Synergistic crosslinking effect of illite and hydroxyethyl cellulose on improving the properties of PVA based hydrogels

Abstract

Flexible solid-state supercapacitors show great potential in the field of portable wearable electronic devices. However, the preparation of hydrogel electrolytes with high mechanical properties and ionic conductivity remains a challenge. In this paper, hydroxyethyl cellulose (HEC) and illite were used as organic and inorganic additives, respectively, for the polyvinyl alcohol (PVA) matrix to prepare composite hydrogels using the cyclic freeze–thaw method. A dense and uniform three-dimensional network structure of hydrogels was formed through the synergistic action of hydrogen bonding between PVA, HEC, and illite. When the contents of HEC and illite were 3 wt% and 6 wt%, respectively, the composite hydrogel (PH₃I₆) exhibited excellent mechanical properties, achieving a fracture stress of 1.41 MPa and a fracture strain of 500%. Its compressive strength was increased by 423% compared to that of the PVA hydrogel. The ionic conductivity of the PH₃I₆ hydrogel electrolyte reached 29.72 mS cm⁻¹. The supercapacitor assembled with the PH₃I₆ hydrogel electrolyte exhibited a specific capacitance of 292 F g⁻¹ and an energy density of 10.14 Wh kg⁻¹ at a current density of 1 A g⁻¹. Furthermore, the PH₃I₆ supercapacitor maintained a capacity retention rate of 96.52% and demonstrated a coulombic efficiency of 99.28% after 10 000 cycles at 1 A g⁻¹. Additionally, this PH₃I₆ supercapacitor possessed excellent flexibility and stability, functioning normally even when subjected to bending, hammering, and puncturing, thereby demonstrating broad application prospects.

---

1. Introduction

With the rapid development of emerging electronic technologies, flexible energy storage devices, especially supercapacitors, have broad application prospects in portable wearable electronic products, foldable screens, smart textiles, and biomedicine.^1–4 Supercapacitors possess high power density, long cycle life, and excellent safety and reliability, making them one of the strong competitors of lithium-ion batteries. As the core component of supercapacitors, electrolytes play a crucial role in ion transport, significantly influencing the electrochemical performance of supercapacitors. However, when traditional liquid electrolytes are used, despite their high ionic conductivity, safety issues such as high leakage risks and electrode corrosion arise, thus hindering their widespread application in flexible devices. Flexible quasi-solid-state electrolytes, on the other hand, serve dual functions as both electrolytes and separators, simplifying the structure of supercapacitors and offering folding and bending capabilities. Therefore, polymer electrolytes, especially hydrogel electrolytes with hydrogels as the polymer skeleton, have attracted widespread attention from researchers.

Hydrogels are three-dimensional networks composed of hydrophilic polymer chains connected through weak physical hydrogen bonds, possessing strong water absorption and retention capabilities, as well as high ionic conductivity, excellent mechanical properties, and flexibility. They can also mitigate the leakage issues associated with traditional liquid electrolytes in supercapacitors, enhancing the stability of supercapacitors. Common hydrogel substrates include polyacrylamide (PAM), polyvinyl alcohol (PVA), and polyacrylic acid (PAA). Among them, PVA is widely used in flexible supercapacitors due to its high ionic conductivity, non-toxicity, degradability, and biocompatibility. There are a large number of hydroxyl groups in the PVA molecule, which can form hydrogen bonds, as well as other non-covalent bonds such as crystal domains and hydrophobic interactions. In addition, PVA chains can also form chain entanglement with salt ions in electrolyte solutions through the Hofmeister effect, making PVA electrolytes relatively stable over a wide pH range. However, problems such as uneven cross-linking and inconsistent internal pore sizes often occur during the preparation of pure PVA hydrogels, leading to reduced mechanical and electrochemical properties and hindering their further development. The current main research directions focus on adding organic and inorganic additives to PVA hydrogels to achieve a balance between the mechanical and electrochemical properties of hydrogel electrolytes.⁵

Cellulose, as the most abundant natural polymer, possesses abundant functional groups and is widely used for enhancing the ionic conductivity and mechanical properties of PVA hydrogels. For example, Zhu et al.⁶ prepared a PVA–CMC hydrogel electrolyte for flexible zinc-ion hybrid capacitors using PVA and carboxymethyl cellulose (CMC), which can provide a maximum energy density of 87.9 Wh kg⁻¹. Subsequently, Zhou et al.⁷ introduced CMC and cellulose nanofibers (CNF) into the PVA matrix to prepare a PVA/CMC/CNF composite hydrogel with an excellent elongation at break of 1293% and an ionic conductivity of 1.17 mS cm⁻¹. Yin et al.⁸ introduced lignocellulose into PVA–PAA dual-network hydrogels, enhancing the adhesion and flexibility of the hydrogels. The hydrogels maintained their original flexibility and appearance even at −20 °C, while ensuring an ionic conductivity of 10 mS cm⁻¹. Adding different types of cellulose to the PVA matrix can endow the hydrogels with various unique properties. Hydroxyethyl cellulose (HEC), a derivative of cellulose ether, is odorless and non-toxic and is widely used in cosmetics, food industry, and coating technology.^9,10 The surface of HEC is rich in hydroxyl groups, which can form hydrogen bonds with PVA molecular chains when added to hydrogels, enhancing the density of its cross-linked network and thereby improving its mechanical and electrochemical properties.

Adding inorganic additives, such as graphene, carbon nanotubes, MXene, etc., to the hydrogel matrix can address the issue of poor electrochemical performance of hydrogels. However, they often exhibit poor compatibility and dispersibility in polymer matrices,¹¹ and due to the complex preparation process or high acquisition cost of these materials, it is difficult to meet the low-cost requirements for commercial production. Clay minerals have abundant reserves and are of low cost, and their mechanical properties, degradability, and conductivity can be significantly improved after being added to hydrogels.¹² Many scholars have incorporated clay into hydrogels and explored their mechanisms of action. For example, Wang et al.¹³ introduced bentonite (BT) nanosheets into cellulose to prepare a composite hydrogel, which achieved an excellent compressive strength of 3.2 MPa and a high ionic conductivity of 89.9 mS cm⁻¹ under the action of Al–O–C cross-linking and hydrogen bonding. Lu et al.¹⁴ used montmorillonite as an additive for PVA hydrogels, obtaining a PVA–MMT hydrogel with higher toughness and thermal stability, and the assembled flexible supercapacitor could provide a high specific capacitance of 161 F g⁻¹.

