Interfacially engineered MXene hydrogels with dual-conductive networks for high-performance multifunctional sensing via a green and sustainable strategy

Abstract

The development of high-performance conductive hydrogels through green and sustainable strategies remains a pivotal challenge in flexible electronics. Conventional methods often rely on toxic crosslinkers, energy-intensive processes, or excessive metal ions, limiting their environmental compatibility and scalability. Furthermore, susceptibility to freezing at sub-zero temperatures poses a major obstacle for their use in extreme climates, typically addressed by organic antifreeze agents that compromise mechanical integrity. Herein, we report a green and efficient strategy for fabricating a multifunctional polyacrylamide/polyvinyl alcohol/CaCl₂/AgNP/proanthocyanidin/MXene (PPCAPM) composite hydrogel using natural proanthocyanidins (PAs) as a multifunctional green mediator. Molecular dynamics simulations reveal that PAs enhance interlayer electrostatic repulsion between MXene nanosheets, effectively suppressing van der Waals-driven restacking and increasing the diffusion coefficient by 48%. Notably, PAs and MXenes synergistically catalyze the in situ reduction of AgNPs at ultralow Ag⁺ concentrations, avoiding the use of conventional toxic reductants. Crucially, the incorporation of CaCl₂ serves as a green and potent antifreeze, enabling exceptional cryoresistance (−40 °C) through a colligative freezing-point depression mechanism, thereby eliminating the need for environmentally harmful organic antifreeze agents. The resulting hydrogel exhibits autonomous moisture retention, high conductivity (1.84 S m⁻¹), excellent impact factor (GF = 3.73), and intrinsic adhesion, enabling high-fidelity monitoring of both large-scale joint movements and subtle physiological signals such as EMG and ECG. This work demonstrates a green and sustainable paradigm for designing multifunctional hydrogel sensors via natural polyphenol-mediated interfacial engineering and ion-regulated antifreezing, offering a promising platform for next-generation wearable diagnostics and human–machine interfaces operable under harsh conditions.

---

Green foundation

1. This work establishes a green paradigm for multifunctional sensors by integrating natural proanthocyanidins (PAs) as a sustainable dispersant, reductant and crosslinker, replacing toxic reagents. It demonstrates a holistic strategy combining biomass-derived mediators and ion-regulated antifreezing to design eco-friendly electronics.

2. We achieved a high-performance MXene hydrogel sensor via PA-mediated interfacial engineering, enabling ultralow Ag⁺ usage (0.02 mmol) and CaCl₂-based green cryoresistance (−40 °C), eliminating organic antifreeze agents. The sensor exhibits excellent conductivity (1.84 S m⁻¹), stretchability (662%), and water retention (80% after 72 h).

3. Further improvements can focus on using fully bio-based solvents, scaling PA extraction from agro-waste, and implementing closed-loop metal recovery to minimize environmental footprint and enhance the sustainability of the synthesis process.

Introduction

The rapid advancement of wearable technologies, driven by the growing demand for personalized healthcare, has propelled significant progress in flexible sensor research. These innovative sensors have shown remarkable potential across diverse applications including human motion monitoring,¹ intelligent robotics,² flexible energy storage devices,³ human–machine interfaces,⁴ and implantable prosthetics.⁵ Among various functional materials, advanced flexible polymer hydrogels have emerged as particularly promising candidates due to their unique combination of tunable mechanical properties and electrical responsiveness.^6–10 However, conventional hydrogel sensors often suffer from limitations such as poor mechanical durability, insufficient sensitivity, and reliance on toxic crosslinkers or energy-intensive fabrication processes (e.g., UV irradiation and prolonged heating). These issues not only hinder their practical applications but also raise environmental and safety concerns.

MXene, a novel two-dimensional material, is distinguished by its exceptional electrical conductivity (exceeding 20 000 S cm⁻¹),¹¹ ultrahigh specific surface area,¹² and versatile surface chemistry with abundant functional groups (–OH, –O, –F, etc.).^13,14 These inherent characteristics endow MXenes with dual advantages for hydrogel-based sensor development: the strong hydrophilicity derived from surface functional groups facilitates homogeneous dispersion in aqueous hydrogel systems, while their rich surface chemistry enables extensive hydrogen bonding interactions with polymer hydroxyl groups.^15–18 Such synergistic interfacial bonding mechanisms significantly enhance the mechanical robustness, electrical performance, and sensing capabilities of MXene-reinforced hydrogels compared to conventional conductive composites. Nevertheless, a critical limitation emerges at elevated MXene concentrations where interlayer van der Waals forces dominate, inducing nanosheet aggregation and inhomogeneous dispersion within the hydrogel matrix.¹⁹ This structural imperfection limits the full potential of MXenes, leading to compromised mechanical integrity and suboptimal sensing performance in practical applications.

To address this nanoscale assembly challenge, innovative surface engineering strategies have been developed to mitigate MXene stacking while retaining functionality. Pioneering work by Pang et al.²⁰ demonstrated a dual-modification approach combining Ag/Cu bimetallic hybridization through hydrothermal synthesis and magnetron sputtering, creating ohmic contacts that enhance electron–hole separation efficiency while physically inhibiting MXene restacking. Jiang's group²¹ implemented a biomolecular functionalization strategy using 4-O-TEMPO-modified MXene edges coupled with polypyrrole integration, achieving concurrent optimization of conductivity maintenance and interlayer spacing control. In a separate advancement, Li et al.²² engineered MXene/NiAl layered double hydroxide (LDH) heterostructures through electrostatic self-assembly, effectively expanding interlayer galleries to reduce mass transfer resistance and promote electron transport kinetics. Although these strategies can effectively solve the problem of MXene stacking, they often require complex operation procedures or the use of hazardous chemicals, which is contrary to the principles of green chemistry.

Among the various green materials, natural polyphenols have emerged as a promising class of sustainable materials.²³ PAs, a subclass of these polyphenols abundantly found in plants, are particularly noteworthy due to their numerous phenolic hydroxyl groups.²⁴ These groups endow PAs with exceptional advantages, including strong metal-chelating capability, outstanding antioxidant activity, and universal adhesion properties, making them ideal for constructing multifunctional materials. Moreover, the biocompatibility and renewable nature of PAs align perfectly with the principles of green chemistry.^25,26 However, the direct use of pure PAs in materials engineering also presents certain challenges, such as potential self-polymerization under ambient conditions and a tendency to cause dark coloration, which may limit optical transparency in some applications. Despite these minor drawbacks, their multifaceted functionalities offer a compelling platform for developing advanced materials through eco-friendly pathways.

