A magnetorheological fluid based infinitely-regulatable triboelectric tactile sensor

Abstract

The trade-off between high sensitivity and wide detection range often limits the application of tactile sensors in complex scenarios. In this work, an adaptive triboelectric pressure sensor (MRF-TPS) is proposed, which offers highly adjustable sensing performance by introducing a magnetorheological fluid as a liquid triboelectric material. Through the reversible phase transition of the magnetorheological fluid, the MRF-TPS can quickly switch between liquid–solid and solid–solid modes. Owing to its exceptional microstructure design and efficient liquid–solid contact electrification, the MRF-TPS achieves an ultrahigh sensitivity of 46.47 kPa⁻¹ in liquid–solid mode. In the solid–solid mode controlled by a magnetic field, the great compressive resistance of the magnetorheological fluid in solid form remarkably extends the detection limit to 362 kPa. Finally, the MRF-TPS is integrated with a visual recognition module to develop an adaptive pressure-sensing system, demonstrating the broad application potential of the MRF-TPS in human–machine interactions and diversified pressure detection.

---

1. Introduction

The rapid advancement of sensing technology and artificial intelligence is propelling society into the era of the Internet of Things (IoT).^1–4 As a prominent topic in the field of sensing, pressure sensors are widely used in machinery manufacturing,^5–7 healthcare^8–12 and human–machine interaction,^13–16 making them highly convenient for industrial development and daily life. Pressure sensors based on various working principles, such as piezoresistive,^17–20 capacitive,^21–24 piezoelectric^25–28 and triboelectric^29–33 sensors, have been developed and applied in different fields. Owing to the coupling effect of contact electrification (CE) and electrostatic induction, triboelectric pressure sensors are able to achieve efficient pressure sensing, with advantages such as high sensitivity and a wide material selection range.^34–36 Sensitivity and detection range, regarded as the two most important parameters of pressure sensors, have widely attracted significant attention from researchers. A key method to improve sensitivity is microstructure surface engineering. Many studies have successfully constructed microstructures on the surface of the triboelectric layer through techniques such as laser scanning,^37–39 lithography^40–42 and 3D printing.^43–45 However, the pursuit of high sensitivity often limits the pressure detection range, making it challenging to achieve both high sensitivity and a wide detection range simultaneously.

The inherent immutability of the CE interface is the main factor for this trade-off problem, which hinders further application of triboelectric pressure sensors. The development of smart materials offers a potential solution, as these materials can regulate their physical and chemical properties in response to external stimuli. For example, a capacitive pressure sensor based on liquid metal has been developed which switches between soft and rigid modes by controlling the phase transition of gallium with respect to temperature change.⁴⁶ However, this temperature-based phase transition is slow and cannot meet real-time response requirements. In addition, a ferrofluid-based triboelectric tactile sensor has been demonstrated, which can regulate the pressure performance through magnets.⁴⁷ Therefore, introducing rheological materials as triboelectric materials presents an opportunity to improve the comprehensive performance of pressure sensors by adjusting the contact electrification interface with an external field.

In this work, we propose a magnetorheological fluid based triboelectric pressure sensor (MRF-TPS) with highly adjustable pressure sensing capabilities. The MRF-TPS can switch between liquid–solid and solid–solid contact modes within milliseconds by real-time control of the morphology of the magnetorheological fluid via magnetic fields. Owing to the micrometer-level microstructure, low Young's modulus of the magnetorheological fluid and efficient charge transfer under liquid–solid CE, the MRF-TPS achieves an ultrahigh pressure sensitivity of 46.47 kPa⁻¹ without a magnetic field. Through the control of the magnetic field, the excellent anti-compression performance of the magnetorheological fluid in solid form greatly extends the detection range of the device to 362 kPa. Finally, we integrate the MRF-TPS with visual sensors to develop an adaptive pressure sensing system. It can adjust and effectively perceive various application scenarios, reflecting the advantages of multimode adjustable pressure sensing and demonstrating the broad application prospects of our device in human–machine interaction and IoT sensing.

2. Results and discussion

2.1 Overview of the MRF-TPS and characterization of the magnetorheological fluid

Fig. 1a shows the overall concept and structure of the MRF-TPS. By introducing magnetorheological fluid as a phase-change triboelectric material, an adaptive triboelectric pressure sensor is prepared, which can respond to diverse sensing scenarios through magnetic field control. Its overall size is 15 × 15 × 8 mm³, with a 10 × 10 × 3 mm³ magnetorheological fluid layer inside. The schematic diagram in Fig. 1a illustrates the particle arrangement inside the magnetorheological fluid, showing the rapid switching between liquid–solid and solid–solid modes under the influence of a magnetic field. In the absence of a magnetic field, the magnetic particles in the magnetorheological fluid are randomly distributed in the carrier liquid, giving a fluid-like appearance (left part of Fig. 1a). When a magnetic field is applied, the randomly distributed magnetic particles are polarized, forming dipoles that align into chain-like structures along the direction of the magnetic induction lines. As the magnetic field intensity increases, the strength of the chain arrangement also increases, transforming the fluid into a solid (right part of Fig. 1a).

