Stretchable and self-healable conductive hydrogel-based multifunctional triboelectric nanogenerator for energy harvesting and dance motion sensing Open Access

The emergence of a new generation of flexible electronic products with excellent electrical performance and adaptable deformation has attracted great attention.^1,2 Wearable flexible electronic devices play an increasingly important role in many fields, such as health care monitoring, human–computer interaction, and personal identification information storage.³ However, the new, flexible electronic devices need high-performance conductive materials, which is different from the requirement of previously used metal electrodes, and, thus, the newly devised conductive materials need to exhibit good compatibility with human skin and tissues.^4–8 In recent years, conductive hydrogel, as a new electrode material, has attracted wide attention because it integrates programmable electrical characteristics, soft and flexible characteristics, and biocompatibility.⁹ When a conductive hydrogel is deformed, the resistance change of the conductive hydrogel is only a few hundred times that of traditional metal conductors, and this meets the requirements of wearable devices in complex deformation environments.^10–14 At the same time, this excellent electrical performance makes them have a high level of robustness and repeatability under a certain range of mechanical elastic deformation. Thus, it is necessary to develop hydrogels with considerable conductivity and good physical and chemical properties, especially as components of wearable devices, to give full play to their excellent electrical properties. In addition, the proposed hybrid conductive hydrogel is composed of a flexible polymer architecture and distributed conductive network. Compared with the previously used inorganic electrode materials, hybrid hydrogels have many advantages, such as excellent skin compatibility and conductivity.^15–19 Owning to its high stability, conductivity, and cytocompatibility, poly(3,4-ethylene dioxythiophene):polyethylene terephthalate (PEDOT:PET) is rapidly gaining recognition in the fields of biology and electronics.²⁰ Furthermore, good fatigue strength and stability are also crucial characteristics of PEDOT:PET-based hydrogels.^21,22 And the strong adhesion could make PEDOT:PET-based hydrogel possible for the gadget to fit properly across the irregular surface of human skin or tissue, which would be beneficial for the conversion of signals and communication stability.²³ Nevertheless, PEDOT:PET has conjugated π bond with rigid characteristics, which will lead to its inherent weak mechanical properties. Therefore, it is necessary to develop interconnection networks to improve the mechanical properties of PEDOT:PET, but this also leads to a decrease in conductivity of the hybrid conductive hydrogel, which brings new challenges for the development of hybrid conductive hydrogels with high strength and high conductivity.^24–27 

In 2021, Liu et al. reported a novel adhesive strain sensor by using acrylic acid (AA), methacrylatoethyl trimethyl ammonium chloride (DMC), and dopamine methacrylate (DMA), with good sensing performance, but the sensitivity and detection range of the device were limited.²⁸ In addition, complex application environments constantly require conductive hydrogels to have excellent mechanical properties and fatigue resistance. However, due to the high water content of soft materials, they are prone to mechanical degradation during use, which may lead to loss of their initial mechanical strength and sensing performance. Inspired by the characteristics of human skin, it is meaningful to develop a hybrid conductive hydrogel with high flexibility and strength.^29,30 Additionally, conductive hydrogels would be able to repair cracks and injuries that were induced by physical loads or stimuli. From previous works, many electronic hydrogels have been developed that show remarkably good electrical performance. However, there is still a long way to go in terms of the development of new high strength and high conductivity hydrogels. On the one hand, it is necessary to promote the research of hybrid conductive hydrogels with multiple properties. On the other hand, it is also important to expand their applications in flexible wearable sensors and human movement energy harvesting to highlight their application value. Significantly, in the era of intelligent education, artificial intelligence technology has been widely used in various human–computer interaction applications. The foundation of human–computer interaction task is human posture sensing, which uses flexible sensors to obtain real-time information of human postures.^31,32 However, in the teaching of dance, sports, and other fields, the diversity of human posture and the complex background of most scenes lead to low accuracy of human posture estimation, and it is difficult to achieve the effect of professional teachers in posture correction. In order to predict accurate human posture information in complex background, high sensitivity wearable sensors are needed. Furthermore, the design of flexible generator based on human motion energy harvesting is also a current research hotspot, and a combination of the excellent characteristics of conductive hydrogels can further promote their development.