Additionally, hydrogel sensors face another critical limitation: inherent water evaporation during prolonged usage.^27–30 Dehydration fundamentally compromises device longevity and performance stability, making hydration maintenance a pivotal factor in practical sensor deployment. Recent investigations have explored innovative hydration preservation strategies through biomimetic designs and solvent engineering. Dai et al.³¹ developed coral-inspired bicontinuous hydrogels (PAD-iP) synthesized via in situ copolymerization of acrylic acid and dimethylphenylpropyl ester within a PEDOT:PSS framework. Integrating glycerol as a hygroscopic agent achieved dual functionality: enhancing water retention (82% initial mass after 7 days) and cryoresistance (−20 °C to 40 °C stability) through hydrogen bond-mediated moisture trapping. Parallel work by Li's group³² developed polyacrylamide/polydopamine hydrogels using deep eutectic solvents (DESs) as polymerization media, achieving remarkable tissue adhesion (20.20 kPa) and low-temperature flexibility via DES-induced hydration shell stabilization. However, these hydration strategies present inherent challenges: foreign solvents (glycerol/DESs) disrupt polymer chain entanglement, reducing crosslinking density. This leads to decreased tensile strength (35% lower than pure hydrogels) and impaired sensing resolution (18% signal drift after hydration cycles).^31,32 While effective, these additives often plasticize the polymer network, compromising its mechanical strength and long-term stability. There is a pressing need for green antifreeze strategies that can endow hydrogels with reliable cryoresistance without sacrificing their mechanical or sensing performance, thereby enabling operation in extreme climates.

In this work, we propose a green and sustainable strategy for constructing multifunctional MXene hydrogels by leveraging the unique properties of PAs. In the designed polyacrylamide/polyvinylalcohol/CaCl₂/AgNP/proanthocyanidin/MXene (PPCAPM) system, PA serves as a multifunctional green mediator that not only acts as an effective dispersant to inhibit MXene restacking through enhanced electrostatic repulsion but also functions as an eco-friendly reductant and stabilizer for the in situ synthesis of silver nanoparticles at low metal ion concentrations. Simultaneously, PA enhances the mechanical robustness and adhesion properties of the hydrogel network through dynamic crosslinking interactions. Furthermore, we introduced CaCl₂ as a green and inexpensive ionic additive to concurrently address the challenges of water retention and freezing tolerance. CaCl₂ enables sustained moisture regulation through the formation of hydration shells and depresses the freezing point via a colligative effect, providing an environmentally benign alternative to conventional organic antifreeze agents. Combined with the dual-conduction mechanism enabled by MXene-AgNPs and Ca²⁺/Cl⁻ ions, the resulting composite achieves integrated functions including autonomous hydration, cryoresistance, adhesion, and dual-mode conduction. This work exemplifies a green and efficient approach for designing high-performance hydrogel sensors via natural polyphenol-mediated interfacial engineering and ion-regulated antifreezing, aligning with the growing demand for sustainable electronics capable of functioning under harsh environmental conditions.

Experimental section

Materials

Polyvinyl alcohol (PVA, polymerization degree = 1700, degree of hydrolysis = 99%), acrylamide (AM, 99%), N,N′-methylenebisacrylamide (MBA, 99%), ammonium persulfate (APS, 99%), and proanthocyanidins (PA, 99%) were purchased from Shanghai McLean Biochemical Technology Co., Ltd. N,N,N′,N′-tetramethylethylenediamine (TEMED, 99%), lithium fluoride (LiF, 99%), hydrochloric acid (HCl, 37%) and calcium chloride anhydrous (CaCl₂, 99%) were obtained from Shanghai Aladdin Biochemical Technology Co., Ltd. Ti₃AlC₂ powder (200 mesh, 99.9%) was supplied by Foshan Xinwen Technology Co., Ltd. Silver nitrate (AgNO₃, 99.8%) was provided by Shantou Xilong Science Co., Ltd. Deionized water was used in all experiments.

Ti₃C₂T_x (MXene) fabrication

The synthesis of Ti₃C₂T_x (MXene) was carried out as illustrated in Fig. S1. Briefly, 10 g of LiF was dissolved in 200 mL of 9 M HCl under constant stirring for 30 min. Subsequently, 10 g of Ti₃C₂T_x powder was gradually added to the LiF/HCl acid solution and stirred at 35 °C for 24 h. After the reaction, the mixture was dispersed in deionized water (DI) and centrifuged at 2500 rpm for 5 min to collect the supernatant containing Ti₃C₂T_x nanosheets. The obtained supernatant was subsequently freeze-dried for further use.

Preparation of AgNPs@PA-MXene-x (AgNPs@PM-x)

MXene aqueous dispersions with precisely controlled concentrations (1, 5, and 10 mg mL⁻¹) were prepared gravimetrically by dispersing 10 mg, 50 mg, and 100 mg of MXene powder in 10 mL of deionized water, respectively. Subsequently, 8 mg of PAs and 0.2 mL of 0.1 M AgNO₃ solution were introduced to each dispersion under vigorous magnetic stirring (500 rpm, 30 min) to achieve homogeneous mixing. The reduction was initiated by ultraviolet irradiation (λ = 365 nm, 15 W) for 15 min, during which PAs and MXene acted synergistically to mediate the nucleation and growth of AgNPs on the MXene surfaces. The resulting nanocomposites, designated as AgNPs@PM-x (where x denotes the initial MXene concentration in mg mL⁻¹), were purified via centrifugation (8000 rpm, 10 min) and redispersed by ultrasonication (40 kHz, 15 min) to ensure colloidal stability.

Preparation of the PAM/PVA/CaCl₂/AgNPs@PM (PPCAPM) hydrogel

A 10 wt% PVA precursor solution was prepared by dissolving 10 g of PVA powder in 90 g of DI water under continuous magnetic stirring (500 rpm) at 90 °C for 2 h. The homogeneous solution was then cooled to 25 °C. Next, a certain amount of anhydrous CaCl₂ was gradually added to 0.8 g of PVA solution under vigorous stirring (800 rpm, 30 min) to ensure complete ion dispersion, forming a translucent homogeneous PVA–CaCl₂ mixture. Functionalized AgNPs@PM-5 was uniformly dispersed into the PVA–CaCl₂ mixture via sequential sonication (40 kHz, 15 min) and mechanical stirring (300 rpm, 20 min), yielding a homogeneous PVA–CaCl₂–AgNPs@PM suspension. Meanwhile, a monomer solution containing AM, MBA, and APS was prepared in DI water. This monomer solution was then thoroughly mixed with the PVA–CaCl₂–AgNPs@PM suspension. The final mixture was transferred into a mold and subjected to free-radical polymerization at 25 °C for 4 h, enabling complete gelation via thermally initiated crosslinking. The resulting PAM/PVA/CaCl₂/AgNPs@PM (PPCAPM) hydrogels with systematically varied compositions (as detailed in Table S1) were equilibrated in DI water for 24 h to remove unreacted monomers. In addition, to investigate the role of CaCl₂ as an antifreeze agent, the prepared PPAPM hydrogel (formulation shown in Table S1) was placed in glycerol for 2 h. After complete soaking, the surface of the hydrogel was washed to obtain PPGAPM (formulation shown in Table S1) as the control group.

