn²⁺ transport kinetics. However, excessive EDTA loading (HPE1C2) leads to the formation of stable Zn–EDTA chelate complexes, which restrict the availability of free Zn²⁺ ions for conduction, thereby reducing the conductivity. This highlights the critical need to optimize the polymer-to-chelate ratio to achieve maximum ion transport. Notably, both HPE1 and HPE1C1 exhibit significantly higher conductivity than the aqueous electrolyte (AE) (Fig. 1a), underscoring the role of the functional groups and confined water within the polymer network in facilitating efficient Zn²⁺ transport. The superior conductivity of HPEs compared to AE can be rationalized by their molecular-level interactions, supported by the density functional theory (DFT) analysis (Fig. 1b). The calculated binding energies of Zn²⁺ with PAA (−16.8 eV) and PVA (−17.7 eV) monomer units are substantially stronger than with the water molecules (−4.6 eV). This indicates that Zn²⁺ preferentially coordinates with the carboxyl (–COOH) and hydroxyl (–OH) groups of the polymer backbone rather than undergoing random hydration, thereby stabilizing the solvation environment and enabling efficient Zn²⁺ hopping along the polymer matrix. Such strong yet dynamic coordination suppresses the uncontrolled hydration effects while enhancing the ion transport pathways.³² Moreover, the incorporation of EDTA provides an additional level of control over the Zn²⁺ solvation. The exceptionally high binding energy of Zn²⁺–EDTA (−41.1 eV) reveals its strong chelating capability, which not only anchors the Zn²⁺ ions but also regulates their transport within the hydrogel matrix. This dual role of EDTA, viz., promoting the dissociation at the optimal concentration while restraining the free ion availability at the excess loading, accounts for the conductivity trends observed experimentally. The DFT-optimized geometry of EDTA (Fig. S2) and the corresponding Zn–O bond lengths in various complexes (Table S1) further corroborate its strong binding affinity and its impact on the Zn²⁺ solvation dynamics.

Fig. 1 (a) The ionic conductivity values corresponding to AE, HPE1, and HPE1C1; (b) the DFT-optimized structures of [H₂O–Zn]²⁺, [CH₂=CH–C(O)–O–Zn]⁺, and [CH₂=CH–O–Zn]⁺, together with the calculated binding energies of the Zn²⁺ with H₂O, PAA and PVA monomers, and EDTA; (c) the ATR-FTIR analysis of HPE0, HPE1, and HPE1C1; (d) the Raman spectra of the water in HPEs.

The mechanical properties of the hydrogel electrolyte were assessed by measuring the tensile strength of the HPEx (Fig. S3a), revealing that the stability of the HPE membrane increases with the addition of PAA, particularly for the HPE1. This improvement can be attributed to the interpenetrating network structure formed between PAA and PVA, making it more resistant to tensile deformation during the cycling. Despite its relatively higher ionic conductivity, the low mechanical strength of HPE2 might make it undesirable for deployment in flexible device applications. Consequently, HPE1 was investigated for further study, incorporating EDTA, owing to its superior mechanical properties and comparable ionic conductivity. The swelling ratio of the PAA : PVA hydrogel was found to increase with the concentration of PAA, reaching a maximum swelling ratio of 290% for a HPE2 (Fig. S3b) due to the hydrophilic carboxyl groups from PAA. The increase in the water retention properties (Fig. S3c) also indicates the same trend and is essential for maintaining a stable ionic conductivity during long-term cycling of the battery. The attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectra of the hydrogels were recorded to confirm the formation of the polymer network and investigate the interactions between the components. The FTIR spectra of HPE1 and HP1C1 presented in Fig. 1c exhibit a weak absorption peak at 1717 cm⁻¹, which is attributed to the stretching vibration of the carbonyl moiety (–C=O) in the acetate group formed by the condensation of PVA and PAA, which is absent in the case of HPE0, where there is no PAA. This peak indicates the successful incorporation of PAA into the hydrogel network. Moreover, the intensity of the –OH peak (typically around 3350 cm⁻¹) decreased significantly after the addition of EDTA, suggests that EDTA interacts with the hydroxyl groups, reducing the free water content in the polymer matrix. A shift in the OH stretching peak to higher wavenumbers further indicates stronger hydrogen bonding between EDTA and water.

The thermogravimetric analysis (TGA) profiles for the PVA polymer membranes with and without PAA, before the electrolyte soaking, were recorded to analyze the thermal stability of the polymers induced by PAA addition (Fig. S4). With the incorporation of PAA to form the interpenetrating network structure, the degradation temperature of the polymer has shifted to much higher temperatures, which is crucial in limiting the thermal degradation and local heat accumulation during the battery operation.

The Raman spectroscopy analysis confirmed the presence of the different forms of water in the hydrogel electrolyte (Fig. 1d). The water molecules generally form hydrogen bonds by donating and accepting hydrogen atoms, with configurations like DA–OH (one donor, one acceptor), DAA–OH (one donor, two acceptors), and DDAA–OH (two donors, two acceptors), representing the different bonding patterns that influence their structure and properties. The Raman spectrum of the HPEs showed two components on the low-frequency side, which are attributed to the strongly hydrogen-bonded water, while the two components on the high-frequency side are assigned to the weakly and non-hydrogen-bonded water.^33,34 The intensity of the free water peak decreased significantly upon the addition of EDTA. The energy dispersive X-ray (EDAX) analysis (Fig. S5) confirms the presence of the elements C, N, O, F, S, and Zn. The incorporation of EDTA into the HPE matrix was further verified by analyzing the X-ray Photoelectron Spectroscopy (XPS) results of HPE1C1. As observed in Fig. S6a and b, the XPS spectra of the HPE1C1 show the characteristic peaks for C 1s and O 1s with their key functional groups such as C–C, C–O, C=O, and O–H. In addition, the deconvoluted N 1s spectra (Fig. S6c) confirm the presence of N–H and C–N, which confirms the EDTA functionalization. The detection of the F 1s, S 2p, and Zn 2p peaks (Fig. S6d–f) suggests the incorporation of these elements within the polymer matrix.

