Eco-designed cellulose-reinforced deep eutectic gels with synergistic mechanical strength, ionic conductivity, and freezing tolerance for flexible electronics

Abstract

Growing interest in flexible electronics is driving the development of deep eutectic gels that combine mechanical adaptability, electrical functionality, and environmental sustainability; however, achieving these properties simultaneously remains challenging. Herein, we propose an eco-designed DEG system reinforced with sugarcane bagasse cellulose, where a three-component choline chloride/acrylic acid/glycerol (ChCl/AA/Gly) deep eutectic solvent (DES) serves simultaneously as a cellulose-processing medium and a polymerization precursor. Owing to the acidity of AA and the strong hydrogen bonding network within the DES, raw bagasse cellulose is directly hydrolyzed into micron-scale cellulose fibers and uniformly dispersed, avoiding harsh pretreatments. Within the gel, ChCl provides mobile ions, AA forms a rigid skeleton, Gly enhances flexibility and ion mobility, and cellulose fibers reinforce the matrix through strong hydrogen bonding. The resulting gels exhibit enhanced tensile strength, broad adhesion capability, favorable ionic conductivity (12.7 mS cm⁻¹ at room temperature), and outstanding freezing tolerance (down to –50 °C). These properties enable versatile applications as flexible strain sensors and supercapacitor electrolytes, delivering a high gauge factor (10.17) and durable cycling stability over 20 000 cycles. This study presents a facile strategy for constructing cellulose-reinforced multi-component eutectic gels, offering a sustainable pathway for advanced flexible electronic materials.

---

Green foundation

1. This work develops an eco-designed strategy to directly integrate sugarcane bagasse cellulose into functional eutectic gels using a deep eutectic solvent (DES) as both the processing medium and the polymerization precursor, eliminating harsh pretreatments and exemplifying biomass valorization and solvent economy.

2. The gel exhibits a satisfactory balance between tensile strength and ionic conductivity, enabling reliable flexible strain sensors with a high gauge factor and durable gel electrolyte for supercapacitors that deliver long-term cycling stability over 20 000 cycles and operate reliably under extreme mechanical deformations and low temperatures. This excellent stability minimizes resource consumption from material degradation or replacement.

3. Future advances in sustainability could be realized by replacing acrylic acid with fully bio-based monomers to achieve a closed-loop sustainable materials platform.

Introduction

Growing interest in wearable and flexible electronics is driving the development of soft materials with mechanical adaptability, electrical functionality, and environmental sustainability.^1–4 Conventional hydrogels, composed mainly of water-swollen polymer networks, are attractive due to their ionic conductivity and high flexibility.^5–7 However, because of their high water content, conventional hydrogels tend to evaporate under heat and freeze in cold conditions, which compromises long-term stability.^8–11

Deep eutectic gels, constructed from deep eutectic solvents (DESs) or their polymerized derivatives, have recently attracted growing attention as next-generation soft electrolytes due to their low-temperature tolerance, electrochemical stability, and ionic transport capability.^12–15 DESs are typically prepared by combining a hydrogen bond donor (HBD) and a hydrogen bond acceptor (HBA) at specific molar ratios, where the extensive hydrogen bonding interactions significantly depress the melting point relative to the individual components.¹⁶ This simple mixing strategy leads to materials with low cost, facile preparation, low volatility, and excellent thermal stability, often regarded as sustainable alternatives to ionic liquids.¹⁷ Importantly, the ionic character of DESs provides intrinsic ionic conductivity and broad compositional tunability, enabling their use in flexible energy and sensing devices.^18–20 Beyond traditional binary DESs, multi-component DESs have emerged as a powerful strategy to further tailor solvent properties. By introducing a third component, often an additional HBD, HBA, or functional additive, DESs offer enhanced hydrogen bonding networks, improved fluidity, and expanded design flexibility. However, deep eutectic gels still face several challenges that limit practical applications. Their mechanical strength is often insufficient, leading to poor durability under repeated deformation.

Biomass-derived cellulose, the most abundant renewable biopolymer, provides a natural reinforcement phase with excellent mechanical properties, low cost, and biodegradability.²¹ Yet, efficient dispersion of cellulose into functional gels typically requires intensive pretreatments, such as chemical modification.²² Remarkably, DESs have been extensively explored for the pretreatment and separation of cellulose from lignocellulosic biomass. Their strong hydrogen bonding capability enables the selective disruption of lignin–hemicellulose networks, thereby facilitating cellulose isolation or modification under relatively mild conditions.^23,24 DESs are viewed as greener alternatives to traditional organic solvents and ionic liquids, offering low toxicity, biodegradability, and facile preparation from inexpensive components.²⁵ Furthermore, DESs can be tailored to enhance cellulose swelling, partial hydrolysis, or fibrillation, thereby improving its accessibility for functionalization or incorporation into composite materials.²⁶ These unique properties make DESs not only valuable processing media for biomass but also promising candidates for directly integrating cellulose into functional gel systems.²⁷ Nevertheless, in gels formed by conventional DES-assisted cellulose processing, increasing crosslinking density can improve mechanical strength but often hinders ion mobility, leading to a trade-off between mechanical robustness and ionic conductivity, which remains a challenge.

