Natural Polymer-Based Mechanically Strong Hydrogel with Fast Self-Healing for Heavy Metal Ions Removal and Supercapacitor Applications

TL;DR

This paper introduces a strong, self-healing hydrogel made from natural polymers that can remove heavy metals and be used in supercapacitors.

Contribution

A novel dual cross-linked hydrogel with fast self-healing and high mechanical strength for multifunctional applications is developed.

Findings

The hydrogel shows a toughness of 137 kJ/m3 and elongation at break up to 1117%.

It effectively removes heavy metals like Cr3+, Ni2+, and Cu2+ with high adsorption capacities.

The material functions as a solid-state electrolyte and separator in flexible supercapacitors with improved performance.

Abstract

Hydrogels have attracted significant interest in multifunctional applications. Among them, self-healing hydrogel stands out for its ability to autonomously repair damage through reversible interactions, yet achieving both rapid self-healing and superior mechanical strength remains challenging. In this study, we report the fabrication of a dual cross-linked hydrogel (PAA-Alg-B) prepared via free radical polymerization of acrylic acid and alginic acid, employing N,N′-methylenebisacrylamide, or vinyl-modified nanocellulose as primary cross-linker, with Fe3+ or borax serving as an additional dynamic cross-linker. The resulting borax based hydrogel (PAA-Alg-B) exhibits remarkable fast self-healing efficiency enabled by reversible borate ester bonds and hydrogen bonding. It demonstrates tunable mechanical strength with toughness of 137 kJ/m3 and elongation at break up to 1117%, alongside exceptional swelling capacity (448 g/g). The adsorption studies reveal high removal efficiencies for heavy metals, with maximum capacities of 87.57 mg/g (Cr3+), 114.02 mg/g (Ni2+), and 99.42 mg/g (Cu2+), governed by chemisorption kinetics. The PAA-Alg-B can also be used as a promising solid-state electrolyte and separator for flexible supercapacitors. Protonic modulation via H2SO4 soaking significantly enhances ionic conductivity, electrochemical performance, and cycling stability. These findings highlight the potential of natural polymer-based, mechanically robust, self-healing hydrogels for sustainable wastewater treatment and advanced energy storage applications.

Figures

Full text

1. Introduction

Hydrogels have gained significant attention in recent decades due to their unique properties, versatility, and flexibility in tailoring their properties and similarity to natural biological tissues [1,2]. Stimuli-responsive smart hydrogels exhibit significant potential for diverse applications, including controlled drug delivery, wound healing, contact lenses, tissue engineering, and energy-related devices. Nevertheless, conventional hydrogel systems are often constrained by their susceptibility to accidental mechanical damage during operation, which compromises structural integrity, reduces functional reliability, and ultimately shortens their service lifespan [3]. Consequently, the development of hydrogels with self-recuperative capabilities upon mechanical damage is imperative. Self-healing hydrogels represent a distinct class of materials capable of autonomously restoring their structural integrity and functional properties following injury, either without external intervention or through the application of specific stimuli. The underlying self-healing mechanisms are broadly categorized into two principal types based on network formation: (i) dynamic covalent bonding and (ii) physical non-covalent interactions [4]. Despite these advances, achieving high mechanical strength alongside efficient self-healing remains a major challenge. Strong covalent crosslinking often restricts chain mobility, reducing healing efficiency, while weak physical interactions fail to dissipate energy effectively under load. Therefore, designing hybrid networks that combine dynamic reversible covalent bonds with hierarchical physical interactions is essential for balancing toughness, elasticity, and damage recovery [5,6].

Borax is an effective crosslinking agent that introduces dynamic borate ester bonds through complexation with cis-diols in polysaccharides such as alginate (Alg) [7,8,9]. These reversible bonds impart malleability, ductility, and self-healing capability to the hydrogel network. Conventional chemical crosslinkers often lead to irregular crosslink distribution and insufficient energy dissipation, resulting in brittle fracture [10,11,12,13,14,15]. To overcome these limitations, bio-friendly vinyl-modified nanocrystalline cellulose (NCC) has also been explored as a multifunctional crosslinker and reinforcement. Surface modification of NCC with vinyl groups enables its covalent integration into the polymer network via free-radical polymerization, improving energy dissipation, load transfer, and enhancing toughness.

Rapid industrialization and urbanization have led to severe contamination of water resources by non-biodegradable, toxic heavy metals, which are prone to bioaccumulation [16,17]. Nickel (Ni^2+^) exposure is associated with carcinogenicity, particularly in the respiratory tract; copper (Cu^2+^) can cause gastrointestinal distress, hepatic and renal damage, and hematological disorders; and chromium (Cr^3+^/Cr^6+^) is linked to dermatitis, bronchial injury, and cancer risk [18,19,20,21]. Among various remediation methods—such as ion exchange, coagulation–flocculation, precipitation, membrane filtration, and adsorption—adsorption stands out for its cost-effectiveness, operational simplicity, and ease of regeneration. Hydrogels are particularly promising sorbents due to their hydrophilicity, swellability, porous structure, and biodegradability. Polysaccharide-containing hydrogels (e.g., Alg) feature abundant carboxylate and hydroxyl groups, which provide effective binding sites for heavy metal uptake. In this work, the PAA-Alg-B hydrogel enriched with electronegative carboxyl groups is prepared as an adsorbent for Cr^3+^, Ni^2+^, and Cu^2+^ removal, with systematic evaluation of adsorption kinetics, isotherms, and reusability [19,20].