In this work, we proposed a double-network hybrid polyacrylamide (PAAM)/poly(acrylic acid) (PAA)/MXene/PEDOT:PET (PPMP) hydrogel with excellent flexibility, self-healing capabilities, and stability. In detail, the basic framework frame consists of PAA and PAAM. The MXene and PEDOT: PET were used to form conductive networks. Furthermore, PPMP conductive hydrogels with MXene/PDMS encapsulation layer can play the role of strain sensor and serve as flexible stretchable triboelectric nanogenerators (FS-TENGs). The PPMP conductive hydrogel-based strain sensor can be used to monitor the middle posture of dance movements, which will provide information for dance training and action correction. Furthermore, we illustrate using a series of demonstrations that FS-TENG devices possess good signal output and energy-harvesting capabilities. Consequently, this research can offer a feasible strategy for dance motion sensors and energy harvesters for posture monitoring and human motion energy harvesting.

Kehui Chemical Co., Ltd. supplied acrylamide monomer (AAM), acrylic acid (AA), N, N′-methylene bisacrylamide (MBA), and ammonium persulfate (APS). Ti₃C₂ MXene were purchased from Beike Nano Co., Ltd. Bayer Chemical Co., Ltd. supplied PEDOT nanotubes and PET solution materials aqueous dispersion. Dow Corning Co., Ltd. supplied the PDMS precursor and curing ingredients. Nanjing Shuguang Chemical Co., Ltd. supplied the single-walled carbon nanotube (length: 5–15 µm).

In this design, MBA was used as cross-linking agent while APS played the role of initiator. Furthermore, MXene and PEDOT:PET were the constituents responsible for the device’s conductivity. PAA is obtained by polymerization of AA and APS in de-ionized (DI) water in 15:1 weight ratio. The PAA was made by polymerizing the solution described above at 45 °C for 3.5 h. Then, the monomer AAM was distributed in the DI water by using a mechanical mixer. Then, the prepared PAA was dissolved in AAM and MBA successively by gentle stirring. APS, PEDOT: PET, and MXene were dissolved in suspension, and then the free radical polymerization of AM was started. The weight ratio of PEDOT to AAM was 25%, while the weight ratio of APS to AAM was just 1.5%. In addition, the weight ratio of MXene to AAM was 0.65%. Finally, to ensure that the sample can be fully gelatinous, the polymerization of AAM monomer was achieved at 75 °C. For comparative experiments, the PAAM hydrogel sample was fabricated without MXene, PEDOT:PET, and PAA. Moreover, the PAAM/PAA hydrogel was produced without MXene and PEDOT.

The MXene/PDMS prepolymer consisted of the precursor, curing agents, MXene, and cyclohexane at a mass ratio of 1:1000:200:60, and it was fabricated by mixing. The MXene/PDMS prepolymer was then poured into a heated at 85 °C for 25 min to produce MXene/PDMS films. Finally, the MXene/PDMS film’s thickness was controlled at 1.5 mm.

The PPMP conductive hydrogel plays the role of a sensing component. Meanwhile, the MXene/PDMS film serves as the substrate and triboelectric layer. Copper wire was used in the electrodes and connected to both ends of the PPMP conductive hydrogel. To produce a sandwich-structured strain sensor, the PPMP conductive hydrogel was placed between two films composed of MXene/PDMS. In detail, the PPMP conductive hydrogel was installed between two MXene/PDMS films to serve as the strain sensor. Moreover, the FS-TENG device consists of three layers: a top MXene/PDMS layer, a middle PPMP conductive hydrogel, and a bottom MXene/PDMS layer. The PPMP conductive hydrogel and external load were connected by a copper wire.