To evaluate the electrochemical performance of the developed hydrogel polymer electrolytes, a series of electrochemical measurements were conducted. The corrosion resistance of the electrolytes was assessed via the linear polarization resistance (LPR) measurement (Fig. 2a) of the Zn‖Zn symmetric system. The corrosion current density of HPE1C1 is significantly lower, while its corrosion potential is more positive, indicating the reduced corrosivity and enhanced stability of the zinc electrode. HPE1C1 suppresses the water-mediated corrosion reactions by limiting the availability of the free water molecules, whereas the presence of the free water in AE leads to severe corrosion due to uncontrolled water activity.^35,36 HPE1 offers moderate protection by partially restructuring the water network through PVA and PAA coordination. The cyclic voltammetry (CV) measurements of the Zn|HPE1C1|SS system reveal a markedly reduced nucleation overpotential (calculated as a′ − a) compared to the Zn|HPE1|SS and Zn|HPE0|SS configurations (Fig. 2b). The lower nucleation overpotential observed for HPE1C1 indicates that the chelating additive facilitates uniform zinc nucleation.^37,38 The linear sweep voltammetry (LSV) studies given in Fig. 2c reveal that HPE1C1 exhibits a significantly increased oxygen evolution reaction (OER) overpotential compared to HPE1 and AE, indicating that the chelating additive effectively suppresses parasitic water-splitting reactions. The chronoamperometry (CA) measurements (Fig. 2d) on Zn‖Zn symmetric cells further confirm deposition behaviours of Zn over the Zn anode. For HPE1C1 and HPE1, the current initially rises and then stabilizes to a steady-state value, characteristic of a 3D diffusion-controlled process. This is directly linked to the ability of the zincophilic groups to reorganize the electrolyte structure, enabling more efficient diffusion pathways.^39,40 However, AE displays a rapid increase in the current, indicating the 2D diffusion kinetics resulting from uncontrolled and non-uniform Zn deposition.^41,42 The zinc ion transference number (t_Zn⁺) was calculated according to eqn (S3) by measuring the I–t curves and EIS spectra of the Zn‖Zn symmetric cells in the aqueous electrolyte and HPE1C1 (Fig. S7). A high transference number of 0.79 in HPE1C1 compared to 0.55 in AE indicates suppressed dendrite growth according to the Sand's model, and improved ion transport channels in the HPE1C1.⁴³

Fig. 2 (a) The linear polarization resistance (LPR) data with the Zn‖Zn symmetric cell by employing the different electrolytes, (b) the cyclic voltammograms (CV) of the Zn‖SS cells in AE, HPE, and HPE1C1, (c) the LSV plots in the Zn‖SS configuration recorded in various electrolytes, and (d) the chronoamperometric test carried out in the Zn‖Zn symmetric cells at a bias potential of 50 mV.

Based on the plating–stripping profiles at 1 mA cm⁻² and 0.50 mA h cm⁻² (Fig. 3a), AE exhibits severe instability with noisy and uneven voltage characteristics, and cell failure in less than 150 cycles as a consequence of the side reactions and dendritic growth. On the other hand, HPE1C1 exhibits extremely uniform Zn plating and stripping, delivering smooth, steady cycling for over 430 h by promoting uniform Zn deposition and successfully inhibiting the water-induced side reactions. The inclusion of EDTA into the polymer matrix substantially enhances the durability, validated by the superior cycle life of HPE1C1 over HPE1. Further, the Zn|HPE1C1|Zn symmetric cell is subjected to plating–stripping at high-capacity conditions (10 mA cm⁻² at a capacity of 5 mA h cm⁻²) that closely simulate the real-world demands for ZMBs, and the corresponding performance profile is given in Fig. 3b. The cell successfully operates for around 100 cycles, revealing the ability of HPE1C1 to support the fast, uniform and reversible zinc deposition. The rate performance of the symmetric cells recorded at 1, 2, 5, and 10 mA cm⁻² at a capacity of 0.5 mA h cm⁻² (Fig. 3c) highlights the limitations of the conventional aqueous electrolytes, which fail at high rates, while the HPE1C1 polymer electrolyte maintains stable operation across all the tested conditions. The atomic force microscopy (AFM) images (Fig. 3d) show rough and uneven zinc deposition in AE, with large surface height variations, suggesting a non-uniform growth and byproduct accumulation. In contrast, HPE1C1 demonstrates a smoother surface with minimal height variation, reflecting the controlled and uniform deposition. The Nyquist plots generated from the electrochemical impedance spectroscopy (Fig. 3e) for the symmetric cell with HPE1C1 polymer electrolyte show a lower charge transfer resistance (R_ct), compared to the higher R_ct observed in the cell with AE, indicating that HPE1C1 enables more efficient and faster Zn²⁺ ion transport at the electrode–electrolyte interface. The equivalent circuit is given in Fig. S8.

Fig. 3 (a) The plating–stripping profiles of the Zn‖Zn symmetric cell recorded at the current density of 1.0 mA cm⁻² and 0.50 mA h cm⁻², (b) the plating–stripping profile of the Zn‖Zn symmetric cell recorded at a current density of 10 mA cm⁻² and 5 mA h cm⁻², (c) the rate performance analysis of the Zn|HPE1C|Zn and Zn|AE|Zn symmetric cells recorded at 1, 2, 5, and 10 mA cm⁻², (d) the AFM images of the Zn foil recorded after the 10 cycles of the plating–stripping, and (e) the EIS spectra of the Zn|HPE1C|Zn and Zn|AE|Zn cell.