Characterization of the PPCAPM hydrogel

The cross-section morphology of the hydrogel was observed using a Zeiss Gemini 300 scanning electron microscope (SEM). The morphology and distribution of AgNPs were observed using a JEOL JEM 2100F transmission electron microscope (TEM). Ultraviolet–visible (UV-Vis) spectra were acquired using a PE Lambda spectrophotometer over a wavelength range of 300–800 nm. Phase composition of the freeze-dried samples was determined by X-ray diffractometry (XRD) using an Ultima IV diffractometer. Fourier transform infrared (FT-IR) spectroscopy was performed using a Nicolet IS50 FT-IR spectrometer (Thermo Fisher Scientific) in the range of 400–4000 cm⁻¹ at a resolution of 4 cm⁻¹ with 32 scans in the attenuated total reflection (ATR) mode. The phase transition temperature of the hydrogel was determined via differential scanning calorimetry (DSC) using a TA Instruments differential scanning calorimeter under a nitrogen atmosphere, with cooling from 20 °C to −80 °C at a rate of 5 °C min⁻¹. Haake Mars 40 was used to test the rheology of the hydrogel. Raman spectra were recorded using a Thermo Scientific DXR Raman spectrometer.

Detailed information related to molecular dynamics simulation, mechanical testing, electrical performance and sensing test, and electrophysiological test is provided in the SI.

Results and discussion

Design of materials

To realize a high-performance yet environmentally benign sensing platform, we designed a composite hydrogel via a green and sustainable dual-phase strategy that synchronizes nanomaterial engineering with polymer network optimization. As illustrated in Fig. 1, the first phase leverages a cooperative redox system between natural PAs and MXenes. The abundant hydroxyl groups on both constituents create electron-rich domains, enabling the in situ reduction and uniform anchoring of monodisperse AgNPs on MXene surfaces at a minimal metal precursor concentration, thereby forming conductive AgNPs@PM Schottky junctions. In the second phase, radical polymerization constructs a multifunctional PAM/PVA matrix. Within this network, AgNPs@PM establishes continuous electron-transport pathways, while the incorporated CaCl₂ introduces complementary ionic conduction through the migration of hydrated Ca²⁺ and Cl⁻ ions. This dual-conduction architecture, coupled with the dynamic nature of the network, is engineered to yield outstanding strain sensitivity.

Fig. 1 Preparation of the PPCAPM composite hydrogel.

Synthesis and characterization of AgNPs@PM-x

The AgNPs@PM-x nanocomposites were fabricated through an integrated green strategy that combines MXene exfoliation with natural polyphenol-assisted dispersion and in situ metal nanoparticle reduction, establishing an environmentally benign pathway for functional nanomaterial preparation. Through selective etching of Ti₃AlC₂ (Fig. S2a) in LiF/HCl solution followed by ultrasonic exfoliation, monolayer Ti₃C₂T_x MXene nanosheets (Fig. S2b and c) were successfully prepared. Comprehensive characterization verified the successful delamination: XRD analysis (Fig. S3a) showed the characteristic (002) peak shifting from 9.5° to 6.6°, corresponding to interlayer spacing expansion from 9.3 Å to 13.4 Å, while the complete disappearance of the (104) peak at 38.8° confirmed thorough aluminum removal. FT-IR spectra (Fig. S3b) identified essential hydrophilic surface groups (–OH at 3353 cm⁻¹ and C–F at 930 cm⁻¹), and XPS analysis (Fig. S4) further corroborated the surface chemistry through characteristic Ti–C, Ti–O, C–O, and C–F bonds, collectively confirming high-quality MXenes suitable for subsequent functionalization.

To sustainably address MXene restacking, we introduced PAs as a natural dispersant. PA modification induced further interlayer expansion, with the (002) diffraction peak shifting from 6.12° to 5.16°, corresponding to a d-spacing increase from 14.4 Å to 17.1 Å, as shown in the XRD patterns in Fig. 2a. This structural evolution stems from electrostatic repulsion and charge-transfer interactions between PA molecules and MXene surface terminations, effectively inhibiting nanosheet aggregation through environmentally friendly mechanisms.^33,34

Fig. 2 In situ reduction of AgNPs by MXene/PAs. (a) XRD patterns of MXene and MXene/PA dispersions. (b) UV-Vis absorption spectra of AgNPs synthesized using different reducing agents. (c) XRD patterns of PAs, MXenes, and various AgNPs@PM. (d–f) TEM images of AgNPs@PM. (g) Elemental mapping images of AgNPs@PM-5.

Capitalizing on this stabilized dispersion, we utilized PAs as a dual-functional agent—serving simultaneously as a dispersant and green reductant for in situ AgNP synthesis. The polyphenolic hydroxyl groups of PAs, synergizing with MXene surface functionalities, enabled controlled reduction of Ag⁺ ions under mild UV irradiation, eliminating conventional toxic reducing agents. Systematic investigation of MXene dispersion concentrations (1, 5 and 10 mg mL⁻¹) revealed concentration-dependent AgNP formation dynamics: UV-Vis spectroscopy (Fig. 2b) displayed characteristic surface plasmon resonance peaks at 420–435 nm for AgNPs@PM-1, AgNPs@PM-5, and AgNPs@PM-10. Peak intensity initially increased (from 1 to 5 mg mL⁻¹) and then decreased (from 5 to 10 mg mL⁻¹), corresponding to optimal AgNP density at 5 mg mL⁻¹ (absorbance = 2.0) versus particle agglomeration at 10 mg mL⁻¹.^35,36 XRD analysis (Fig. 2c) confirmed face-centered cubic (FCC) Ag crystalline structures in AgNPs@PM-1 and AgNPs@PM-5, exhibiting distinct (111), (200), (220), and (311) planes at 38.1°, 44.3°, 64.4°, and 77.3°, respectively. In contrast, AgNPs@PM-10 showed suppressed crystallinity due to kinetic overgrowth. TEM characterization (Fig. 2d–f) quantitatively verified this trend: AgNPs@PM-1 displayed moderately dispersed particles (50 ± 8 nm), AgNPs@PM-5 achieved optimal monodisperse particles (18.7 ± 3 nm), and AgNPs@PM-10 exhibited severe aggregation. Elemental mapping (Fig. 2g) and HRTEM (Fig. S5) of AgNPs@PM-5 further confirmed uniform Ag distribution and lattice fringes (d-spacing = 0.316 nm, corresponding to the (111) FCC plane), validating successful synthesis. This concentration-dependent behavior arises from competing reduction kinetics: moderate MXene concentrations (5 mg mL⁻¹) balance nucleation sites and reaction rates, while an excessive MXene concentration (10 mg mL⁻¹) accelerates reduction, promoting particle coalescence.³⁷ The optimized AgNPs@PM-5 nanocomposite establishes efficient MXene–AgNP–MXene electron transport pathways, while maintaining excellent dispersion integrity, representing a sustainable nanomaterial preparation route that minimizes chemical waste, utilizes natural products as green alternatives to hazardous reagents, and employs energy-efficient processes for environmentally conscious material design.