In order to examine the morphology of PPMP conductive hydrogels, completely swollen cylinders of PPMP hydrogels were first frozen, cryonic broken into minute, thin strips in nitrogen gas, and finally vacuum freeze-dried at 45 °C for 96 h. The morphologies were analyzed with a scanning electron microscope and an energy-dispersive x-ray spectrometer (EDX). Using a spectrometer and a method known as attenuated total reflection-Fourier transform infrared (ATR-FTIR) spectroscopy, analyses of the chemical structures of freeze-dried PPMP hydrogels were carried out. The MCT-2150 tensile tester was used to measure the device’s mechanical property. Moreover, the electrical characteristics of the PPMP conductive hydrogel were measured using the Keithley-6514, including the V_oc and I_sc of the FS-TENG device. The electroconductivity of the PPMP conductive hydrogel was determined using the formula σ = 1/ρ = L/(RS), where ρ, L, R, and S stand for the resistivity, length, resistance, and the cross-sectional area of the PPMP conductive hydrogel, respectively.

The preparation process of a PPMP conductive hydrogel is depicted in Fig. 1(a), which illustrates the self-repair mechanism of PPMP conductive hydrogels. Furthermore, we marked the PAAM, PAAM/PAA, and PAAM/PAA/MXene/PEDOT:PET as M, MA, and PPMP, respectively. The PPMP conductive hydrogel was fabricated by solution polymerization method, where MXene and PEDOT:PET as conductive materials were added on the cross cooperative network. In addition, the addition of PAA can serve as the basic framework and help to uniformly disperse MXene and PEDOT:PET, which will enhance the flexibility and conductivity of the PPMP hydrogel. After first dissolving acrylamide (AAM) in de-ionized water, polyacrylamide and the chemical cross-linking agent N, N′-methylene bisacrylamide (MBA) were added to the mixture in order to produce a homogeneous solution. The combination was then subjected to continuous agitation for a period of 3 h after the addition of the MXene and PEDOT: PET suspensions. Furthermore, adding ammonium persulfate (APS) to the mixture and then heating it to 72 °C to get the hybrid double-network conductive hydrogel. This hydrogel was composed of PAAM and PAA, in addition to MXene and PEDOT: PET chains that were randomly distributed throughout its structure. Figure 1(b) illustrates the scanning electron microscope (SEM) image of a dried PPMP conductive hydrogel. As shown in Figs. 1(c)–1(f), the energy-dispersive x-ray spectroscopy (EDS) mapping pictures demonstrate a homogeneous distribution in the PPMP conductive hydrogel, such as that of C, O, and S elements. Figure 1(g) illustrates the picture of PPMP conductive hydrogel, and a coin is used to compare dimensions. Hydrogel samples of M, MA, and PPMP were subjected to attenuated total reflection-Fourier transform infrared spectroscopy, as illustrated in Fig. S1 of the supplementary material.

To develop the PPMP hydrogel’s mechanical characteristics, we compared the tensile strengths of three conductive hydrogels. As shown in Fig. 2(a), the MA and PPMP hydrogels indicate better strength owing to the hybrid double networks structure. In addition, it is noteworthy that PPMP conductive hydrogels possess higher breaking strength compared with MA hydrogels. However, since both MXene and PEDOT:PET are rigid, this will lead to a decrease in the elongation at break of PPMP conductive hydrogels. From the results, the fracture strength of PPMP conductive hydrogels is found to reach 180 kPa and the fracture elongation of PPMP conductive hydrogels is found to reach 500%. In addition, we also developed the compressive performance of PPMP conductive hydrogels. Figure 2(b) shows that the compressive stress of MA and PPMP hydrogels is higher than that of M hydrogels, indicating that they are more resistant to compression. Furthermore, the PEDOT: PET, and MXene combination considerably raises the compressive stress of PPMP hydrogels in comparison to MA hydrogels when the compressive strain is greater than 91%. Furthermore, continuous loading cycles under 80% compression strain were also used to examine the compressive behavior of the PPMP hydrogel, as shown in Fig. 2(c). Only after the first loading cycle is the recovery rate able to reach 82%. However, in the following nine loading cycles, the recovery rate almost remains unchanged, indicating that PPMP conductive hydrogels have a stable network structure. The tensile properties of PPMP conductive hydrogels are demonstrated in Fig. 2(d). The PPMP hydrogels’ reliable self-adhesive property enables good adherence to a wide variety of substrates without needing additional tape, which simplifies signal detecting processes and stabilizes the signal. To develop the adhesion characteristics of PPMP hydrogels, we design a lap shear test to develop the PPMP hydrogel adhesive performance on various materials’ surface, as shown in Figs. 2(e)–2(h). Additionally, the viscosity of the proposed PPMP hydrogel is reversible, and the PPMP hydrogel demonstrates persistent viscosity over five cycles, as illustrated in Fig. 2(g). The self-healing hydrogels are able to resist a certain amount of deformation and cure themselves after being damaged, making them excellent for wearable applications since they can repair themselves after being damaged. Tensile testing can be used to quantitatively examine the relationship between self-healing time and healing qualities, as shown in Figs. 2(i) and 2(j). From the results, the effectiveness of the healing was considerably increased in proportion to the lengthened duration of the process. When the time allowed for healing was increased from 6 to 12 h, the results were as expected: The stress/strain self-healing efficiency could reach 77% and 87%, respectively. On the other hand, there was not any substantial increase in healing time when compared to 15 h. The PPMP hydrogel was cut in half and reattached to itself without any outside stimulation to test its self-healing abilities.