Illite is a 2 : 1 type mineral composed of an Al–O octahedral layer sandwiched between two Si–O tetrahedral layers, with the chemical formula K_1–1.5Al₄[Si_6.5–7Al_1–1.5O₂₀](OH)₄. Due to its excellent polymer compatibility, thermal stability, and ion adsorption capacity, it finds applications in various fields such as the rubber and plastics industry, refractory materials, and wastewater treatment.^15–17 In the field of hydrogel electrolytes, the clay additives currently used by researchers mainly include bentonite and halloysite,^18,19 with very little research on illite. The good dispersibility and suspension properties of illite in the aqueous phase are beneficial for its good dispersion in the PVA polymer matrix. A large number of Al–OH groups and Si–OH groups on the surfaces of illite's Al–O octahedral and Si–O tetrahedral layers can form hydrogen bonds with the hydroxyl groups on the PVA and HEC molecular chains, acting as inorganic cross-linking agents and contributing to the improvement of the mechanical and electrochemical properties of the hydrogel. In addition, the surface of illite carries a large number of negative charges, which can form an electric double layer structure in the aqueous phase, promoting efficient ion transport in an electric field, thereby enhancing the charge storage capacity of supercapacitors assembled with hydrogel electrolytes. However, research reports on the addition of illite to hydrogels to enhance the physical and electrochemical properties of the matrix are very rare.

In this study, a composite hydrogel electrolyte composed of PVA, HEC, and illite was prepared using the cyclic freeze–thaw method. During the synthesis process, the addition amounts of HEC and illite were optimized to obtain good mechanical properties. Furthermore, the effects of HEC and illite on the surface morphology, mechanical properties, thermal stability, and ionic conductivity of the hydrogel were investigated. In addition, the flexible supercapacitor assembled based on the PH₃I₆ hydrogel electrolyte exhibited excellent electrochemical performance, good flexibility, and acid–base resistance, indicating broad application prospects.

2. Experimental section

2.1. Materials

Illite was sourced from Antu County, Jilin Province, China, and was purified using a simple gravity separation method before use. Polyvinyl alcohol (PVA), hydroxyethyl cellulose (HEC), and sulfuric acid (H₂SO₄) were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd, while potassium hydroxide (KOH) was purchased from MACKLIN Reagent Co., Ltd. Acetylene black (AB) and polyvinylidene fluoride (PVDF) were purchased from Tianjin Avexin Co., Ltd. All chemical reagents were of analytical grade and could be used without further purification. The water used throughout the experiment was deionized water (DI) (18.0 MΩ cm⁻¹), which was prepared in the laboratory.

2.2. Preparation of hydrogels

The preparation of hydrogels was carried out using the cyclic freeze–thaw method. Firstly, 0.24 g of illite was dispersed in 20 mL of deionized water and stirred for 48 hours, followed by ultrasonic treatment for 3 hours to obtain a uniform dispersion. Meanwhile, 0.12 g of HEC was added to 20 mL of deionized water and stirred for 1 hour until it dissolved. The solution was then stirred in a water bath at 65 °C for 1 hour until it reached a transparent state, yielding the HEC solution. Subsequently, 4 g of PVA was added to the illite dispersion and stirred in a water bath at 90 °C for 1 hour. The resulting solution was then mixed and stirred with the HEC solution for 1 hour to achieve uniform dispersion. The solution was treated under vacuum for 30 minutes to remove any bubbles and then transferred to a glass mold. It was first frozen at −20 °C for 20 hours and then thawed at room temperature (20 °C) for 4 hours. After repeating this freeze–thaw cycle three times, the PH₃I₆ hydrogel was obtained, where the addition of HEC accounted for 3% of the PVA mass and the addition of illite accounted for 6% of the PVA mass. Without adding illite, PH_x hydrogels could be obtained by controlling the amount of HEC added, where x represents the percentage of added HEC mass to PVA mass. Based on the addition of 3% HEC, PH₃I_y hydrogels with different illite contents could be obtained by adjusting the amount of illite added, where y represents the percentage of added illite mass to PVA mass. Finally, the hydrogel was immersed in 1 mol L⁻¹ sulfuric acid to obtain a hydrogel electrolyte. To evaluate the stability of the hydrogel in an alkaline KOH electrolyte, the PH₃I₆ hydrogel was soaked in a 6 M KOH electrolyte to obtain a hydrogel electrolyte containing KOH electrolyte, and relevant electrochemical performance tests were conducted.

2.3. Preparation of electrodes and supercapacitors

Activated carbon, acetylene black, and PVDF were mixed in a mass ratio of 8 : 1 : 1, and an appropriate amount of NMP was added to prepare the electrode slurry. The slurry was coated on the surface of carbon cloth and then dried under vacuum at 80 °C for 12 hours to obtain electrodes. The mass of active material on each electrode was measured. Two carbon electrodes with the same mass of active material were placed on both sides of the hydrogel electrolyte to assemble a sandwich-structured supercapacitor.

2.4. Characterization

Using a DX-2700BH X-ray diffractometer, Cu Kα radiation (λ = 1.5406 Å) was employed to perform X-ray diffraction (XRD) analysis on the hydrogel that was freeze-dried using a vacuum freeze dryer (LGJ-10) and then flattened, with a step scanning rate of 0.05° s⁻¹.

The surface morphology of the hydrogel was observed using a JSM-6700F scanning electron microscope (SEM) and an EDS elemental composition analyzer was used to analyze its elemental distribution.

Different hydrogels were tested using a Nexus Fourier transform infrared spectrometer (FTIR) in ATR mode.

The hydrogel sample was tested using an HTC-3 thermogravimetric differential thermal analyzer in an air atmosphere for thermogravimetric (TG) analysis, with a heating rate of 10 °C min⁻¹. The temperature range was from room temperature to 700 °C.

The mechanical properties of hydrogels were tested using a universal testing machine (ZQ-990LB), with the hydrogel samples being stretched at a rate of 100 mm min⁻¹. Prior to conducting the hydrogel tensile test, the samples were shaped into a dumbbell shape using a fixed mold, with the stretchable area measuring 30 × 4 × 1.5 mm. To test the compressive strength, cylindrical hydrogel samples with a diameter of 10 mm and a height of 12 mm were compressed at a rate of 10 mm min⁻¹.

The electrolyte uptake rate of the hydrogel was calculated using formula (1):

where W₀ (g) is the mass of the hydrogel after thorough drying, and W₁ (g) is the mass of the dried hydrogel after fully absorbing the electrolyte solution.

