A sustainable biopolymer binder enables the fabrication of high-performance β-MnO₂ cathodes for aqueous zinc-ion storage

Abstract

Rechargeable aqueous zinc-ion batteries (ARZIBs) have gained considerable attention as sustainable energy storage systems due to their inherent safety, environmental friendliness, and low cost. Among various cathode candidates, β-MnO₂ is particularly attractive owing to its structural stability and abundance. However, its practical application is hindered by the dissolution of Mn²⁺ ions during cycling, which leads to poor long-term performance. In this study, β-MnO₂ was synthesized via a hydrothermal method and integrated into electrodes using both conventional PVDF and a novel water-based, cross-linked binder system composed of xanthan gum and citric acid (c-XG-CA). The c-XG-CA binder, abundant in hydroxyl, carboxyl, and acetyl groups, was shown to enhance Mn²⁺ adsorption capacity, improve electrode adhesion, and increase hydrophilicity compared to PVDF. The formation and stability of the cross-linked structure, along with its manganese ion adsorption behavior, were verified through FTIR and DFT analyses. Electrochemical evaluations revealed that the β-MnO₂-c-XG-CA cathode achieved superior cycling stability (73% capacity retention after 200 cycles at C/2) and higher diffusion coefficients. Post-cycling XRD and SEM characterization studies indicated the formation of reversible Zn–buserite and Zn_x(OTf)_y(OH)_2x−y·nH₂O phases. These findings demonstrate that the c-XG-CA binder offers significant structural and electrochemical advantages, making it a promising alternative to conventional binders for high-performance ARZIBs.

---

Introduction

To date, many different cathode materials have been used in rechargeable aqueous zinc-ion batteries (ARZIBs). Among these, MnO₂ stands out due to its advantages such as straightforward synthesis, low cost, and the existence of various polymorphic forms (α, β, γ, δ and λ), which endow it with diverse electrochemical properties. The first commercially successful system utilizing MnO₂ as a cathode was the Zn‖MnO₂ primary battery with an alkaline electrolyte (i.e. %30 KOH), which was introduced in 1952 and is widely used in household electronics.¹ However, the cycle life of this battery system is quite limited (20–30 cycles).² The main reason for this limitation is the formation of irreversible products such as ZnO and Mn(OH)₂ on the surface of zinc and MnO₂ electrodes during the charge/discharge processes. To overcome this problem, rechargeable Zn‖MnO₂ battery systems have been developed in the following years using aqueous electrolyte containing slightly acidic ZnSO₄.³ These systems have attracted considerable attention due to their longer cycle life and have led to an increase in studies on ARZIBs since 2012.⁴

Various MnO₂ polymorphs have been extensively investigated as cathode materials for ARZIBs, with β-MnO₂ standing out as the most thermodynamically stable form, despite possessing the narrowest tunnel structure (∼[1 × 1], 2.3 × 2.3 Å).⁵ However, the Jahn–Teller effect, which occurs during the reversible intercalation of Zn²⁺ into the β-MnO₂ structure, leads to the dissolution of Mn²⁺ ions into the electrolyte environment, which reduces the cycle stability.⁶ To overcome this problem, various strategies have been developed, as exemplified by using electrolyte additives and coating the cathode surface with materials such as carbon, graphene oxide (GO), conductive polyacrylonitrile (PANI) or tannic acid.^7–9 For example, the cycling stability of the β-MnO₂ electrode synthesized in nanorod form by the hydrothermal method was significantly enhanced by the addition of Mn(CF₃SO₃)₂ into Zn(CF₃SO₃)₂ containing aqueous electrolyte. This electrolyte additive was shown to suppress Mn²⁺ dissolution and maintain electrode integrity by forming a uniform and porous MnO_x layer on the cathode surface. Thus, the Zn‖β-MnO₂ cell exhibited high capacity (225 mAh g⁻¹ at 0.65C), high-rate performance (100 mAh g⁻¹ at 32.50C) and long cycle life.¹⁰ In another study, β-MnO₂ was processed with expanded graphite and its surface was coated with a thin carbon layer to prevent Mn dissolution. The resulting β-MnO₂/C cathode exhibited a capacity of 130 mAh g⁻¹ at a current density of 300 mA g⁻¹ for 400 cycles in 3 M Zn(CF₃SO₃)₂ + 0.1 M M MnSO₄ aqueous electrolyte.¹¹ Similarly, for the β-MnO₂/GO composite synthesized by the hydrothermal method, it was reported that GO suppressed the dissolution of Mn²⁺ ions and thus 130 mAh g⁻¹ capacity was maintained even after 4000 cycles at 4C in electrolyte containing 3 M ZnSO₄ + 0.2 M MnSO₄.¹² The β-MnO₂/GO/PANI composite electrode obtained by aniline polymerization presented a capacity retention rate of 82.7% after 600 cycles in 2 M ZnSO₄ electrolyte.¹³ In a more recent study, β-MnO₂ was coated with tannic acid, and it was reported that tannic acid chelates Mn²⁺ ions thanks to its structure rich in hydroxyl groups inhibiting Mn²⁺ dissolution and providing improved electrochemical performance in 3 M Zn(CF₃SO₃)₂ aqueous electrolyte containing 0.1 M Mn(CF₃SO₃)₂.⁹

Although research in the ARZIB field has mainly focused on components such as the cathode, anode and electrolyte, the role of binders has remained comparatively underexplored, despite their significant influence on the electrochemical performance and cycling stability of electrode materials.¹⁴ These binders are expected to adhere strongly to the current collector and be electrochemically stable over the applied voltage range of the cells.¹⁵ Binders can be divided into traditional, biopolymer, or conductive polymer; however, conventional binders such as PVDF are restricted due to the toxicity and environmental impact of N-methyl-2-pyrrolidone (NMP) as a solvent; therefore, biopolymers are preferred.¹⁶ Studies on the implementation of environmentally friendly and water-based binder materials instead of polyvinylidene fluoride (PVDF), one of the traditional binders commonly used in ARZIB systems, are still limited. These water-based binders containing functional groups such as –NH₂, –OH, C–O–C, –COOH and –C=O have been shown to effectively reduce the dissolution of the Mn-based cathode by forming strong interactions with Mn²⁺ ions.^17–19 To date, various water-based binders including sodium alginate (SA), hydroxyethyl cellulose (HEC), carboxymethyl cellulose (CMC), and calcium crosslinked SA have been systematically explored in ARZIBs employing MnO₂ as the cathode material. The observed enhancements in electrochemical performance have been attributed to the effective integration of these binders, which contribute to improved structural stability and interfacial adhesion.^20–23 For example, the MnO₂ cathode prepared with hydrophilic CMC reached a reversible capacity of 100 mAh g⁻¹ at a current density of 0.2 A g⁻¹. In contrast, an analogous cell assembled with a PVDF binder exhibited a complete capacity fade, reaching zero after approximately 200 cycles. These findings reveal that the hydrophilic nature of the binder significantly improves the performance and cycle stability of MnO₂ cathodes.²¹ Recent advances in aqueous zinc-based batteries demonstrate that optimizing electrolytes (e.g., magnesium aluminosilicate-based colloids and chlorine-containing systems) and designing functional polymer binders (polyimide-based polymers rich in the sulfonic acid group (R–SO₃H)) can effectively regulate ion solvation, suppress active material dissolution, and enhance long-term cycling stability across various cathode chemistries.^24–26

Herein, β-MnO₂ was initially synthesized via a hydrothermal method and the electrochemical performance of the cathode prepared using water-based, cross-linked binders was systematically investigated. In this study, for the first time, the effect of a cross-linked xanthan gum (XG)-citric acid (CA) (denoted as c-XG-CA) binder, prepared based on the reported method, on battery performance was evaluated. Xanthan gum is an acidic polysaccharide derived from sources such as sugar cane and corn, comprising trisaccharide side chains linked to glucose units in the backbone through α-1,3 bonds with mannose (β-1,4) and glucuronic acid (β-1,2). Previous reports have demonstrated its use as an electrolyte component to stabilize zinc surfaces.^27,28 In these studies, this biopolymer has primarily been used as an electrolyte or an interfacial modifier. However, to date, there has been no systematic investigation into its application as a water-based binder in ARZIBs. Therefore, this work represents one of the first comprehensive studies evaluating the binder functionality of this biopolymer in ARZIB systems.