In this work, we developed a sugarcane bagasse cellulose-reinforced eutectic gel system with synergistic mechanical strength, ionic conductivity, interfacial adhesion, and freezing tolerance. Notably, the three-component choline chloride/acrylic acid/glycerol (ChCl/AA/Gly) DES was selected and serves concurrently as a polymerization medium, ionic conductor, and cellulose-processing agent. Owing to the acidity of acrylic acid and the strong hydrogen-bonding network within the DES, raw sugarcane bagasse cellulose is directly hydrolyzed into micron-scale fibers (CMFs) without harsh chemical pretreatments. In this integrated system, ChCl provides mobile ions for conductivity, AA forms a rigid polymer network skeleton, Gly enhances flexibility and ion mobility, and cellulose fibers act as a reinforcing phase through extensive hydrogen bonding with DES components, enhancing mechanical strength and interfacial adhesion. Besides, the incorporation of sugarcane bagasse cellulose, which originated from abundant agricultural residue, imparts sustainability and biomass valorization. As a consequence, the tailored eutectic gel ensures enhanced tensile strength, strong adhesion, excellent ionic conductivity, and freezing tolerance, exhibiting outstanding performance in versatile electronics applications, including information transmission through Morse code signaling, strain-responsive monitoring of human motion, and electrolyte integration in flexible supercapacitors for stable energy storage.

Materials and experimental

Preparation and mechanical properties

Sugarcane bagasse cellulose (delignified and hemicellulose-removed) was obtained from Guangxi Nanning East Asia Sugar Group (China). Choline chloride (ChCl, AR), acrylic acid (AA, AR), and glycerol (Gly, AR) were obtained from Aladdin Reagent Co., Ltd (Shanghai, China). Activated carbon (AC) and Ketjen Black were obtained from Guangdong Canrd New Energy Technology Co., Ltd (Guangdong, China).

Synthesis of EGC gels

Although the bagasse cellulose used was delignified and hemicellulose-removed, the subsequent DES treatment enables direct hydrolysis of cellulose without further harsh chemical pretreatment, which can still be regarded as a mild approach. First, ChCl, AA, and Gly were mixed at a molar ratio of 1 : 2.0: 1.2 and heated at 60 °C under stirring until a transparent deep eutectic solvent formed. Then, bagasse cellulose (mass gradients provided in Table S1) was added and stirred at 90 °C for 4 h, followed by probe sonication (20% power, 5 min) to obtain a uniform CMF dispersion. Next, the photoinitiator (1 mol% relative to AA) was introduced and stirred at 60 °C for 15 min. Finally, the dispersion was exposed to a 150 W UV lamp for 10 min, yielding EGC gels.

Characterization studies

Fourier transform infrared (FTIR, Thermo Fisher Nicolet iS50, 4000–400 cm⁻¹) and X-ray diffraction (XRD, Bruker D8 ADVANCE) were used for structural analyses. Cellulose morphologies were examined by scanning electron microscopy (SEM, JSM-7600Fs). Mechanical properties were tested on a universal testing machine (UTM6503), including tensile strength and adhesive performance (lap shear test at 50 mm min⁻¹). The sensing performance of the gels was evaluated using a Digital Graphical Sampling Multimeter (Keithley 7510). The gels were integrated into circuits and attached to human joints (fingers, wrists, and elbows) or mounted on the universal testing machine for tensile and compressive cycles. Deformation was induced either through human motion or programmed machine control, and the corresponding real-time resistance changes were recorded. The human motion monitoring tests were conducted in accordance with The Code of Ethics of the World Medical Association, and informed consent was obtained from the volunteer.

Electrochemical performance tests

The ionic conductivity of the gels was measured on a CHI 760E electrochemical workstation using electrochemical impedance spectroscopy (EIS) in the frequency range of 0.1 Hz–100 kHz. For supercapacitor assembly, activated carbon (AC) and Ketjen Black served as the electrode active material and conductive agent, respectively. Cyclic voltammetry (CV) of the supercapacitor was performed using a CHI 760E within 0.0–1.8 V at scan rates ranging from 5 to 100 mV s⁻¹. Galvanostatic charge–discharge (GCD) tests were conducted at current densities of 1.0–10.0 A g⁻¹. Long-term cycling stability was evaluated using a CT3001A LAND battery test system at 5.0 A g⁻¹ within the same voltage window.