Molecular dynamics (MD) simulations were employed to elucidate the green dispersion mechanism of PAs toward MXene nanosheets in an aqueous environment. The structures of MXene, PA and water molecules are shown in Fig. S6. Comparative analysis of pure MXenes and the MXene/PA composite over a 1 ns trajectory revealed fundamentally different behaviors. While the pure MXene system rapidly transitioned from a dispersed state to a clear aggregation due to dominant van der Waals forces (Fig. 3a and b), the MXene/PA system maintained a homogeneous distribution throughout the simulation. This enhanced stability was attributed to the strong electrostatic repulsion introduced by the negatively charged PA molecules, which effectively counteracts the interlayer attractive forces and suppresses restacking.

Fig. 3 Molecular dynamics simulation: (a and b) changes in the dispersion state of MXene and MXene/PAs in aqueous solution; (c and d) mean square displacement variations of MXene and MXene/PAs in aqueous solution (M₁, M₂, M₃ and M₄ represent four layers of MXenes, respectively); and (e and f) changes in the centroid distances between the four MXene layers and the MXene/PA layer in aqueous solution (M₁–M₂ represents the change in the centroid distance between M₁ and M₂, and the same applies to the other values).

Further quantitative analysis corroborates this mechanism. The mean square displacement (MSD) analysis showed that the MXene/PA system possesses a 48% higher diffusion coefficient than pure MXene (Fig. 3c and d), indicating significantly improved mobility and dispersion stability. Structurally, the interlayer centroid distances in the MXene/PA system were maintained at 70–90 Å, a substantial expansion compared to the 65–75 Å range observed in the aggregated pure MXene system (Fig. 3e and f).

These simulation results provide a theoretical foundation for our green strategy, showing at the molecular level how a natural polyphenol can serve as a highly effective and sustainable dispersant. By leveraging electrostatic repulsion, PAs successfully overcome the key challenge of MXene restacking without the need for complex synthetic polymers or harsh chemical treatments, offering an eco-friendly pathway to stable nanomaterial dispersions for high-performance composites.

Preparation and characterization of the PPCAPM composite hydrogel

The PPCAPM composite hydrogel was fabricated through an environmentally benign process by integrating the optimized AgNPs@PM-5 Schottky junction into a CaCl₂-modified polyacrylamide-polyvinyl alcohol (PAM-PVA) matrix, forming a dual-conductive network that synergistically combines MXene/AgNP electronic transport with Ca²⁺/Cl⁻ ionic conduction. Structural evolution analysis (Fig. 4a) revealed sequential network optimization through green modification strategies: the pristine PAM-PVA hydrogel exhibited a lamellar morphology, while MXene incorporation induced hydrogen-bond crosslinking between its surface hydroxyls and polymer chains, transforming the architecture into a uniform 3D network (pore size = 28 ± 4 μm). The addition of CaCl₂, serving as a green ionic crosslinker and antifreeze, further densified the matrix and reduced the pore size to 18 ± 3 μm. Subsequent PA-enhanced hydrogen bonding yielded ultradense nanopores (5.2 ± 0.8 μm) in the final PPCAPM hydrogel, a microstructure critical for achieving high strain-responsive conductivity.

Fig. 4 Characterization of the composite hydrogel. (a) Cross-sectional SEM images. (b) Elemental mapping of the PPCAPM hydrogel cross-section. (c) FT-IR spectra. (d) XRD patterns.

Elemental mapping (Fig. 4b) confirmed the homogeneous distribution of Ag (3.8 wt%), F (from MXene), and Ca (2.1 wt%) throughout the hydrogel cross-section, validating the successful and uniform integration of multifunctional components. FT-IR analysis (Fig. 4c) showed hydrogen-bond reinforcement, evidenced by the shifts of the O–H/N–H stretching band from 3420 to 3350 cm⁻¹ and the C=O amide I vibration from 1630 to 1615 cm⁻¹, indicating significantly strengthened interfacial interactions between PVA, MXene, and PAM chains. XRD patterns (Fig. 4d) illustrated the phase evolution, displaying the characteristic MXene (002) peak at 5.6° (d = 15.8 Å), Ag (111) peak at 38.1°, and CaCl₂ (200) peak at 40.3°, alongside a broad polymer amorphous halo at 24°.

This composite structured hydrogel overcomes traditional limitations through its dual-conduction mechanism and green composition design. The Ca²⁺ hydration shells enable exceptional cryogenic performance without organic antifreeze agents, while the MXene–AgNP percolation networks maintain high electromechanical sensitivity. The all-aqueous fabrication process and the use of biocompatible components establish PPCAPM as a sustainable platform for flexible electronics, achieving simultaneous environmental resilience and high sensing performance through green materials engineering.

Mechanical properties of the PPCAPM composite hydrogel

The mechanical robustness of hydrogels is paramount for their practical application in flexible sensors. The PPCAPM composite hydrogel exhibits exceptional mechanical versatility, sustaining large deformations including stretching, twisting, and compression while maintaining structural integrity under substantial loads such as a 20 g weight (Fig. 5a). These properties originate from a rationally designed hierarchical network achieved through green and orthogonal modifications.

Fig. 5 Mechanical properties of the composite hydrogel. (a) Digital photos showing the flexibility of PPCAPM: stretching, twisting, loading 20 grams of weights, and compression recovery. (b) Stress–strain curves. (c) Fracture stress and modulus. (d) Elongation at break and work of fracture of hydrogels. (e and f) Hysteretic stress–strain curves of PPCAPM under cyclic loading at different strains. (g) Cyclic stress–strain response of PPCAPM during 10 consecutive loading–unloading tests.