Experimental

Synthesis of β-MnO₂

β-MnO₂ was synthesized as follows according to the method described in the literature.^29,30 First, 8 mmol of manganese sulfate monohydrate (MnSO₄·H₂O) (Sigma-Aldrich, 99% purity) and 12 mmol of ammonium persulfate ((NH₄)₂S₂O₈) (Sigma-Aldrich, 98% purity) were dissolved in 15 mL of distilled water. The resulting solution was stirred with a magnetic stirrer for about 15 minutes at room temperature until a homogeneous mixture was obtained. The resulting homogeneous solution was subsequently transferred to a Teflon-lined autoclave and maintained at 120 °C for 12 hours. The dark brown product formed after the reaction was washed several times with distilled water and ethanol and then dried in an oven at 80 °C to obtain the final β-MnO₂ product.

Material characterization

X-ray diffraction (XRD, Bruker D8) analyses were performed in the range of 10°–90° to determine the phase structure of the synthesized β-MnO₂. Scanning electron microscopy (SEM, SEM-Philips XL 30) and energy dispersive X-ray spectroscopy (EDS) analyses were applied to investigate the structural properties and surface morphology. The functional groups contained in the binders were characterized by Fourier transform infrared spectroscopy (FTIR, PerkinElmer 100). The mechanical properties of the binder films were evaluated by Atomic Force Microscopy (AFM, Veeco NanoScope IV) analysis at room temperature and in air. In addition, the thermal strength of the binders and their structural changes after crosslinking were investigated by Thermogravimetric Analysis (TGA, TA Instruments Q500) by heating from room temperature to 800 °C under a nitrogen atmosphere. The wettability properties of the electrode surfaces were determined using a KSV CAM 200 contact angle meter. The Highest Occupied Molecular Orbital (HOMO) and Lowest Unoccupied Molecular Orbital (LUMO) energy levels of binders (PVDF, XG and c-XG-CA) were calculated using the B3LYP/6-31G+(d,p) level of theory with the Gaussian 16 software package.³¹

Binder preparation and their Mn²⁺ ion adsorption capacity

For the preparation of the binder, 180 mg xanthan gum (Alfasol) and 10 mg citric acid (Sigma-Aldrich, 99.5% purity) were dissolved in 30 mL distilled water. After 12 hours of stirring, the homogenized solution was poured into a Petri dish. The electrode was kept at 165 °C for 7 min to induce cross-linking between xanthan gum and citric acid. For the 4% of PVDF solution, PVDF was dissolved in N-methyl-2-pyrrolidone (NMP) solvent and poured into a Petri dish and dried at 80 °C. In this study, 4.2 mg of cross-linked xanthan gum-citric acid (c-XG-CA) film and 7.4 mg of PVDF film were used as adsorbents. Mn²⁺ reference solutions were prepared by dissolving MnSO₄ salt in distilled water. All prepared reference solutions were analyzed by Electron Paramagnetic Resonance (EPR, JEOL JES-FA300) spectroscopy. Samples of 10 μL of 10 mM, 13 mM, 15 mM, 21 mM and 23 mM Mn²⁺ solutions with different concentrations were transferred to quartz capillary tubes and EPR measurements were performed at 9.8 GHz microwave frequency. The EPR signals of these five different concentrations were integrated and the calibration curve was plotted with the data obtained. For the adsorption studies, the binder films were kept in 20 mM MnSO₄ solution for 24 hours and then the adsorption capacities of binder films were calculated. Furthermore, the presence and distribution of manganese on the binder surface were confirmed by energy dispersive X-ray spectroscopy (EDS) analysis. The chemical changes in the binder structures before and after adsorption were analyzed by Fourier Transform Infrared Spectroscopy (FTIR). Total energies of divalent cation Mn²⁺ coordinated with binder functional groups were determined by geometric optimizations using the three-parameter Becke model with the Lee–Yang–Parr modification (B3LYP) and the 6-311++G(3df,3pd) basis set using the GAUSSIAN16 software package.^32,33 Binding energies were determined using the equation:²¹ ΔE_B = ΔE_binder-Mn − ΔE_binder − ΔE_Mn, where ΔE_B is the binding energy, ΔE_binder-Mn is the total energy of the functional groups of the binder and Mn²⁺, ΔE_binder is the total energy of the PVDF binder and functional groups of c-XG-CA (pyruvate, acetyl, and carboxyl) and ΔE_Mn is the total energy of Mn²⁺.

Electrochemical analyses

The working electrode was coated on a graphite current collector by preparing a slurry containing 80 wt% β-MnO₂, 10 wt% Ketjen black and 10 wt% binder and then dried in a vacuum oven at 80 °C for 24 hours. The crosslinking process of XG-CA was carried out by first coating the electrode slurry onto a graphite plate and drying it at room temperature, followed by placing it in a muffle furnace at 165 °C for 10 min to achieve crosslinking. The electrode thickness was 0.177 mm (the thickness of the graphite plate is 0.519 mm) and the active material loading was 1.5 mg for a 1 cm² electrode area. For the electrochemical tests, two-electrode Swagelok-type cells were assembled with the β-MnO₂ cathode, a zinc metal anode (Alfa Aesar, thickness: 0.25 mm, 99.98% metal basis, area: 3.8 cm²), a GF/C glass microfiber separator and aqueous electrolytes containing 1 M ZnSO₄ or 1 M Zn(CF₃SO₃)₂. Galvanostatic charge–discharge (GCD) cycling tests were performed using different binder systems (c-XG-CA, PVDF) in the potential range of 0.8–1.9 V. The rate performance tests were conducted by applying different current densities (C/5, 1C, 3C, 5C, and 10C) (1C corresponds to 616.6 mAh g⁻¹). The Galvanostatic intermittent titration technique (GITT) was applied to determine the diffusion coefficient of the active material. In the GITT measurement, at a current density of C/5, starting from the open circuit voltage (OCV), 5-minute discharge steps were followed by a 1-hour resting time for the cell after each step. Electrochemical impedance spectroscopy (EIS) measurements were performed at open circuit voltage with a 5 mV amplitude over a frequency range of 1 MHz–10 mHz.

Results and discussion

Characterization of β-MnO₂

The phase and crystallinity of the synthesized MnO₂ active material were characterized using X-ray diffraction (XRD). In the obtained XRD pattern (Fig. 1a), the diffraction peaks observed at 2θ = 28.7°, 37.4°, 41.0°, 42.8°, 46.1°, 56.7°, 59.3°, 65.0° and 67.1° correspond to the (110), (101), (200), (111), (210), (211), (220), (002) and (310) planes, respectively. These reflections match well with the standard β-MnO₂ phase (JCPDS no: 24-0735).^10,30 In addition, some low intensity peaks indicated by the # symbol point to the presence of an α-MnO₂ phase, which was identified using JCPDS card number 44-0141.³⁴ The β-MnO₂ crystal structure exhibits a tetragonal unit cell composed of MnO₆ octahedral units, with the lattice parameters a = b = 4.4 Å and c = 2.874 Å.³⁵ The interplanar spacing corresponding to the (101) plane was determined to be 2.40 Å, further confirming the formation of the β-MnO₂ phase. This phase is typically obtained by the oxidation of Mn²⁺ ions by S₂O₈²⁻ and described by the following reaction:²⁹

MnSO₄ + (NH₄)₂S₂O₈ + 2H₂O → MnO₂ + (NH₄)₂SO₄ + 2H₂SO₄

Fig. 1 (a) Structure and morphology of β-MnO₂, (a) XRD patterns of β-MnO₂, and (b) EDS spectrum of β-MnO₂ and the SEM image embedded in it.