2.5. Electrochemical testing

The supercapacitor was tested using an electrochemical workstation (CHI660E) for cyclic voltammetry (CV), galvanostatic charge–discharge (GCD), cycle performance, and ionic conductivity. Among them, ionic conductivity was measured by clamping a hydrogel electrolyte between two stainless steel sheets and conducting an electrochemical impedance spectroscopy (EIS) test within a frequency range of 10⁻² Hz to 10⁶ Hz. The calculation formulas for ionic conductivity (σ), specific capacitance (C), energy density (E), and power density (P) are as follows:where L (cm) is the thickness of the hydrogel electrolyte, R (Ω) is the intercept of the EIS curve with the real axis, and S (cm²) is the contact area between the stainless steel sheet and the hydrogel electrolyte.where I (A) is the discharge current in the GCD test, Δt (s) is the discharge time after subtracting the voltage drop, m (g) is the mass of the active material on the surface of a single electrode, and ΔV (V)is the discharge voltage excluding the voltage drop.where E (Wh kg⁻¹) represents the energy density, C (F g⁻¹) denotes the specific capacitance, and ΔV (V) signifies the discharge voltage excluding the voltage drop.where P (W kg⁻¹) represents the power density, and Δt (s) denotes the discharge time excluding the voltage drop.

3. Results and discussion

3.1. Characteristics of hydrogels

Fig. 1 shows the SEM images of PVA, PI₆, PH₃, and PH₃I₆ hydrogels. It can be observed that the PVA hydrogel has a smooth surface with fewer pores, and its pore size is approximately 15 µm. In contrast, the PI₆ hydrogel prepared by adding illite to the PVA matrix exhibits an increased number of pores, and the skeleton structure of the hydrogel can be observed, indicating that the addition of illite enhances the cross-linking density of PVA. The PH₃ hydrogel prepared by adding HEC to the PVA matrix presents a denser porous network structure, with a large number of smaller pores. When both illite and HEC are added to the PVA matrix, the obtained PH₃I₆ hydrogel exhibits numerous dense and uniformly distributed micropores, indicating that PVA achieves the highest cross-linking density at this time. A denser hydrogel cross-linking network is beneficial for stress dispersion and ion transport, endowing the hydrogel with better mechanical properties and ionic conductivity.^20,21 Fig. S1 shows an Al energy spectrum diagram of the PH₃I₆ hydrogel, revealing that the Al element is uniformly distributed in the PH₃I₆ hydrogel, confirming that illite is evenly distributed within it.

Fig. 1 SEM images of (a) PVA, (b) PI₆, (c) PH₃, and (d) PH₃I₆.

Fig. 2a shows the XRD patterns of PVA, PI₆, PH₃, and PH₃I₆ hydrogels. Among them, the characteristic peak of the pure PVA hydrogel is located at 19.5°, exhibiting a broad band with amorphous characteristics. However, after adding HEC to the PVA matrix, the intensity of this characteristic peak significantly increased, indicating an enhancement in the crystallinity of the hydrogel. This is due to the formation of hydrogen bonds between PVA molecular chains and HEC molecular chains, which promotes the formation of PVA crystal domains.^22,23 After adding illite to the PVA matrix, the XRD pattern of the prepared PI₆ hydrogel clearly shows the characteristic peaks of illite at 8.84° (d = 10.00 Å), 17.65° (d = 5.02 Å), 26.67° (d = 3.34 Å), and 45.19° (d = 2.01 Å), with no appearance of other new diffraction peaks, indicating that the addition of illite to the PVA matrix is a physical binding process. When both illite and HEC are added to the PVA matrix, in addition to the characteristic diffraction peaks of illite in the XRD pattern of the PI₆ hydrogel, the intensity of the PVA characteristic peak at 19.5° increases compared to that of the PVA hydrogel, also indicating an enhancement in the crystallinity of the PVA hydrogel. This suggests that the addition of illite does not disrupt the ability of PVA molecular chains and HEC molecular chains to form hydrogen bonds.

Fig. 2 (a) XRD patterns of illite, HEC, and PVA, PH₃, PI₆, and PH₃I₆ hydrogels and (b) FTIR spectra of PVA, PH₃, PI₆, and PH₃I₆ hydrogels.

Fig. 2a shows the FTIR spectra of the prepared hydrogels. All hydrogels exhibit strong and broad absorption peaks at 3700 cm⁻¹ to 3000 cm⁻¹, representing the stretching vibration of –OH in H₂O.²⁴ Compared to PVA hydrogels, the addition of 3% HEC or 6% illite to the PVA matrix results in a red shift of this characteristic peak, indicating the formation of new hydrogen bonds between HEC and illite and PVA, which reduces the electron density of –OH in H₂O.²³ The number of hydrogen bonds formed can be measured by the degree of red shift of this characteristic peak. It can be seen that the PH₃I₆ hydrogel forms the most hydrogen bonds, followed by PH₃ and PI₆. The absorption peaks at 2945 cm⁻¹ and 2916 cm⁻¹ in the FTIR spectrum of PVA are attributed to the asymmetric stretching vibration of C–H.²⁵ The sharp peak at 1666 cm⁻¹ represents the stretching vibration of C=O in carboxyl groups,²⁶ and the absorption peak at 1421 cm⁻¹ is caused by the symmetric bending vibration of CH₂.²⁷ In the FTIR spectrum of PVA, the absorption peak at 1001 cm⁻¹ is attributed to the stretching vibration of C–O.²⁸ This characteristic adsorption peak exhibits a blue shift after the addition of HEC or illite to the PVA matrix, due to the restricted stretching of C–O caused by the hydrogen bonds formed between PVA and HEC or illite, leading to a shift of the characteristic peak to higher wavenumbers. The change in the absorption peak at 1001 cm⁻¹ can also be used to determine the amount of newly formed hydrogen bonds.⁶ The blue shift amount of C–O stretching vibration decreases from high to low in the order of PH₃I₆, PH₃, and PI₆, indicating that the order of newly formed hydrogen bonds decreases from high to low in the same order after adding fillers to the PVA matrix, which is consistent with the previously mentioned results.