Beyond environmental remediation, mechanically resilient and self-healing hydrogels with high ionic conductivity are well-suited for flexible supercapacitors—a class of energy storage devices that require stretchability, bendability, and operational stability under mechanical deformation [21,22,23,24,25]. The dynamic borate ester network in Alg promotes self-healing, while the PAA/Alg polyelectrolyte framework supports ion transport. Additionally, the incorporation of vinyl-functionalized NCC contributes to mechanical reinforcement and stable network architecture [26,27,28,29,30,31,32]. The hydrogel can be configured as a gel electrolyte or electrode binder, and, where appropriate, combined with conductive fillers (e.g., carbon-based nanomaterials or metal oxides) to enhance electronic pathways [33,34,35,36,37,38]. This multifunctional design enables a single hydrogel platform to adsorb heavy metals and serve as a component in flexible supercapacitors, bridging environmental and energy applications [39,40].

This work presents a comprehensive strategy for designing multifunctional hydrogels that integrate self-healing, mechanical robustness, heavy-metal adsorption, and energy storage performance, addressing critical challenges in sustainable environmental and energy technologies. By leveraging dynamic borate ester bonds [41,42,43], natural NCC-based nano crosslinking, and PAA/Alg polyelectrolyte networks, the proposed hydrogel system offers a scalable, bio-friendly solution for next-generation smart materials.

2. Experimental Section

2.1. Materials

The primary chemicals used in this research included acrylic acid (AA) purchased from LOBA Chemie (Mumbai, India) and alginic acid (Alg) from Tokyo Chemical Industry Co., Ltd. (TCI), Tokyo, Japan. Additional reagents comprised borax anhydrous, tetramethylethylenediamine (TEMED), nickel(II) nitrate hexahydrate, ammonium persulfate (APS), 3-methacryloxypropyltrimethoxysilane (MPTS), acetic acid, and copper(II) sulfate pentahydrate, all obtained from Sigma-Aldrich (Merck KGaA), St. Louis, MO, USA. Microcrystalline cellulose (MCC) was procured from Qualikems Fine Chem Pvt. Ltd., Vadodara, Gujarat, India, while sodium hydroxide and ethanol were supplied by Merck KGaA, Darmstadt, Germany. N,N-methylenebisacrylamide (BIS) was purchased from Acros Organics, Fair Lawn, NJ, USA, chromium(III) nitrate from Uni Chem Inc., South Kearny, NJ, USA, and sulfuric acid from RCI Labscan Ltd., Bangkok, Thailand. All chemicals and reagents were of analytical grade and used without further purification. Deionized water was used as the solvent for preparing most solutions in this study.

2.2. Preparation of NCC from MCC

In a 500 mL round-bottom flask, 10 g of microcrystalline cellulose was added with 200 mL 63.4% (w/w) H_2_SO_4_. The mixture, along with a magnetic stirrer, was placed in an oil bath at 60 °C for 1 h. Cellulose and H_2_SO_4_ mixture was hydrolyzed with continuous stirring. After 90 min, the flask was placed in an ice bath for cooling. The ultrasonication was applied for 15 min. The mixture was then allowed to settle until the top layer was clear, and was washed with distilled water several times until the mixture turned into a white suspension. The suspension was then transferred into centrifuge tubes, and centrifugation was held at 4000 rpm for 10 min to remove excess acid and water-soluble fragments. The cellulose suspension was then washed with deionized water. The process was repeated several times to reduce the acid content. The suspension was then freeze-dried, and an NCC powder sample was obtained [36].

2.3. Modification of NCC by MPTS

To silanize NCC, MPTS was used. A 10% (w/w) MPTS solution was made for hydrolysis in a 9:1 vol. percent ethanol-water mixture [36]. The addition of a 0.1 mol L^−1^ acetic acid solution brought the pH of the mixture to levels of 3–4. The hydrophilic NCC was then introduced to the mixture. The resultant mixture was stirred for 30 min at 500 rpm while heated in an oil bath at 60–70 °C. The mixture was then centrifuged to separate the particles. To remove any remaining reactants, the silanized NCC product was subsequently placed in deionized water for 24 h. To remove the effects of physical adsorption, the product was rinsed several times with deionized water. To remove excess moisture, the silanized NCC was allowed to dry for 2 days at room temperature. For an additional 4 h, the dry product was heated to 80 °C.

2.4. Synthesis of PAA-Alg-B Hydrogel

The PAA-Alg-B hydrogel was synthesized via a one-pot free radical polymerization process, using ammonium persulfate (APS) as the initiator, tetramethylethylenediamine (TEMED) as the accelerator, and borax as the crosslinker (Table 1). Initially, a fixed molar concentration of acrylic acid (AA), 1 wt% alginic acid (Alg) relative to the monomer, and 0.3 M NaOH were mixed in distilled water under continuous magnetic stirring to obtain a clear solution. Once a transparent solution was achieved, borax (0.05–0.2 M) was added and stirred for 30 min. Subsequently, 1 mM N,N-methylenebisacrylamide (BIS) and 0.22 M APS were added, followed by 5 μL TEMED. The solution was transferred to a test tube for nitrogen (N_2_) bubbling to remove dissolved oxygen, then poured into a mold formed by two glass plates separated by a 2 mm silicone spacer. The mold was placed in an oven at 55 °C for 4 h, during which the hydrogel formed.

2.5. Synthesis of PAA-Alg-MNCC-B Hydrogel

The PAA-Alg-MNCC-B hydrogel was synthesized using the same method, except that a predetermined amount of MNCC was first mixed with deionized water under continuous stirring before the remaining components were added to the reaction mixture. In this formulation, the conventional crosslinker BIS was completely omitted.