Characterization of binders

Fourier transform infrared (FTIR) spectroscopy was used to analyze the functional groups in the xanthan gum, citric acid, XG-CA, and c-XG-CA binder film and determine the presence of different types of bonds, including covalent ester bonds, ionic bonds, and hydrogen bonds. XG possesses a cellulose-like backbone, with trisaccharide side chains composed of β-D-mannose-(1,4)-β-D-glucuronic acid-(1,2)-β-D-mannose units attached to every alternate glucose residue.³⁶ The mannose residues linked to the backbone are acetylated, which is essential for the formation of the double helix structure. However, the electrostatic repulsion forces between the charged side chains prevent their full integration into the helical structure. As a result, fibrils are formed by self-assembly of the double helices, and a network structure develops through interactions between the helical ends.³⁶ Consequently, XG adopts a double helix structure and contains hydroxyl, ester and ionized groups. The presence of these functional groups was verified by the FTIR spectrum of XG. As seen in Fig. 2a, the absorption bands at 3313, 2988, 1722, 1603, 1260 and 1035 cm⁻¹ in the FTIR spectrum are attributed to O–H stretching vibration, C–H anti-symmetric stretching vibration, C=O stretching vibration, C–O axial deformation of the enol group, CH₂ variable angle vibration and C–O stretching vibration, respectively.³⁷ As shown in Fig. 2a, the FTIR spectrum of CA exhibits characteristic absorption bands in the 1696 cm⁻¹ and 1745 cm⁻¹ regions corresponding to carbonyl (C=O) stretching vibrations of aliphatic carboxylic acid groups. As a result of the heat treatment applied to the XG-CA film at 165 °C, a condensation reaction occurred between the carboxyl groups of CA and the hydroxyl groups of XG (denoted as c-XG-CA). This crosslinking process was evidenced by the appearance of a new peak at a wave number of 1733 cm⁻¹ in the FTIR spectrum, corresponding to the formation of ester bonds (Fig. 2a).³⁸ After heat treatment, the absorption band of the C=O group shifted from 1696 cm⁻¹ to 1745 cm⁻¹, while the C–O stretching vibration of the acetyl group shifted from 1035 cm⁻¹ to 1016 cm⁻¹. Furthermore, the acid/ester ratio showed a small change from 1.12 to 0.74 in the presence of citric acid.³⁹ This change indicates that the added crosslinker stabilizes the amounts of carbonyl groups. On the other hand, the FTIR spectrum of PVDF powder is presented in Fig. 2b. The bands at 872, 1181 and 1403 cm⁻¹ are attributed to the non-polar α-phase, while the bands at 840 and 1066 cm⁻¹ correspond to the polar β-phase. The β-phase of PVDF is characterized by CH₂ bending deformations, CF₂ tensile vibrations and C–C stress bands, revealing its polarity properties. In contrast, the α-phase is characterized by C–C tensile vibrations, indicative of its non-polar crystal structure.⁴⁰

Fig. 2 FTIR spectra of (a) XG powder, CA powders, XG-CA film and c-XG-CA film, (b) PVDF powder, (c) TGA curves of XG, XG-CA, c-XG-CA and PVDF films, and (d) comparison of HOMO and LUMO levels of binders.

As seen in Fig. 2c, XG-CA and c-XG-CA samples exhibited different thermal degradation behaviors in different temperature ranges. It was observed that the amount of water retained on the surface of the c-XG-CA film decreased after heat treatment. TGA analysis revealed that the XG-CA sample exhibited a three-step weight loss behavior, whereas the c-XG-CA sample displayed a two-step degradation profile. In the second stage (200–422 °C), 68% and 63% weight loss occurred for c-XG-CA and XG-CA, respectively. In the third stage (400–650 °C), only the XG-CA sample showed a weight loss of 22%, which was attributed to the lack of crosslinking. Similarly, the TGA curve of cross-linked Fe-XG has been reported to show degradation over a wider temperature range in the final stage.⁴¹ It was also found that the PVDF binder showed thermal stability at higher temperatures compared to c-XG-CA.

DFT calculations were carried out to evaluate the electrochemical stability of PVDF, XG and c-XG-CA binders. The Highest Occupied Molecular Orbital (HOMO) and Lowest Unoccupied Molecular Orbital (LUMO) energy levels were calculated using the B3LYP/6-31G+(d,p) level of theory with the Gaussian 16 software package.³¹ A high HOMO level indicates that the material is more prone to oxidation in an anodic environment, while a low LUMO level indicates a higher tendency for reduction in cathodic environments. Since it is energetically favorable to donate electrons from a high-lying HOMO and accepts them into a low-lying LUMO, a narrow HOMO–LUMO gap implies lower kinetic stability and higher chemical reactivity.⁴² Binders exhibited HOMO values below −4.46 eV, indicating electrochemical stability during charge–discharge cycles.⁴³ The HOMO–LUMO gap (Fig. 2d) of XG is 5.24 eV, which increased to 6.08 eV upon crosslinking, suggesting enhanced structural stability. Moreover, c-XG-CA exhibited a lower HOMO–LUMO gap (ΔE = 6.08 eV) indicating higher electronic conductivity. These findings demonstrate that the synthesized c-XG-CA binders are promising candidates for ARZIBs due to their structural integrity within the operating potential window.³²

Atomic Force Microscopy (AFM) analysis was performed to study the surface morphologies of β-MnO₂ electrodes prepared using two different binders.⁴⁴ As shown in Fig. S1a and b, β-MnO₂ electrodes with PVDF and c-XG-CA binders exhibited a surface roughness of approximately 41.7 nm and 32.3 nm, respectively. The lower surface roughness of the β-MnO₂ electrode prepared with the c-XG-CA binder indicates a more homogeneous and stable structure compared to the electrode with the PVDF binder. In addition, to evaluate the binder strength, the electrodes were subjected to a peel-off test with double-sided tape (Fig. S2a and b). After the tape that adhered to the electrode surface was removed, the amount of material remaining on the tape was converted to black and white format and analyzed using Python software. When the black pixel counts were calculated, a value of 16.79 was obtained for the β-MnO₂ electrode with the PVDF binder (Fig. S2a) and 3.34 was obtained for the electrode with the c-XG-CA binder (Fig. S2b). In addition, a large portion of the electrode with the c-XG-CA binder remained on the graphite current collector after peeling, indicating that this binder provides higher adhesion strength. The contact angles were determined to be 84° for the β-MnO₂ electrode containing PVDF (Fig. S3a) and 26° for the electrode with the c-XG-CA binder (Fig. S3b). These results reveal that the electrode with the c-XG-CA binder has better wettability and exhibits hydrophilic properties.

