Contact-induced continuous electricity generation by ion-electron positive feedback coupled transport for self-powered ionic touch panel Authors: Kedong Shang, Jiahao Fang, Xiaobo Pu, Peng Wang, Yao Chen, Hai Liu, Ning Zhang, Junjie Hao, Yong Zhang, Bingjun Yu, Linmao Qian, Tingting Yang Published: 2026 ABSTRACT AbstractIonic touch panels are regarded as a key platform for future human-computer interaction and meta-universe due to their stretchable, transparent and skin-fitting properties. Inspired by the fact that human skin relies on ionic current to sense contact position information, we have investigated an ionogel based closed-loop electrical system that also converts contact into ionic current to form a self-powered single-layer ionic touch panel. Benefiting from the slowed charge transfer dynamics, the positive feedback coupling of the electrical double layer, and the high-density charge characteristics, the device generates an approximately steady-state electrical signal when touched. It is clearly different from the pulsed electrical phenomenon of conventional contact electrification devices. When a finger touches the touch panel, the voltage/current signal amplitude at each corner electrode of the ionogel has been proven to express the touch position. The continuity of the electrical signal ensures high-resolution recognition of the touch trajectory without the need for further contact separation. With the advantages of good transparency, large stretchability, self-power, single-layer structure, fast response and high resolution, we expect this emerging ionic touch panel to be an ideal candidate for a variety of human-computer interaction applications. Subject terms: Electronic properties and materials, Physics Introduction Integrated touch panels have evolved into various types, including resistive1–4, capacitive5–7, triboelectric8–11, surface acoustic wave12–14, and infrared-based systems15–17. Among these, ionogels have emerged as key materials for next-generation flexible touch panels due to their stretchability, high transparency, and skin-friendliness. The rapid advancement of ionic touch panel technology is expected to enable a wide range of applications, such as human-computer interaction11,18–20, autonomous driving20–22, and metaverse interfaces19,22–24. An ideal touch panel for these applications should possess high transparency, excellent stretchability, and the ability to continuously detect touch positions with high precision. Moreover, it is desirable for the panel to operate without an external power supply, thereby enhancing the flexibility, durability, and lightweight nature of the sensing system. However, both ionic resistive and ionic capacitive touch panels are limited by issues arising from external power supply dependence5,25,26. Ion triboelectric touch panels, while capable of energy harvesting, generate only transient pulsed electrical signals during contact-separation events, making them suitable solely for point-contact sensing and unsuitable for accurately tracking continuous sliding motions27–29. Furthermore, most existing ionic touch panels feature complex, multi-layered structures with stacked layers and electrode arrays. These multilayer interfaces are susceptible to delamination and light diffraction, compromising deformability, optical transparency, and overall reliability30,31. And a comparison of the device characteristics based on the different mechanisms mentioned above is presented in Table S1. To address these challenges, developing an ionic touch panel that outputs near-steady-state electrical signals upon touch and operates with a single-layer structure would significantly broaden its practical applicability. Contact charging is a common phenomenon on ionogel surfaces. When an ionogel comes into contact with another material, ions or electrons can be transferred across the interface22,32,33. If anions (or electrons) and cations (or holes) exhibit differing tendencies to transfer, this asymmetric carrier migration leads to a net charge transfer at the contact interface34–38. Although this phenomenon has been extensively studied, its underlying mechanism remains poorly understood due to the complex coupling between ionic and electronic processes. Nevertheless, it is widely accepted that the contact electrification response decays rapidly—within tens of milliseconds to seconds—owing to fast charge transfer kinetics and limited interfacial charge37,39–42. Therefore, achieving a touch-induced steady-state electrical signal remains a significant challenge. In this work, we present an ionic touch panel with a configuration comprising fixed metal corner electrodes, an ionogel layer, and a movable metal electrode, operating via a triboelectric-potentiometric hybrid sensing mechanism. Contact between the movable electrode and the ionogel surface closes the originally open circuit into a complete electrical loop, transforming the conventional interfacial effect of dynamic electrical double-layer modulation into a bulk effect. This bulk effect, combined with the slow ion migration dynamics of nano-confined ionogel, enables the output of steady-state electrical signals. Through a series of control experiments, we investigate the coupled transport behavior of electrons and ions underlying this bulk effect and elucidate design principles for the ionogel. The resulting touch panel features a simple single-layer structure and stable signal output, enabling reliable, continuous, and accurate recognition of various touch interactions, including short-term point contact, prolonged point contact, and continuous sliding contact. The design of ionic touch panel and performance impact factors