Fig. 1 Overall concept of the MRF-TPS and characterization of the magnetorheological fluid. (a) Structure of the MRF-TPS and the mechanism of mode switching, illustrated by the arrangement of magnetic particles inside the magnetorheological fluid. (b) Hysteresis loop of the magnetorheological fluid. (c) Relationship between the yield stress and magnetic field intensity of the magnetorheological fluid. (d) Viscosity of the magnetorheological fluid under zero magnetic field, with the inset showing the relationship between the shear stress and shear rate.

The performance of triboelectric pressure sensors is heavily affected by the physical and chemical properties and surface roughness of materials. In this case, the magnetorheological fluid based on carbonyl iron powder is chosen as the triboelectric material. Owing to its reversible liquid–solid phase transition behavior controlled by a magnetic field, the MRF-TPS can switch between several modes within milliseconds. In the liquid–solid mode (without a magnetic field), the MRF-TPS achieves a high sensitivity due to the efficient charge transfer under liquid–solid contact electrification. When a magnetic field is applied, the stability of the MRF-TPS is significantly improved by the prominent yield behavior of the magnetorheological fluid in its solid form, thereby effectively expanding the detection range of pressure sensing and making the device more adaptable to various applications.

Fig. 1b shows the hysteresis loop of this magnetorheological fluid, indicating its superparamagnetism with almost zero residual magnetism and a saturation magnetization intensity of 1.76 T. The relationship between the yield stress and magnetic field intensity of the magnetorheological fluid is shown in Fig. 1c. The yield stress increases with the increasing magnetic field, presenting a good linear relationship (k represents the slope of the straight line, and R² represents the coefficient of determination). Fig. 1d shows the rheological behavior of the magnetorheological fluid in the absence of a magnetic field, revealing the relationship between the shear rate and fluid viscosity. The inset shows the curves of the shear rate and shear stress, both of which indicate that the magnetorheological fluid conforms to the characteristics of a Bingham fluid.

2.2 Operational principle of the MRF-TPS and performance improvement through surface microstructure engineering

Fig. 2a explains the operational principle and performance enhancement of the MRF-TPS through surface microstructure engineering. The cross-sectional schematic diagram of the MRF-TPS in Fig. 2a shows the use of silicone rubber and magnetorheological fluid as triboelectric layers. The working mechanism of this device, as illustrated in Fig. 2b, operates in single electrode mode. Due to the higher electronegativity of silicone rubber, it easily captures electrons from the magnetorheological fluid during contact under pressure (Fig. 2b-i), leading to the generation of triboelectric charges at the interface (Fig. 2b-ii). As the pressure is released, the separation of the two materials causes an induced potential to develop across the bottom electrode, producing a measurable electrical signal (Fig. 2b-iii). When the pressure is applied again, the open-circuit voltage gradually decreases to zero (Fig. 2b-iv). The above mechanism involves the coupling of contact electrification and the electrostatic induction effect, which drives the MRF-TPS to convert mechanical energy into electrical signals and achieve pressure sensing. During this process, the CE between the silicone rubber and magnetorheological fluid plays a decisive role in the sensing performance of the MRF-TPS.

Fig. 2 Operational principle of the MRF-TPS and performance improvement through surface microstructure engineering. (a) Cross-sectional schematic diagram of the MRF-TPS. (b) Working mechanism of the MRF-TPS under cyclic pressure. (c) Optical images of the microstructure on silicone rubber, with the insets showing the results of the water contact angle tests. (d) Schematic diagram of the microstructure fabrication process. Relationships and linear fits between the relative variations in voltage and pressure of (e) 100#, (f) 300# and (g) 600# MRF-TPSs. (h) The summarized variation in sensitivity in the low-pressure region of different MRF-TPSs.