The FESEM analysis of the cycled Zn anodes provides visible evidence of the role of EDTA in modulating Zn deposition behavior. In the Zn|AE|Zn cell (Fig. 4a and b), the Zn surface exhibits irregular and aggregated deposits, indicating uncontrolled Zn nucleation and growth. In contrast, Zn anodes cycled with HPE1 (Fig. 4c and d) and HPE1C1 (Fig. 4e and f) show smooth, uniform deposition, attributed to the zincophilic functional groups and chelating interactions that promote homogeneous ion flux and nucleation. The cross-sectional images further reveal that the Zn deposits in AE are loosely bound to the Zn anode surface, which may fall off during the cycling (Fig. 4g). On the other hand, HPE1C1 (Fig. 4h) enables compact and uniform Zn deposits. The elemental mapping confirms the uniform distribution of the C, O, F, S, and Zn in the HPE1C1 system, supporting the presence of a stable and uniform SEI layer. The XRD analysis of the cycled Zn anodes (Fig. S9) confirms the presence of the zinc byproducts in AE, as evidenced by the additional peaks, while these byproducts are significantly reduced in HPE1 and nearly absent in HPE1C1. The dominant zinc metal peaks of the Zn foil cycled in the presence of HPE1C1 highlight its ability to suppress the parasitic reactions effectively.

Fig. 4 The post-cycling FESEM images of the Zn foil obtained after the 50 plating–stripping cycles of the cells: (a) and (b) Zn|AE|Zn, (c) and (d) Zn|HPE1|Zn, and (e) and (f) Zn|HPE1C1|Zn; (g) the cross-section image and elemental mapping of the Zn foil recovered from the Zn|AE|Zn cell; (h) the cross-section image and elemental mapping of the Zn foil recovered from the Zn|HPE1C1|Zn cell.

Overall, HPE1C1 shows superior electrochemical stability and structural uniformity, attributed to its enhanced electrolyte formulation, which mitigates the byproduct formation and promotes dendrite-free zinc deposition.

Synthesis and characterization of the NMO cathode

The sodium-intercalated manganese oxide (NMO) cathode was successfully synthesized through a hydrothermal method, as illustrated in the schematic diagram in Fig. S10a. The Field Emission Scanning Electron Microscopy (FESEM) image of the NMO cathode, shown in Fig. S10b, reveals a nanoplate-like morphology consistent with the layered structure characteristic of birnessite-type manganese oxides, resulting in a high surface area and providing abundant electrochemically active sites. Further analysis using Transmission Electron Microscopy (TEM) revealed the presence of nanopores within the NMO plates (Fig. S10c), ensuring that the Zn²⁺ ions can easily diffuse in and out of the cathode material during the charge–discharge cycles. The XRD patterns (Fig. S10d) display peaks at 12.2°, 24.7°, 36.7°, 41.9°, and 65.9°, corresponding to the (001), (002), (100), (101), and (020) planes of δ-MnO₂, verifying a crystalline, layered structure suitable for the Zn²⁺ intercalation. In addition, the Barrett–Joyner–Halenda (BJH) pore distribution (Fig. S10e) indicates a micro-mesoporous architecture (2–30 nm), which aids in the electrolyte penetration and ion diffusion within the cathode. The X-ray Photoelectron Spectroscopy (XPS) analysis was performed to investigate the surface chemistry and oxidation states of the Mn in the NMO cathode. The Mn 2p spectrum, as indicated by Fig. S10f, reveals the presence of multiple oxidation states, with the main peak corresponding to the Mn⁴⁺ at 643.3 eV. Additionally, the Mn³⁺ and Mn²⁺ peaks were observed at 642.2 and 640.8, indicating a mixed valence state for the Mn ions.⁴⁴ This mixed valence is typical of the birnessite-type MnO₂ and is known to enhance the electronic conductivity and charge storage capacity of the material.⁴⁵ The O 1s spectrum given in Fig. S10g shows three distinct oxygen species: lattice oxygen (O_latt), surface oxygen (O_surf), and oxygen vacancies (O_vac). The O_latt content (54.39%) is responsible for maintaining the overall stability of the MnO₂ structure, while the O_vac (15.47%) and O_surf (30.15%) contribute to enhanced ion transport and electrochemical reactivity.^46,47 The oxygen vacancies create active sites for the Zn²⁺ ion diffusion and can improve the ionic conductivity. The presence of the mixed oxidation states of Mn (Mn³⁺, Mn²⁺, Mn⁴⁺) improves the electronic conductivity of MnO₂, facilitating faster charge and discharge processes. The combination of these structural and surface features makes MnO₂ an ideal cathode material for realizing highly efficient zinc-ion batteries with long cycle life. The presence of the pre-intercalated Na ions can be confirmed from the Na 1s XPS spectra of the NMO cathode, as shown in Fig. S11.