2.6. Synthesis of PAA-Alg-MNCC-Fe3+ Hydrogel

For the synthesis of PAA-Alg-MNCC-Fe^3+^ hydrogel, first, AA, Alg, MNCC, and TEMED were dispersed in 3 mL water by magnetic stirring and sonication for 30 min. Separately, KPS and FeCl_3_·6H_2_O (at varying ratios: 1%, 0.75%, 0.5%, 0.25% of monomer) were dissolved in 1 mL of water. Both solutions were nitrogen-purged for 30 min to remove oxygen. After purging, the initiator solution was slowly mixed with the monomer mixture under an ice bath to control heat, as gelation occurs rapidly. The mixture was then transferred to a glass mold and allowed to polymerize for ~48 h at 45–50 °C.

3. Characterizations

3.1. Chemical Structure Analysis

The chemical structure of the PAA-Alg-B hydrogel was characterized using Fourier Transform Infrared Spectroscopy (FTIR). FTIR measurements were performed with an (ATR PRO 4X, JASCO Corporation, Tokyo, Japan) accessory in the range of 4000–500 cm^−1^ at a resolution of 2 cm^−1^. Prior to analysis, the hydrogel samples were dried at 45 °C for 48 h and examined directly in ATR mode without KBr pellets. The spectra were recorded in transmittance mode as a function of wavenumber.

3.2. Mechanical Properties Measurement

Mechanical properties of the hydrogels were evaluated using a Universal Testing Machine (100P250-12 System, TestResources, Inc., Shakopee, MN, USA) at room temperature. For tensile testing, hydrogels with a thickness of 2 mm were cut into rectangular specimens (length: 12 mm, width: 9 mm). The extension rate was set at 100 mm·min^−1^. Stress–strain curves were recorded, and the stress (σ) and strain (ε) were calculated using the following equations:

$$[eqn]$$

where F is the applied load, w is the original width, and t is the original thickness of the specimen.

$$[eqn]$$

where Δl is the change in length and l_0_ is the initial gauge length.

The Young’s modulus was determined from the slope of the initial stress–strain curve, and fracture toughness was calculated by integrating the area under the curve.

3.3. Self-Healing Performance Analysis

Self-healing ability was assessed through macroscopic observation and tensile testing. Rectangular hydrogel samples (2 mm thickness) were cut into two pieces; one piece was immersed in a methylene blue (MB) solution for visual contrast. The two pieces were immediately brought into contact and incubated at ambient conditions for a predetermined time. Tensile tests were performed on healed samples at an extension rate of 100 mm·min^−1^. Healing efficiency (HE, %) was calculated using:

$$[eqn]$$

where $[eqn]$ and $[eqn]$ represent the tensile strength of healed and original hydrogels, respectively.

3.4. Water Absorbency Determination

Swelling behavior was determined gravimetrically. Freshly prepared hydrogel samples were weighed and immersed in distilled water at room temperature for 30 h until equilibrium was reached. Excess surface water was removed, and swollen samples were weighed. The swelling degree (Q) was calculated as:

$$[eqn]$$

where $[eqn]$ and $[eqn]$ are the hydrated and initial weights, respectively. Equilibrium water absorbency (EWA) was calculated by replacing $[eqn]$ with $[eqn]$ in the above equation.

3.5. Metal Ion Adsorption Study

Adsorption studies were conducted under shaking conditions using hydrogel samples in solutions containing chromium (Cr^3+^), nickel (Ni^2+^), and copper (Cu^2+^) ions at varying concentrations (200–1000 ppm) [34]. Ammonium hydroxide reacts with Cu^2+^ ions to form a colored complex. A standard solution of the Cu complex is prepared and a calibration curve is plotted. Residual metal ion concentrations were measured at regular intervals using a UV–Visible spectrophotometer. Cr^3+^ and Ni^2+^ solutions were analyzed directly. The adsorption capacity of heavy metals at different time intervals was calculated using the following equation

$$[eqn]$$

Additionally, the adsorption capacity at equilibrium (q_e_) was calculated using:

$$[eqn]$$

where $[eqn]$ and $[eqn]$ are the initial concentration at time t and the equilibrium concentrations (ppm), respectively, $[eqn]$ is the solution volume (L), and $[eqn]$ is the hydrogel mass (g). Adsorption performance was evaluated by varying the initial ion concentration and contact time. Statistical analysis was performed using Origin 8.5.

3.6. Electrochemical Measurements

Electrochemical performance was measured in a two-electrode symmetric configuration using an electrochemical analyzer (CHI660E, CH Instruments, Inc., Austin, TX, USA). Cyclic Voltammetry (CV) was conducted at scan rates of 10–150 mV·s^−1^ within a potential window of 0–1 V [35]. Galvanostatic Charge–Discharge (GCD) was performed at current densities of 0.5–10 mA·cm^−2^. Specific capacitance (C) was calculated using:

$$[eqn]$$

where $[eqn]$ is discharge current, $[eqn]$ is discharge time, $[eqn]$ is active material mass, and $[eqn]$ is a potential window. Electrochemical Impedance Spectroscopy (EIS) was recorded over 0.01 Hz–100 kHz with a 10 mV amplitude at open circuit potential. Nyquist plots were analyzed to determine the charge-transfer resistance (Rct) and ion-diffusion characteristics.

4. Results and Discussion

In this study, a simple and efficient strategy was developed to have self-healing, superabsorbent, adhesive, and conductive properties in a single hydrogel system. The hydrogel was synthesized via a one-pot free-radical polymerization approach.