3.2. Physical properties of hydrogels

The mechanical properties of hydrogels are one of their core properties, directly determining their safety and stability during application. To investigate the impact of the addition of HEC and illite on the mechanical properties of hydrogels, tensile and compressive tests were first conducted on composite hydrogels with different HEC contents. The stress–strain curves are shown in Fig. S2a and b. With the increase in HEC content, the fracture strength, fracture strain, and compressive strength of the hydrogels first increase and then decrease, reaching a maximum at an HEC addition of 3%. The addition of an appropriate amount of HEC can enhance the cross-linking density of the hydrogel network, thereby promoting energy dissipation and improving mechanical properties. However, excessive HEC addition can lead to local entanglement of PVA chains, resulting in uneven cross-linking of the hydrogel network and a decrease in mechanical properties.²⁹ Based on the optimal HEC addition of 3%, the tensile and compressive properties of PH₃I₆ hydrogels with different illite additions were tested, and the results are shown in Fig. S2c and d. With the increase in illite content, the fracture strength, fracture strain, and compressive strength of the hydrogels first increase and then decrease, following the same trend as hydrogels with HEC addition. When the illite content reaches 6%, the mechanical properties of PH₃I₆ reach their optimum. When an appropriate amount of illite is added, it can form hydrogen bonds with HEC and PVA molecular chains, thereby increasing the compactness of the hydrogel structure and enhancing the tensile and compressive strengths. However, excessive addition of illite may lead to its agglomeration, reducing dispersibility and damaging the cross-linking network of the hydrogel, thereby decreasing its mechanical properties.³⁰

Fig. 3a shows the stress–strain curves of PVA, PI₆, PH₃, and PH₃I₆ hydrogels. The tensile fracture stress of the pure PVA hydrogel is only 0.23 MPa, with a fracture strain of 103%. The fracture stress and strain of composite hydrogels PH₃ and PI₆, obtained by adding 3% HEC and 6% illite to the PVA matrix, respectively, have been improved to some extent. After simultaneously adding 3% HEC and 6% illite to the PVA matrix, the fracture stress of the prepared composite hydrogel PH₃I₆ increases by 513%, reaching 1.41 MPa, and the fracture strain increases by 385%, reaching 500%, achieving a higher level compared to other PVA-based hydrogels (Table S1). Fig. 3b shows the stress–strain curves during the compression process of various hydrogels. It can be concluded that compared to the pure PVA hydrogel, adding 3% HEC and 6% illite to the PVA matrix separately improves the compressive strength of the hydrogel, but the improvement is not significant. The compressive strength of PH₃ increases by 110% compared to PVA, reaching 4.01 MPa, while the compressive strength of PI₆ increases by 32% compared to PVA, reaching 2.52 MPa. When 3% HEC and 6% illite are simultaneously added to the PVA matrix, the compressive strength of the composite hydrogel PH₃I₆ increases from 1.91 MPa of PVA to 9.99 MPa, achieving a significant improvement of 423%. The reason for this enhancement in mechanical properties is shown in Fig. 3c. In pure PVA, only hydrogen bonds are formed between PVA molecular chains through –OH. When HEC is added to the PVA matrix, in addition to the hydrogen bonds between PVA molecular chains, hydrogen bonds can also form between –OH in PVA molecular chains and –OH in HEC, cross-linking the HEC chains with the PVA chains to form a double network structure, increasing the cross-linking density, forming a dense hydrogel network, and promoting the improvement of the tensile and compressive strength of the hydrogel.³¹ For illite, its lamellar structure contains a large number of Al–O octahedral and Si–O tetrahedral layers, where Al–O and Si–O bonds can simultaneously form hydrogen bonds with –OH in HEC and PVA molecular chains. Therefore, illite layers serve as cross-linking sites in the hydrogel network, connecting HEC chains with PVA chains, further increasing the cross-linking density, and forming a much denser network structure. Therefore, the synergistic effect between HEC and illite when added to PVA makes the mechanical properties of PH₃I₆ significantly better than those of PH₃ and PI₆.

Fig. 3 (a) Tensile stress–strain curves of PVA, PI₆, PH₃, and PH₃I₆ hydrogels, (b) compressive stress–strain curves of PVA, PI₆, PH₃, and PH₃I₆ hydrogels, and (c) synthesis and structure diagram of the PH₃I₆ hydrogel.

To visually demonstrate the excellent mechanical properties of the PH₃I₆ hydrogel, destructive tests were conducted on PH₃I₆, as shown in Fig. 4. A 1.5 mm thick hydrogel sheet was punctured using an iron rod with a diameter of 8 mm. It can be seen that even under highly concentrated stress, the hydrogel remained unbroken despite significant deformation and could recover its original state after the stress was removed (Fig. 4a). When a cylindrical hydrogel with a diameter of 20 mm was cut using a knife, only a very faint trace was left on its surface, with no fracture or gap observed (Fig. 4b). When the same iron rod with a diameter of 8 mm was used to apply pressure to a cylindrical hydrogel with a diameter of 20 mm, the hydrogel quickly regained its original shape after the iron rod was removed, with no damage observed on the surface (Fig. 4c). In stark contrast, when the same destructive tests were conducted on the pure PVA hydrogel, it was evident that PVA performed poorly (Fig. S3). The above destructive experimental results demonstrate that the PH₃I₆ hydrogel exhibits excellent flexibility and mechanical properties.

Fig. 4 Destructive experiment of PH₃I₆ hydrogel under different conditions: (a) puncturing with an iron rod, (b) cutting with a knife, (c) compressing with an iron rod.

To analyze the thermal stability of hydrogels, thermogravimetric analysis was used to characterize different hydrogels, as shown in Fig. 5a. Within the tested temperature range (room temperature to 700 °C), the hydrogels exhibit two thermal weight loss stages. The first weight loss stage occurs at approximately 25–110 °C, during which water molecules mainly evaporate. The second weight loss stage occurs at approximately 240–500 °C, primarily due to the decomposition of PVA macromolecular chains, followed by further degradation and carbonization of the resulting small molecular chains.³² Additionally, the addition of HEC increases the cross-linking strength and density of the hydrogel due to the hydrogen bonding between PVA chains and HEC chains. The denser hydrogel network effectively inhibits the escape of water molecules, resulting in lower weight loss values for PH₃ compared to PVA in both stages of thermal weight loss. The thermal decomposition temperature of illite is much higher than 700 °C, and no mass loss occurs throughout the entire experimental temperature range. Meanwhile, illite can form hydrogen bonds with the hydroxyl groups in polymer chains through the abundant functional groups on its layered surface, delaying the degradation of polymer chains. Under the synergistic cross-linking effect of HEC and illite, more hydrogen bonds are formed in PH₃I₆, thus achieving optimal thermal stability.

Fig. 5 (a) TG curves of PVA, PI₆, PH₃, and PH₃I₆ hydrogels, and (b) electrolyte uptake rates of PVA, PI₆, PH₃, and PH₃I₆ hydrogel electrolytes.