Results and discussion

Preparation and mechanical properties

Fig. 1 illustrates the preparation process of the sugarcane bagasse cellulose-reinforced eutectic gel (EGC). First, a three-component ChCl/AA/Gly DES was synthesized by heating at 60 °C under continuous stirring. This DES was then employed to directly downsize raw bagasse cellulose into micron-scale cellulose fibers (CMF) via ultrasonication, yielding a homogeneous DES/CMF dispersion. Finally, the eutectic gel reinforced with CMF was obtained through photoinitiated in situ polymerization. Detailed procedures are provided in the SI. In this DES system, acrylic acid (AA) and glycerol (Gly) act as hydrogen bond donors (HBDs), while choline chloride (ChCl) functions as the hydrogen bond acceptor (HBA). Importantly, the synergistic interactions among the three components endow the DES with multifunctional roles. In detail, ChCl provides abundant ionic species that facilitate efficient ion transport, AA contributes a rigid framework for the gel network, and Gly imparts flexibility while further promoting ionic conductivity.²⁸ Therefore, in this design, the pure eutectic gel matrix ensures high ionic conductivity, while the uniformly dispersed CMF imparts effective mechanical reinforcement through strong hydrogen bonding interaction. The physical photograph of the as-prepared EGC gel highlights its high transparency, which also confirms the uniform dispersion of CMF in the gel matrix.

Fig. 1 Schematic illustration of the preparation route and composition of the EGC gel.

Fig. 2a presents photographs of bagasse cellulose processed with deionized water (i) and with DES (ii), where the DES/CMF dispersion is homogeneous, whereas the deionized water/bagasse cellulose system shows obvious delamination and sedimentation. Besides, Fig. 2b shows their optical microscopy images. The DES-processed sample exhibits significantly shortened cellulose fibers (∼200 µm), which is further confirmed by SEM observations (Fig. S1). This difference arises from the acidity of AA combined with the strong hydrogen bonding interactions in the DES, which cleave bagasse cellulose into shorter segments, while simultaneous hydrogen bonding interactions between CMF and DES ensure uniform dispersion in the medium.^27,29 Such efficient dispersion is essential for achieving uniform reinforcement in the final eutectic gel, thereby ensuring reliable mechanical and functional performance.

Fig. 2 (a) Dispersion photos of the deionized water-processed (i) and DES-processed (ii) bagasse cellulose. (b) Length of the deionized water-processed (i) and DES-processed (ii) bagasse cellulose. (c) XRD patterns of CMF and original bagasse cellulose. (d) Stress–strain curves, (e) EIS plots, and (f) ionic conductivity of the gels with different added contents of CMF. (g) Frequency dependence of the stored modulus (G′) and loss modulus (G″) of the EG and EGC-1 gels. (h) Cyclic stress–strain profiles of the EGC-1 gel. (i) Adhesive strength of the EGC-1 gel to different substrates.

Fig. 2c exhibits the XRD patterns of the hydrolyzed CMF and the original bagasse cellulose. Both samples exhibit characteristic peaks at the (110), (200), and (004) planes, confirming that the DES treatment does not alter the crystalline structure of bagasse cellulose.³⁰ This preservation arises because DES primarily interacts with the amorphous regions of cellulose, leaving the crystalline domains intact. Complementarily, the FTIR spectrum (Fig. S2) shows the disappearance of the C=C bond in the EGC gel, indicating successful polymerization of the AA monomer matrix. Besides, ChCl and Gly are physically trapped within the poly(acrylic acid) (PAA)/cellulose matrix through extensive hydrogen bonding, as illustrated in Fig. 1.

To investigate the influence of bagasse cellulose content on ionic conductivity and mechanical performance, a series of gels with varying cellulose loadings was prepared (see Table S1 for details). Fig. 2d compares the mechanical performance of these gels. The cellulose-free gel (EG) exhibits an ultrahigh maximum strain of 2517.8% but a relatively low maximum stress of 13.1 kPa. In contrast, the EGC-1 gel with only 0.05 wt% cellulose achieves a markedly enhanced maximum stress of 66.9 kPa, while maintaining a high strain of 2028.0%. With further cellulose incorporation, the stress continues to increase but at the expense of strain. At 1.0 wt% cellulose, the EGC-5 gel delivers the highest stress (348.8 kPa) but a much-reduced strain of 188.8%. These results indicate that moderate cellulose addition reinforces the gel network by serving as a stress-transfer skeleton and hindering crack propagation. However, excessive cellulose leads to fiber agglomeration and interfacial defects, which disrupt the uniformity of the gel matrix and compromise its ductility.