Electrochemical performance of the β-MnO₂ cathode

Effect of salt on electrochemical performance. Electrochemical performance was comparatively evaluated using 1 M ZnSO₄ (ionic conductivity: 50 mS cm⁻¹ and pH: 5.16) and 1 M Zn(CF₃SO₃)₂ (ionic conductivity: 55.5 mS cm⁻¹ and pH: 5.37) electrolytes. Cyclic voltammetry (CV) curves of β-MnO₂ with the PVDF binder in aqueous 1 M ZnSO₄ and 1 M Zn(CF₃SO₃)₂ electrolytes at a scan rate of 0.1 mV s⁻¹ are presented in Fig. 3a and b, respectively. After the first cycle, in both electrolytes, redox pairs were observed, indicating two-stage charge storage.⁴⁵ The first reduction peak occurred at a lower potential (0.93 V vs. Zn²⁺/Zn) in Zn(CF₃SO₃)₂ electrolyte. In the second cycle, reduction peaks were recorded at 1.37/1.21 V and 1.34/1.08 V in ZnSO₄ and Zn(CF₃SO₃)₂ electrolytes, respectively. The significant difference observed between the first and subsequent cycles is ascribed to the phase transition in the electrode material.¹⁰ The discharge/charge plateaus obtained at 1C current density are shown in Fig. 3c and d. The flat discharge/charge plateaus observed at 1.2 V and 0.93 V in ZnSO₄ and Zn(CF₃SO₃)₂ electrolytes, respectively, are in good agreement with the CV curves. In the following cycles, the steep plateaus, which become more pronounced especially in ZnSO₄ electrolyte, indicate the reversible intercalation of first H⁺ and then Zn²⁺ ions.^45,46 The β-MnO₂ cathode prepared with the PVDF binder exhibited a better performance with an initial capacity of 171 mAh g⁻¹ in Zn(CF₃SO₃)₂ electrolyte and maintained a discharge capacity of 84 mAh g⁻¹ after 100 cycles, indicating higher stability compared to the value of 49 mAh g⁻¹ in ZnSO₄ electrolyte (Fig. 3e and f). This performance difference is attributed to the fact that CF₃SO₃⁻ anions limit the solvation effect by reducing the water molecules around the Zn²⁺ ions, thus facilitating ion transport.⁴⁷ Zn²⁺ ions are densely surrounded by water molecules in the ZnSO₄ electrolyte forming [Zn(H₂O)₆]²⁺ hydrated complexes due to the high polarity of aqueous electrolytes. Thus, these complexes need to surpass a certain energy barrier to reduce the solvation effect and enable efficient ion transport.⁴⁸

Fig. 3 β-MnO₂-PVDF electrode CV curves in (a) 1 M ZnSO₄ electrolyte and (b) 1 M Zn(CF₃SO₃)₂ electrolyte. Voltage profiles in (c) 1 M ZnSO₄ electrolyte and (d) 1 M Zn(CF₃SO₃) electrolyte. Cycling performances at 1C current density in (e) 1 M ZnSO₄ electrolyte and (f) 1 M Zn(CF₃SO₃) electrolyte (note: 1C is 616.6 mAh g⁻¹).

Effect of binder on electrochemical performance. In the previous section, the effect of zinc salt on the electrochemical performance of β-MnO₂ was investigated and it was concluded that Zn(CF₃SO₃)₂ electrolyte performed better than ZnSO₄ electrolyte. Therefore, the experiments were carried out in 1 M Zn(CF₃SO₃)₂ aqueous electrolyte to investigate the impact of the binder on the electrochemical performance. The galvanostatic discharge/charge curves of the β-MnO₂ cathode prepared with PVDF and c-XG-CA binders at C/2 current density are shown in Fig. 4a and b. The initial discharge profiles of all cathodes displayed characteristic voltage plateaus around 1 V vs. Zn²⁺/Zn, indicating similar redox behavior. Notably, the choice of binder did not influence the reduction and oxidation peak positions of β-MnO₂. At C/2, the β-MnO₂-c-XG-CA cathode maintained a discharge capacity of 157 mAh g⁻¹ after 200 cycles, outperforming β-MnO₂-PVDF (76 mAh g⁻¹) (Fig. 4c and d). As a result, after 200 cycles, β-MnO₂-c-XG-CA and β-MnO₂-PVDF retained 73% and 37% of their initial discharge capacities, respectively. The β-MnO₂-c-XG-CA cathode exhibited superior performance compared to β-MnO₂-PVDF in 1 M aqueous Zn(CF₃SO₃)₂ electrolyte. Subsequently, comparative analyses were conducted to assess the effect of the crosslinked binder structure (Fig. S4). In contrast to the non-crosslinked counterpart (XG-CA without heat treatment), c-XG-CA exhibited significantly improved capacity retention, which can be attributed to the enhanced structural integrity and electrode stability provided by the binder. This enhanced performance is attributed to the c-XG-CA binder, which contains a high density of polar functional groups (hydroxyl, ester, and ionizable groups). During electrochemical processes, it is well established that Mn²⁺ ions tend to dissolve into the electrolyte, leading to capacity fading and structural degradation. The chelation of Mn²⁺ by specific functional groups has been shown to significantly suppress this dissolution. The ability of various functional groups to chelate Mn²⁺ ions was evaluated, and the following binding energy trend was proposed: amide < ester < lithium carbonate < carboxylic acid < ether ≈ nitrile < alcohol < fluoroalkane.^49–52 As a result, the superior performance of the β-MnO₂-c-XG-CA cathode can be attributed to its high chelation capacity for Mn²⁺ ions.

Fig. 4 Voltage profiles of (a) β-MnO₂-c-XG-CA and (c) β-MnO₂-PVDF; cycling performances of (b) β-MnO₂-c-XG-CA and (d) β-MnO₂-PVDF cycled at C/2 current density (note: 1C is 616.6 mAh g⁻¹).

In addition to the electrodes with a 10 wt% binder, electrodes containing different binders (5 wt% and 15 wt%) were also prepared to determine the optimum ratio. The 15 wt% binder electrode consisted of 80% β-MnO₂ and 5 wt% Ketjen black, whereas the 5 wt% binder electrode consisted of 80% β-MnO₂ and 15 wt% Ketjen black. Fig. S5 shows the electrochemical performance of β-MnO₂ electrodes with different binder contents. Among these, the 10 wt% binder electrode (80 wt% β-MnO₂ and 10 wt% Ketjen black) exhibited the highest capacity at 1C (Fig. S5a), superior high-rate performance (Fig. S5b), and the best capacity recovery upon returning to 1C. The electrochemical performances of β-MnO₂-c-XG-CA and β-MnO₂-PVDF cathodes were investigated at different current densities. As illustrated in Fig. 5a, the β-MnO₂-c-XG-CA cathode exhibited discharge capacities of 239, 127, 87, 70 and 55 mAh g⁻¹ at current densities of C/5, 1C, 3C, 5C and 10C, respectively. When the current density was adjusted back to 1C, the β-MnO₂-c-XG-CA cathode exhibited a reversible electrochemical performance by exhibiting a discharge capacity of 117 mAh g⁻¹. On the other hand, the β-MnO₂-PVDF cathode exhibited lower discharge capacities at the same current densities. Besides, when the current density was set at 1C, the β-MnO₂-PVDF cathode showed poorer reversibility, exhibiting a very low discharge capacity (68 mAh g⁻¹) compared to the initial value (83 mAh g⁻¹). Then, the long cycle performance of the β-MnO₂-c-XG-CA cathode was evaluated at 1C current density and exhibited a discharge capacity of 92 mAh g⁻¹ after 500 cycles, indicating a capacity retention rate of 62% (Fig. 5b). However, even after extended cycling beyond 800 cycles, the electrode was still able to deliver around 60 mAh g⁻¹, despite showing a relatively faster capacity fade (Fig. S6). To evaluate the thermal stability of the binder, an electrochemical cycling test was conducted at 55 °C (Fig. S7), where the c-XG-CA binder containing electrode exhibited accelerated capacity fading, indicating its limited stability under elevated temperature conditions. Although the β-MnO₂-c-XG-CA cathode exhibited improved performance in 1 M Zn(CF₃SO₃)₂ aqueous electrolyte, the capacity decay that occurred over time may be due to the dendrite structures, as revealed in the SEM images (Fig. S8a), in contrast to the smooth morphology of the pristine Zn surface (Fig. S8b). It was demonstrated that designing a new electrolyte system incorporating chaotropic additives can improve plating/stripping kinetics, reduce water reactivity and the tendency for the hydrogen evolution reaction (HER), thereby enabling the optimized Zn/MnO₂ batteries to exhibit high performance.⁵³

Fig. 5 (a) Electrochemical performances of β-MnO₂-c-XG-CA and β-MnO₂-PVDF cathodes at different current densities. (b) Long cycle performance of the β-MnO₂-c-XG-CA cathode at 1C current density (note: 1C is 616.6 mAh g⁻¹).