Silicone rubber is selected as the solid triboelectric material because of its rapid solidification and low elastic modulus. To further improve performance, various cubic microstructures are constructed on the surface of the silicone rubber using a simple template imprinting method, as shown in Fig. 2c and d. Three types of wire mesh (100#, 300#, and 600#) are used to create varying microstructure sizes, each of which is confirmed via optical microscopy. The corresponding square aperture side lengths are 150 μm, 50 μm and 25 μm (Fig. S1†). This proves that the microstructure of the wire mesh can be effectively transferred to the surface of the silicone rubber through this method, forming an array of cubic microstructures. The water contact angle of raw silicone rubber is 107° (Fig. S2†). The insets in Fig. 2c show the water contact angles of 100#, 300# and 600# silicone rubber, which are 115°, 126° and 139°, respectively. Therefore, the microstructure can prominently increase the hydrophobicity of silicone rubber. In addition, it can also improve surface roughness and significantly enhance the CE between interfaces. Subsequently, the influence of the microstructure on improving the pressure sensing sensitivity is explored. Fig. 2e–g present the relationships between the relative voltage changes (V − V₀)/V₀ and the pressures of the three MRF-TPSs, 100#, 300# and 600#, where the slope of the fitted curve represents the pressure sensing sensitivity. The sensitivity of the MRF-TPS can be divided into two regions.

In the low-pressure region, owing to the dominant increase in the contact area and the rapid reduction in the interface distance, the device possesses high sensitivity. Whereas in the high-pressure region, microstructure deformation reaches its limit, leading to saturation of the contact area and then a reduction in sensitivity. Fig. 2h summarizes the sensitivity of the MRF-TPS without a microstructure (Fig. S3†) and with a microstructure in the low-pressure region. As the microstructure size decreases, the sensitivity of the MRF-TPS obviously improves. The sensitivity in the high-pressure region is similar (Fig. S4†). Notably, the 600# MRF-TPS has the best performance with an ultrahigh sensitivity of 46.47 kPa⁻¹ in the pressure range of 0–0.28 kPa. It still retains a sensitivity of 1.42 kPa⁻¹ in the range of 0.28–22.38 kPa. Compared to the MRF-TPS without a microstructure, the sensitivity is increased by 1721%. These experimental results are consistent with other studies.^48–50 The synergistic effect of the 25 μm microstructure, efficient charge transfer at the liquid–solid interface and low modulus of the magnetorheological fluid in the liquid state contribute to ultrahigh pressure sensing sensitivity. Therefore, subsequent experiments are conducted to study the relevant performance of the 600# MRF-TPS, which is abbreviated as MRF-TPS.

2.3 Characterization of the basic sensing performance of the MRF-TPS

The magnetorheological fluid exhibits a Bingham fluid-like behavior, rapidly transforming into a solid state under the effect of a magnetic field, as shown in Fig. S5.† Therefore, the presence or absence of a magnetic field determines whether the MRF-TPS operates based on liquid–solid or solid–solid CE. Fig. 3a compares electron transfer models of different interfaces to demonstrate the CE phenomenon between the magnetorheological fluid and silicone rubber. According to Wang's electron transfer theory, the overlapping electron clouds of surface atoms during interface contact between different materials create conditions for electron transfer. Without confinement by a magnetic field, the deep overlap of electron clouds between iron carbonyl compounds and organosilicon polymers at the liquid–solid interface reduces the interatomic potential barrier, facilitating significant electron transitions. In the solid–solid contact mode under a magnetic field, electron transfer is still the primary CE mechanism, but the higher interatomic potential barrier at the solid–solid interface partially inhibits electron transfer. Fig. 3b shows schematic diagrams of the CEs at different interfaces. Without a magnetic field, the liquid–solid CE between the magnetorheological fluid and silicone rubber results in a larger contact area. Upon applying a magnetic field, the microstructure deformation of the silicone rubber becomes more pronounced in the solid–solid CE. At a pressure of 30 kPa, a finite element simulation reveals that the shape of the MRF-TPS in the liquid–solid mode changes significantly, whereas the device in the solid–solid mode can withstand greater pressure, indicating a noticeable improvement in its modulus (Fig. 3c).

Fig. 3 Sensing performance of the MRF-TPS. (a) CE between different interfaces in the electron cloud model. (b) Schematic diagram of the CE between different interfaces. (c) Finite element analysis simulations of the MRF-TPS in the liquid–solid and solid–solid modes under a pressure of 30 kPa. (d) Output voltages of the MRF-TPS under various static pressures in the liquid–solid and solid–solid modes. (e) Response and recovery time of the MRF-TPS in the liquid–solid and solid–solid modes. (f) Periodic loading–unloading process of the MRF-TPS in the solid–solid mode for 2000 cycles.