Fig. 2e and f present the ionic conductivity of the gels at room temperature as a function of cellulose content. With increasing cellulose addition, the electrochemical impedance rises, leading to a gradual decline in ionic conductivity. The cellulose-free EG gel exhibits the highest conductivity of 17.1 mS cm⁻¹. At a low cellulose loading of 0.05 wt% (EGC-1), the conductivity decreases slightly to 12.7 mS cm⁻¹, while at 1.0 wt% cellulose (EGC-5), it drops sharply to 3.2 mS cm⁻¹. Such reduction arises from the dense hydrogen bonding network formed between cellulose and the gel matrix, which restricts ion transport. These results highlight the trade-off between enhanced mechanical strength and compromised ionic transport introduced by cellulose. To balance these effects for flexible electronic applications, the EGC-1 gel (0.05 wt% cellulose) was selected for subsequent characterization, as it achieves a desirable combination of conductivity and mechanical reinforcement. The aforementioned FTIR spectrum and optical images are based on this optimized composition.

Fig. 2g presents the rheological properties of the EG and EGC-1 gels at room temperature. In both cases, the storage modulus (G′) consistently exceeds the loss modulus (G″) across the tested shear frequency range, confirming their elastic solid-like behavior and the integrity of their internal networks. Notably, the EGC-1 gel exhibits higher values of both G′ and G″ compared to the EG gel, demonstrating that the introduction of cellulose enhances mechanical stability through hydrogen bonding interactions.³¹ The cyclic stress–strain response of the EGC-1 gel under 100% fixed strain is shown in Fig. 2h. After the first loading–unloading cycle, the strain cannot fully return to its initial state, which is attributed to the irreversible breakage of some weak hydrogen bonds that cannot be reformed immediately. Nevertheless, even after 100 continuous cycles, the gel maintains a favorable stress level, underscoring its stable and durable mechanical performance.

Fig. S3a demonstrates the adhesive performance of the EGC-1 gel, which can tightly and stably adhere to diverse substrates, including a 100 g metal weight, rubber, copper foil, titanium foil, and wood, highlighting its excellent adhesion capability. To further evaluate this property, lap shear tests were carried out on various substrates. As shown in Fig. 2i and Fig. S3b, the EGC-1 gel exhibits strong adhesion not only to surfaces rich in active functional groups (e.g., titanium foil and glass) but also to low-energy surfaces such as PTFE, which typically resist adhesion. This universal adhesion originates from the abundant hydrogen bonding interactions provided by the DES matrix and cellulose micron fibers, which enable intimate interfacial contact with chemically diverse substrates. Moreover, the gel maintains favorable repeatability. Even after three consecutive lap shear tests on titanium foil, the adhesive strength remains around 1.07 MPa. Such robust and reusable adhesion is particularly advantageous for flexible electronic devices, where stable electrode–electrolyte interfaces and device integration under repeated use are essential.

Sensing performance

Based on its favorable ionic conductivity and strong adhesive properties, the EGC-1 gel was first explored as a flexible sensor material. Fig. 3a illustrates the conductivity of the gel in a closed circuit with an LED light. When the EGC-1 gel is connected to the circuit, the LED lights up, while cutting the gel disconnects the circuit and extinguishes the light. Remarkably, when the two gel segments are simply brought back into contact, the circuit is restored and the LED re-lights, even without complete self-restoration of the gel, demonstrating the excellent ionic conductivity and interfacial connectivity of the EGC-1 gel. Moreover, when tested under simulated low-temperature conditions, the LED remains lit, confirming the gel's stable ionic conduction at sub-zero temperatures. The complete circuit photographs are displayed in Fig. S4. To further highlight its functionality, the EGC-1 gel was employed for Morse code signaling, which is often used to convey information where spoken language or complex text is impractical.^32,33 As shown in Fig. 3b–f, by controlling the pressing duration of a finger on the EGC-1 gel, distinct output signals corresponding to different Morse codes were successfully generated, enabling the communication of messages such as ‘SOS’, ‘HELLO’, ‘GOODJOB’, and ‘ICIFP’. These results confirm that the EGC-1 gel can serve as an effective and low-temperature-tolerant ionic conductor for flexible information transmission devices.

Fig. 3 (a) Ionic conductivity of the EGC-1 gel under self-healed and low-temperature conditions. (b) Illustration of the Morse code. Sensing output signals of the EGC-1 gel corresponding to Morse codes: (c) “SOS”, (d) “HELLO”, (e) “GOODJOB”, and (f) “ICIFP”.