Systematic evaluation of hydrogels with progressively integrated components reveals distinct mechanical enhancement pathways (Fig. 5b–d). MXene incorporation (PPM) significantly reinforced the base matrix, increasing tensile strength 4-fold (to 124.8 kPa) and elastic modulus 3.5-fold (to 18.4 kPa) while preserving high extensibility (627.1%), with toughness quintupling to 368.5 kJ m⁻³. That is probably because MXene nanosheets form strong hydrogen bonds with polymer chains through their surface functional groups, promoting efficient stress transfer. During stretching, the sliding and orientation of MXene nanosheets, together with the reversible rupture and reorganization of interfacial hydrogen bonds, provide effective energy dissipation mechanisms. Subsequent CaCl₂ addition (PPCM) maintained these mechanical gains (127 kPa strength, 18.9 kPa modulus), confirming its compatibility as a green ionic additive. The introduction of natural PAs in PPAPM/PPCAPM further enhanced mechanical performance through hydrogen-bond reinforcement, elevating strength to 177–191 kPa and modulus to 25.5–25.7 kPa without compromising stretchability (642.5–662.1% elongation). In addition, we further verified the cross-linking effect of PAs by dynamic rheology (Fig. S7). The storage modulus (G′) of PPCAPM was greater than that of PPCM, indicating that PAs played a key role in the dynamic crosslinking process. This result is consistent with the mechanical property data. The chemical orthogonality between PA and CaCl₂ modifications is evidenced by nearly identical fracture work values. This staged optimization—utilizing MXenes for structural reinforcement, PAs for interfacial bonding, and CaCl₂ for ionic functionality—yielded PPCAPM's optimal mechanical profile (191 kPa strength, 662.1% strain and 25.7 kPa modulus).

Cyclic loading–unloading tests demonstrated PPCAPM's exceptional resilience and rapid self-recovery (Fig. 5e–g). The hydrogel exhibited significant strain-dependent energy dissipation through reversible hydrogen bond rupture-reformation and MXene nanosheet sliding mechanisms. Under 400% strain cycling, an initial modulus reduction (38.2% from cycle 1 to 2) indicated temporary network disengagement, while subsequent cycles showed outstanding stability with 96.4% modulus retention and minimal hysteresis loop area reduction (3.2%) over 10 cycles. The invariant loop superposition after cycle 2 (R² = 0.998) validated rapid self-recovery within 30 seconds. This robust mechanical performance, achieved through natural component integration and aqueous processing, positions PPCAPM as an ideal sustainable platform for demanding sensing applications requiring both environmental compatibility and mechanical reliability.

Water retention and antifreeze properties of the PPCAPM composite hydrogel

The environmental adaptability of hydrogel sensors was fundamentally enhanced through the sustainable integration of CaCl₂ as a multifunctional green additive, effectively overcoming the inherent limitations of conventional hydrogels in moisture retention and cryogenic resistance. Under controlled conditions (25 °C, 45% RH), PPCAPM hydrogels exhibited superior water-holding capacity, with only 26.56% mass loss over 72 h and minimal volumetric change, significantly outperforming the CaCl₂-free PPAPM hydrogel which suffered 74.02% water loss and 68% height reduction (Fig. 6a and b). This enhanced hydration stability originates from the Ca²⁺-mediated formation of tightly bound hydration shells that restructure hydrogen-bonding networks and effectively inhibit water evaporation. Thermal stability tests across a wide temperature range (−20 °C to 70 °C) further demonstrated PPCAPM's remarkable resilience, maintaining less than 30% dehydration after 72 h at 50 °C while effectively preventing ice crystallization below −20 °C through colligative freezing-point depression (Fig. 6c).^38–41

Fig. 6 Water retention and frost resistance of the composite hydrogel. (a) Visual evolution of water loss in PPAPM and PPCAPM hydrogels over 72 h. (b) Water loss rates of PPAPM and PPCAPM hydrogels over 72 h. (c) Water loss of the PPCAPM hydrogel at different temperatures over 72 h. (d) Raman spectra of H₂O and CaCl₂/H₂O solutions. (e) DSC thermograms of the PPAPM hydrogel and PPCAPM hydrogel. (f) Stress–strain curves of the PPCAPM hydrogel after 4 h at 25 °C and −40 °C. (g) Cyclic stress cycle curves of the PPCAPM hydrogel after 4 h at −40 °C. (h) Digital photos of the mechanical performance of PPAPM and PPCAPM hydrogels after being placed at −40 °C for 4 h.

The exceptional cryogenic performance of PPCAPM hydrogels was engineered through ion-regulated hydrogen-bond modulation, leveraging the Hofmeister effect to suppress ice formation via Ca²⁺-mediated water structure reorganization. Raman spectroscopy (Fig. 6d) revealed a characteristic blue shift in the O–H stretching region (3431.77 to 3465.92 cm⁻¹), indicating strengthened covalent hydrogen bonds and weakened non-covalent interactions that collectively depress the freezing point. DSC analysis provided insights into the freezing resistance mechanism. As shown in Fig. 6e, the phase transition temperature progressively decreased with increasing CaCl₂ content: 0 °C for PPAPM, −18 °C for PPC₁APM, −40 °C for PPCAPM, and −41 °C for PPC₅APM. Notably, the freezing point depression plateaued beyond a certain CaCl₂ concentration, with PPCAPM and PPC₅APM exhibiting similar values. Considering this trade-off, PPCAPM was identified as the optimal formulation. Remarkably, after 4 h at −40 °C, PPCAPM maintained 97% tensile strain retention (648% vs. initial 663%) and 95% fracture energy (Fig. 6f), while the strain of PPGAPM decreased significantly (Fig. S8). Cyclic tensile tests under cryogenic conditions demonstrated exceptional fatigue resistance with 94% stress retention over five cycles and less than 7% variation in the hysteresis loop area (Fig. 6g), attributable to Ca²⁺-stabilized bound water layers that preserve polymer chain mobility.

Practical evaluation confirmed these findings: after 4 h at −40 °C, PPAPM underwent complete mechanical failure, freezing into a brittle state, while PPCAPM retained full flexibility with reversible twisting and stretching capability (Fig. 6h). Electrical performance showed equally dramatic differences, with PPCAPM maintaining 93% of its room-temperature conductivity while PPGAPM only 79% (Fig. S9 and S10). Furthermore, the LED lamp connected by PPCAPM maintained normal brightness, while the one connected by PPAPM was almost extinguished (Fig. S11). This comprehensive cryoresilience stems from Ca²⁺-mediated suppression of ice nucleation and preservation of ion mobility through hydration shell stabilization. The green integration of CaCl₂ as a biocompatible and inexpensive antifreeze alternative positions PPCAPM as a breakthrough candidate for sustainable electronics capable of reliable operation in extreme environments, particularly for polar-region biomedical monitoring applications where both environmental compatibility and performance are paramount.