The basic performance of the MRF-TPS is subsequently tested in both liquid–solid mode (H = 0 kA m⁻¹) and solid–solid mode (H = 144 kA m⁻¹). Fig. 3d illustrates the relationship between the open circuit voltage signal and static pressure of the MRF-TPS in both modes, showing a consistent increase in output voltage as the pressure increases from 0.25 kPa to 218.59 kPa. In addition, the MRF-TPS demonstrates excellent response and recovery time, reaching 153 ms and 136 ms in the liquid–solid mode, respectively (Fig. 3e). Benefiting from the thorough and rapid contact separation in the solid–solid mode, the response time improves to 64 ms, and the recovery time is 97 ms. The repeatability and reliability of the MRF-TPS are verified for 2000 load-unload cycles, as shown in Fig. 3f. The device retained strong repeatability in both modes (Fig. S6†).

2.4 Influence of the magnetic field on the pressure sensing performance of the MRF-TPS

The effect of the magnetic field intensity on the sensing performance of the MRF-TPS is further studied under magnetic fields of 0, 48, 96 and 144 kA m⁻¹. The relationship between the relative voltage change (V − V₀)/V₀ of the MRF-TPS and pressure under the four magnetic fields is shown in Fig. 4a. The slope of the fitted curve represents the pressure sensing sensitivity of each region. Similarly, the sensitivity under each magnetic field can be divided into two regions. The data for 100# and 300# MRF-TPSs are shown in Fig. S7 and S8,† respectively. Fig. 4b shows a comparison of the sensitivity under different magnetic fields in both regions. The inset shows the amplification of the high-pressure region. A sensitivity comparison of the 100# and 300# MRF-TPSs is shown in Fig. S9 and S10,† respectively. It shows that as the magnetic field intensity increases, the sensitivity of both regions decreases to varying degrees due to the change in the morphology of the magnetorheological fluid. Without a magnetic field, the magnetorheological fluid has a lower Young's modulus and deforms more under low pressure. In addition, the fluid morphology allows for a more thorough CE with silicone rubber. After being subjected to pressure, it can quickly immerse the cubic microstructure to generate a larger contact area, thereby achieving a significant increase in its charge transfer and an ultrahigh sensitivity of 46.47 kPa⁻¹. The high-pressure region also shows an overall downward trend, but the peaks can be observed at moderate magnetic field intensities (48 or 96 kA m⁻¹). Under stronger magnetic fields, the surface of the magnetorheological fluid develops spike-like microstructures, as shown in Fig. 4c. The inset shows an optical image of the microstructure. When the magnetic field is strong enough, the magnetic particles inside overcome gravity and surface tension to form protrusions along the direction of the magnetic induction line. It has a positive impact on the sensitivity improvement, whereas the liquid–solid phase transition of the magnetorheological fluid inhibits the improvement. The combined effect of these two factors leads to the phenomenon that the sensitivity in the high-pressure region first increases and then decreases with increasing magnetic field.

Fig. 4 Effect of the magnetic field on the pressure sensing performance of the MRF-TPS. (a) Relationships and linear fits between the relative variations in the voltage and pressure of the 600# MRF-TPS under different magnetic fields. (b) Summarized variation of sensitivity in two regions under different magnetic fields. (c) Schematic diagram of the microstructure of the magnetorheological fluid, with the inset showing the optical image. (d) Summary of the variation in the detection range under different magnetic fields. (e) Stress–strain curves of the MRF-TPS compressed under different magnetic fields. (f) Young's modulus of the MRF-TPS under different magnetic fields. (g) Comparison of the sensitivity and detection range of the MRF-TPS with those of recent methods.^51–57

The detection range of the MRF-TPS is significantly extended with increasing magnetic field intensity, as shown in Fig. 4d. The detection limit increases from 22.4 kPa (without a magnetic field) to 362.1 kPa at 144 kA m⁻¹, a 1616% increase. This enhancement is attributed to changes in the physical properties of the magnetorheological fluid. As the magnetic field intensity increases, the magnetorheological fluid gradually transforms from a liquid to a solid. Moreover, its Young's modulus and yield stress continue to increase, enhancing the stability of the device structure. The stress–strain curves of the MRF-TPSs compressed under different magnetic fields are also tested, as shown in Fig. 4e. The initial overlap of the curves is mainly caused by the cavity inside the device. As the intensity increases, the yield stress generated also increases when MRF-TPS is compressed. Additionally, as shown in Fig. 4f, the overall Young's modulus of the MRF-TPS is calculated and compared through stress–strain curve fitting. The results indicate that the magnetic field can effectively improve the overall modulus of the device. These factors work together to remarkably extend the pressure detection range of the device.