To further explore its practical potential in flexible sensing, the EGC-1 gel was applied to human motion monitoring. Fig. 4a–e present the output signals generated by finger, wrist, and elbow movements. As shown in Fig. 4a, stable relative resistance changes (ΔR/R₀) are obtained by repeatedly pressing and releasing the EGC-1 gel with fingers. When adhered to the finger, the EGC-1 gel sensor responds sensitively to bending angles of 30°, 60°, and 90°, with the ΔR/R₀ value increasing proportionally with the degree of deformation (Fig. 4b). This demonstrates that the EGC-1 gel can effectively translate mechanical strain into electrical signals. Furthermore, the sensor exhibits rapid response and reliable signal stability. As shown in Fig. 4c, the ΔR/R₀ value increases or decreases immediately with changes in finger angle and remains constant when the finger is held still. Beyond fine movements, the EGC-1 gel also shows robust performance in monitoring larger-scale motions such as wrist flexion and elbow bending (Fig. 4d and e), where distinct and repeatable resistance changes are observed. Overall, the findings demonstrate the EGC-1 gel's high sensitivity, stability, and adaptability, underscoring its potential for real-time motion sensing in wearable electronic devices.

Fig. 4 Output signals of the EGC-1 gel originating from (a) repeated finger pressing and releasing, (b) bending the finger at different angles repeatedly, (c) bending the finger at different angles held for a period, and (d) wrist and (e) elbow movement. (f) GF of the EGC-1 gel at different tensile strains. (g) Long-term sensing performance of the EGC-1 gel under a fixed tensile strain of 50%.

To further assess sensitivity, the EGC-1 gel sensor's gauge factor (GF) was measured, representing how ΔR/R₀ varies with applied strain.³⁴ As shown in Fig. 4f, ΔR/R₀ increases progressively with tensile strain from 0% to 300%. The calculated GF values are 5.77, 8.10, and 10.17 in the strain ranges of 0–100%, 100–200%, and 200–300%, respectively, demonstrating that the EGC-1 gel sensor possesses high sensitivity across a wide deformation range and is well-suited for strain-sensing applications. In addition, the fatigue resistance of the sensor was examined under cyclic loading. Fig. 4g shows that under a fixed tensile strain of 50%, the EGC-1 gel sensor maintains stable output signals over 3000 cycles. Even after repeated stretching, the gel retains excellent mechanical integrity and ionic conductivity, highlighting its robust fatigue resistance. These results confirm that the EGC-1 gel sensor combines high sensitivity with long-term durability, both of which are essential for reliable operation in wearable and flexible electronics.

Electrochemical performance

Supercapacitors, as emerging electrochemical energy storage devices, can be charged and discharged within seconds and are distinguished by their high power density and ultra-long cycling life.³⁵ Among their key components, gel electrolytes are particularly critical, as they directly govern the electrochemical performance and mechanical stability of flexible supercapacitors.^36,37 Given its favorable ionic conductivity, robust mechanical properties, and strong interfacial adhesion, the EGC-1 gel was selected as the electrolyte to assemble a flexible supercapacitor, hereafter referred to as EGC-SC. The electrochemical properties of EGC-SC were then systematically investigated to evaluate its practical applicability in flexible energy storage devices.

Fig. S5a presents the linear sweep voltammetry (LSV) results, which were used to determine the electrochemical window of EGC-SC. The LSV curve remains relatively smooth within 0.0–1.8 V, indicating stable operation. This result was further verified by cyclic voltammetry (CV) measurements under different voltage windows (Fig. 5a). When the applied voltage exceeds 1.8 V, the current of EGC-SC increases sharply and evolves into an asymmetric shape, suggesting electrolyte decomposition. Thus, the safe operating voltage window of EGC-SC is established at 0.0–1.8 V.

Fig. 5 (a) CV curves at 5 mV s⁻¹ with different potential ranges. (b) CV curves with the scan rates increasing from 5 mV s⁻¹ to 100 mV s⁻¹. (c) Specific capacity at different current densities. (d) Long-term cycling performance at a current density of 5 A g⁻¹. (e) GCD curves of the supercapacitor with different bending angles. (f) CV curves of the supercapacitor with different loads. (g) Practical application photos of the supercapacitor powering a timing device. (h) Ionic conductivity of the EGC-1 gel and (i) capacitance retention of the EGC-SC at different temperatures.

Furthermore, CV measurements of EGC-SC were performed at scan rates from 5 to 100 mV s⁻¹ (Fig. 5b). The CV curve exhibits an approximately rectangular shape at 5 mV s⁻¹, characteristic of ideal capacitive behavior and stable operation of the symmetric supercapacitor device. With increasing scan rate, the curves show only slight distortion, and even at 100 mV s⁻¹, they retain a nearly rectangular profile without significant deformation. This demonstrates that EGC-SC maintains efficient charge storage and rapid ion transport at a high scan rate, fulfilling the requirements of a solid-state electrolyte for high-power supercapacitors.