Self-adhesion properties of the PPCAPM composite hydrogel

The development of intrinsic self-adhesion in hydrogel sensors represents a significant advancement toward sustainable wearable electronics, eliminating the need for external adhesives that not only compromise signal fidelity but also generate additional consumable waste. Our PPCAPM hydrogel achieves multi-substrate adhesion through dynamically reversible hydrogen-bond networks formed between its abundant polar groups (PAM amides and PVA hydroxyls) and various material surfaces, utilizing a green bonding mechanism that operates without chemical crosslinkers or synthetic adhesives. Experimental validation confirmed robust adhesion to diverse substrates including plastics, glass, wood, metals, ceramics, PTFE, and PE (Fig. 7a), with quantified shear adhesion strengths (Fig. 7b) of 16.1 kPa (paper), 11.7 kPa (rubber), 4.7 kPa (glass), and 1.9 kPa (plastic) (Fig. 7c). Remarkably, the hydrogel maintained over 50% of its initial adhesion strength after five attachment–detachment cycles across all tested substrates (Fig. 7d), demonstrating exceptional reusability that extends device lifespan and reduces material waste. We further evaluated the adhesion stability of the PPCAPM hydrogel under low-temperature conditions. After storage at −40 °C for 4 h, the hydrogel underwent five consecutive attachment–detachment cycles and retained excellent adhesion, demonstrating performance comparable to that at room temperature (Fig. S12). This recyclable bonding mechanism (Fig. 7e), relying on reversible interfacial hydrogen bonding rather than permanent covalent linkages, enables conformal contact maintenance during dynamic strain sensing while ensuring gentle detachment without substrate damage or residue left behind. The synergistic combination of tissue-compliant adhesion, bonding stability, and environmental benignity makes PPCAPM an ideal platform for next-generation sustainable wearable systems requiring high-fidelity physiological monitoring and minimal environmental impact.

Fig. 7 Self-adhesive properties of the PPCAPM hydrogel. (a) Photos of the PPCAPM composite hydrogel adhered to diverse substrates. (b) Shearing adhesion test diagram. (c) Stress–displacement curves from lap-shear tests on paper, plastic, rubber, and glass. (d) Adhesion strength and reusability of the PPCAPM hydrogel on different substrates. (e) The adhesion mechanism of the PPCAPM hydrogel to various substrates.

Sensing performance of the PPCAPM composite hydrogel

The integration of exceptional mechanical robustness, cryogenic stability, and intrinsic adhesion establishes the PPCAPM hydrogel as an ideal platform for high-performance flexible sensing, with systematic characterization confirming its superior electromechanical properties. Strain sensitivity analysis (Fig. 8a) revealed a progressively increasing gauge factor across three deformation regimes: GF = 1.15 (0–50% strain), 2.26 (100–200% strain), and 3.73 (200–600% strain), showing enhanced sensitivity under large deformations. This nonlinear response originates from the strain-induced separation of MXene nanosheets, which progressively disrupts the percolation network and reduces conductive pathways. Control experiments (Fig. S13) verified MXene's crucial role in sensitivity enhancement, while consistent relative resistance changes (ΔR/R₀) across incremental strains (Fig. 8b and c) and stable cyclic loading responses (Fig. 8d and e) confirmed excellent signal reproducibility. Furthermore, PPCAPM exhibited excellent strain-sensing performance and stability even after being placed at −40 °C for 4 h (Fig. S14 and S15).

Fig. 8 Sensing performance of the PPCAPM composite hydrogel. (a) Strain-dependent gauge factor evolution (0–600% strain range). (b and c) Relative resistance change (ΔR/R₀) versus strain/time curves (0–100% and 100–600% strain, respectively). (d and e) Cyclic (ΔR/R₀) response under varying strain amplitudes (0–100% and 100–600% strain, respectively). (f) ΔR/R₀ characteristics during progressive extension–release cycles. (g) Dynamic response time characterization (activation/recovery). (h) Strain-rate independence evaluation across different stretching speeds. (i) Long-term cycling stability assessment.

With a conductivity of 1.84 S m⁻¹, PPCAPM surpasses existing hydrogel sensors in critical metrics including operational strain range (600%), sensitivity gradient, and electrical stability, as quantitatively benchmarked against previous systems (Fig. S16).^42–45 The hydrogel also exhibited outstanding electromechanical stability, maintaining consistent electrical responses during 0–200% flexion–extension cycles (Fig. 8f) and demonstrating rapid response dynamics with 240 ms activation and 280 ms recovery times (Fig. 8g). Notably, the sensor maintained strain-rate independence (2–8 mm s⁻¹) in relative resistance profiles (Fig. 8h) and showed remarkable durability with less than 8% signal drift and 95% waveform consistency after 400 continuous loading cycles at 50% strain (Fig. 8i).

These exceptional performance metrics stem from two synergistic mechanisms enabled by our green design: (1) dynamic reconfiguration of reversible hydrogen bonds that maintain structural integrity and (2) fully recoverable conduction pathways in the MXene–AgNP network. The combination of sub-second responsiveness, mechanical endurance, and self-adhesive operation—all achieved through sustainable material choices and aqueous processing—establishes PPCAPM as a viable platform for continuous biomechanical monitoring in athletic training, rehabilitation therapy, and human–machine interfaces, demonstrating that environmental compatibility and high performance can be simultaneously realized in advanced sensor design.

The PPCAPM hydrogel was fabricated into a self-adhesive wearable strain sensor that achieves conformal epidermal contact without auxiliary fixation, leveraging its integrated multifunctionality for real-time motion tracking. When attached to finger joints, the sensor accurately discriminates incremental flexion angles (30°–90°) through characteristic relative resistance changes (Fig. 9a). Similarly, it reveals distinct kinematic signatures during wrist and elbow flexion–extension cycles (Fig. 9b and c) and captures dynamic whole-body movement patterns associated with walking, running, sitting, and jumping with minimal signal interference (Fig. 9d–g). More notably, epidermal placement on the laryngeal region enables noninvasive monitoring of swallowing physiology (Fig. 9h) and phonation mechanics, where reproducible signal waveforms from repeated utterances demonstrate its potential for speech recognition and disorder diagnosis (Fig. 9i). This multimodal sensing capability, combined with the material's green attributes and self-adhesive properties, shows the platform's broad utility in kinesiology studies, rehabilitation assessment, and human–machine interfaces.