The relative resistance variation (ΔR/R₀) is usually used to evaluate the resistance change trend of the sensor under different strains. This change trend is not affected by the resistance value of the resistance and is an indicator for evaluating the sensor performance. Initially, the ΔR/R₀ of the PPMP hydrogel against the applied stresses was examined, where ΔR and R₀ represent the resistance variation and the initial resistance, respectively. As shown in Fig. 3(a), the response time of the PPMP hydrogel was around 0.52 and 0.84 s. The ability to react quickly is useful for capturing precise data on human mobility in real time. Due to the flexibility of human movement, low and high strain detection in hydrogel-based sensors is essential. Figures 3(b) and 3(c) depict the ΔR/R₀–time curve of the PPMP hydrogel during the modest tensile stresses (6%–64%) and large tensile strains (100%–300%). The PPMP hydrogel has excellent repeatability and strain sensing stability over a large range of tensile strain. As shown in Fig. 3(d), the PPMP hydrogel possesses excellent compatibility and recoverability.

Human posture evaluation is the guiding basis in the field of dance teaching. Artificial intelligence coaches can correct students’ actions through human action recognition in sports and dance, guide wrong actions, and help avoid sports injuries. However, some professional detection equipment is often expensive, so flexible sensor-based human posture detection method is the key to solve this problem. Therefore, we develop a PPMP hydrogel-based sensor, as shown in Fig. 4(a). The PPMP hydrogel was encapsulated between two pieces of stretchy MXene/PDMS films to create the dance posture monitoring sensor (size: 1 × 3 cm²) to detect delicate human dance postures. The MXene/PDMS film serves as the triboelectric layer, which can obtain electrons from the skin’s surface. The PPMP hydrogel-based sensors require direct contact with the human body for sensitive monitoring. Hence, tape is employed to adhere the sensor to the appropriate body component. When the human body moves, the MXene/PDMS film rubs against the skin, producing triboelectric charges, which generate electrical sensing signals through the hydrogel electrode. As shown in Fig. 4(c), PPMP hydrogel-based sensors are sensitive enough to pick up on facial expression changes. Facial expressions, including smiling, frowning, and blinking, are the primary means of expression and communication in dance performance. In detail, an adhesive hydrogel sensor placed on the forehead may detect the consistent and repeatable resistance signals caused by transitions between neutral and frowning facial expressions. Moreover, resistance signals can be used to track the little twitches of the muscles that accompany microexpressions. Figure 4(d) shows the change in resistance signal of PPMP conductive hydrogel-based strain sensor by sticking the PPMP conductive hydrogel-based strain sensor to the human shoulder under the arm swing angle ranging from 30° to 90°. As depicted in Fig. 4(e), facial expressions can be detected by connecting the PPMP hydrogel strain sensor installed on human cheek. Moreover, the PPMP conductive hydrogel-based sensors can also help detect the bending angles of wrist joint as shown in Fig. 4(f).