Galvanostatic charge–discharge (GCD) tests of EGC-SC were conducted at different current densities (Fig. S5b). With the increase of current density from 1.0 to 10.0 A g⁻¹, the charge–discharge profiles consistently show nearly symmetrical isosceles triangular shapes, confirming the stable and reversible operation of the supercapacitor device. The specific capacitances calculated from the GCD curves (formula provided in the SI) are 152 F g⁻¹ at 1.0 A g⁻¹, 126 F g⁻¹ at 5.0 A g⁻¹, and 93 F g⁻¹ at 10.0 A g⁻¹ (Fig. 5c). Although the capacitance gradually decreases with increasing current density, the device still retains 83% and 61% of its initial capacitance at high current densities of 5.0 and 10.0 A g⁻¹, respectively, demonstrating favorable rate performance and fast ion transport of EGC-SC. Additionally, the EGC-SC achieves an energy density of 16.9 Wh kg⁻¹ at a power density of 463.4 W kg⁻¹ and maintains 9.9 Wh kg⁻¹ at 5940.1 W kg⁻¹ (Fig. S6). A long-term cycling test of the EGC-SC was conducted at 5 A g⁻¹. The EGC-SC maintains 78.05% capacitance after 20 000 charge–discharge cycles while maintaining a high coulombic efficiency of 99.3%, demonstrating excellent electrochemical reversibility and long-term stability (Fig. 5d).

Considering that flexible electronic devices often operate under mechanical deformation, the electrochemical performance of the EGC-SC was evaluated at various bending angles. As seen in Fig. 5e and Fig. S7a–c, both GCD and CV tests were performed with fixed bending angles of 45°, 60°, and 90°. In all cases, the GCD curves retain nearly ideal isosceles triangular shapes, while the CV curves maintain approximately rectangular profiles, indicating stable capacitive behavior. Interestingly, the specific capacitance shows a slight increase with increasing bending angle, likely due to improved electrode–electrolyte contact. These results confirm that the EGC-SC operates reliably under bending deformation, ensuring mechanical tolerance and stable functionality in flexible electronic applications.

The pressure tolerance of the EGC-SC was further investigated under heavy loading conditions. As shown in Fig. 5f, CV curves recorded at 40 mV s⁻¹ under applied weights of 50 g and 100 g reveal that the EGC-SC continues to operate normally. With increasing load, the enclosed CV area expands, attributed to the compression of the EGC-1 gel, which reduces the electrode spacing, shortens ion transport pathways, and enhances ionic conduction, thereby improving charge storage capability. Furthermore, the CV curves recorded under 50 g and 100 g loading (Fig. S7d and S7e) consistently maintain nearly rectangular and symmetric shapes, confirming the electrochemical stability of the EGC-SC under mechanical pressure. As shown in Fig. 5g, the EGC-SC continues to power the electronic timer even after being struck with a hammer, demonstrating its mechanical robustness and reliable device integration. Under normal conditions, the EGC-SC is able to sustain stable operation of the timer for up to 220 minutes. These results highlight the robustness of the EGC-SC, making it highly suitable for practical applications in wearable electronics that must withstand pressing or impact forces during operation.

Benefiting from the wide working temperature window of the eutectic gel, the EGC-SC exhibits reliable performance even under extremely low-temperature conditions. To evaluate this, the EGC-1 gel was placed in a low-temperature reactor with controlled environmental temperature and connected to an electrochemical workstation to measure its conductivity. As shown in Fig. S8 and Fig. 5h, the impedance gradually increases as the temperature decreases. The EGC-1 gel exhibits a conductivity of 12.7 mS cm⁻¹ at 30 °C, which decreases to 8.9 mS cm⁻¹ at −10 °C and remains at 4.5 mS cm⁻¹ even at −50 °C, confirming its ability to operate in subzero environments. Compared with the reported DES-based gel for flexible electronics, EGC-1 achieves an optimal balance between ionic conductivity and mechanical performance (Table S2). Further CV measurements of the EGC-SC at low temperatures (Fig. S9) revealed that, although the enclosed CV area decreases, the curves retain an approximately rectangular shape, indicating stable capacitive behavior. The calculated specific capacitance retention is 86.7% at −10 °C and 59.0% at −50 °C relative to room temperature (30 °C) (Fig. 5i). Notably, the device is still able to power an electronic timer under freezing conditions, as illustrated by the inset of Fig. 5i, underscoring its excellent anti-freezing capability and reliable operation in extreme environments. Notably, although complete delamination is not intended due to the designed high adhesion, the device enables module-level reuse that extends service life and reduces material consumption (Fig. S10).