To more intuitively demonstrate the advantage of adjustable performance, we set the maximum detection range of pressure sensing as the x-axis and the highest sensitivity as the y-axis, comparing the MRF-TPS with several related studies in Fig. 4g.^51–57 The comparison shows that other systems excel either in sensitivity or detection range but not both. However the MRF-TPS benefits from magnetic field regulation, allowing it to switch quickly between multiple modes. The liquid–solid mode achieves a sensitivity of 46.47 kPa⁻¹ (0–0.28 kPa), while the solid–solid mode allows for a detection range of up to 362.1 kPa (0.22 kPa⁻¹).

2.5 Application of the MRF-TPS as an adaptive pressure sensing system

We demonstrated the application of the MRF-TPS as an adaptive pressure-sensing system, assisted by visual recognition, taking advantage of highly adjustable sensing performance of the MRF-TPS under magnetic field control. Fig. 5a outlines the flow chart of the visual-recognition-assisted adaptive pressure sensing system based on the MRF-TPS. It consists of two interconnected parts: robot vision and pressure perception. In the robot vision part, feature learning is used to endow the system with the ability to recognize and distinguish objects. In the subsequent pressure perception part, the mode of the MRF-TPS is controlled on the basis of customized classification to achieve adaptive perception in diverse scenarios. Fig. 5b shows the workflow of the entire system, including the visual recognition module K210 and the pressure perception module MRF-TPS, where the magnetic field is generated through an energized flexible coil based on liquid metal. For example, by defining hands as class 1 and pills as class 2, K210 can classify objects on the basis of their features after training. When the object is identified as class 1, the MRF-TPS adjusts to the solid–solid mode. If the object is identified as class 2, the MRF-TPS remains in the liquid–solid mode.

Fig. 5 Demonstration of the MRF-TPS as an adaptive pressure sensing system. (a) Flow chart of the adaptive pressure sensing system assisted by visual recognition. (b) Schematic diagram of the adaptive pressure sensing system. (c) Application of the MRF-TPS in the liquid–solid and solid–solid modes.

Fig. 5c shows the representative applications of our adaptive pressure sensing system in different scenarios.

In the solid–solid mode, the overall modulus of the device increases, and some common scenes in human–machine interaction are simulated via a hand model equipped with the MRF-TPS, such as shaking, poking and beating actions. Each action generates distinct waveforms and voltage peaks, allowing the system to recognize different interactions based on the voltage characteristics. The shaking action is relatively slow, and the low-frequency voltage peak has a longer holding time. The poking action has a faster response time, whereas the voltage signal of the beating action is represented as continuous peaks with higher frequency and amplitude. Therefore, the recognition of human–machine interaction actions can be achieved through the characteristics of voltage peaks. In the liquid–solid mode, the device's low modulus and efficient liquid–solid CE allow it to detect tiny pressures, such as the weight of a pill (3 Pa) or continuous water droplets. These results showcase the powerful adaptive capabilities of the MRF-TPS in diverse sensing environments, highlighting its broad application potential.

3. Conclusion

We have developed a highly adaptive triboelectric pressure sensor (MRF-TPS) based on magnetorheological fluid, which enables tunable pressure sensing performance through magnetic field regulation. An efficient microstructure manufacturing method is adopted to imprint regular microstructure arrays on silicone rubber using wire mesh. By addressing the trade-off between high sensitivity and a wide detection range, the MRF-TPS achieves ultrahigh sensitivity (46.47 kPa⁻¹) in liquid–solid mode and a detection range of 362 kPa in solid–solid mode. Furthermore, by integrating the MRF-TPS with the visual recognition module, an adaptive pressure sensing system assisted by visual recognition is developed, which achieves adaptive pressure sensing for complex scenarios, demonstrating the broad potential of the MRF-TPS for human–machine interaction and IoT pressure detection applications.

4. Experimental section

4.1 Materials

Magnetorheological fluid was purchased from Bohai Technology Co., Ltd. Silicone rubber (Ecoflex 00-50, Smooth-On), Galinstan (Geratherm Medical) and wire mesh were purchased from authorized resellers. The cylindrical magnets comply with the N35 standard, with a radius of 1 cm and a height of 0.4 cm. K210 was purchased from Yabo Intelligence. All reagents were used as received without further processing.