Fig. 5b shows the electrolyte uptake rate of various hydrogels. It can be observed that the electrolyte uptake rate of the pure PVA hydrogel is the lowest, only 83.8%. However, after adding 6% illite to the PVA matrix, the electrolyte uptake rate of the hydrogel increases to 112.6%. This is due to the fact that after the addition of illite, the –OH groups in illite, as well as Al–O and Si–O bonds, can form hydrogen bonds with water molecules, promoting the absorption and storage of electrolyte.³³ After adding HEC to the PVA matrix, the –OH groups contained in the HEC chains can also form hydrogen bonds with water molecules, facilitating the absorption of electrolyte by the hydrogel, resulting in an electrolyte uptake rate of 114.6% for PH₃. When both HEC and illite are added to the PVA matrix, both HEC and illite can utilize their own hydrophilic groups to enhance the adsorption of aqueous electrolyte by forming hydrogen bonds with water molecules. At the same time, hydrogen bonds can be formed between PVA, HEC, and illite, resulting in a more developed cross-linking network within the PH₃I₆ hydrogel and the formation of numerous dense and uniformly distributed microporous structures (Fig. 1). This further enhances the adsorption capacity of the hydrogel for electrolyte, resulting in an electrolyte uptake rate of 177.3% for the PH₃I₆ hydrogel. The high electrolyte uptake rate of the hydrogel is beneficial for constructing more ion transport pathways, promoting the improvement of ionic conductivity of the hydrogel electrolyte.³⁴

3.3. Electrochemical properties of hydrogels

The ionic conductivity of hydrogel electrolytes is a key parameter for measuring their electrochemical performance. Firstly, the Nyquist plots of composite hydrogels with different HEC additions were measured using electrochemical impedance spectroscopy, as shown in Fig. S4a. The ionic conductivity was calculated using formula (2), as shown in Fig. S4b. From Fig. S4a, it can be seen that as the content of HEC in the PVA matrix gradually increases, the bulk impedance of the hydrogel electrolyte first decreases and then increases, while the ionic conductivity of the hydrogel electrolyte first increases and then decreases (Fig. S4b). The ionic conductivity of the PH₃ electrolyte is the highest, and further addition of HEC actually leads to a decrease in the ionic conductivity of the hydrogel electrolyte. This is because an appropriate amount of HEC is beneficial for forming a more uniform and denser hydrogel network, which facilitates ion transport. Excessive HEC leads to excessive cross-linking of the hydrogel network, forming an overly dense and non-porous structure locally, resulting in blocked ion transport pathways and reduced ionic conductivity.^34,35 Based on the addition of 3% HEC, a series of PH₃I_y composite hydrogels were prepared by adding different amounts of illite. Their ionic conductivity is shown in Fig. S4d. It can be seen that when 6% illite is added, the prepared PH₃I₆ achieves the highest ionic conductivity, reaching 29.72 mS cm⁻¹, thereby ranking among the forefront within the realm of PVA-based hydrogel electrolytes (Table S1). The improvement in the ionic conductivity of PH₃I_y due to an appropriate amount of illite is attributed to the abundant hydroxyl groups on the surface of illite, which form hydrogen bonds with PVA and HEC molecules, increasing the cross-linking density of the polymer network, enhancing the porosity of the hydrogel, providing more ion transport sites and channels, and thus improving the ionic conductivity of the hydrogel electrolyte. However, excessive illite can aggregate and hinder ion migration, reducing the ionic conductivity.²²

The three-dimensional network composed of hydrogel polymer chains and the large number of water molecules contained in the network pores endow hydrogels with excellent ion transport properties. Meanwhile, the different crosslinking states of polymer chains affect the ionic conductivity of hydrogel electrolytes. As shown in Fig. 6, the bulk impedance of the PVA hydrogel electrolyte is the highest, at 12.11 Ω, while the bulk impedance of PI₆ and PH₃ is 4.89 Ω and 2.65 Ω, respectively. PH₃I₆ has the lowest bulk impedance, reaching 1.76 Ω. The corresponding ionic conductivities of these hydrogel electrolytes are 4.36 mS cm⁻¹ (PVA), 10.71 mS cm⁻¹ (PI₆), 19.91 mS cm⁻¹ (PH₃), and 29.72 mS cm⁻¹ (PH₃I₆). After adding HEC and illite to the PVA matrix, the ionic conductivity of the hydrogel electrolyte has been significantly improved. This is due to the difference in cross-linking network density between different hydrogels. A higher polymer network cross-linking density can generate a denser pore-rich structure (Fig. 1), which can accommodate more ions and provide more boundary sites and ion transport channels for ion transfer, thereby enhancing the ionic conductivity. When HEC and illite are simultaneously added to the PVA matrix, it is equivalent to introducing an inorganic cross-linking agent, illite, into the PVA–HEC hydrogel network, further enhancing the cross-linking degree of the hydrogel molecular chains and forming more microporous structures, as shown in the SEM image of Fig. 1. This provides more space for electrolyte storage and more transport channels for ion conduction, thereby significantly improving the ionic conductivity of the PH₃I₆ hydrogel electrolyte. The ionic conductivity of the PH₃I₆ hydrogel electrolyte is increased by 582% compared to the PVA hydrogel electrolyte. In addition, the addition of illite has a unique effect on the improvement of the ionic conductivity of hydrogel electrolytes. The zeta potential of illite is −21.81 mV, indicating that the surface of illite carries a negative charge. It can attract cations in the electrolyte and form cation-rich areas on the surface of illite, while anion-rich areas are formed at locations far from illite. Under the driving of an electric field, cations migrate between illite sites in a “jumping migration” manner, forming a fast track for cation migration on the surface of illite. Due to the good dispersibility of illite in the PH₃I₆ hydrogel, as shown in Fig. S1, and its participation in the formation of the rich cross-linked network of PH₃I₆ together with HEC, the unique surface charge characteristics of illite effectively promote the transport of cations in the PH₃I₆ hydrogel electrolyte.

Fig. 6 (a) EIS curves of PVA, PI₆, PH₃, and PH₃I₆ hydrogel electrolytes and (b) ionic conductivities of different hydrogel electrolytes.

Fig. 7a depicts the Nyquist plots of supercapacitors assembled with different hydrogel electrolytes. In the high-frequency region, the intersection point of the curve with the real axis represents the equivalent series resistance (R_s), the diameter of the semicircle denotes the charge transfer resistance (R_ct), and the slope represents the Warburg impedance (Z_w). The supercapacitor assembled with the PVA hydrogel electrolyte exhibits an R_s of 5.34 Ω and an R_ct of 8.86 Ω (Table S2). When only illite is added to the PVA matrix, the R_s of the obtained PI₆ hydrogel electrolyte decreases to 3.68 Ω, and the R_ct decreases to 4.63 Ω. This is attributed to the hydrogen bonding between illite and PVA, which enhances the cross-linking of the hydrogel network and promotes the formation of a porous structure. This porous structure can adsorb more electrolyte, providing additional paths for ion transport and improving the compatibility between the hydrogel and the electrode. Similarly, the introduction of HEC into the PVA matrix alone can also reduce the R_s (0.92 Ω) and R_ct (1.56 Ω) of the prepared PH₃ hydrogel electrolyte. This is because HEC can also form hydrogen bonding with PVA, increasing the porosity of the hydrogel and the electrolyte uptake rate, thereby enhancing the ionic conductivity of the hydrogel electrolyte and the compatibility between the hydrogel electrolyte and the electrode. When illite and HEC play a synergistic cross-linking role in the PVA matrix, the PH₃I₆ hydrogel obtains a richer mesh structure, significantly improving the electrolyte uptake rate and ionic conductivity of the hydrogel electrolyte. Consequently, the R_s (0.61 Ω) and R_ct (0.51 Ω) of the PH₃I₆ supercapacitor decrease to 11.4% and 5.7% of those of the PVA supercapacitor, respectively, which is beneficial for enhancing the electrochemical reaction efficiency.