Conclusions

In summary, this work presents a facile and eco-designed strategy for constructing multifunctional deep eutectic gels reinforced with DES-assisted hydrolyzed bagasse cellulose. The three-component ChCl/AA/Gly DES serves as both a green polymerization medium and a biomass-processing agent that directly converts raw bagasse into micron-scale cellulose fibers without additional modification. The polymerization process operates under mild conditions, minimizing energy use and waste. This integrated process exemplifies biomass valorization and solvent economy, fully consistent with green chemistry principles. Compared to traditional nanocellulose-reinforced gels, which tend to be more expensive due to the higher cost of nanocellulose, this method demonstrates satisfactory techno-economic feasibility and sustainability. The resulting cellulose-reinforced eutectic gels achieve a satisfactory synergy between mechanical strength and ionic conductivity, exhibiting enhanced tensile strength, strong adhesion, efficient ion transport, and remarkable low-temperature tolerance down to −50 °C. When applied in flexible devices, the gels deliver a high gauge factor up to 10.17, enabling Morse code-based information transmission, strain-responsive human motion sensing with stable and repeatable signals, and energy storage in flexible supercapacitors. The EGC-SC maintains 78.05% capacitance after 20 000 charge–discharge cycles with 99.3% coulombic efficiency, retaining robust electrochemical performance even at extremely low temperatures. Overall, this study presents a facile strategy for constructing cellulose-reinforced multi-component eutectic gels, offering a sustainable pathway for advanced flexible electronic materials.

Author contributions

Xiangyu Lin: writing – original draft, data curation, and conceptualization. Jie Li and Ziming Zhu: investigation, methodology, formal analysis, and data curation. Fei Fu: writing – original draft and investigation. Yuandong Xu: supervision and writing – review & editing. He Liu: conceptualization and funding acquisition. Xu Xu: writing – review & editing.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information is available, including detailed experimental and characterization section; SEM images of the processed bagasse cellulose; FTIR spectrum; adhesion ability to different materials; LSV profile and GCD curves at different current densities; CV curves of the EGC-SC at different bending angles and with different loads; EIS plots of the EGC-1 gel and CV curves of the EGC-SC at different temperatures. See DOI: https://doi.org/10.1039/d5gc05362e.

Acknowledgements

We gratefully acknowledge the National Natural Science Foundation of China (No. 32494794) for the financial support.