4.2 Fabrication of the MRF-TPS

The organic silicone rubber shell and triboelectric layer were formed by mixing Ecoflex 00-50 A and B proportionally and curing them at room temperature in acrylic templates. Wire mesh of different specifications was fixed in the center of the template to imprint microstructures with different parameters on the surface of the silicone rubber. Acrylic templates were made from acrylic sheets with a laser cutting machine and acrylic adhesive. Copper foil was adhered to the interior of the silicone rubber chamber and served as the bottom electrode. Subsequently, the magnetorheological fluid was evenly dropped onto the surface of the copper electrode as a positive liquid friction layer. Finally, a specialized adhesive was used to tightly bond the silicone rubber triboelectric layer with the outer shell and seal it as a whole to fabricate the MRF-TPS. The magnetic field that regulates the performance of the MRF-TPS was provided by magnets and energized flexible liquid metal coils. The coil was made by filling a homemade silicone rubber template with a gallium indium tin alloy.

4.3 Characterization and measurement

The surface microstructure morphology of the silicone rubber and wire mesh was characterized and photographed via optical microscopy (Leica). The contact angle of the silicone rubber triboelectric layer was measured with a contact angle measuring instrument (OCA11). The output performance of the MRF-TPS was tested via a programmable electrometer (Keithley 6514), and real-time collection of electrical signal data was achieved via LabView 2013. A vertical stepper motor (HC14-10) was used to control the stroke. Varying vertical pressure was applied and the pressure value was measured using a pressure gauge (DS2-2000 N-XD).

Data availability

All relevant data are within the manuscript and its additional files.

Author contributions

Xin Chong: investigation, formal analysis, methodology, visualization, writing-original draft, writing-review & editing. Zhenqiu Gao, Zifan Jiang, Ao Wang, Jia Shia and Lanyue Shen: investigation, visualization and writing-review & editing. Zhen Wen and Xuhui Sun: conceptualization, methodology, supervision, funding acquisition, writing-review & editing.

Conflicts of interest

The authors declare no conflicts of interest. They have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was supported by the National Key R&D Program of China (No. 2023YFB3208100), the National Natural Science Foundation of China (No. 62174115 and No. U21A20147), Natural Science Foundation of Jiangsu Province of China (No. BK20240152), the Natural Science Foundation of the Jiangsu Higher Education Institutions of China (No. 24KJA430011), the Collaborative Innovation Center of Suzhou Nano Science & Technology, the 111 Project, and the Joint International Research Laboratory of Carbon-Based Functional Materials and Devices.