Fig. 7 (a) Nyquist plot of PVA, PI₆, PH₃, and PH₃I₆ hydrogel-based supercapacitors, (b) CV curves of different supercapacitors at 200 mV s⁻¹, (c) GCD curves of different supercapacitors at a current density of 1 A g⁻¹, (d) comparison of specific capacitance of different supercapacitors at different current densities, (e) Ragone plot of the PH₃I₆ hydrogel-based supercapacitor, and (f) the cycling stability and coulomb efficiency curve of the PH₃I₆ hydrogel-based supercapacitor at a current density of 1 A g⁻¹.

The PVA, PI₆, PH₃, and PH₃I₆ hydrogel electrolytes were assembled into supercapacitors, and their cyclic voltammetry (CV) curves obtained at different scan rates are shown in Fig. S5a–d. Compared to the supercapacitors assembled with PVA, PI₆, and PH₃ hydrogel electrolytes, the PH₃I₆ supercapacitor exhibits a quasi-rectangular shape at all scan rates, even at high scan rates, with no significant distortion. This indicates that the supercapacitor exhibits ideal electric double layer capacitance (EDLC) behavior, rapid reversible response, and excellent rate performance.³⁶ When comparing the CV curves of the four-hydrogel electrolyte-assembled supercapacitors at a scan rate of 200 mV s⁻¹ (Fig. 7b), it can be observed that the supercapacitor assembled with the PH₃I₆ hydrogel electrolyte has the largest CV curve area, while the PVA supercapacitor has the smallest CV curve area. Since the area of the CV curve corresponds to the charge storage capacity of the supercapacitor, it can be found that the PH₃I₆ supercapacitor has the best capacitive performance, which is attributed to the high electrolyte uptake rate and ionic conductivity of the PH₃I₆ hydrogel, as well as the good electrode compatibility of the PH₃I₆ hydrogel electrolyte.

Fig. S6 shows the GCD curves of supercapacitors assembled with PVA, PI₆, PH₃, and PH₃I₆ hydrogel electrolytes under different current densities. It can be observed that the GCD curves of supercapacitors assembled with different hydrogel electrolytes exhibit a symmetrical isosceles triangle shape under different current densities. When comparing the GCD curves of supercapacitors assembled with different hydrogel electrolytes under the same current density (Fig. 7c), it is found that the areas of the isosceles triangle curves are different. According to formula (3), the specific capacitances of the electrodes of supercapacitors assembled with PVA, PI₆, PH₃, and PH₃I₆ hydrogel electrolytes are 40 F g⁻¹, 86 F g⁻¹, 122 F g⁻¹, and 292 F g⁻¹, respectively. The specific capacitance of the supercapacitor assembled with the PH₃I₆ hydrogel electrolyte is 6.3 times higher than that of the PVA supercapacitor. The variation in specific capacitance of supercapacitors assembled with different hydrogel electrolytes under different current densities is shown in Fig. 7d. As the current density increases, the specific capacitance of the capacitors gradually decreases. However, the PH₃I₆ supercapacitor maintains a higher capacity retention rate of 90.6% even at a high current density of 4 A g⁻¹. As shown in Fig. 7e, the PH₃I₆ supercapacitor can provide energy densities of 10.42 Wh kg⁻¹ and 9.44 Wh kg⁻¹ at power densities of 125 W kg⁻¹ and 1000 W kg⁻¹, respectively, reaching a higher level compared to previously reported EDLC-type PVA-based supercapacitors (Table S3).^12,37–41 Table S4 presents the specific capacitance, power density, and energy density data at different current densities, to visually demonstrate the excellent capacitive performance of the PH₃I₆ capacitor.

Fig. 7f shows the capacitance retention and coulombic efficiency of the supercapacitor assembled with the PH₃I₆ hydrogel electrolyte after 10 000 charge–discharge cycles at a current density of 1 A g⁻¹. It was found that the supercapacitor still maintained a higher capacity retention rate of 96.52% after 10 000 cycles. The figure also displays the GCD curves of the supercapacitor for the first and last 5 cycles, both of which maintain an isosceles triangle shape. Furthermore, after 10 000 GCD cycles, the supercapacitor exhibits a higher coulombic efficiency of 99.28%, indicating excellent cycling stability.

To investigate the flexibility and mechanical stability of the assembled supercapacitors using the PH₃I₆ hydrogel electrolyte, the assembled flexible supercapacitors were bent at different angles, and their electrochemical performance changes were tested. As shown in Fig. 8a, at a scan rate of 200 mV s⁻¹, the CV curves of the supercapacitors tested at different bending angles still maintain a quasi-rectangular shape with good symmetry. At the same time, it can be observed that as the bending angle increases, the area of the quasi-rectangular CV curve slightly decreases, but this does not affect the electric double layer capacitance (EDLC) behavior and rapid reversible response of the supercapacitor.³⁶Fig. 8b shows the changes in EIS of the supercapacitors at different bending angles. When the bending angle is less than 180°, the R_s of the supercapacitors barely changes, but after the bending angle reaches 180°, R_s increases from 0.64 Ω (0°) to 0.89 Ω (180°). The R_ct of the supercapacitors shows an increasing trend with the increase in bending angle, especially when the bending angle reaches a large value of 180°. The increase in R_s and R_ct values of the supercapacitors during bending originates from the emergence of fine wrinkles on the surface of the carbon electrode,⁴² reducing the interfacial contact with the hydrogel electrolyte. However, the changes in R_s and R_ct are both within 1 Ω, indicating that the bending angle has a very limited impact on the internal ion transport and charge transfer of the flexible supercapacitor. Fig. 8c shows the changes in GCD curves of the flexible supercapacitors at different bending angles under a current density of 1 A g⁻¹. The bending angle has a minimal impact on the symmetry of the GCD curve, and the GCD curves tested at all bending angles still exhibit a standard equilateral triangle shape. The specific capacitance corresponding to different bending angles calculated using formula (3) is shown in Fig. 8d. Although the increase in bending angle leads to a decrease in the specific capacitance of the supercapacitor, the decrease is not significant. Even when the bending angle reaches a severe 180°, the flexible supercapacitor still maintains 94% of its specific capacity. The above results demonstrate that the PH₃I₆ hydrogel-based flexible supercapacitor exhibits excellent flexibility and mechanical stability, maintaining stable electrochemical performance under different bending conditions.