References

1. Y. Zuo, W. Zhang, M. Wei, P. Zhang, S. Zhao, P. Pei, L. Qiu, H. Wang, Z. Meng and K. Wang, Energy Storage Mater., 2022, 53, 136–147.
2. H. C. Ates, A. K. Yetisen, F. Güder and C. Dincer, Nat. Electron., 2021, 4, 13–14.
3. Z. J. Chen, T. Y. Shen, M. H. Zhang, X. Xiao, H. Q. Wang, Q. R. Lu, Y. L. Luo, Z. Jin and C. H. Li, Adv. Funct. Mater., 2024, 34, 2314864.
4. M. Babu, Z. Lautman, X. Lin, M. H. B. Sobota and M. P. Snyder, Annu. Rev. Med., 2024, 75, 401–415.
5. M. Zhang, T. Xu, K. Liu, L. Zhu, C. Miao, T. Chen, M. Gao, J. Wang and C. Si, SusMat, 2025, 5, e255.
6. Z. Wang, S. Wang, L. Zhang, H. Liu and X. Xu, Research, 2024, 7, 0298.
7. X. Yan, X. Lin, H. Liu, J. Lu, H. Wang, X. Huang, H. Liu and X. Xu, Carbohydr. Polym., 2024, 344, 122552.
8. Y. Duan, T. Lv, K. Dong, F. Zheng, X. Li, Y. Qi, Z. Chen, W. Tang, Q. Sun, S. Cao and T. Chen, Chem. Eng. J., 2023, 474, 145551.
9. Y. Zhang, X. Chen, W. Li, B. Wang, Q. Zhang, Z. Zhao, Y. Wang, Y. Guan and Y. Zhang, Adv. Funct. Mater., 2025, e11554.
10. P. Sun, J. Chen, Y. Li, X. Tang, H. Sun, G. Song, X. Mu, T. Zhang, X. Zha, F. Li, Y. Gao, S. Cong and Z. Zhao, InfoMat, 2023, 5, 12363.
11. Q. Liu, T. Wang, C. Wang and D. Wu, Chem. Eng. J., 2024, 481, 148292.
12. Y. Ma, Y. Yang, T. Li, S. Hussain and M. Zhu, Green Chem., 2024, 26, 3627–3669.
13. B. B. Hansen, S. Spittle, B. Chen, D. Poe, Y. Zhang, J. M. Klein, A. Horton, L. Adhikari, T. Zelovich, B. W. Doherty, B. Gurkan, E. J. Maginn, A. Ragauskas, M. Dadmun, T. A. Zawodzinski, G. A. Baker, M. E. Tuckerman, R. F. Savinell and J. R. Sangoro, Chem. Rev., 2021, 121, 1232–1285.
14. H. Qin, X. Hu, J. Wang, H. Cheng, L. Chen and Z. Qi, Green Energy Environ., 2020, 5, 8–21.
15. A. P. Abbott, G. Capper, D. L. Davies, R. K. Rasheed and V. Tambyrajah, Chem. Commun., 2003, 153, 70–71.
16. E. L. Smith, A. P. Abbott and K. S. Ryder, Chem. Rev., 2014, 114, 11060–11082.
17. Y. Li, H. Li, X. Fan, M. Cai, X. Xu and M. Zhu, ACS Sustainable Chem. Eng., 2022, 10, 7253–7264.
18. Q. Zeng, X. Lai, H. Li, Z. Chen, X. Zeng and L. Zhang, Adv. Funct. Mater., 2024, 34, 2411029.
19. Y. Lai, X. Zhong, S. Zhang, G. Kim, J. Bae, Y. Ding and G. Li, Chem. Mater., 2025, 37, 7534–7555.
20. M. H. Dokoohaki, A. R. Zolghadr and A. Klein, Ind. Eng. Chem. Res., 2022, 61, 11464–11473.
21. X. Sun, Y. Zhu, J. Zhu, K. Le, P. Servati and F. Jiang, Adv. Funct. Mater., 2022, 32, 1526311.
22. W. Yu, C. Wang, Y. Yi, W. Zhou, H. Wang, Y. Yang and Z. Tan, Cellulose, 2019, 26, 3069–3082.
23. T. Xiao, M. Hou, X. Guo, X. Cao, C. Li, Q. Zhang, W. Jia, Y. Sun, Y. Guo and H. Shi, Renewable Sustainable Energy Rev., 2024, 192, 114243.
24. G. Wu, Y. Cheng, C. Huang, C. Yong and Y. Fu, Green Chem., 2025, 27, 1556–1569.
25. X. Shi, Z. Wang, S. Liu, Q. Xia, Y. Liu, W. Chen, H. Yu and K. Zhang, Nat. Sustain., 2024, 7, 315–325.
26. S. Wang, L. Zhang, K. A. Khan, M. W. Ullah, Y. Fan and Z. Wang, Mater. Sci. Eng., R, 2025, 165, 101025.
27. X. Sun, Y. Zhu, J. Zhu, K. Le, P. Servati and F. Jiang, Adv. Funct. Mater., 2022, 32, 2202533.
28. J. Li, X. Lin, S. Wang, L. Chen, H. Liu, H. Wang, Q. Li and X. Xu, J. Energy Storage, 2025, 127, 117126.
29. C. C. Okwuwa, F. Adam and M. E. Ries, Cellulose, 2025, 32, 5335–5359.
30. L. Zhang, L. Chen, S. Wang, S. Wang, D. Wang, L. Yu, X. Xu, H. Liu and C. Chen, Nat. Commun., 2024, 15, 3859.
31. X. H. Shao, X. Yang, Y. Zhou, Q. C. Xia, Y. P. Lu, X. Yan, C. Chen, T. T. Zheng, L. L. Zhang, Y. N. Ma, Y. X. Ma and S. Z. Gao, Soft Matter, 2022, 18, 2814–2828.
32. Z. Ma, J. Zhang, Z. Ma, M. Lou, P. Zou, H. Wang and L. Jia, J. Mater. Chem. A, 2025, 13, 9418–9427.
33. Z. Wei, L. Jia, J. Yu, H. Xu, X. Guo, T. Xiang and S. Zhou, Mater. Horiz., 2025, 12, 2604–2618.
34. G. Zeng, S. Chen, Y. Zhou, G. Li, G. Li, J. Zhu, C. Xu and Y. Xin, ACS Appl. Mater. Interfaces, 2025, 17, 38532–38544.
35. Y. Wang, S. Sun, X. Wu, H. Liang and W. Zhang, Nano-Micro Lett., 2023, 15, 78.
36. X. Liang, C. Liu, S. Liao, S. X. Yao and M. He, ACS Appl. Mater. Interfaces, 2024, 16, 4661–4670.
37. X. Gao, X. Liu, A. Mariani, G. A. Elia, M. Lechner, C. Streb and S. Passerini, Energy Environ. Sci., 2020, 13, 2559–2569.

---