The fabricated PAA-Alg-B hydrogel was primarily constructed through dual crosslinking mechanisms. First, AA underwent polymerization facilitated by the chemical crosslinker BIS. Concurrently, the cis-diol groups of Alg interacted with borax to form dynamic covalent borate ester bonds. Additionally, extensive intra- and intermolecular hydrogen bonding occurred among AA, Alg, and borate species, as illustrated in Figure 1.

Borax has vacant molecular orbitals of boron, which enable electron acceptance and subsequent reaction with the diol structures of Alg. This interaction results in the formation of borate ions that bridge two Alg diol units, making reversible crosslinks. The dynamic nature of these borate ester bonds endows the hydrogel with self-healing capacity. Incorporating Alg into the PAA-Alg-B system further enhances mechanical strength while improving self-adhesion and water absorbency.

The successful gelation of PAA-Alg-B_0.10_ hydrogel was confirmed by ATR-FTIR analysis (Figure 2). The broad band at 3445 cm^−1^ corresponds to O–H stretching vibrations. Characteristic peaks associated with borax and borate structures were observed: asymmetric B–O–C stretching at 1448 and 1394 cm^−1^, B–O stretching from residual borate ions at 814 cm^−1^, and bending vibrations of B–O–B linkages within the borate network at 654 cm^−1^. These spectral features provide strong evidence for the formation of borate ester crosslinks and the presence of a borate network within the PAA-Alg-B_0.10_ hydrogel matrix.

To investigate the influence of boronate ester crosslinking on the mechanical properties of the hydrogel, a series of PAA-Alg-B samples were prepared with varying borax concentrations. PAA-Alg and PAA-Alg-MNCC-Fe^3+^ hydrogels were also prepared for comparison. Mechanical performance, including tensile strength, elongation at break, toughness, and Young’s modulus, was evaluated through tensile stress–strain measurements [31]. The corresponding values were derived from the stress–strain curves shown in Figure 3, Figure 4 and Figure 5 and summarized in Table 2. The PAA-Alg hydrogel prepared using only MNCC as primary crosslinker shows high flexibility, with a toughness of 908 kJ/m^3^ and an elongation at break of 1414.6%, but it has very low Young’s modulus (11.55 kPa) and tensile strength (0.8 kPa) (Figure 3). This softness is due to weak entanglement between PAA and Alg chains, allowing them to slide easily.

Incorporating Fe^3+^ ions into the PAA-Alg hydrogel greatly improves its mechanical properties (Figure 3). The PAA-Alg-MNCC-Fe^3+^-0.3% sample reaches a tensile strength of 57 kPa, mainly due to strong coordination bonds between Fe^3+^ and carboxylate groups, which act as sacrificial bonds during deformation. However, increasing Fe^3+^ reduces elongation at break because additional cross-linking restricts chain mobility. At concentrations above 0.3%, uneven Fe^3+^ distribution leads to reduced tensile strength and premature fracture.

When monomer and Fe^3+^ concentrations are fixed, varying the MNCC content significantly affects mechanical behavior. The PAA-Alg-Fe^3+^-MNCC-0.75% hydrogel shows the best performance, with high tensile strength (90 kPa), toughness (589 kJ/m^3^), and elongation at break (1313%), due to uniform nanoscale cross-linking (Figure 4). Beyond this optimal MNCC level, excessive cross-linking increases rigidity and decreases toughness, stretchability, and tensile strength. The self-healing ability of the PAA-Alg-MNCC-Fe^3+^ hydrogel arises from dynamic and reversible non-covalent interactions, primarily metal–ligand coordination and hydrogen bonding. When the hydrogel is cut, these dynamic bonds break at the fracture surface. Upon bringing the cut pieces back into contact, the reversible bonds readily re-form, restoring the network structure. The coordination bonds form through interactions between Fe^3+^ ions and the carboxylate (COO^−^) groups of PAA and alginate. Additionally, both intermolecular and intramolecular hydrogen bonds among PAA, alginate, and MNCC strengthen network cohesion and promote efficient bond reformation at the damaged interface, thereby enhancing the hydrogel’s self-healing capability.

In contrast, for PAA-Alg-B hydrogels, the tensile strength showed a distinct trend with borax concentration (Figure 5). It increased gradually with borax content up to 0.10 M, reaching a maximum of 24.14 kPa. Beyond this concentration, tensile strength declined with further borax addition. This behavior can be attributed to excessive crosslinking at higher borax levels, leading to localized stress concentrations and premature fracture of the hydrogel network.

Similarly, elongation at break decreased consistently with increasing borax concentration. This reduction in flexibility is explained by the formation of a dense crosslinked network, where borate ions (B(OH)4^−^) interact with the hydroxyl groups of Alg chains, generating numerous crosslinking points. While these interactions enhance rigidity and strength up to an optimal level, excessive crosslink density restricts polymer chain mobility, reducing extensibility and toughness. Thus, the mechanical properties of PAA-Alg-B hydrogels are highly dependent on the balance between network integrity and chain flexibility, governed by borax concentration.

Compared with the PAA-Alg-Bx hydrogel, the tensile strength of all the PAA-Alg-MNCC-Fe^3+^ hydrogel increases. The conventional BIS crosslinker fixed the polymer chain into shorter chain so that it cannot dissipate energy under applied stress. On the other hand, MNCC-based hydrogel, due to the presence of nano-crosslinker polymer chains in the network, can relax under applied stress. In addition, PAA-Alg-MNCC-Fe^3+^ hydrogels have strong electrostatic and hydrogen-bonding interactions between PAA and the Alg polymer chain with MNCC, and a metal–ligand coordination bond between -COO-/ Fe^3+^ ions.