References

1. Q. Mao, Z. Liao, J. Yuan and R. Zhu, Nat. Commun., 2024, 15, 6871–6883.
2. X. Yang, W. Chen, Q. Fan, J. Chen, Y. Chen, F. Lai and H. Liu, Adv. Mater., 2024, 36, 2402542–2402569.
3. H. C. Ates, P. Q. Nguyen, L. Gonzalez-Macia, E. Morales-Narvaez, F. Guder, J. J. Collins and C. Dincer, Nat. Rev. Mater., 2022, 7, 887–907.
4. H. Niu, H. Li, S. Gao, Y. Li, X. Wei, Y. Chen, W. Yue, W. Zhou and G. Shen, Adv. Mater., 2022, 34, 2202622–2202632.
5. Y. Gao, H. Li, S. Chao, Y. Wang, L. Hou, T. Bai, J. Bai, X. Man, Z. Cui, N. Wang, Z. Li and Y. Zhao, ACS Nano, 2024, 18, 16958–16966.
6. S. H. Shin, S. Ji, S. Choi, K. H. Pyo, B. Wan An, J. Park, J. Kim, J. Y. Kim, K. S. Lee, S. Y. Kwon, J. Heo, B. G. Park and J. U. Park, Nat. Commun., 2017, 8, 14950–14957.
7. D. Yang, H. K. Nam, T. D. Le, J. Yeo, Y. Lee, Y. R. Kim, S. W. Kim, H. J. Choi, H. C. Shim, S. Ryu, S. Kwon and Y. J. Kim, ACS Nano, 2023, 17, 18893–18904.
8. K. Meng, X. Xiao, W. Wei, G. Chen, A. Nashalian, S. Shen, X. Xiao and J. Chen, Adv. Mater., 2022, 34, 2109357–2109379.
9. M. Chao, L. He, M. Gong, N. Li, X. Li, L. Peng, F. Shi, L. Zhang and P. Wan, ACS Nano, 2021, 15, 9746–9758.
10. Y. Fang, Y. Zou, J. Xu, G. Chen, Y. Zhou, W. Deng, X. Zhao, M. Roustaei, T. K. Hsiai and J. Chen, Adv. Mater., 2021, 33, 2104178–2104185.
11. C. Ning, K. Dong, R. Cheng, J. Yi, C. Ye, X. Peng, F. Sheng, Y. Jiang and Z. L. Wang, Adv. Funct. Mater., 2020, 31, 2006679–2006687.
12. Z. Song, J. Yin, Z. Wang, C. Lu, Z. Yang, Z. Zhao, Z. Lin, J. Wang, C. Wu, J. Cheng, Y. Dai, Y. Zi, S. Huang, X. Chen, J. Song, G. Li and W. Ding, Nano Energy, 2022, 93, 106798–106810.
13. S. Li, H. Yu, G. Pan, H. Tang, J. Zhang, L. Ye, X. Zhang and W. Ding, IEEE Trans. Robot., 2024, 40, 4506–4525.
14. S. Li, Z. Wang, C. Wu, X. Li, S. Luo, B. Fang, F. Sun, X. Zhang and W. Ding, IEEE J. Sel. Top. Signal Process., 2024, 18, 267–287.
15. M. Zhu, Z. Sun, Z. Zhang, Q. Shi, T. He, H. Liu, T. Chen and C. Lee, Sci. Adv., 2020, 8693–8706.
16. F. Wen, Z. Sun, T. He, Q. Shi, M. Zhu, Z. Zhang, L. Li, T. Zhang and C. Lee, Adv. Sci., 2020, 7, 2000261–2000275.
17. W. Chen and X. Yan, J. Mater. Sci. Technol., 2020, 43, 175–188.
18. Q. Zheng, J.-h. Lee, X. Shen, X. Chen and J.-K. Kim, Mater. Today, 2020, 36, 158–179.
19. Y. Gao, C. Yan, H. Huang, T. Yang, G. Tian, D. Xiong, N. Chen, X. Chu, S. Zhong, W. Deng, Y. Fang and W. Yang, Adv. Funct. Mater., 2020, 30, 1909603–1909610.
20. H. Liu, X. Chen, Y. Zheng, D. Zhang, Y. Zhao, C. Wang, C. Pan, C. Liu and C. Shen, Adv. Funct. Mater., 2021, 31, 2008006–2008017.
21. Y. Xiong, Y. Shen, L. Tian, Y. Hu, P. Zhu, R. Sun and C.-P. Wong, Nano Energy, 2020, 70, 104436–104446.
22. S. R. A. Ruth, L. Beker, H. Tran, V. R. Feig, N. Matsuhisa and Z. Bao, Adv. Funct. Mater., 2019, 30, 1903100–1903111.
23. J. Qin, L. J. Yin, Y. N. Hao, S. L. Zhong, D. L. Zhang, K. Bi, Y. X. Zhang, Y. Zhao and Z. M. Dang, Adv. Mater., 2021, 33, 2008267–2008297.
24. K. H. Ha, W. Zhang, H. Jang, S. Kang, L. Wang, P. Tan, H. Hwang and N. Lu, Adv. Mater., 2021, 33, 2103320–2103331.
25. Z. Yang, S. Zhou, J. Zu and D. Inman, Joule, 2018, 2, 642–697.
26. C. Dagdeviren, Y. Su, P. Joe, R. Yona, Y. Liu, Y. S. Kim, Y. Huang, A. R. Damadoran, J. Xia, L. W. Martin, Y. Huang and J. A. Rogers, Nat. Commun., 2014, 5, 4496–4505.
27. S. Min, D. H. Kim, D. J. Joe, B. W. Kim, Y. H. Jung, J. H. Lee, B. Y. Lee, I. Doh, J. An, Y. N. Youn, B. Joung, C. D. Yoo, H. S. Ahn and K. J. Lee, Adv. Mater., 2023, 35, 2301627–2301637.