Fabrication and evaluation of the full cell based on the hydrogel polymer (HPE) in conjunction with the NMO cathode

The electrochemical performance of the NMO cathode was evaluated using a full cell configuration using AE (Zn|AE|NMO) and HPE1C1 (Zn|HPE1C1|NMO). The CV profiles recorded at 1 mV s⁻¹ (Fig. 5a) reveal distinct redox peaks corresponding to the typical reversible insertion/extraction of the Zn²⁺ ions and H⁺, in agreement with prior studies.^48,49 The narrower voltage gap (0.33 V) observed in the Zn|HPE1C1|NMO system compared to the Zn|AE|NMO system (0.39 V) suggests enhanced redox kinetics and improved reversibility in the presence of the HPE1C1 electrolyte, attributed to the optimized ionic conductivity. The galvanostatic discharge/charge (GCD) curves recorded at a current density of 0.1 A g⁻¹ (Fig. 5b) exhibit discharge capacities of approximately 192 and 170 mA h g⁻¹ for the Zn|AE|NMO and Zn|HPE1C1|NMO cells, respectively. The higher OCV in the HPE-based full cell may arise from the modified Zn²⁺ solvation and interfacial chemistry that shifts the electrode equilibrium potentials, raising the cell voltage compared to the aqueous electrolyte.⁵⁰ The rate capability results of Zn|AE|NMO (Fig. 5c) reveal a significant drop in the capacity from ∼188.1 mA h g⁻¹ at 0.1 A g⁻¹ to 23.1 mA h g⁻¹ at 1 A g⁻¹, indicating poor rate capability. However, the Zn|HPE1C1|NMO cell exhibits better rate performance (185.4 at 0.1 A g⁻¹, 164.2 at 0.25 A g⁻¹, 147.7 at 0.50 A g⁻¹, 126.5 at 1.0 A g⁻¹, and recovers to 178.4 at 0.1 A g⁻¹) compared to AE. In the MnO₂-based cathodes, evaluated through the GCD test over 500 cycles, as given in Fig. 5d for Zn|AE|NMO, over the course of 500 cycles, the capacity drops significantly to around 54% of the initial capacity, indicating poor capacity retention and rapid degradation of the active material (NMO). This drastic capacity loss can be attributed to a combination of the structural instability, dissolution/loss of the active material, and side reactions at the electrode–electrolyte interface. The long-term cycling stability of the cells reflected the deteriorating interfacial kinetics and electrode integrity. On the other hand, the Zn|HPE1C1|NMO cell exhibited a capacity retention of around 81% (Fig. 5e). The stable cycling performance highlights the beneficial contribution of the HPE1C1 electrolyte during the repeated discharge–charge cycles, where the dissolution of active Mn species in the aqueous electrolyte is effectively mitigated.

Fig. 5 The electrochemical performance of the full cells employing AE and HPE1C1: (a) the CV plots of Zn|AE|NMO and Zn|HPE1C1|NMO with a scan rate 1.0 mV s⁻¹; (b) the galvanostatic discharge/charge curves of the Zn|AE|NMO and Zn|HPE1C1|NMO full cells at 0.10 A g⁻¹; (c) the evaluation of rate capability; (d) the long term cycling stability of the Zn|AE|NMO full cell recorded at a current density of 1.0 A g⁻¹; and (e) the cycling performance of the Zn|HPE1C1|NMO full cell recorded at a current density of 1.0 A g⁻¹.

The EIS measurements conducted at the end of the 1st, 10th, and full cells before cycling (pristine) provide further insights into the degradation mechanisms. The Nyquist plots of Zn|AE|NMO, as represented in Fig. 6a, show a progressive increase in the charge transfer resistance (R_ct) with cycling (the equivalent circuit model given in Fig. S12).^51,52 The capacity loss primarily results from the self-discharge, structural deterioration caused by the repeated ion insertion and extraction, and the dissolution of the Mn into the electrolyte. Contrary to this, for the Zn|HPE1C1|NMO cell (Fig. 6b), the R_ct remains far less than that of Zn|AE|NMO, indicating better stability of the NMO cathode in the HPE1C1-based battery.

Fig. 6 The EIS plots of the (a) Zn|AE|MnO₂, and (b) Zn|HPE1C1|MnO₂ cells; the charging profiles before and after the 3 days OCV rest test of the (c) Zn|AE|MnO₂, and (d) Zn|HPE1C1|MnO₂ cells; (e) the capacity loss contribution due to the various processes occurring in the system.

To evaluate the role of the chelating additive in the HPE1C1 in minimizing the MnO₂ dissolution, a 72 h open-circuit voltage (OCV) rest test was conducted. The fully charged cells with aqueous electrolyte and HPE1C1 were held at OCV for 72 h, and their recoverable charging capacity was subsequently measured. The decline in the recharge capacity following the OCV rest can be mainly attributed to the Mn dissolution. On the other hand, the capacity loss during the continuous galvanostatic charge–discharge cycling predominantly arises from the structural changes leading to structural damage as well as Mn dissolution.⁵³ Hence, the extent of the Mn dissolution can be inferred from the overall recharge capacity loss post-OCV rest. As exhibited in Fig. 6c following the OCV test, the capacity loss for the Zn|AE|MnO₂ cell reached 16.58%, whereas the Zn|HPE1C1|MnO₂ (Fig. 6d) counterpart displayed a much lower loss of only 5.58%. Also, the Zn|HPE1C1|MnO₂ cell exhibited a notably lower capacity fade of just 0.09% per cycle (Fig. 5e) over the 100 charge–discharge cycles, whereas the Zn|AE|MnO₂ system experienced a significantly higher loss of 0.54% per cycle (Fig. 5d). From the above two observations, the capacity loss due to the Mn dissolution can be calculated as 16.04% in AE and 5.71% in HPE1C1, and is shown in Fig. 6e.

To further demonstrate the practical relevance of the electrolyte design, the full cell tests were carried out at the cathode mass loadings of 2.0 and 4.0 mg cm⁻² using both the HPE1C1 and AE. As shown in Fig. S13, the NMO cathodes tested with the aqueous electrolyte exhibit rapid capacity fading, retaining only 45% (2.0 mg cm⁻²) and 43% (4.0 mg cm⁻²) after 500 cycles. In contrast, the HPE1C1 enables markedly improved stability, with 60% (2.0 mg cm⁻²) and 61% (4.0 mg cm⁻²) retention, confirming that the chelating electrolyte suppresses the Mn dissolution and inhibits the structural degradation of the NMO cathode even under the higher loading conditions. Furthermore, the quantitative analysis of the OCV test data (Fig. S14) shows that the cells with AE suffer higher irreversible losses compared to the HPE1C1-based cells. These results collectively highlight that the HPE1C1 not only prolongs the cycling stability but also mitigates the rest-induced capacity loss by effectively stabilizing the cathode–electrolyte interface.