Human skin is stimulated by touch and converts the different stimuli into a series of ionic currents that are transmitted via nerve fibers to the brain for analysis to recognize the touch location and pressure (Fig. 1a). Inspired by the skin, researchers can also convert force signals into electrical signals and realize self-powered tactile perception through traditional triboelectric and piezoelectric principles. However, as shown in Fig. 1b, c, existing triboelectric and piezoelectric sensors can only respond to dynamic touch stimuli, i.e., producing electrical output (especially current output) only at the moment of contact or separation, resulting in the inability to accurately recognize a stable touch over a long period of time. Sensors based on chemical-battery reactions enable detection of dynamic and static mechanical stimuli, but have not yet been utilized to develop touch-screen devices. The iontronic devices developed in recent years use ions as the main charge carriers and are closer to the response mode of the biological nervous system. The iontronic tactile sensors have the advantages of high sensitivity, low power consumption and biocompatibility. However, existing ion-triboelectric and ion-piezoelectric devices are also only suitable for dynamic measurements. Ion-piezocapacitive and ion-piezoresistive devices can be used for static measurements but still inevitably rely on an external power supply. So far, there is no self-powered ionic touch panel that can continuously detect the touch position with high precision. The conceptualization and electrical output of the ionic touch panel. a The sensing system of the human skin that senses external stimuli. b Contact response characteristics of triboelectric device, piezoelectric device and the device. c Schematic of the response mechanism of triboelectric and piezoelectric sensors. d Structural components of the device. e Voltage and current output of the device for long-term contact response Here, we developed an iontronic touch panel. In a closed loop, the aluminum sheet gently touches the touch panel (2 kPa), and the Cu corner electrode end can detect about 280 mV voltage and 27 nA current for up to 3 h, as shown in Fig. 1d, e. The main component of the touch panel is a piece of [EMIM][TFSI]/TPU composite ionogel film. The TPU polymer chain consists of hard segment and soft segment, where the hard segment ensures the strength of the material and the soft segment ensures the flexibility and stretchability of the material43,44. As shown in Fig. 1d, e, when the movable electrode (e.g., Al) contacts the ionogel surface, the originally open circuit is closed, forming a complete electrical loop comprising the movable electrode, ionogel, corner electrode, and external resistor. Prior to contact, a stable electric double layer (EDLC) exists at the corner electrode/ionogel interface due to equilibrium ion redistribution. Upon contact, a new dynamic electric double layer (EDLM) instantaneously forms at the movable electrode/gel interface as a result of contact electrification and asymmetric ion/electron transfer. Critically, the surface properties and work function of the movable electrode differ from those of the fixed corner electrode, creating an intrinsic asymmetry between EDLM and EDLC. This asymmetry generates a built-in potential difference that drives continuous charge migration through the closed external circuit. The sustained steady-state output arises from three synergistic factors. First, the positive feedback coupled ion-electron transport ensures that electron flow through the external circuit continuously reinforces ion redistribution at both EDL interfaces: the EDLC interface experiences cation inflow while the EDLM interface undergoes cation outflow, mutually sustaining the charge imbalance. Second, the nano-confined structure of the ionogel imparts exceptionally slow ion migration dynamics (ion diffusion coefficient of 2.94 × 10⁻¹⁵ m²/s, Fig. 2i), which prolongs the charge redistribution process and prevents rapid signal decay. Third, the work function difference between the electrodes determines the driving force for electron transfer. As shown in Fig. 4d, e, the output amplitude follows the reverse order of work functions (Cu < Fe < Ti < Al < Zn), confirming that a larger work function mismatch enhances the built-in potential and signal amplitude. Thus, the stable electrical output is not a transient contact electrification pulse but rather a quasi-steady-state signal sustained by the positive feedback loop, retarded ion dynamics in the nano-confined ionogel, and asymmetric EDL stabilization. This mechanism distinguishes our device from conventional triboelectric or piezoelectric systems that produce only transient outputs. Effect of electrical circuit connections, material composition and electrode selection on the output performance of the iontronic devices. a–c Schematic of the iontronic devices with different electrical circuit connections: closed-loop and grounding (a), closed-loop (b), open-loop (c). d–f Response of the iontronic devices with different electrical circuit connections: closed-loop and grounding (d), closed-loop (e), open-loop (f). g Schematic representation of the transport mode of anions and cations in ionic liquid and ionogel. h Microstructural morphology of ionogel. i Electrochemical impedance spectra (Nyquist plot) of ionic liquid and ionogel. j–l, Effects of different corner electrode and movable electrode pairings on the voltage output of the ionic liquid (IL) device, including Cu-Al electrode pair (j), Au-Pt electrode pair (k), and Au-Au(l) electrode pair Characterization and analysis of the ion-electron positive feedback