One of the main reasons for the poor cycling performance of β-MnO₂ electrode materials in aqueous acidic media is the dissolution of Mn²⁺ ions into the electrolyte due to the Jahn–Teller effect.⁵⁴ Mn²⁺ (total spin S = 5/2) and Mn⁴⁺ (S = 3/2) ions exhibit paramagnetic properties and can be detected by Electron Spin Resonance (ESR) spectroscopy.⁵⁵ On the other hand, the Mn³⁺ ion (S = 2) is diamagnetic and therefore “silent” in terms of ESR and cannot be directly detected by this technique.⁵⁶ In this context, Mn²⁺ ions dissolved from β-MnO₂ into the electrolyte were detected using ESR spectroscopy. Fig. S9 presents the ESR analysis results performed by sampling 10 μL from 1 mL electrolyte at the end of the first discharge cycle at a C/20 current density, aimed at quantifying Mn²⁺ ions dissolved in the 1 M Zn(CF₃SO₃)₂ electrolyte. The ESR spectra exhibit the characteristic signal of Mn²⁺ ions released into the electrolyte medium from β-MnO₂-c-XG-CA and β-MnO₂-PVDF cathodes. Notably, the dissolution of Mn²⁺ ions was detected in the electrodes prepared with the β-MnO₂-c-XG-CA binder. This indicates that while the binder could not entirely suppress manganese dissolution, the electrode still demonstrated improved cycling stability.

Mn²⁺ ion adsorption capacities of binders. To examine the reason why the c-XG-CA binder exhibits superior electrochemical performance compared to the PVDF binder, both c-XG-CA and PVDF films were prepared and the Mn²⁺ ion adsorption capacities of these films were investigated. First, MnSO₄ solutions were prepared in distilled water at concentrations of 10 mM, 13 mM, 15 mM, 21 mM, 23 mM in order to create a calibration curve. The characteristic peaks corresponding to Mn²⁺ in each solution are shown in Fig. 6a. Fig. 6b displays the ESR signals magnified within the 260 and 400 mT range, along with the integrated spectra from Fig. 6a. As shown in Fig. 6c, the intensity of the integrated signals increases with the Mn²⁺ concentration, ranging from 10 mM to 23 mM. The area under each curve was calculated and plotted against the concentration values, yielding a highly linear calibration curve (R²: 0.997). The amount of adsorbent is 4.2 mg and 7.4 mg for c-XG-CA and PVDF, respectively. These films were kept in 23 mM MnSO₄ solution for 24 hours in a way that would be compatible with the electrochemical performance period. Following a 24-hour adsorption period, 10 μL of the supernatant was sampled and subjected to ESR analysis. These results are given in Fig. 6d. The ESR signal of the sample obtained from the solution in which the PVDF film was immersed was observed to be relatively higher. The integrated ESR spectra are presented in Fig. 6e. Based on the calibration curve, the Mn²⁺ concentration in the PVDF-containing solution was determined to be 18.5 mM, while the corresponding value for the solution containing the c-XG-CA film was determined to be 8.94 mM. Consequently, the Mn²⁺ adsorption capacity of the PVDF film was calculated to be 0.33 mg g⁻¹, whereas that of the c-XG-CA film was found to be 1.84 mg g⁻¹. These results clearly demonstrate that the c-XG-CA film displays a significantly higher adsorption capacity for Mn²⁺ ions, which contributes to its enhanced electrochemical performance. Moon et al. implemented the ESR technique to demonstrate the dissolution of Mn²⁺ ions during discharge, and to mitigate capacity loss, they introduced a porous carbon interlayer, which significantly enhanced the electrochemical performance of the MnO₂ electrode.⁵⁶

Fig. 6 (a) ESR spectra of MnSO₄ solutions prepared at different concentrations, (b) the integrated ESR spectra of the Mn²⁺ reference solutions, obtained through integration of the data presented in (a), (c) stack plot of ESR spectra shown in (b), (d) calibration curve, (e) ESR spectra of MnSO₄ solution after the adsorption process and (f) integrated ESR spectra illustrated in (e).

The surface morphologies and elemental analysis of PVDF and c-XG-CA binder films before and after adsorption were analyzed by SEM and EDS methods, respectively. SEM images shown in Fig. S10a and b and S11a and b reveal that both films have a homogeneous and smooth surface. EDS elemental analysis confirmed the presence of manganese on the surface of PVDF and c-XG-CA films after adsorption. In both cases, EDS analysis shows that the c-XG-CA binder (Fig. S10c) and PVDF films (Fig. S11c) do not contain manganese prior to adsorption. Especially in the EDS analysis of the c-XG-CA film after adsorption, it was determined that the amount of manganese on the surface (2 wt%) (Fig. S10d) was higher than that on the PVDF film (0.44 wt%) (Fig. S11d). These findings are consistent with the results of ESR analysis. FTIR spectra were evaluated to investigate changes in the functional groups of the binder films following Mn²⁺ adsorption. As shown in Fig. S12a, the intensity of the CF₂ stretching peak at 1165 cm⁻¹ in the PVDF film decreased after adsorption. Similarly, in the c-XG-CA film, a reduction in the intensity of the C=O band at 1733 cm⁻¹, attributed to the acetyl group, was observed after adsorption (Fig. S12b). These observations indicate that deprotonation of the carboxyl groups of D-glucuronic acid and pyruvic acetal acids in the structure of XG, used as a binder in ARZIBs, occurred upon soaking in an electrolyte containing zinc sulfate and manganese sulfate. This deprotonation caused the broadening of the –OH intensity observed in the FTIR spectrum.²⁸

DFT calculations were performed to evaluate the interactions between Mn²⁺ ions and functional groups of binders. Fig. S13 illustrates the comparison of the binding energies of Mn²⁺ with different binder components, including PVDF and c-XG-CA. As seen in Fig. S13, DFT calculations revealed that c-XG-CA binder functional groups, particularly carboxylic groups (−29.8 eV), exhibit stronger Mn²⁺ binding than pyruvate (−25.9 eV), acetyl (−27.1 eV), and PVDF (−2.36 eV), with all interactions being energetically favourable. Compared to PVDF, the functional groups and multidimensional main-chain interactions of c-XG-CA more effectively adsorb Mn²⁺, thereby reducing its loss from the cathode and improving cycle stability. Consistent with Bergmann et al., cations are proposed to bind to pyruvate residues at side-chain ends, forming intramolecular complexes within the same xanthan molecule.⁵⁷