Physical characterization of the cathode after the cycling was conducted to identify the underlying causes of the performance degradation. As shown in Fig. 7a, the FESEM images of the cathode cycled with HPE1C1 for 50 cycles retained their original plate-like morphology, indicating improved structural stability. In Fig. 7b, the cathode cycled in an aqueous electrolyte shows significant damage. Prolonged Zn²⁺ insertion/extraction induces volumetric fluctuations, generating internal stress that leads to microcracking and is marked with yellow lines in Fig. 7b. This structural deterioration accelerates the active material degradation, increases the interfacial resistance, disrupts the electrical connectivity, and degrades the electrochemical performance, ultimately shortening the battery's cycle life. The structural and chemical evolution of the cathode material after the 50 charge–discharge cycles was investigated using XRD. The XRD analysis of the cycled cathode (Fig. S15a) reveals the disappearance of the original crystalline phases after the cycling, indicating the potential structural disorder induced by the repeated Zn²⁺ intercalation and deintercalation or loss of the active material. The XPS analysis (Fig. S15b) of the Mn 2p spectra shows a clear increase in the Mn²⁺ signal, indicating a partial reduction of the Mn³⁺ and Mn⁴⁺ states to the Mn²⁺ state. This reduction is likely caused by the disproportionation reactions and dissolution of the Mn species into the electrolyte, which is known to occur under prolonged cycling conditions. These changes strongly agree with the capacity fading and increase in the R_ct over the cycling.

Fig. 7 Morphological analysis via FESEM of NMO cathode (a) after the cycling in HPE1C1, and (b) after the cycling in AE, (c) the FESEM image of the Zn anode recovered from the Zn|HPE1C1|NMO cell after the 50 cycles, and (d) the FESEM image of the Zn anode obtained from the Zn|AE|NMO cell after the 50 cycles, (e) the Mn 2p XPS spectra of the Zn obtained after the 50 charge–discharge cycles from Zn|HPE1C1|NMO and Zn|AE|NMO, and (f) the Mn and Zn ion concentrations obtained from the ICP-OES analysis.

To comprehensively investigate the viability of HPE1C1 in trapping the Mn ions, the EDAX mapping of the post-cycled HPE1C1 was carried out after recovering the hydrogel polymer from the cell and washing it thoroughly with DI water, as shown in Fig. S16. The obtained results indicate that, in addition to Zn, C, O, N, and S, Mn is also present in the hydrogel, indicating that Mn is trapped within the polymer. This has been further verified by the presence of Mn in the XPS spectrum of the recovered HPE1C1, in addition to the Zn (Fig. S17a and b). The dissolution of the Mn from the cathode also contributes to the cathode–anode cross-talk by depositing on the anode surface, further degrading the overall cell performance. Interestingly, the morphology of the Zn foil cycled as a part of the Zn|AE|NMO and Zn|HPE1C1|NMO cells exhibits extremely different characteristics. Fig. 7c depicts the uniform morphology of the Zn foil recovered after the cycling in the Zn|HPE1C1|NMO cell, whereas the Zn foil cycled in the Zn|AE|NMO cell (Fig. 7d) shows random, non-uniform deposition of the Zn during the continuous charge–discharge cycling. The deposition of the dissolved Mn over the Zn anode can be verified from the Mn 2p XPS spectra given in Fig. 7e. The Zn anode cycled in AE clearly indicates the presence of the Mn deposited over its surface, whereas the peak corresponding to the Mn is absent in the case of the system based on HPE1C1, indicating the efficiency of HPE1C1 in preventing the Mn from reaching the Zn anode. The XPS depth profiling data over the sputtering times from 0 s to 360 s were recorded. The Zn foil cycled with NMO in AE (Fig. S18a) further reveals that Mn 2p peaks persist through the surface and subsurface of the AE-cycled Zn foil, while the HPE1C1 sample (Fig. S18b) shows negligible Mn signals at all depths, confirming its ability to suppress the Mn deposition and associated byproduct formation. The F 1s XPS spectra (Fig. S19a) show that the SEI layer on the Zn foil from the Zn|HPE1C1|NMO full cell contains both the inorganic species, such as ZnF₃ and CF₃, while the C 1s spectra (Fig. S19b) confirm the presence of the organic components like C–C, C–O, and CF₃. This inorganic–organic rich SEI provides several advantages: the inorganic ZnF₃ offers high chemical and electrochemical stability, protecting the electrode from the side reactions, while the organic species (C–O, C–C) enhance the flexibility and compatibility with the electrolyte. The combination leads to a robust, uniform, and ionically conductive SEI that suppresses dendrite growth and improves the cycling stability. The XPS analysis (Fig. S20) at the higher cathode loadings (2.0 and 4.0 mg cm⁻²) reveals pronounced Mn deposition on the Zn anodes from the AE-based cell compared to the HPE1C1-based cells, which show no detectable Mn signal, which correlates with their superior stability.

For further confirmation about the coordinating ability towards the Mn that arises from the EDTA incorporated in the polymer system, an inductively coupled plasma-optical emission spectrometer (ICP-OES) analysis was employed. HPE0, HPE1, and HPE1C1 were immersed in a solution consisting of 10 ppm of dissolved Zn and Mn ions for 24 h, and the solution after the uptake of the electrolyte was analyzed using ICP-OES to obtain the amount of the ions remaining in the solution. As observed from Fig. 7f, the concentration of the Mn retained is the least in the solution in which HPE1C1 was immersed, indicating the Mn coordinating ability of the EDTA-functionalized HPE.