Fig. 9 PPCAPM hydrogel sensor for real-time monitoring of human motion. Changes in relative resistance during (a) finger, (b) wrist, and (c) elbow movements. Relative resistance changes caused by knee joint movements during (d) walking and (e) running. Relative resistance variations associated with (f) sitting down and (g) standing up. Relative resistance fluctuations caused by laryngeal movement during (h) swallowing and (i) speech.

PPCAPM composite hydrogel sensor health monitoring system

The PPCAPM composite hydrogel sensor exhibits significant clinical potential by enabling high-fidelity acquisition of both cardiorespiratory and neuromuscular signals, leveraging its conformal adhesion and sustainable material composition for diagnostic-grade monitoring. Integrated systems positioned on the left chest and wrist successfully captured real-time electrocardiogram (ECG) and electromyogram (EMG) signals (Fig. 10a), displaying characteristic waveforms with clearly defined P-waves, QRS complexes, and T-waves essential for detecting arrhythmias and ischemic conditions (Fig. 10b). Quantitative cardiac interval analysis revealed resting values of 0.7 s (85 bpm) that decreased to 0.4 s (150 bpm) during exercise, providing valuable data for real-time training optimization and cardiovascular risk assessment. The system maintained exceptional signal stability with less than 5% variation during 12 hours of continuous monitoring (Fig. 10c), demonstrating reliability for extended clinical use.

Fig. 10 PPCAPM composite hydrogel sensor for monitoring electrophysiological signals. (a) Working diagram of the PPCAPM hydrogel sensor. (b) ECG signals at rest and during movement. (c) Long-term ECG signal monitoring over 12 h. (d) Comparison of EMG signals between a PPCAPM sensor and a commercial electrode. (e) EMG signal changes with increasing grip strength.

For neuromuscular assessment, the sensor accurately detected dynamic EMG patterns during fist clenching and relaxation cycles, showing strong correlation with commercial systems (Pearson's r = 0.97) (Fig. 10d). Progressive grip strength testing from 5 kg to 40 kg produced dose-dependent increases in EMG amplitude (ΔV = 0.08 mV to 0.32 mV), validating its capability to quantitatively track rehabilitation progress following musculoskeletal injuries (Fig. 10e). This dual-modality monitoring platform combines millisecond-scale temporal resolution with long-term stability, establishing a green and accessible alternative to conventional clinical sensors for telemedicine and point-of-care diagnostics in resource-limited settings.

Conclusions

In summary, we have successfully developed a high-performance multifunctional sensor based on a MXene composite hydrogel through a green and sustainable fabrication strategy. Guided by molecular dynamics simulations, natural PAs were employed as an eco-friendly multifunctional mediator, which not only effectively inhibited MXene restacking via enhanced electrostatic repulsion but also enabled the in situ synthesis of silver nanoparticles under mild conditions with minimal metal consumption. Furthermore, the incorporation of CaCl₂ as a green and inexpensive ionic additive endowed the hydrogel with remarkable moisture retention and antifreeze performance, maintaining excellent mechanical and electrical properties even at −40 °C without relying on conventional organic antifreeze agents. The optimized PPCAPM hydrogel exhibited outstanding comprehensive performance, including high conductivity, large stretchability, and stable adhesion, enabling accurate monitoring of human motions from large-scale joint movements to subtle physiological signals such as EMG and ECG. This work provides a green and feasible paradigm for the design of next-generation wearable sensing systems via natural polyphenol-mediated interfacial engineering and ion-regulated environmental adaptability, showcasing significant potential for sustainable electronics in extreme condition applications.

Author contributions

Runfeng Zhang: data curation, formal analysis, investigation, methodology, visualization, and writing – original draft. Yan Zhang: investigation, methodology and visualization. Zongmao Lv: data curation and visualization. Hai Yu: data curation and formal analysis. Jie Liu: supervision and writing – review & editing. Xuejing Zheng: conceptualization, project administration, resources, supervision, and writing – review & editing. Keyong Tang: supervision. All authors analyzed the experimental data and discussed the results and reviewed the manuscript.

Conflicts of interest

There are no conflicts to declare.

Data availability

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary information (SI). Supplementary information is available. Details of the molecular dynamics simulation, mechanical testing, electrical performance and sensing tests, and electrophysiological tests are provided in the Supporting Information. See DOI: https://doi.org/10.1039/d5gc06447c.

Acknowledgements

This work was supported by the Hubei Key Laboratory of Plasma Chemistry and Advanced Materials (2025P03) and the Opening Fund of the CAS Key Laboratory of Engineering Plastics, Institute of Chemistry, China (No. 20200010).