28. Z. Yi, Z. Liu, W. Li, T. Ruan, X. Chen, J. Liu, B. Yang and W. Zhang, Adv. Mater., 2022, 34, 2110291–2110302.
29. K. Dong, X. Peng, J. An, A. C. Wang, J. Luo, B. Sun, J. Wang and Z. L. Wang, Nat. Commun., 2020, 11, 2868–2878.
30. C. Chen, L. Chen, Z. Wu, H. Guo, W. Yu, Z. Du and Z. L. Wang, Mater. Today, 2020, 32, 84–93.
31. Y.-W. Cai, X.-N. Zhang, G.-G. Wang, G.-Z. Li, D.-Q. Zhao, N. Sun, F. Li, H.-Y. Zhang, J.-C. Han and Y. Yang, Nano Energy, 2021, 81, 105663–105671.
32. C. Ning, R. Cheng, Y. Jiang, F. Sheng, J. Yi, S. Shen, Y. Zhang, X. Peng, K. Dong and Z. L. Wang, ACS Nano, 2022, 16, 2811–2821.
33. X. Qu, J. Xue, Y. Liu, W. Rao, Z. Liu and Z. Li, Nano Energy, 2022, 98, 107324–107332.
34. S. Wang, L. Lin and Z. L. Wang, Nano Energy, 2015, 11, 436–462.
35. Z. L. Wang, Rep. Prog. Phys., 2021, 84, 96502–96570.
36. X. Wang, Y. Zhang, X. Zhang, Z. Huo, X. Li, M. Que, Z. Peng, H. Wang and C. Pan, Adv. Mater., 2018, 30, 1706738–1706745.
37. Z. Yan, L. Wang, Y. Xia, R. Qiu, W. Liu, M. Wu, Y. Zhu, S. Zhu, C. Jia, M. Zhu, R. Cao, Z. Li and X. Wang, Adv. Funct. Mater., 2021, 31, 2100709–2100717.
38. S. Zhang, M. Chi, J. Mo, T. Liu, Y. Liu, Q. Fu, J. Wang, B. Luo, Y. Qin, S. Wang and S. Nie, Nat. Commun., 2022, 13, 4168–4177.
39. R. Wang, S. Gao, Z. Yang, Y. Li, W. Chen, B. Wu and W. Wu, Adv. Mater., 2018, 30, 1706267–1706274.
40. Y. Zhang, Z. Zhou, L. Sun, Z. Liu, X. Xia and T. H. Tao, Adv. Mater., 2018, 30, 1805722–18057230.
41. D. Kim, S.-B. Jeon, J. Y. Kim, M.-L. Seol, S. O. Kim and Y.-K. Choi, Nano Energy, 2015, 12, 331–338.
42. Q. Zhou, K. Lee, K. N. Kim, J. G. Park, J. Pan, J. Bae, J. M. Baik and T. Kim, Nano Energy, 2019, 57, 903–910.
43. P. Zhu, B. Zhang, H. Wang, Y. Wu, H. Cao, L. He, C. Li, X. Luo, X. Li and Y. Mao, Nano Res., 2022, 15, 7460–7467.
44. S. A. Basith and A. Chandrasekhar, Nano Energy, 2023, 108, 108183–108194.
45. B. Chen, W. Tang and Z. L. Wang, Mater. Today, 2021, 50, 224–238.
46. S. Lee, S. H. Byun, C. Y. Kim, S. Cho, S. Park, J. Y. Sim and J. W. Jeong, Adv. Mater., 2022, 34, 2204805–2204816.
47. J. Liu, Z. Wen, H. Lei, Z. Gao and X. Sun, Nanomicro Lett., 2022, 14, 88–98.
48. G. Yao, L. Xu, X. Cheng, Y. Li, X. Huang, W. Guo, S. Liu, Z. L. Wang and H. Wu, Adv. Funct. Mater., 2019, 30, 1907312–1907320.
49. Z. Zhao, L. Zhou, S. Li, D. Liu, Y. Li, Y. Gao, Y. Liu, Y. Dai, J. Wang and Z. L. Wang, Nat. Commun., 2021, 12, 4686–4693.
50. Y. E. Shin, Y. J. Park, S. K. Ghosh, Y. Lee, J. Park and H. Ko, Adv. Sci., 2022, 9, 2105423–2105433.
51. H. Wu, R. Luo, Z. Li, Y. Tian, J. Yuan, B. Su, K. Zhou, C. Yan and Y. Shi, Adv. Mater., 2024, 36, 2307546–2307560.
52. T. Wang, X. Shang, H. Wang, J. Wang and C. Zhang, Compos. Appl. Sci. Manuf., 2024, 181, 108163–108170.
53. M. Liu, X. Pu, C. Jiang, T. Liu, X. Huang, L. Chen, C. Du, J. Sun, W. Hu and Z. L. Wang, Adv. Mater., 2017, 29, 1703700–1703708.
54. B. Ji, Q. Zhou, B. Hu, J. Zhong, J. Zhou and B. Zhou, Adv. Mater., 2021, 33, 2100859–2100869.
55. S. J. Park, J. Kim, M. Chu and M. Khine, Adv. Mater. Technol., 2017, 3, 1700158–1700164.
56. Y. Mao, B. Ji, G. Chen, C. Hao, B. Zhou and Y. Tian, ACS Appl. Nano Mater., 2019, 2, 3196–3205.
57. J. Luo, F. R. Fan, T. Zhou, W. Tang, F. Xue and Z. L. Wang, Extreme Mech. Lett., 2019, 2, 28–36.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ta07129h

---