While EDTA is known to form complexes with both Zn²⁺ and Mn²⁺, its behavior in the HPE is governed by relative binding affinities and reversible coordination dynamics. The complexation ability of EDTA was further confirmed by changes in the ¹H NMR spectra compared to free EDTA (Fig. S21 and S22). The coordination to Zn²⁺ shows distinct methylene environments for the chelated and non-chelated EDTA. Complexation of Zn²⁺ by EDTA results in methylene protons experiencing different chemical environments, giving rise to the peak splitting. The paramagnetic Mn²⁺, on the other hand, causes severe broadening that largely obscures the fine structure. The ¹³C NMR spectra (Fig. S23) of EDTA show the sharp, well-resolved carbonyl and methylene carbons, whereas the complexation with Zn²⁺ produces separate chelated vs. free signals with modest downfield shifts. The Mn²⁺ complex shows severe broadening of the carbonyl signal, due to the enhanced paramagnetic relaxation effects arising from the proximity of Mn²⁺, which forms the coordinate bonds with the carbonyl oxygen atoms. The binding of Zn²⁺–EDTA (−41.1 eV) was found to be stronger than that of Mn²⁺–EDTA (−39.0 eV, optimized geometry, Fig. S24), indicating a preferential coordination of Zn²⁺ with EDTA. This selective affinity facilitates the Zn²⁺ transport while effectively suppressing the Mn-ion cross-talk within the electrolyte.⁵⁴ The calculated bond lengths of (Zn²⁺–EDTA)²⁻ and (Mn²⁺–EDTA)²⁻ complexes are summarized in Table S2.

To assess the impact of the dissolved Mn²⁺ on the Zn anode, electrochemical analysis was performed using an aqueous electrolyte containing 1 mM Mn²⁺ (AEMn). The CV profiles (Fig. S25a) for the Zn|AEMn|SS system reveal a nucleation overpotential of 81 mV, which is slightly higher than that of Zn|AE|SS, indicating that the presence of the Mn²⁺ influences the Zn plating process by increasing the energy barrier for nucleation. The LPR measurements (Fig. S25b) indicate that the Zn|AEMn|Zn system exhibits a more negative corrosion potential along with a higher corrosion current density compared to Zn|AE|Zn, suggesting that the Mn²⁺ accelerates parasitic side reactions, leading to faster Zn degradation. The impact of the Mn²⁺ on Zn reversibility was further examined through the plating–stripping analysis (Fig. S25c) at a current density of 1.0 mA cm⁻² and a capacity of 1.0 mA h cm⁻². The Zn|AEMn|Zn cell exhibited rapid failure after only 30 cycles, underscoring the adverse effect of the Mn²⁺ contamination on the Zn cycling stability. This failure can be attributed to the accumulation of the Mn-related side reactions, which promote the uneven Zn deposition, increase the interfacial impedance, and trigger the formation of the passivation layers or dendritic structures that disrupt the stable Zn stripping and plating. Additionally, the plating–stripping overpotential for Zn|AEMn|Zn was measured at 0.07 V (Fig. S25d), significantly higher than the 0.052 V observed for Zn|AE|Zn. This increase in the overpotential suggests that the Mn²⁺ not only destabilizes the Zn deposition kinetics but also enhances the polarization, leading to increased energy losses during the cycling. The impact of the externally added Mn on the Zn dissolution–deposition behavior was confirmed with the aid of the FESEM analysis of the Zn foil recovered after the plating–stripping cycling in AE_Mn (Fig. S26), and the Zn foil shows uneven Zn deposition and Zn dendrite formation. These findings highlight the critical challenge posed by the Mn dissolution in aqueous Zn-ion batteries, as it accelerates the Zn anode degradation, reduces the cycling stability, and compromises the overall battery performance. The results emphasize the necessity of the electrolyte engineering strategies, such as the chelating additives or protective interphases, to mitigate the Mn-induced side reactions and enhance the Zn anode reversibility for long-term battery operation. Finally, to demonstrate the feasibility of employing the HPE1C1 hydrogel polymer electrolyte for the flexible device application, a flexible device was fabricated. The CV profile (Fig. S27a) of the quasi-solid-state AZMB cell exhibits a reasonable performance as compared to the coin cell assembly, and the GCD profile (Fig. S27b) shows a discharge capacity of around 120 mA h g⁻¹ at 0.1 A g⁻¹.

The inclusion of the EDTA chelating additive in HPE1C1 leads to remarkable improvements in the ionic conductivity, corrosion resistance, zinc nucleation behavior, and electrochemical stability. By simultaneously suppressing the water-associated side reactions and enhancing the ion transport properties, HPE1C1 represents a significant advancement in the electrolyte design for ZMBs. The synergistic effects of the additive's coordination chemistry and its impact on the electrolyte's structural and electrochemical properties underscore its potential for enabling high-performance, durable energy storage systems.

Conclusions

This study presents a dual-functional hydrogel polymer electrolyte (HPE1C1) with EDTA as an additive in the PVA : PAA polymer matrix, offering a novel approach to overcoming the critical challenges in the zinc metal batteries (ZMBs). The hydrogel matrix provides high ionic conductivity, mechanical robustness, and water retention, while the incorporation of EDTA enables dual stabilization: coordinating Zn²⁺ to promote uniform, dendrite-free plating/stripping, and chelating Mn²⁺ to suppress the cathode–anode cross-talk. The DFT calculations elucidate the molecular origins of these effects, revealing that zincophilic –COO⁻ and –OH groups exhibit much stronger Zn²⁺ binding than water, thereby enhancing conductivity, and that the preferential Zn²⁺–EDTA interaction over Mn²⁺ selectively dampens the Mn²⁺ transport. The high density and reversible coordination capability of these functional groups enable efficient Zn²⁺ transport (t_Zn⁺ = 0.79). Electrochemical analyses demonstrate superior cycling stability with 81% capacity retention after 500 cycles, high coulombic efficiency (>99%), reduced polarization, and uniform Zn deposition. Even under high cathode loadings (2.0–4.0 mg cm⁻²), Mn cross-talk is effectively suppressed. The synergistic contributions of dendrite suppression, Mn²⁺ chelation, and reduced water activity underscore HPE1C1 as a high-performance electrolyte, offering a unified strategy to simultaneously overcome multiple intrinsic limitations of aqueous ZMBs.