The ability of hydrogels to self-heal after mechanical damage or disintegration is a remarkable feature that significantly enhances their reliability and service life [35]. Rapid, efficient self-healing is essential for maintaining structural integrity and functional performance in practical applications. To achieve this, the PAA-Alg-B_0.10_ hydrogel was specifically designed and evaluated for its self-healing capability.

The self-healing property of this hydrogel arises from dynamic, reversible interactions within its network, primarily hydrogen bonds and borate ester linkages. These reversible bonds enable the hydrogel to autonomously repair damage without external intervention. To confirm this property, both macroscopic visual inspection and mechanical testing were performed, as illustrated in Figure 6 and Figure 7.

Initially, the hydrogel samples—one dyed with methylene blue (MB) and the other undyed—were cut into two separate pieces. The cut surfaces were then brought into contact and incubated under ambient conditions without external stimuli. As shown in Figure 6, the two segments rejoined immediately after intimate contact, forming a unified structure (within 5 min). The healed hydrogel exhibited excellent integrity, capable of being stretched and manipulated similarly to the original sample. Remarkably, the repaired hydrogel could support a 20 g weight while maintaining its shape, demonstrating strong interfacial adhesion and network recovery. The rapid healing occurred without external stimuli or disturbances, confirming the efficiency of dynamic bonding interactions in restoring the hydrogel’s structural integrity. To complement visual observations, tensile and macro-stress tests were conducted to quantify the recovery of mechanical properties after healing. The results, presented in Figure 7, demonstrate that the healed hydrogel retained a significant portion of its original tensile strength and elongation at break, validating the robustness of the self-healing mechanism.

The self-healing efficiency (HE) of the PAA-Alg-B_0.10_ hydrogel was quantified by measuring its tensile strength (Figure 7). HE was calculated by comparing the tensile strength of the healed hydrogel to that of the original, uncut sample. The HE of the hydrogel is 83.13%, indicating that it effectively restored mechanical integrity after damage (Table 3). This comparison shows that our hydrogel’s healing efficiency is comparable to or better than other reported systems, even though the healing time is similar (24 h). The underlying mechanism of this self-healing behavior is the presence of dynamic borate ester bonds within the hydrogel network. After several hours of contact between the cut surfaces, numerous hydroxyl groups and B(OH)4^−^ ions at the interface interact to form new borate ester linkages. This complexation primarily occurs between the hydroxyl groups on alginate chains and borate ions, enabling rapid and efficient reformation of the network. Consequently, the hydrogel exhibits autonomous self-repair without external stimuli, ensuring durability and reliability for practical applications.

Effective adhesion is essential for hydrogels used in wearable strain sensors, as it ensures stable contact with substrates and accurate transmission of electrical signals during sensing tests. The presence of Alg in the hydrogel matrix contributes significantly to its adhesive performance. The fabricated PAA-Alg-B_0.10_ hydrogel exhibited strong self-adhesive properties across a wide range of substrates.

As illustrated in Figure 8, the hydrogel adhered firmly to a range of materials, including glass, wood, rubber, polypropylene, Teflon, skin, graphite electrodes, and plastics. Notably, the hydrogel demonstrated excellent tissue-adhesive behavior, as evidenced by its ability to bond directly between human skin (finger) and a glass test tube without additional adhesives. This property is particularly advantageous for wearable and biomedical applications, where intimate contact with soft tissues is essential. The exceptional adhesion performance of the PAA-Alg-B_0.10_ hydrogel can be attributed to the incorporation of borax, which promotes non-covalent interactions, primarily hydrogen bonding, between the hydrogel network and various substrate surfaces. These dynamic interactions enable strong yet reversible adhesion, enhancing the hydrogel’s versatility for practical applications in flexible electronics and healthcare devices.

Swelling capacity and water content are critical parameters for hydrogels for wastewater treatment, as they directly influence ion diffusion and adsorption efficiency. Figure 9 shows that all synthesized hydrogel samples exhibited high water uptake, exceeding 448 g/g, indicating excellent water permeability and potential to facilitate the penetration of foreign metal ions into internal adsorption sites.

The equilibrium swelling degree of the hydrogels varied with borax concentration, measured as 142 g/g, 322 g/g, 448 g/g, 312.5 g/g, and 277 g/g for hydrogels containing 0 M, 0.05 M, 0.10 M, 0.15 M, and 0.20 M borax, respectively. The swelling ratio increased with borax content up to 0.10 M, reaching a maximum of 448 g/g, and then gradually decreased at higher borax concentrations. This trend can be explained by the progressive formation of crosslinks between borax and Alg as borax content increases. At higher concentrations, the increased number of borate–Alg interactions reduces the availability of free hydroxyl groups, slightly enhancing the hydrogel’s hydrophobicity and limiting hydrogen bonding with water molecules. Consequently, fewer unoccupied hydroxyl groups remain to absorb water.

Additionally, excessive borax loading introduces structural constraints within the polymer network, which restricts water penetration and swelling. Figure 9b presents the swelling kinetics curve, where the ordinate represents relative weight change, calculated as:

$$[eqn]$$

where W0 and W_t_ denote the initial dry weight and the weight after soaking for a given time, respectively. The hydrogels exhibited gradual swelling over time, reaching equilibrium after approximately 27 h. This behavior reflects the influence of the crosslinked polymer network, which governs water uptake and expansion, resulting in a progressive yet stable swelling rate at equilibrium.