Charge storage processes of β-MnO₂ cathodes. Based on CV analysis of β-MnO₂-c-XG-CA and β-MnO₂-PVDF electrodes, the reaction kinetics at various scan rates ranging from 0.1 mV s⁻¹ to 1 mV s⁻¹ were studied. The charge storage processes of β-MnO₂ cathodes involve both the diffusion of ions at the cathode-electrolyte interface and the redox reaction rate of β-MnO₂. According to the CV profiles in Fig. S14a and S15a, at the lowest scan rate of 0.1 mV s⁻¹, β-MnO₂-PVDF and β-MnO₂-c-XG-CA electrodes possess 1.34/1.07 and 1.37/1.11 reduction pairs and 1.53/1.60 and 1.52/160 oxidation pairs, respectively. The two-step electrochemical processes observed at both electrodes are associated with H⁺ and Zn²⁺ intercalation. The first reduction peak during discharge is due to H⁺ intercalation, while the subsequent peak is due to Zn²⁺ intercalation.¹⁹ With increasing scan rates, the two redox peaks merge into a broader single peak (Fig. S14a and S15a); this behavior provides insights into changes in reaction kinetics and the charge storage mechanism. The reaction kinetics of β-MnO₂-c-XG-CA and β-MnO₂-PVDF electrodes were studied by analyzing the relationship between current (i) and the scan rate (v): i = av^b.^22,58 Here, a and b are adjustable parameters. The value of b is determined from the slope of the linear plot between log(i) and log(v). b Values close to 1.0 indicate a capacitive-controlled process, while b values close to 0.5 indicate a diffusion-controlled redox process.²² The b value of the reduction peak of β-MnO₂-PVDF in Fig. S14b is 0.68 and the b value of the oxidation peak is 0.63, indicating a capacitive-controlled process. The b value of the reduction peak of β-MnO₂-c-XG-CA is 0.51, while the b value of the oxidation peak is 0.55 (Fig. S15b), indicating that a diffusion-controlled behavior is dominant. Similarly, the b values close to 0.5 for the MnO₂ electrode prepared with hydrophilic carboxymethyl cellulose and hydroxyethyl cellulose binders (b values higher than 0.5 for the MnO₂ electrode with the PVDF binder) indicated that the diffusion of ions at the cathode-electrolyte interface dominated.⁵⁹ The contributions of capacitive-controlled and diffusion-controlled processes to the total charge storage are determined according to the following equation: i(V) = k₁v + k₂v^1/2,⁵⁹ where i(V) is the current value corresponding to a constant potential and k₁v and k₂v^1/2 are capacitive-controlled and diffusion-controlled processes, respectively.⁶⁰ Fig. S14c and S15c demonstrate that the capacitive contribution increases with an increasing scan rate for both electrodes. This indicates that the capacitive-controlled storage mechanism becomes more efficient for energy storage at higher scan rates through rapid ion transfer of Zn²⁺ and H⁺ ions and efficient charge–discharge processes. At the same scan rate of 0.5 mV s⁻¹, the capacitive contribution rates of β-MnO₂-PVDF and β-MnO₂-c-XG-CA electrodes were calculated to be 67% (Fig. S14d) and 54% (Fig. S15d), respectively. Thus, ion diffusion processes play a more dominant role in the electrochemical process of the β-MnO₂-c-XG-CA electrode.

The galvanostatic intermittent titration technique (GITT) was used to measure the diffusion coefficient. The diffusion coefficient of Zn²⁺ was calculated based on the following equation:⁶¹

where t is the duration of the current pulse (s), τ is the relaxation time (s) and ΔE_s is the steady state potential change with the current pulse (V). ΔE_t is the potential change (V) during the steady current pulse after the IR drop is removed. Furthermore, M_B, V_M, m_B and A represent the molecular weight (g mol⁻¹), molar volume (cm³ mol⁻¹), amount of active material (g) and electrode area (cm²) of MnO₂, respectively. Fig. S16a and b show the GITT curves of β-MnO₂-c-XG-CA and β-MnO₂-PVDF electrodes in the 5th discharge cycle, respectively. The Zn²⁺ diffusion coefficient can be divided into two main stages during discharge, representing the intercalation of H⁺ and Zn²⁺ ions, depending on the energy barriers required. In particular, a decrease of the diffusion coefficient from 0.94 V to 0.8 V was observed for both electrodes. In an agreement with the CV curve, the second reduction peak was attributed to Zn²⁺ ion intercalation. Faster diffusion coefficient values up to 0.94 V support this phenomenon. The comparative values of the diffusion coefficient with respect to voltage given in Fig. S16c are obtained in the range of 2.3 × 10⁻⁸–8.2 × 10⁻⁹ and 1.5 × 10⁻⁸–4.3 × 10⁻¹⁰ cm² s⁻¹ for β-MnO₂-c-XG-CA and β-MnO₂-PVDF, respectively. As observed, the diffusion coefficient values of β-MnO₂-c-XG-CA are higher, aligning with those reported for MnO₂ electrodes prepared using hydrophilic binders.²² These findings indicate that electrode transport kinetics are strongly influenced by ion–binder interactions during the discharge process.

Electrochemical impedance spectroscopy (EIS) measurements were subsequently performed for β-MnO₂-c-XG-CA and β-MnO₂-PVDF cathodes. Fig. S17a and Fig. S17b depict the Nyquist plots of β-MnO₂-PVDF and β-MnO₂-c-XG-CA cathodes, respectively. The EIS plots consist of a semicircle in the high frequency region and an upward sloping line in the low frequency region. The EIS data of both electrodes fit using an equivalent circuit placed inside the Nyquist plot. Here, R1, Q1, R2, Q2 and R3 represent the internal resistance of the cell, polarization of the surface layer, charge transfer resistance, electrical double layer capacitance of the electrode interface and solid electrolyte interfacial formation, respectively.⁶² From the fitted EIS curves, the R_total of β-MnO₂-PVDF was found to be 166.10 Ω, while the R_total of β-MnO₂-c-XG-CA was determined to be 78.26 Ω. In addition, the R2 value expressing the charge transfer resistance is lower in the β-MnO₂-c-XG-CA (R2: 66.95) electrode than in the β-MnO₂-PVDF (R2: 156.5) electrode.

To date, the XRD technique used in studies to understand the charge storage mechanism of β-MnO₂ in Zn(CF₃SO₃)₂ aqueous electrolyte has reported the formation of new phases after the discharge processes. These new peaks were reported to originate from the Zn–buserite phase.⁹ In this study, it was observed that new peaks appeared after the discharge of the β-MnO₂-c-XG-CA electrode in 1 M Zn(CF₃SO₃)₂ electrolyte in the experiment performed at C/10 current density (Fig. 7a). The peak observed at around 10° in the XRD spectrum of the β-MnO₂-c-XG-CA electrode corresponds to the c-XG-CA film. After fully discharging β-MnO₂-c-XG-CA to 0.8 V (vs. Zn²⁺/Zn) in 1 M Zn(CF₃SO₃)₂ electrolyte at a current density of C/10, new diffraction peaks appeared at 6.6°, 13.2°, 19.8° and 33.1°, highlighted with a green background. These peaks indicate the formation of the Zn–buserite phase (Zn_0.5MnO_x·nH₂O) and Zn_x(OTf)_y(OH)_2x−y·nH₂O, agreeing with the literature.⁶¹ The diffraction peaks at 6.6° and 19.8° are assigned to crystallographic planes (002) and (003) of the layered Zn–buserite phase formed after the intercalation of Zn²⁺ into β-MnO₂-c-XG-CA, respectively.⁶³ The peaks in the orange background are due to the graphite current collector. Other peaks attributed to Zn_x(OTf)_y(OH)_2x−y·nH₂O precipitation indicate H⁺ intercalation.^64,65Fig. 7b shows the SEM image of the β-MnO₂-c-XG-CA electrode. The rod-like morphology of β-MnO₂ is distinguished. Furthermore, the characteristic flake-like morphology of the Zn_x(OTf)_y(OH)_2x−y·nH₂O precipitate, formed on the electrode surface as a discharge product, is revealed in the SEM image presented in Fig. 7c.

Fig. 7 (a) XRD spectrum of the β-MnO₂-c-XG-CA electrode, the XRD spectrum of the β-MnO₂-c-XG-CA electrode after first discharge at C/10 current density, and the XRD spectrum of the c-XG-CA film, (b) SEM image of the β-MnO₂-c-XG-CA electrode, and (c) SEM image of the β-MnO₂-c-XG-CA electrode after first discharge at C/10 current density.

Overall, a comparison of our c-XG-CA binder with other reported eco-friendly binders for MnO₂-based cathodes (Table S1) reveals that, without any electrolyte additives such as MnSO₄, c-XG-CA delivers a discharge capacity of 157 mAh g⁻¹ at C/2 with 73% retention after 200 cycles in 1 M Zn(CF₃SO₃)₂ electrolyte. This performance clearly exceeds that of PVDF under identical conditions (76 mAh g⁻¹, 37% retention) and is competitive with or superior to that of other green binders reported in additive-containing electrolytes. The superior performance of c-XG-CA is attributed to its high Mn²⁺ binding energy and multidimensional interaction capability, which effectively suppress manganese dissolution and enhance cycle stability.