As shown in Fig. 5(a), the PPMP hydrogel-based strain sensor can also be built as the FS-TENG device, where stretchable MXene/PDMS film serves as the triboelectric layer. The operation mechanism of the FS-TENG device is depicted in Fig. 5(b). The separated triboelectric pair is in equilibrium at the initial state [Fig. 5(b1)]. When the nylon film contacts the MXene/PDMS layer, the surfaces of the two triboelectric materials will generate an equal amount of triboelectric charges [Fig. 5(b2)]. When the nylon membrane is separated from the MXene/PDMS membrane, the positive charge in the PPMP conductive hydrogel electrode will be driven to the MXene/PDMS membrane interface. This will help balance the mismatched negative charges. Then, due to electrostatic induction, electrons move from the hydrogel electrode to the nylon film, and a current is produced [Fig. 5(b3)]. When the MXene/PDMS sheet and nylon film are completely separated to a certain distance, the system will be in static equilibrium and no current will be generated in the circuit [Fig. 5(b4)]. When the nylon film is close to the MXene/PDMS film again, the electrons in the circuit will move in the opposite direction to balance the potential difference generated on the surface of the triboelectric material [Fig. 5(b5)]. From the results in Figs. 5(c) and 5(d), the FS-TENG can obtain an open-circuit voltage (V_oc) of 169.2 V and a short-circuit current (I_sc) of 9.6 µA. As depicted in Fig. 5(e), the V_oc increased as the uniaxial strain increased, which may be attributed to increased contact area between the MXene/PDMS film and nylon film, which suggests that the stretchability of FS-TENG is beneficial to enhancing its energy-harvesting capacity. Figure S2 shows the reliability of the FS-TENG device.

To test the efficacy of FS-TENG as a self-powered system, we connected it to a capacitor and a thermohygrometer over the size of 2 × 4 cm². As shown in Fig. 6(a), FS-TENG could create electricity by being pushed regularly by a nylon film at a force of around 4 Hz, with the power being stored in a capacitor and the direction of the FS-TENG current being controlled by a rectifier bridge. As shown in Fig. 6(b), as the storage capacity of a commercial capacitor increases, the charging rate drops correspondingly. The capacitor’s increasing voltage indicated that the energy generated by FS-TENG can be stored effectively. Electricity stored in capacitors is not only more concentrated and powerful but also capable of being stored for longer, making it ideal for powering a wide range of electrical devices. The electric energy generated by FS-TENG can power a hygrothermograph through the charging/discharging circuit illustrated in Fig. 6(c). The capacitor could be charged to 7.2 V in 265 s, allowing it to power a hygrometer thermometer, as illustrated in Figs. 6(d) and 6(e). As long as the FS-TENG was held down, the capacitor was gradually charged back up to its full 7.2 V capacity. Based on these findings, the FS-TENG shows promise as a self-powered source for energy harvesting.

In summary, we designed a hybrid PPMP conductive hydrogel with various excellent features, including conductivity, extensibility, self-healing, and stability. Furthermore, the proposed PPMP conductive hydrogel with a MXene/PDMS encapsulation layer can play the role of a strain sensor and serve as a flexible stretchable FS-TENG. The PPMP conductive hydrogel-based strain sensor possesses excellent stability and endurance for about 1000 cycles and may be used as a dance posture sensor to provide motion information during dance training. Moreover, the FS-TENG (size: 2 × 4 cm²) used can generate an V_oc of 169 V and a I_sc of 9.6 µA. In addition to this, the FS-TENG is able to charge a wide range of capacitors and supply power to more compact electronic devices such as the commercial hygrothermograph. This research offers a feasible strategy for both dance posture sensing and human motion energy harvesting.

In the supplementary material, see Sec. I for a detailed characteristic peak analysis and Sec. II for more information on the reliability of the FS-TENG device.

This work was supported by the Social Science Foundation of Hunan Province (Grant No. 20JD053).

The authors have no conflicts to disclose.

Pin Qin: Conceptualization (lead); Data curation (lead); Formal analysis (lead); Funding acquisition (lead); Investigation (lead); Methodology (lead); Project administration (lead); Resources (lead); Software (lead); Supervision (lead); Validation (lead); Visualization (lead); Writing – original draft (lead); Writing – review & editing (lead).

The data that support the findings of this study are available within the article and its supplementary material.