The adsorption performance of the PAA-Alg-B_0.10_ hydrogel for heavy metals (Cr^3+^, Ni^2+^, and Cu^2+^) was evaluated through batch experiments conducted at 30 °C [34]. The effect of initial heavy metal concentration on the adsorption performance of the synthesized PAA-Alg-B_0.10_ hydrogel was investigated, as shown in Figure 10a–c. The equilibrium adsorption capacity for Cr^3+^, Ni^2+^, and Cu^2+^ increased significantly with rising initial concentrations. Specifically, the adsorption amount rose from 23.92 mg/g (Cr^3+^), 34.28 mg/g (Ni^2+^), and 25.06 mg/g (Cu^2+^) at 200 ppm to 87.57 mg/g (Cr^3+^), 114.02 mg/g (Ni^2+^), and 99.42 mg/g (Cu^2+^) at 800 ppm, 1000 ppm, and 1000 ppm, respectively. These results demonstrate the hydrogel’s strong affinity for heavy metal ions, suggesting its potential for wastewater treatment applications. However, practical implementation would require further evaluation under real wastewater conditions, accounting for factors such as competing ions, pH variability, and regeneration efficiency to ensure sustainable, cost-effective performance.

This trend can be attributed to the enhanced driving force for mass transfer at higher metal ion concentrations, which effectively overcomes the resistance to ion transport between the ions in solution and the active sites on the hydrogel surface. At lower concentrations, the limited ratio of metal ions to available adsorption sites results in reduced uptake. Conversely, at higher concentrations, the stronger concentration gradient accelerates diffusion and promotes efficient coordination bonding and surface complexation, thereby increasing adsorption capacity until saturation is reached. The effect of contact time on metal ion uptake was studied over a range of 5–160 min, as shown in Figure 10d. Initially, rapid adsorption occurred due to the abundance of vacant active sites and functional groups on the hydrogel surface, enabling efficient binding of metal ions. Over time, the adsorption rate gradually decreased, and equilibrium was reached.

As illustrated in Figure 10d, the hydrogel exhibited the fastest adsorption for Cu^2+^, achieving equilibrium within 30 min, indicating its strong affinity for this ion. In contrast, Ni^2+^ and Cr^3+^ required approximately 140 min to reach equilibrium. The adsorption capacity increased significantly during the initial phase, ranging from 34.92 to 81.32 mg/g for Cu^2+^, 78.01 to 114.02 mg/g for Ni^2+^, and 0.709 to 64.7 mg/g for Cr^3+^, after which it remained constant. This plateau suggests that the adsorption sites became saturated, primarily due to coordination bonding and surface complexation between metal ions and functional groups on the hydrogel. Once equilibrium is reached, any further increase in contact time does not change the adsorption capacity, which remains constant. This indicates that dynamic equilibrium has been established for all the studied heavy metal ions.

The rapid initial uptake can be attributed to the large number of accessible sites for metal-ion attachment. As adsorption progressed, the number of vacant sites decreased, slowing the rate until equilibrium was achieved. At this stage, a phase transition occurs where metal ions occupy nearly all active sites, and repulsive forces between adsorbed ions and those in the bulk solution further inhibit additional adsorption. This behavior reflects the typical adsorption mechanism governed by surface saturation and electrostatic interactions.

In this study, two widely used kinetic models—the pseudo-first-order (PFO) and pseudo-second-order (PSO) models—were applied to describe the adsorption behavior of heavy metal ions on the PAA-Alg-B_0.10_ hydrogel. The linear form of the PFO can be expressed as follows.

$$[eqn]$$

The linear form of the PSO kinetic model can be expressed as follows.

$$[eqn]$$

where q_e_ (mg g^−1^) is the equilibrium adsorption capacity, q_t_ (mg g^−1^) is the adsorption capacity at time t, k1 (min^−1^) is the pseudo-first-order rate constant, and k2 (g mg^−1^ min^−1^) is the PSO rate constant.

The fitting curves for the adsorption kinetics are presented in Figure 11, and the corresponding parameters are summarized in Table 4. The comparison of correlation coefficients (R^2^) revealed that the PSO model exhibited significantly higher R^2^ values than the PFO model for all three metal ions (Cr^3+^, Ni^2+^, and Cu^2+^), with values approaching unity. This indicates that the PSO model provides a more accurate representation of the adsorption process.

Furthermore, the equilibrium adsorption capacities calculated from the PFO model ( $[eqn]$ )—36.14 mg/g (Cr^3+^), 53.20 mg/g (Ni^2+^), and 175.91 mg/g (Cu^2+^)—differed substantially from the experimental values ( $[eqn]$ ). In contrast, the PSO model yielded calculated values of 89.77 mg/g (Cr^3+^), 118.76 mg/g (Ni^2+^), and 84.25 mg/g (Cu^2+^), which were in close agreement with experimental results. This strong correlation confirms that the adsorption process follows the PSO kinetic model. The dominance of the PSO model suggests that chemisorption is the primary mechanism governing metal ion uptake. Chemisorption involves electron sharing or exchange between the adsorbent and adsorbate, forming strong coordination bonds. In this context, the number of active sites on the hydrogel surface is directly proportional to its adsorption capacity, consistent with the principles of chemical adsorption.