Conclusions

In this study, a β-MnO₂ cathode material was synthesized via a hydrothermal method and its electrochemical performance was optimized using various polymeric binders and two different electrolyte salts (1 M ZnSO₄ and 1 M Zn(CF₃SO₃)₂). The use of Zn(CF₃SO₃)₂ electrolyte enhanced the specific capacity due to its ability to reduce the solvation effect of Zn²⁺ ions, thereby facilitating ion transport. Environmentally friendly and cost-effective polymeric binder xanthan gum (XG) was cross-linked via thermal treatment at 165 °C to improve electrode integrity. FTIR analysis confirmed the formation of covalent cross-links between hydroxyl groups of XG, ester groups of XG and carboxyl groups of citric acid. Density Functional Theory (DFT) calculations demonstrated that the cross-linked binder structures maintained structural stability within the operating potential window, highlighting their promise for aqueous rechargeable zinc-ion batteries (ARZIBs). The cathode fabricated with this novel binder, c-XG-CA, exhibited superior performance owing to its abundant hydroxyl, carboxyl, and acetyl groups that enabled strong Mn²⁺ ion adsorption. DFT calculations confirm that c-XG-CA interacts with Mn²⁺ much more strongly than PVDF, explaining its superior ability to suppress manganese dissolution. Furthermore, the helical structure of XG provided robust interfacial interactions among electrode components, ensuring homogeneous slurry distribution and enhanced adhesion through increased contact points between electrode layers. Compared to PVDF, the synthesized binders displayed greater hydrophilicity. Electrochemical Impedance Spectroscopy (EIS) revealed that the c-XG-CA-based cathode had significantly lower charge-transfer resistance, while GITT analysis showed higher diffusion coefficients, both contributing to enhanced electrochemical performance. Finally, SEM analysis identified flake-like discharge precipitation on the β-MnO₂-c-XG-CA electrode surface, indicating the formation of a new phase associated with the intercalation of Zn²⁺ and H⁺ ions during the discharge process. These findings collectively show that the c-XG-CA binder, due to its structural, chemical, and electrochemical advantages, represents a highly promising alternative to conventional binders for high-performance and sustainable aqueous zinc-ion battery systems.

Author contributions

Selin Sariyer: investigation, visualization, writing original draft. Rezan Demir-Cakan: writing – review & editing, supervision, resources, project administration.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the SI or can be obtained by the lead contacts upon reasonable request. Supplementary information: in particular, the SI includes GITT, EIS, and contact angle measurements, providing for comprehensive evaluation of the results. See DOI: https://doi.org/10.1039/d5se00939a.

Acknowledgements

The authors gratefully acknowledge the financial support provided by the Bosphorus Program (Project No: TUBITAK 119N054), jointly funded by the Scientific and Technological Research Council of Türkiye (TUBITAK) and the French Ministry for Europe. The authors would like to thank Dr Zeynep Erdöl for the X-ray diffraction (XRD) measurements and Mehmet Emre Aköz for the Electron Spin Resonance (ESR) analyses. The numerical calculations reported in this paper were fully performed at TUBITAK ULAKBIM, High Performance and Grid Computing Center (TRUBA resources).