References

1. Q. Shen, M. Xie, S. Wang, L. Wang and G. Song, Int. J. Biol. Macromol., 2025, 309, 142903.
2. Q. Wang, Z. Zhou, J. He, L. Zhuo, C. Zhu, W. Qian, W. Shi and D. Sun, Microsyst. Nanoeng., 2025, 11, 55.
3. W. Li, S. Wu, M. Kang, X. Zhang, X. Zhong, H. Qiao, J. Chen, P. Wang and L. Tao, J. Mater. Sci. Technol., 2024, 201, 130–138.
4. R. D. Salian, S. Mishra, C. C. Gowda, R. K. Barik, A. K. Singh and C. S. Tiwary, Small Methods, 2025, 2500068.
5. P. Glass, D. Rhoades, G. Bohannon, R. I. Joh, I. Pretzer-Aboff, S. H. Park and D. Joung, Biosens. Bioelectron., 2025, 273, 117161.
6. Y. Han, Y. Liu, Y. Liu, D. Jiang, Z. Wu, B. Jiang, H. Yan and Z. Toktarbay, Adv. Compos. Hybrid Mater., 2025, 8(1), 154.
7. J. Ma, J. Zhu, S. Zhou, C. Zhao, C. Liu, Z. Xin, J. Cai, J. He, P. Feng, L. Guo and X. Tao, Int. J. Biol. Macromol., 2025, 302, 140430.
8. L. Wang, J. Wei, M. You, Y. Jin, D. Li, Z. Xu, A. Yu, J. Li and C. Chen, Carbohydr. Polym., 2025, 354, 123345.
9. H. Zeng, H. Ma, L. Xu, J. Gao, M. Yan and Q. Wang, Int. J. Biol. Macromol., 2025, 297, 139847.
10. Z. Zhao, H. Yuan, J. Wang, J. Tian, H. Gao and Y. Nie, Chem. Eng. J., 2025, 505, 159540.
11. L. Yao, L. Qian, W. Song, S. Zhang, Y. Zhang, L. Zhang, X. Li, G. Yan and V. Nica, ACS Appl. Mater. Interfaces, 2024, 16(36), 48147–48162.
12. L. Lu, Z. Dong, F. Tijiptoharsono, R. J. H. Ng, H. Wang, S. D. Rezaei, Y. Wang, H. S. Leong, P. C. Lim, J. K. W. Yang and R. E. Simpson, ACS Nano, 2021, 15(12), 19722–19732.
13. C. Zhang, Y. Ma, X. Zhang, X. Zhang, S. Abdolhosseinzadeh, H. Sheng, W. Lan, A. Pakdel, J. Heier and F. Nuesch, Energy Environ. Mater., 2020, 3, 29–55.
14. R. Zhang, S. Jiao, X. Zheng, J. Liu and K. Tang, Mater. Today Commun., 2025, 46, 112608.
15. Y. Wu, J. Li, S. Zhang, S. Chang, S. Chang, W. Sun, P. Wang and X. Huang, Chem. Eng. J., 2025, 522, 167125.
16. Y. Wang, M. Huang, J. Lu, H. Zhang, L. Wang, Z. Feng and Y. Wang, Sens. Actuators, B, 2025, 433, 137533.
17. W. Guo, Y. Jiang, B. Wang and M. Ma, J. Mater. Sci. Technol., 2026, 246, 299–313.
18. Z. Yang, Y. Guo, W. Guo, M. Zhao, H. Wang, B. Wei, Y. Miao and K. Guo, Small, 2025, 21(14), 2409621.
19. D. H. Ho, Y. Y. Choi, S. B. Jo, J.-M. Myoung and J. H. Cho, Adv. Mater., 2021, 33(47), 2005846.
20. Y. Pang, Z. Liu, X. Chen, Y. Zhao, D. Wei, J. Wang and C. Yao, J. Mater. Chem. A, 2025, 13, 14113–14126.
21. S. Jiang, Y. Shi, L. Lu, C. Hua, Y. Song and J. Li, J. Power Sources, 2025, 641, 236895.
22. Y. Li, W. Yu, H. Lv, L. Jia, H. Shi, Z. Bian and H. Wang, Sep. Purif. Technol., 2025, 364, 132392.
23. P. Yang, J. Zhang, R. Zhang, G. Duan, Y. Li and Z. Li, Green Chem., 2024, 26, 3329–3337.
24. A. Rauf, M. Imran, T. Abu-Izneid, I. Ul-Haq, S. Patel, X. Pan, S. Naz, A. S. Silva, F. Saeed and H. A. R. Suleria, Biomed. Pharmacother., 2019, 116, 108999.
25. C. Chen, S. A. Althawab and J. M. Awika, Food Chem., 2025, 478, 143513.
26. Z. Huang, B. Guo, J. Pan, D. Gong and G. Zhang, Food Chem., 2025, 484, 144354.
27. J. Luo, Y. Hu, S. Luo, X. Wang, S. Chen and H. Qin, Mater. Today Commun., 2025, 44, 111893.
28. J. Wang, Y. Bi, J. Liang, Z. Lu, K. Liu, Y. Liu, C. Jiang, Z. Yu, K. Zhang, X. Peng, K. Dong and Y. Xia, Nano Energy, 2024, 124, 109500.
29. W. Wu, L. Shi, K. Qian, J. Zhou, T. Zhao, S. Thaiboonrod, M. Miao and X. Feng, J. Colloid Interface Sci., 2024, 654, 1260–1271.
30. Y. Zhang, R. Zhang and Y. Tao, Carbohydr. Polym., 2023, 321, 121300.
31. T. Dai, Y. Lin, Q. Yin, Q. Ji, J. Wang and H. Jia, J. Colloid Interface Sci., 2025, 684, 575–585.
32. Q. Li, B. Tian, G. Tang, H. Zhan, J. Liang, P. Guo, Q. Liu and W. Wu, J. Mater. Chem. A, 2024, 12(6), 3589–3600.
33. R. Chen, H. Tang, Y. Dai, W. Zong, W. Zhang, G. He and X. Wang, ACS Nano, 2022, 16(11), 19124–19132.
34. H. Yan, Z. Yang, C. Xu, J. Li, Y. Liu, R. Zheng, H. Yu, L. Zhang and J. Shu, Energy Storage Mater., 2022, 48, 74–81.
35. L. Zhou, H. Gu, C. Wang, J. Zhang, M. Lv and R. He, Colloids Surf., A, 2013, 430, 103–109.
36. Y. Bai, L. Yan, J. Wang, Z. Yin, N. Chen, F. Wang and Z. Tan, Mater. Des., 2016, 103, 315–320.
37. J. Li, B. Tian, T. Li, S. Dai, Y. Weng, J. Lu, X. Xu, Y. Jin, R. Pang and Y. Hua, Int. J. Nanomed., 2018, 13, 1411–1424.
38. H. Fang, Z. Tang, X. Liu, Y. Huang and C. Sun, Mol. Liq., 2019, 279, 485–491.
39. C. Sun, J. Chen, Y. Gong, X. Zhang and Y. Huang, J. Phys. Chem. B, 2018, 122(3), 1228–1238.
40. J. Wei, G. Wei, Z. Wang, W. Li, D. Wu and Q. Wang, Small, 2020, 16(44), 2004091.
41. Y. Gong, Y. Zhou, H. Wu, D. Wu, Y. Huang and C. Sun, J. Raman Spectrosc., 2016, 47(11), 1351–1359.
42. R. Wang, S. H. Kim, F. Sun, X. Zheng, F. Jiang, X. Wang, B. Diao, H. Zhang, X. Li, R. Li, S. W. Joo, C. Cong and S. Li, ACS Appl. Electron. Mater., 2025, 7(3), 1217–1229.
43. C. Xiang, C. Wen, Z. Wang, Y. Tian, Y. Li, Y. Liao, M. Liu, Y. Zhong, Y. Lin, C. Ning, L. Zhou, R. Fu and G. Tan, ACS Appl. Mater. Interfaces, 2025, 17(5), 8327–8339.
44. D. Zhao, J. Luo, K. Fang, C. Huang, X. Zhou and K. Jiang, Int. J. Biol. Macromol., 2025, 310, 143438.
45. Y. Zhao, X. Song, J. Chen, Y. Chen, X. Wang and F. Wang, Polymer, 2025, 327, 128384.

---