To predict the adsorption process and understand the interaction between metal ions and the hydrogel adsorbent, two widely used isotherm models—Langmuir and Freundlich—were applied. These models are commonly employed to describe short-term, mono-component adsorption behavior of metal ions. The Langmuir model assumes monolayer adsorption on a homogeneous surface with finite, identical sites. Its linear form is expressed as:

$$[eqn]$$ $$[eqn]$$

where q_max_ (mg g^−1^) is the maximum monolayer adsorption capacity, C_e_ (mgL^−1^) is the equilibrium concentration of the metal ion solution, q_e_ (mg g^−1^) is the equilibrium adsorption of metal ions, and K_L_ (Lmg^−1^) is the Langmuir adsorption constant.

The Freundlich model is more suitable for describing chemisorption and adsorption on heterogeneous surfaces, particularly in aqueous systems and wastewater treatment. Its logarithmic form is given by:

$$[eqn]$$ $$[eqn]$$

K_F_ (mg g^−1^) is a constant that indicates the adsorption capacity of the adsorbent, 1*/n*, ranging between 0 and 1, indicates the adsorption intensity or surface heterogeneity, and C_e_ (mg L^−1^) is the equilibrium concentration of the metal ion solution.

The fitting plots for Langmuir and Freundlich models for Cr^3+^, Ni^2+^, and Cu^2+^ adsorption by PAA-Alg-B_0.10_ hydrogel are shown in Figure 12, and the corresponding parameters are summarized in Table 5. From the slope and intercept of the Langmuir plot ( $[eqn]$ vs. $[eqn]$ ), $[eqn]$ and $[eqn]$ were calculated, while $[eqn]$ and $[eqn]$ were obtained from the Freundlich plot ( $[eqn]$ vs. $[eqn]$ ).

The correlation coefficients ( $[eqn]$ ) indicate the suitability of each model. For Ni^2+^ and Cu^2+^, the Langmuir model exhibited higher $[eqn]$ values (0.9714 and 0.9969, respectively) compared to the Freundlich model, suggesting monolayer adsorption on a relatively homogeneous surface. Conversely, for Cr^3+^, the Freundlich model provided a better fit ( $[eqn]$ ), implying adsorption occurred on a heterogeneous surface with possible multilayer formation.

The favorability of adsorption was assessed using the Langmuir separation factor ( $[eqn]$ ) and the Freundlich constant ( $[eqn]$ ): $[eqn]$ values between 0 and 1 indicate favorable adsorption. In this study, all $[eqn]$ values fell within this range, confirming that Langmuir adsorption was favorable for all three metal ions. $[eqn]$ suggests good adsorption capacity across the entire concentration range. All calculated $[eqn]$ values exceeded 1, indicating that the Freundlich model predicts favorable adsorption for Cr^3+^, Ni^2+^, and Cu^2+^ over the studied concentration range. The results demonstrate that Ni^2+^ and Cu^2+^ adsorption is best described by the Langmuir model (monolayer adsorption), while Cr^3+^ adsorption aligns more closely with the Freundlich model (heterogeneous, multilayer adsorption).

The initial compressive strength of PAA-Alg-B_0.10_ hydrogel was relatively low compared to its strength after heavy metal adsorption. As shown in Figure 13b, both compressive strength and toughness increased significantly following metal ion uptake. This enhancement can be attributed to the formation of additional physical crosslinks within the hydrogel network. The carboxyl and hydroxyl groups on the polymer chains interact with heavy metal ions (Cr^3+^ and Cu^2+^) via complexation, effectively reinforcing the network and improving its mechanical performance.

Interestingly, the PAA-Alg-B_0.10_ hydrogel loaded with Cr^3+^ exhibited markedly higher compressive strength and toughness compared to the hydrogel loaded with Cu^2+^. This difference can be explained by the distinct adsorption mechanisms and interaction behaviors of the metal ions with the hydrogel surface. Adsorption isotherm analysis revealed that Cr^3+^ adsorption followed the Freundlich model, indicating multilayer adsorption on a heterogeneous surface. This multilayer binding likely led to more extensive crosslinking, resulting in a stiffer, mechanically stronger hydrogel. In contrast, Cu^2+^ adsorption conformed to the Langmuir model, which assumes monolayer adsorption on a homogeneous surface. While this interaction also contributed to improved strength through coordination with carboxyl groups, the extent of reinforcement was less pronounced than in the case of Cr^3+^. These findings demonstrate that heavy metal adsorption not only enhances the hydrogel’s functional properties for wastewater treatment but also significantly improves its mechanical robustness through ion-induced secondary crosslinking.

The PAA-Alg-B hydrogels were also tested as a solid-state electrolyte and separator for flexible supercapacitors (FSCs). Circular hydrogel disks (10 mm × 2 mm) were sandwiched between activated carbon nanosheet (ACNS) electrodes to fabricate FSC devices. Hydrogel electrolytes were prepared by soaking in varying H_2_SO_4_ concentrations (0.1–1.0 M), and the resulting FSC devices were categorized as follows: FSC-01: ACNS/PAA-Alg-B/ACNS, FSC-02: ACNS/PAA-Alg-B@0.1 M H_2_SO_4_/ACNS, FSC-03: ACNS/PAA-Alg-B@0.5 M H_2_SO_4_/ACNS, and FSC-04: ACNS/PAA-Alg@1.0 M H_2_SO_4_/ACNS

The capacitive behavior of the assembled FSCs was first evaluated by CV at a scan rate of 150 mV s^−1^ (Figure 14). FSC-01 exhibits the smallest enclosed CV area and a relatively distorted pseudo-rectangular profile, indicating limited charge propagation and sluggish ion transport within the pristine hydrogel electrolyte under high-rate operation. This behavior is consistent with the low concentration of mobile ionic carriers in the undoped hydrogel, leading to increased polarization and insufficient ion diffusion into the porous ACNS electrode. In contrast, the acid-doped devices (FSC-02 to FSC-04) display markedly enlarged CV areas with improved rectangularity, confirming that proton incorporation enhances ionic conductivity and facilitates rapid formation of the electric double layer throughout the porous carbon network. FSC-03 exhibits the most symmetric and near-ideal rectangular CV shape, suggesting highly reversible electric double-layer capacitive (EDLC) behavior and optimized charge transport. FSC-04 shows a smaller proportional increase in CV response than FSC-03, suggesting diminishing returns at high acid concentrations (Figure 14).