References

1. W. J. Wruck, B. Reichman, K. R. Bullock and W. H. Kao, J. Electrochem. Soc., 1991, 138, 3560–3567.
2. K. Kordesch, J. Gsellmann, M. Peri, K. Tomantschger and R. Chemelli, Electrochim. Acta, 1981, 26, 1495–1504.
3. T. Yamamoto and T. Shoji, Inorg. Chim. Acta, 1986, 117, L27–L28.
4. C. Xu, B. Li, H. Du and F. Kang, Angew. Chem., Int. Ed., 2012, 51, 933–935.
5. Y. Xu, G. Zhang, J. Liu, J. Zhang, X. Wang, X. Pu, J. Wang, C. Yan, Y. Cao, H. Yang, W. Li and X. Li, Energy Environ. Mater., 2023, 6, e12575.
6. L. Dai, Y. Wang, L. Sun, Y. Ding, Y. Yao, L. Yao, N. E. Drewett, W. Zhang, J. Tang and W. Zheng, Adv. Sci., 2021, 8, 2004995.
7. J. Huang, Z. Wang, M. Hou, X. Dong, Y. Liu, Y. Wang and Y. Xia, Nat. Commun., 2018, 9, 2906.
8. S. Islam, M. H. Alfaruqi, J. Song, S. Kim, D. T. Pham, J. Jo, S. Kim, V. Mathew, J. P. Baboo, Z. Xiu and J. Kim, J. Energy Chem., 2017, 26, 815–819.
9. S. Paik, I. Choi, S. Lee and K. W. Nam, ACS Appl. Mater. Interfaces, 2024, 16, 50775–50784.
10. N. Zhang, F. Y. Cheng, J. Liu, L. Wang, X. Long, X. Liu, F. Li and J. Chen, Nat. Commun., 2017, 8, 1–9.
11. W. Jiang, X. Xu, Y. Liu, L. Tan, F. Zhou, Z. Xu and R. Hu, J. Alloys Compd., 2020, 827, 154273.
12. S. Ding, M. Zhang, R. Qin, J. Fang, H. Ren, H. Yi, L. Liu, W. Zhao, Y. Li, L. Yao, S. Li, Q. Zhao and F. Pan, Nano-Micro Lett., 2021, 13, 173.
13. J. Mao, F. F. Wu, W. H. Shi, W. X. Liu, X. L. Xu, G. F. Cai, Y. W. Li and X. H. Cao, Chin. J. Polym. Sci., 2020, 38, 514–521.
14. M. Zhou, X. Zhou, Y. Yang, H. Yin, Y. Lei, S. Liang and G. Fang, Chem. Eng. J., 2024, 497, 154916.
15. Q. Ni, B. Kim, C. Wu and K. Kang, Adv. Mater., 2022, 34, 2108206.
16. N. A. Salleh, S. Kheawhom, N. A. A. Hamid, W. Rahiman and A. A. Mohamad, J. Mater. Res. Technol., 2023, 23, 3470–3491.
17. M. H. Ryou, S. Hong, M. Winter, H. Lee and J. W. Choi, J. Mater. Chem. A, 2013, 1, 15224–15229.
18. M. Shin, W. J. Song, J. G. Han, C. Hwang, S. Lee, S. Yoo, S. Park, H. K. Song, S. Yoo, N. S. Choi and S. Park, Adv. Energy Mater., 2019, 9, 1900570.
19. C. Zou, Y. Huang, L. Zhao, W. H. Ren, Z. Zhao, J. Liu, X. Li, M. Wang, B. Guo and Y. Lin, ACS Appl. Mater. Interfaces, 2022, 14, 14226–14234.
20. H. Dong, R. Liu, X. Hu, F. Zhao, L. Kang, L. Liu, J. Li, Y. Tan, Y. Q. Zhou, D. J. L. Brett, G. He and I. P. Parkin, Advanced Science, 2023, 10, 2205084.
21. N. Jaikrajang, W. Kao-Ian, T. Muramatsu, R. Chanajaree, T. Yonezawa, Z. Y. Al Balushi, S. Kheawhom and R. Cheacharoen, ACS Appl. Energy Mater., 2021, 4, 7138–7147.
22. S. Lu, Y. Gu, G. Cheng, Y. Mei, D. Wu, Y. Gao, X. Yu, L. Wu, Y. Su and H. Wang, J. Mater. Chem. A, 2024, 12, 24647–24655.
23. D. Xie, J. Zhao, Q. Jiang, H. Wang, H. Huang, P. Rao and J. Mao, Chemphyschem, 2022, 23, e202200106.
24. X. Liang, Q. Dong, S. Guo, C. Zeng, Z. Chen, B. Zhang, C. Zhou, J. Zhang, Z. Xing, X. Tian and X. Shi, Adv. Energy Mater., 2025, 2500673.
25. X. Shi, C. Zhou, F. Yang, L. Shan, B. Tang, J. Zhang, Q. Nan, Y. Xie, J. Li, H. Li and X. Tian, ACS Energy Lett., 2024, 9, 1063–1072.
26. C. Zhou, Z. Xu, Q. Nan, J. Zhang, Y. Gao, F. Li, Z. Zhao, Z. Xing, J. Li, P. Rao, Z. Kang, X. Shi and X. Tian, Adv. Energy Mater., 2025, 15, 2405387.
27. J. Jang, J. Chun and C. Jo, Energy Storage Mater., 2023, 62, 102948.
28. S. Zhang, N. Yu, S. Zeng, S. Zhou, M. Chen, J. Di and Q. Li, J. Mater. Chem. A, 2018, 6, 12237–12243.
29. L. Li, T. K. A. Hoang, J. Zhi, M. Han, S. Li and P. Chen, ACS Appl. Mater. Interfaces, 2020, 12, 12834–12846.
30. X. Wang and Y. Li, J. Am. Chem. Soc., 2002, 124, 2880–2881.
31. J. X. Xue, S. X. Jia, T. Q. Xiang, J. J. Zhou and L. Li, ACS Appl. Mater. Interfaces, 2024, 16, 38458–38465.
32. S. Choi, J. K. Seo, H. Park, J. H. Park, S. W. Kim, S. W. Shin, J. A. Bang, K. N. Jung and B. G. Kim, Adv. Funct. Mater., 2025, 35, 2417179.
33. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Petersson, H. Nakatsuji, X. Li, M. Caricato, A. V. Marenich, J. Bloino, B. G. Janesko, R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell, J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, T. A. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo, R. Cammi, J. W. Ochterski, R. L. Martin, K. Morokuma, O. Farkas, J. B. Foresman and D. J. Fox, Gaussian 16, Revision A.03, Gaussian Inc., Wallingford CT, 2016.
34. D. K. Walanda, G. A. Lawrance and S. W. Donne, J. Power Sources, 2005, 139, 325–341.
35. W. Liu, X. Zhang, Y. Huang, B. Jiang, Z. Chang, C. Xu and F. Kang, J. Energy Chem., 2021, 56, 365–373.
36. Y. K. Jeong, T. W. Kwon, I. Lee, T. S. Kim, A. Coskun and J. W. Choi, Energy Environ. Sci., 2015, 8, 1224–1230.
37. J. He, H. Zhong, J. Wang and L. Zhang, J. Alloys Compd., 2017, 714, 409–418.
38. N. Reddy and Y. Q. Yang, Food Chem., 2010, 118, 702–711.
39. V. B. Bueno, R. Bentini, L. H. Catalani and D. F. S. Petri, Carbohydr. Polym., 2013, 92, 1091–1099.
40. S. Rajeevan, S. John, D. Ponnamma and S. C. George, J. Energy Storage, 2022, 56, 105919.
41. E. Vásquez, S. V. Piguillem Palacios, S. Rubio, J. R. A. Diaz, H. A. Baldoni, E. Vega and M. A. Masuelli, Acad. J. Chem., 2020, 5, 31–40.
42. S. G. Patnaik, R. Vedarajan and N. Matsumi, J. Mater. Chem. A, 2017, 5, 17909–17919.
43. J. Jian, Q. Chen, H. Sun, R. Li, Y. Hou, Y. Liu, J. Liu, H. Xie and J. Zhu, J. Energy Chem., 2024, 97, 228–236.
44. S. W. Wang, Q. Liu, C. Zhao, F. Lv, X. Qin, H. Du, F. Kang and B. Li, Energy Environ. Mater., 2018, 1, 28–40.
45. M. Liu, Q. Zhao, H. Liu, J. Yang, X. Chen, L. Yang, Y. Cui, W. Huang, W. Zhao, A. Song, Y. Wang, S. Ding, Y. Song, G. Qian, H. Chen and F. Pan, Nano Energy, 2019, 64, 103942.
46. W. Sun, F. Wang, S. Y. Hou, C. Y. Yang, X. L. Fan, Z. H. Ma, T. Gao, F. D. Han, R. Z. Hu, M. Zhu and C. S. Wang, J. Am. Chem. Soc., 2017, 139, 9775–9778.
47. W. Sun, F. Wang, B. Zhang, M. Zhang, V. Küpers, X. Ji, C. Theile, P. Bieker, K. Xu, C. Wang and M. Winter, Science, 2021, 371, 46–51.
48. N. Zhang, F. Y. Cheng, Y. Liu, Q. Zhao, K. Lei, C. Chen, X. Liu and J. Chen, J. Am. Chem. Soc., 2016, 138, 12894–12901.
49. T. Dong, P. Mu, S. Zhang, H. Zhang, W. Liu and G. Cui, Electrochem. Energy Rev., 2021, 4, 545–565.
50. J. Heo, S. Chong, S. Kim, R. Kim, K. Shin, J. Kim and H. T. Kim, Batteries Supercaps, 2021, 4, 1881–1888.
51. C. Li, S. Nie and H. Li, Chem.–Eur. J., 2024, 30, e202303733.
52. T. Qin, H. Y. Yang, Q. Li, X. Q. Yu and H. Li, Ind. Chem. Mater., 2024, 2, 191–225.
53. I. Aguilar, J. Brown, L. Godeffroy, F. Dorchies, V. Balland, F. Kanoufi and J. M. Tarascon, Joule, 2025, 9, 1–15.
54. A. Banerjee, Y. Shilina, B. Ziv, J. M. Ziegelbauer, S. Luski, D. Aurbach and I. C. Halalay, J. Am. Chem. Soc., 2017, 139, 1738–1741.
55. Y. Shilina, B. Ziv, A. Meir, A. Banerjee, S. Ruthstein, S. Luski, D. Aurbach and I. C. Halalay, Anal. Chem., 2016, 88, 4440–4447.
56. H. Moon, K. H. Ha, Y. Park, J. Lee, M. S. Kwon, J. Lim, M. H. Lee, D. H. Kim, J. H. Choi, J. H. Choi and K. T. Lee, Advanced Science, 2021, 8, 2003714.
57. D. Bergmann, G. Furth and C. Mayer, Int. J. Biol. Macromol., 2008, 43, 245–251.
58. D. Wang, Q. Zhang, J. Liu, E. Y. Zhao, Z. W. Li, Y. Yu, Z. Y. Guo, L. Wang and S. Q. Zhang, J. Mater. Chem. A, 2020, 8, 9693–9700.
59. W. Lu, J. L. Shen, P. Zhang, Y. J. Zhong, Y. Hu and X. W. Lou, Angew. Chem., Int. Ed., 2019, 58, 15441–15447.
60. G. Fang, S. Liang, Z. Chen, P. Cui, X. Zheng, A. Pan, B. Lu, X. H. Lu and J. Zhou, Adv. Funct. Mater., 2019, 29, 1905267.
61. Z. Q. Wang, X. D. Liu, H. M. Chen, X. Y. Zhu, L. Y. Song, Y. G. Yang, J. Bai, M. J. Kim, W. M. Lau, F. Rosei and D. Zhou, J. Mater. Chem. A, 2023, 11, 1380–1393.
62. A. Uctepe, E. Demir, B. Tekin, B. Dursun, O. Ozturk, O. Sel and R. Demir-Cakan, Solid State Ionics, 2020, 354, 115409.
63. N. Li, G. Li, C. Li, H. Yang, G. Qin, X. Sun, F. Li and H. M. Cheng, ACS Appl. Mater. Interfaces, 2020, 12, 13790–13796.
64. S. Sariyer, N. M. Keppetipola, O. Sel and R. Demir-Cakan, Chemsuschem, 2025, 18, 1–12.
65. L. Wang, K. W. Huang, J. Chen and J. Zheng, Sci. Adv., 2019, 5, eaax4279.

---