Fig. 8 (a) CV curves at 200 mV s⁻¹, (b) EIS curves, (c) GCD curves at 1 A g⁻¹, and (d) specific capacitance of the PH₃I₆ hydrogel-based supercapacitor at different bending angles.

Furthermore, Fig. S7 shows the LSV curve of the supercapacitor assembled with the PH₃I₆ hydrogel electrolyte at 5 mV s⁻¹. It is found that within the voltage range of −1.7 V to 1.7 V, the LSV curve exhibits a horizontal straight line, indicating that the capacitor exhibits stable EDLC behavior in this range, proving that the PH₃I₆ hydrogel electrolyte has a wide electrochemical stability window.⁴³

Both H₂SO₄ and KOH are commonly used liquid electrolytes for supercapacitors. The aforementioned experiments have verified that the PH₃I₆ hydrogel exhibits excellent electrochemical performance after adsorbing H₂SO₄. Fig. S8a and b show the CV and GCD curves of a supercapacitor assembled using the PH₃I₆ hydrogel electrolyte after adsorbing KOH. It can be observed that when the PH₃I₆ hydrogel absorbs KOH solution to form a hydrogel electrolyte, the CV curve of the assembled supercapacitor still maintains a quasi-rectangular shape, and even at higher scan rates, there is no significant shape distortion (Fig. S8a). The GCD curve shown in Fig. S8b exhibits a standard regular triangle shape with almost no voltage drop, demonstrating good electrochemical reversibility and excellent EDLC behavior. At a current density of 1 A g⁻¹, a higher electrode specific capacitance of 254 F g⁻¹ is still maintained. The above data indicate that PH₃I₆ can also achieve good performance in alkaline electrolyte environments.

To verify the practical application effect of the PH₃I₆ hydrogel electrolyte, a flexible supercapacitor was assembled using the PH₃I₆ hydrogel electrolyte. Fig. 9 showcases that the supercapacitor can light up an LED normally under bending, hammering, puncturing, and heavy pressure (SI video). Fig. S9 displays the CV and GCD curves under bending, hammering, poking, and pressing, proving the good practicality of the PH₃I₆ hydrogel electrolyte in flexible devices.

Fig. 9 Photos of the PH₃I₆ hydrogel-based supercapacitor illuminating an LED light under different conditions: (a) bending, (b) hammering with a hammer, (c) poking with a screwdriver, and (d) applying pressure with a heavy object.

4. Conclusions

In summary, by introducing organic additive HEC and inorganic additive illite into the PVA matrix and using the cyclic freeze–thaw method, a composite hydrogel was obtained. Due to the possibility of formation of hydrogen bonds between illite and HEC chains and the fact that illite could serve as a cross-linking site to connect PVA chains and HEC chains, the cross-linking density between PVA chains and HEC chains was increased, forming a denser and more uniform network structure. This was conducive to stress dispersion, thereby enhancing the mechanical properties of the hydrogel. After adding 3% HEC and 6% illite, compared to the pure PVA hydrogel, the fracture stress of the PH₃I₆ hydrogel increased by 513%, the fracture strain increased by 385%, and the compressive strength increased by 423%. The enhancement of the cross-linking density of the hydrogel by HEC and illite also contributed to the formation of a large number of uniformly distributed micropores on the surface and inside the hydrogel, which enabled the PH₃I₆ hydrogel to achieve a high electrolyte absorption rate (177.3%) and promoted the improvement of its ionic conductivity. Furthermore, the surface of illite possessed a fast track for cation migration, which further enhanced the ionic conductivity, resulting in the PH₃I₆ hydrogel electrolyte achieving a high ionic conductivity of 29.72 mS cm⁻¹. When the PH₃I₆ hydrogel electrolyte was incorporated into a supercapacitor, the device exhibited standard EDLC behavior and excellent electrochemical reversibility under higher scanning rates and current densities. The supercapacitor showed an excellent energy density of 10.14 Wh kg⁻¹. After 10 000 charge–discharge cycles at 1 A g⁻¹, the PH₃I₆ supercapacitor maintained a higher capacity retention rate of 96.52% and a coulombic efficiency of 99.28%, demonstrating excellent cycling stability. Its superior capacitive performance was attributed to the high electrolyte absorption rate and ionic conductivity of the PH₃I₆ electrolyte. Furthermore, the supercapacitor remained stable after bending, hammering, and puncturing. The supercapacitor could also achieve good performance in alkaline electrolyte environments. Therefore, this study provided a promising approach for high-performance and flexible energy storage devices for wearable electronics and other advanced applications.

Conflicts of interest

The authors declare no competing financial interests.

Data availability

All the available data are contained in the main manuscript and its supplementary information (SI). Supplementary information: EDS mapping of PH₃I₆; mechanical properties of PH_x and PH₃I_y; destructive experiment of PVA; ionic conductivity of PH_x and PH₃I_y; CV and GCD curves of PVA, PI₆, PH₃, and PH₃I₆ supercapacitors; voltage stability window of PH₃I₆ supercapacitor; CV and GCD curves of PH₃I₆ immersed in 6 M KOH; CV and GCD curves of PH₃I₆ hydrogel-based supercapacitor under different conditions; comparison tables of PH₃I₆ hydrogel and PH₃I₆ supercapacitor performance; specific capacitance, energy density, and power density of supercapacitor under different current densities. See DOI: https://doi.org/10.1039/d5nj03978a.

Acknowledgements

This work was supported by the National Natural Science Foundation of China [NSFC, grant number 41702036], the Science and Technology Research Project of Jilin Provincial Education Department [grant number JJKH20211080KJ], and the Project of Science and Technology Department (Jilin Province, grant no. 20220203110SF).