Author contributions

A. B. and K. S. formulated the research idea and designed the study. The A. B. performed the experiments and prepared the initial draft. K. S. provided overall conceptual guidance and project supervision. F. N. F. carried out the computational studies and analysis, with the supervision of G. G. S. D. contributed to the scientific discussions and assisted in revising the manuscript. All the authors have approved the final version of the manuscript.

Conflicts of interest

The authors declare no conflict of interest.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). See DOI: https://doi.org/10.1039/d5ta07773g.

Acknowledgements

A. B. and K. S. acknowledge the Council of Scientific & Industrial Research (CSIR), New Delhi, India, for the financial support. K. S. acknowledges CSIR, New Delhi, India, for funding through the project FBR060309. A. B. acknowledges Dr Sapna Ravindranathan for her help in carrying out the NMR analysis.

References

1. Y. Zhao, J. Zhu, H. Lv and C. Zhi, Device, 2024, 2, 100556.
2. B. Sambandam, V. Mathew, F. Ahmad Nurul, S. Kim, M. Song and J. Kim, ACS Energy Lett., 2024, 9, 3058–3065.
3. J. Wang, Y. Zhou, Y. Zhuo, K. Fang, S. Wang, B. Zhao, J. Zhou and H. Wang, Chem. Sci., 2025, 16, 552–574.
4. F. Huang, W. Xu, Y. He, D. Li, Y. Tan and H. He, Battery Energy, 2025, 4, e70036.
5. N. A. Thieu, W. Li, X. Chen, S. Hu, H. Tian, H. N. N. Tran, W. Li, D. M. Reed, X. Li and X. Liu, Batteries, 2023, 9, 41.
6. Y. Yuan, S. D. Pu, M. A. Pérez-Osorio, Z. Li, S. Zhang, S. Yang, B. Liu, C. Gong, A. S. Menon, L. F. J. Piper, X. Gao, P. G. Bruce and A. W. Robertson, Adv. Mater., 2024, 36, 2307708.
7. X. Zhou, T. Ruan, J. Xu, C. Li, S. Huang, J. Zhou, S. Lu, R. Song and R. Li, RSC Adv., 2024, 14, 23023–23036.
8. C. Zhai, D. Zhao, Y. He, H. Huang, B. Chen, X. Wang and Z. Guo, Batteries, 2022, 8, 153.
9. T.-H. Wu, Y.-Q. Lin, Z. D. Althouse and N. Liu, ACS Appl. Energy Mater., 2021, 4, 12267–12274.
10. C. Zhan, T. Wu, J. Lu and K. Amine, Energy Environ. Sci., 2018, 11, 243–257.
11. D. Wang, S. Hu, T. Li, C. Chang, S. Li, S. Guo, H. Li, Q. Liu, J. Gong, J. Zhou and C. Han, Nat. Commun., 2025, 16, 259.
12. S. Chen and C. Zhi, Commun. Chem., 2025, 8, 1–10.
13. W. Lv, J. Tao, D. Huang, J. Zhu, B. Wang, X. Yin, J. Xu and H. He, Adv. Funct. Mater., 2025, e13857.
14. F. Li, H. Zhang, X. Liu, B. Xu and C. Wang, Batteries, 2024, 10, 420.
15. N. Hu, W. Lv, H. Tang, H. Qin, Y. Zhou, L. Yi, D. Huang, Z. Wu, J. Liu, Z. Chen, J. Xu and H. He, J. Mater. Chem. A, 2023, 11, 14921–14932.
16. H. Tang, N. Hu, L. Ma, H. Weng, D. Huang, J. Zhu, H. Yang, Z. Chen, X. Yin, J. Xu and H. He, Adv. Funct. Mater., 2024, 34, 2402484.
17. N. Hu, J. Tao, Y. Tan, H. Song, D. Huang, P. Liu, Z. Chen, X. Yin, J. Zhu, J. Xu and H. He, Adv. Energy Mater., 2024, 14, 2404018.
18. L. Du, X. Hou, X. Yang, V. Siozios, B. Yan, X. Ju, E. Paillard, M. Winter, T. Placke and J. Li, Batteries Supercaps, 2022, 5, e202100376.
19. F. Wu, F. Du, P. Ruan, G. Cai, Y. Chen, X. Yin, L. Ma, R. Yin, W. Shi, W. Liu, J. Zhou and X. Cao, J. Mater. Chem. A, 2023, 11, 11254–11263.
20. Y. Yang, T. Wang, Y. Guo, P. Liu, X. Han and D. Wu, Mater. Today Chem., 2023, 29, 101384.
21. J. Li, A. Azizi, S. Zhou, S. Liu, C. Han, Z. Chang, A. Pan and G. Cao, eScience, 2025, 5, 100294.
22. K. Lu, T. Jiang, H. Hu and M. Wu, Front. Chem., 2020, 8, 546728.
23. W. Cai, X. Zhang, G. Li and L. Chen, Chem. Eng. J., 2024, 484, 149390.