The trends were further corroborated using GCD analysis (Figure 15). All devices display quasi-triangular charge–discharge profiles, confirming predominantly capacitive energy storage. However, FSC-01 shows the shortest discharge time and the most pronounced initial IR drop, indicating high equivalent series resistance (ESR) and inefficient charge delivery through the electrolyte/electrode interface. The acid-doped FSCs exhibit significantly prolonged discharge durations and suppressed IR drops, demonstrating reduced ESR and improved conduction within the hydrogel network. FSC-03 delivers the longest discharge time with the smallest IR drop and highly linear charge–discharge slopes, indicating efficient ion transport, improved electrode wetting, and superior charge–discharge reversibility. Increasing the acid concentration to 1.0 M (FSC-04) does not yield a proportional improvement in discharge time, again supporting a non-linear relationship between proton concentration and effective charge storage (Figure 15).

The areal capacitance values calculated from the GCD profiles at 1.5 mA cm^−2^ are summarized in Figure 16. FSC-01 exhibits a low capacitance of 68.7 mF cm^−2^, reflecting limited use of the porous electrode surface due to poor ionic conductivity in the undoped hydrogel. Upon acid doping, the capacitance increases substantially to 140.5 mF cm^−2^ for FSC-02 and further to 183.8 mF cm^−2^ for FSC-03, confirming that moderate protonation significantly enhances ion accessibility and charge storage efficiency. FSC-04 shows only a marginal increase (188.8 mF cm^−2^) relative to FSC-03, suggesting that further increases in proton concentration do not translate into additional effective capacitance. This plateau behavior is consistent with saturation of accessible adsorption sites within the porous carbon framework and/or increased polarization effects at high ionic strength.

EIS analysis supported these findings: FSC-01 displays the largest real-axis intercept in the high-frequency region, confirming the highest ESR. It also shows the largest semicircle diameter, corresponding to elevated charge-transfer resistance (R_ct), and the most pronounced deviation from a vertical line in the low-frequency region, indicating non-ideal capacitive behavior and hindered ion diffusion. In contrast, FSC-02 to FSC-04 exhibit progressively reduced ESR and smaller semicircle diameters, demonstrating improved electrolyte conductivity and faster interfacial charge propagation. Furthermore, the low-frequency region becomes increasingly vertical for the acid-doped FSCs, indicating more ideal capacitive behavior due to improved ion transport. However, FSC-04 shows no significant impedance improvement over FSC-03, reinforcing that excessive acid concentration does not further optimize charge storage kinetics. (Figure 17).

PAA-Alg-B hydrogel demonstrates strong potential as a self-healing solid-state electrolyte for flexible supercapacitors. Acid doping enhances ionic conductivity and redox activity, with 0.5 M H_2_SO_4_ (FSC-03) providing the best combination of capacitance, low resistance, rate capability, and stability. These results highlight the importance of optimizing ionic content for durable, high-performance flexible energy storage systems.

5. Conclusions

In this study, we successfully synthesized a self-healing hydrogel with high mechanical strength by incorporating metal ions or borax as an additional dynamic cross-linker alongside a primary cross-linker, such as BIS or MNCC. The use of BIS as a primary cross-linker provides considerable mechanical strength with borax; however, incorporating MNCC results in an even superior mechanical performance. The formation of reversible borate ester bonds with the di-hydroxyl groups of alginate, along with inter- and intramolecular hydrogen bonding, imparted excellent self-healing capability and high mechanical toughness. These dynamic bonds acted as sacrificial linkages, enabling efficient energy dissipation and remarkable extensibility. Mechanical strength and toughness increased with borax concentration up to an optimum level, after which excessive crosslinking reduced flexibility. The hydrogel also demonstrated favorable swelling and self-adhesive properties, making it suitable for practical applications. Its abundant hydroxyl and carboxyl groups facilitated strong coordination with heavy metal ions, enabling efficient removal of Cr^3+^, Ni^2+^, and Cu^2+^ from aqueous solutions. Adsorption capacity increased with contact time and initial concentration, while kinetic analysis confirmed pseudo-second-order behavior, indicating chemisorption as the dominant mechanism. Isotherm modeling showed that Ni^2+^ and Cu^2+^ adsorption followed the Langmuir model, suggesting monolayer adsorption, whereas Cr^3+^ adsorption followed the Freundlich model, indicating heterogeneous multilayer adsorption. The hydrogel was also evaluated as a solid-state electrolyte and separator for flexible supercapacitors. Doping with H_2_SO_4_ markedly improves ionic conductivity and redox activity, resulting in near-ideal CV profiles, extended discharge durations with minimal IR drop, and moderate charge-transfer resistance, while exhibiting capacitive behavior in EIS analysis. The combination of self-healing capability, mechanical resilience, efficient heavy metal adsorption, and excellent electrochemical performance positions the PAA-Alg-B hydrogel as a multifunctional material for sustainable technologies, including wastewater treatment and flexible energy storage systems.