References

1. Y. Xiong, Y. Liu, J. H. Yang, Y. F. Wang, N. Xu, Z. L. Wang and Q. J. Sun, Nano Energy, 2024, 129, 109956.
2. C. Li, X. Yang, Y. Wang, J. Liu and X. Zhang, Adv. Funct. Mater., 2024, 34, 2410659.
3. J. Lv, D. Zhang, X. Li, Y. Miao, Y. Wang, Y. Wang and X. Zhang, Nat. Commun., 2025, 16, 7871.
4. X. Yao, S. Zhang, N. Wei, L. Qian, H. Ding, J. Liu, W. Song and S. Coseri, Susmat, 2024, 4, e249.
5. J. Chong, C. Sung, K. S. Nam, T. Kang, H. Kim, H. Lee, H. Park, S. Park and J. Kang, Nat. Commun., 2023, 14, 2206.
6. X. Zhu, C. Ji, Q. Meng, H. Mi, Q. Yang, Z. Li, N. Yang and J. Qiu, Small, 2022, 18, 2200055.
7. Q. Zhou, H. Abushammala, D. Gao, P. Xu, D. Niu, W. Yang and P. Ma, Carbohydr. Polym., 2025, 357, 123425.
8. Y. Yan, C. He, L. Zhang, H. Dong and X. Zhang, Int. J. Biol. Macromol., 2023, 224, 143–155.
9. L. Motelica, D. Ficai, G. Petrisor, O.-C. Oprea, R.-D. Trusca, A. Ficai, E. Andronescu, A. Hudita and A. M. Holban, Pharmaceutics, 2024, 16, 1225.
10. C. Yang, D. Zhang, D. Wang, H. Luan, X. Chen and W. Yan, ACS Appl. Mater. Interfaces, 2023, 15, 5811–5821.
11. K. Shu, Y.-X. Huang, J.-B. Yu, X. Yang, M.-D. Luo and X.-P. Chen, Carbohydr. Polym., 2024, 346, 122638.
12. M. Ding, H. Su, Y. Li, K. Yang, L. Dang, F. Li and B. Xue, Mater. Today Chem., 2022, 26, 101049.
13. S. Wang, L. Yu, S. Wang, L. Zhang, L. Chen, X. Xu, Z. Song, H. Liu and C. Chen, Nat. Commun., 2022, 13, 1–11.
14. C. Lu and X. Chen, Nano Lett., 2020, 20, 1907–1914.
15. Z. Wang, S. Wang, X. Yu, H. Zhang and S. Yan, ACS Omega, 2021, 6, 19017–19025.
16. M. V. Vasic, L. Radovanovic, L. Pezo and Z. Radojevic, J. Therm. Anal. Calorim., 2023, 148, 1783–1803.
17. T. Zhang, W. Wang, Y. Zhao, H. Bai, T. Wen, S. Kang, G. Song, S. Song and S. Komarneni, Chem. Eng. J., 2021, 420, 127574.
18. P. Xu, C. Wang, B. Zhao, Y. Zhou and H. Cheng, J. Power Sources, 2021, 506, 230196.
19. X. Yan, X. Lin, H. Liu, J. Lu, H. Wang, X. Huang, H. Liu and X. Xu, Carbohydr. Polym., 2024, 344, 122552.
20. Y. Zhang, Y. Sun, J. Nan, F. Yang, Z. Wang, Y. Li, C. Wang, F. Chu, Y. Liu and C. Wang, Small, 2024, 20, 2309900.
21. G. Ji, M. Sun, M. Li and J. Zheng, Adv. Funct. Mater., 2025, 2500110.
22. J. Hu, Y. Wu, Q. Yang, Q. Zhou, L. Hui, Z. Liu, F. Xu and D. Ding, Carbohydr. Polym., 2022, 275, 118697.
23. C. Tang, Y. Yao, M. Li, Y. Wang, Y. Zhang, J. Zhu, L. Wang and L. Li, Adv. Funct. Mater., 2025, 35, 2417207.
24. S. Wu, M. Hua, Y. Alsaid, Y. Du, Y. Ma, Y. Zhao, C.-Y. Lo, C. Wang, D. Wu, B. Yao, J. Strzalka, H. Zhou, X. Zhu and X. He, Adv. Mater., 2021, 33, 2007829.
25. D. Jeong, C. Kim, Y. Kim and S. Jung, Eur. Polym. J., 2020, 127, 109586.
26. Y. Cheng, X. Ren, G. Gao and L. Duan, Carbohydr. Polym., 2019, 223, 115051.
27. M. A. Saadiah, D. Zhang, Y. Nagao, S. K. Muzakir and A. S. Samsudin, J. Non-Cryst. Solids, 2019, 511, 201–211.
28. M. Rosi, F. Iskandar, M. Abdullah and K. Khairurrijal, Int. J. Electrochem. Sci., 2014, 9, 4251–4256.
29. J. Shojaeiarani, D. S. Bajwa and S. Chanda, Compos., Part C: Open Access, 2021, 5, 100164.
30. B. Li, J. Bailey, S. Yoon, P. Nockemann, T. Liu and B. Chen, J. Power Sources, 2025, 626, 235746.
31. Z. Ma, Z. Zhang, H. Wang, Z. Ma, T. Yi, Y. Li, K. Yang, F. Li and B. Xue, Mater. Today Chem., 2024, 37, 101996.
32. M. Hou, B. Xia, R. Qu, J. Dong, T. Li, S. Wang, Y. Wang and W. Dong, Int. J. Biol. Macromol., 2024, 273, 132939.
33. F. Qiu, Y. Huang, G. He, C. Luo, X. Li, M. Wang and Y. Wu, Electrochim. Acta, 2020, 363, 137241.
34. Q. Zhang, L. Zhao and F. Ran, Renewable Energy, 2022, 194, 80–88.
35. W. Li, S.-M. Li, M.-C. Kang, X. Xiong, P. Wang and L.-Q. Tao, Int. J. Biol. Macromol., 2024, 254, 127434.
36. Z. Zhao, H. Wang, H. Huang, L. Li and X. Yu, Colloids Surf., A, 2021, 626, 127125.
37. X. Bin, M. Sheng and W. Que, Front. Chem., 2024, 12, 1482072.
38. C. Peng, X. Huang, M. Zhao, S. Li, S. Liao and S. Yu, Carbon, 2024, 229, 119497.
39. H. Peng, Y. Lv, G. Wei, J. Zhou, X. Gao, K. Sun, G. Ma and Z. Lei, J. Power Sources, 2019, 431, 210–219.
40. K. Sun, S. Cui, X. Gao, X. Liu, T. Lu, H. Wei, H. Peng and G. Ma, J. Energy Storage, 2023, 61, 106658.
41. F. Wang and X. Maimaitiyiming, Int. J. Biol. Macromol., 2024, 282, 137293.
42. T. Zhao, Q. Xin, G. Yang, S. Liang and J. Lin, Int. J. Biol. Macromol., 2025, 314, 144273.
43. Z. Qi, R. Ren, J. Hu, Y. Chen, Y. Guo, Y. Huang, J. Wei, H. Zhang, Q. Pang, X. Zhang and H. Wang, Small, 2024, 20, 2400369.

---