24. W. Xu, Q. Xie, X. Chen, M. Wu, J. Zhang and Q. Ru, Ionics, 2023, 29, 2783–2791.
25. K. Xie, K. Ren, C. Sun, S. Yang, M. Tong, S. Yang, Z. Liu and Q. Wang, ACS Appl. Energy Mater., 2022, 5, 4170–4178.
26. H. Peng, D. Wang, F. Zhang, L. Yang, X. Jiang, K. Zhang, Z. Qian and J. Yang, ACS Nano, 2024, 18, 21779–21803.
27. S. S. More, N. D. Khupse, J. D. Ambekar, M. V. Kulkarni and B. B. Kale, Energy Fuels, 2022, 36, 4999–5008.
28. Y. Zeng, J. Yang, X. Shen, R. Li, Z. Chen, X. Huang, P. Zhang and J. Zhao, J. Power Sources, 2022, 541, 231681.
29. X. Yu, Y. Song and A. Tang, J. Power Sources, 2021, 507, 230295.
30. J. Cao, D. Zhang, R. Chanajaree, Y. Yue, Z. Zeng, X. Zhang and J. Qin, Adv. Powder Mater., 2022, 1, 100007.
31. L. Song, Y. Li, Y. Fan, Z. Liao, R. Zhang, X. Che, Z. Yang, H. Fan and L. Ma, Electrochim. Acta, 2025, 513, 145587.
32. K. Xie, K. Ren, Q. Wang, Y. Lin, F. Ma, C. Sun, Y. Li, X. Zhao and C. Lai, eScience, 2023, 3, 100153.
33. S. Dilwale, P. P. Puthiyaveetil, A. Babu and S. Kurungot, Small, 2024, 20, 2311923.
34. Q. Sun, Molecules, 2022, 27, 7009.
35. D. Xu, X. Ren, H. Li, Y. Zhou, S. Chai, Y. Chen, H. Li, L. Bai, Z. Chang, A. Pan and H. Zhou, Angew. Chem., 2024, 136, e202402833.
36. Y. Xia, R. Tong, J. Zhang, M. Xu, G. Shao, H. Wang, Y. Dong and C.-A. Wang, Nano-Micro Lett., 2024, 16, 82.
37. W. Shi, Z. Song, W. Zhang, S. Lian, F. Huang, Q. An and Q. Li, Energy Environ. Sci., 2024, 17, 7372–7381.
38. L. Zhang, J. Wu, T. Lu, X. Li, H. Wu, T. Chen, Y. Zhang, J. Wei, M. Hu, X. Zheng, H. Gao, Z.-D. Huang, Q. Wang, S. Wang and Z. Jin, Chem. Eng. J., 2024, 496, 153815.
39. M. Zhu, H. Wang, H. Wang, C. Li, D. Chen, K. Wang, Z. Bai, S. Chen, Y. Zhang and Y. Tang, Angew. Chem. Int. Ed., 2024, 63, e202316904.
40. A. Bayaguud, X. Luo, Y. Fu and C. Zhu, ACS Energy Lett., 2020, 5, 3012–3020.
41. Y. He, R. Zhang, P. Zou, R. W. Chu, R. Lin, K. Xu and H. L. Xin, J. Am. Chem. Soc., 2025, 147, 6427–6438.
42. H. Lu, X. Zhang, M. Luo, K. Cao, Y. Lu, B. B. Xu, H. Pan, K. Tao and Y. Jiang, Adv. Funct. Mater., 2021, 31, 2103514.
43. J. Yu, M. Li, X. Kong, T. Wang, H. Zhang, X. Zhu, J. Zhao, Z. Ma and H. Yang, J. Mater. Chem. A, 2025, 13, 16182–16192.
44. S. Dang, Y. Wen, T. Qin, J. Hao, H. Li, J. Huang, D. Yan, G. Cao and S. Peng, Chem. Eng. J., 2020, 396, 125342.
45. W. T. Osowiecki, S. W. Sheehan, K. J. Young, A. C. Durrell, B. Q. Mercado and G. W. Brudvig, Dalton Trans., 2015, 44, 16873–16881.
46. Y. Wang, Y. Zhang, G. Gao, Y. Fan, R. Wang, J. Feng, L. Yang, A. Meng, J. Zhao and Z. Li, Nano-Micro Lett., 2023, 15, 219.
47. X.-Z. Zhai, J. Qu, S.-M. Hao, Y.-Q. Jing, W. Chang, J. Wang, W. Li, Y. Abdelkrim, H. Yuan and Z.-Z. Yu, Nano-Micro Lett., 2020, 12, 56.
48. W. Sun, F. Wang, S. Hou, C. Yang, X. Fan, Z. Ma, T. Gao, F. Han, R. Hu, M. Zhu and C. Wang, J. Am. Chem. Soc., 2017, 139, 9775–9778.
49. M. Chen, W. Zhou, A. Wang, A. Huang, J. Chen, J. Xu and C.-P. Wong, J. Mater. Chem. A, 2020, 8, 6828–6841.
50. Z. Zheng, H. Cao, W. Shi, C. She, X. Zhou, L. Liu and Y. Zhu, Polymers, 2023, 15, 212.
51. A. R. C. Bredar, A. L. Chown, A. R. Burton and B. H. Farnum, ACS Appl. Energy Mater., 2020, 3, 66–98.
52. C. Bischoff, O. Fitz, C. Schiller, H. Gentischer, D. Biro and H.-M. Henning, Batteries, 2018, 4, 44.
53. L. Zhang, J. Hu, B. Zhang, J. Liu, H. Wan, L. Miao and J. Jiang, J. Mater. Chem. A, 2021, 9, 7631–7639.
54. M. Tang, X. Zhao, R. Han, Y. Wang, Y. Ding, Z. Si, B. Li, D. Zhou and F. Kang, Angew. Chem., Int. Ed., 2025, 64, e202421574.

---