coupled transport mechanism in the iontronic device. a Schematic diagram of the device output performance test using an Au-Au electrode pair. b, Voltage output performance of the device under Point touch mode under (a) test condition. c Current output performance of the device under Point touch mode under (a) test condition. d Comparison of the output performance of five different metals as movable electrodes, with the fixed corner electrode being copper. e Surface work function of these five different metals. f Schematic representation of the characterization of the anion ([TFSI-]) and cation ([EMIM+]) distribution states at locations of P1-P4 before and after the loop close using Raman mapping analysis. g Raman mappings of cation ([EMIM+]: 2800–3050 cm−1) distribution at P4 (corner position of the zinc mesh aperture) before and after the loop close. h Raman mappings of cation ([EMIM+]: 2800–3050 cm−1) distribution at P3 (center position of the zinc mesh aperture) before and after the loop close. i Raman mappings of cation ([EMIM+]: 2800–3050 cm−1) distribution at P2 (corner position of the copper mesh aperture) and P1 (center position of the copper mesh aperture) before and after the loop close In order to explore the role of closed-loop electrical connection design, three electrical circuit connection modes including grounding & closed-loop, closed-loop, and open-loop, were designed and compared, respectively, as shown in Fig. 2a–c. In the experiments, the metals copper and aluminum were used as fixed corner electrode and movable electrode, respectively. And two contact modes were used during the experiment. Point contact mode 1: move the movable electrode from left to right to contact each of the 6 equally spaced points on the ionogel membrane for 2 s, then return and repeat the procedure in reverse. Prolonged contact mode 2: the movable electrode is in continuous contact with the gel membrane at a fixed position for more than 600 s. As shown in Fig. 2d, e and Fig. S1a, b, for the two electrical circuit connection modes, closed-loop & grounding and closed-loop, the voltage/current response value is highly consistent with the change pattern of the contact position. In contrast, When the loop is not closed (Fig. 2f and Fig. S1c), the voltages and currents barely respond, indicating that closed-loop connection is necessary. When the loop is not closed, EDLC and EDLM can only transfer charge through the ionogel itself. Both EDLC and EDLM can also realize charge transfer via an external circuit when the loop is closed, i.e., converting the traditional dynamic EDL modulation effect at the interface into a bulk effect. The related charge transfer process and intrinsic mechanism will be discussed and analyzed in depth later. Throughout the ion-electron coupled transport loop, we hypothesize that the ion transport kinetics within the ionogel is a key factor in whether the electrical signal exhibits a steady-state output. Therefore, an attempt was made to replace the ionogel (Fig. 2g) with an ionic liquid (IL, [EMIM][TFSI]) for touch experiments. Unlike pure ionic liquids, ionogel exhibits a homogeneous nano-confined channel structure (Fig. 2h and inset) and extremely slow ion transport dynamics due to ion-polymer backbone interactions. As shown in Fig. 2i and Fig. S2, through electrochemical impedance test and corresponding equivalent circuit analysis, the ion diffusion coefficient of ionogel was obtained to be 2.94 × 10−15, which is about 15.2% of that of the pure ionic liquid (1.94 × 10−14). Interestingly, as the ion diffusion coefficient increases, the electrical signal output stability of the pure ionic liquid device with the application of a 600 s static touch stimulus (Fig. 2j and Fig. S3) decays dramatically within the first 200 s. The phenomenon of the electrical signal decay with time is originated from the ionic-triboelectric mechanism. Because after replacing the Cu-Al electrode pair with the inert metal Au-Pt (Fig. 2k) or Au-Au electrode pair (Fig. 2l), there is still a clear electrical signal and a similar decaying trend. In contrast, the Cu-IL-Al device shows steady state signal at 200–600 s, which is presumed to be caused by a slow and stable chemical reaction of galvanic cell. The electrical polarity of the ionic triboelectricity and the galvanic cell reaction is consistent. Since both ionic-triboelectric mechanism and potentiometric reaction are extremely charge-dense power generation phenomena, combined with slow charge transfer kinetics, it is understandable that the iontronic device could produce a small, persistent and stable current output at the nA level33,43,45. Furthermore, as can be seen from Fig. 2j–k, differences in pulse amplitude and decay time are observed when different electrode materials are used. This may be attributed to the combined effects of the electron transfer driving force, asymmetry in the interfacial electrostatic double layer, and the galvanic cell reaction. To be specific, the work function difference (analysis below) between the movable and corner electrodes determines the initial electron flow direction and magnitude upon contact. A larger work function mismatch drives stronger electron transfer through the external circuit, enhancing the built-in potential between EDLM and EDLC, thereby increasing the signal amplitude. For decay time, materials with lower work functions exhibit sustained signals due to continuous electron supply maintaining EDL asymmetry. In contrast, when using inert metals such as Au (Fig. 4b–c), although pure ionic-triboelectric still generates an asymmetric EDL, the absence of additional driving force from galvanic reactions results in faster signal decay as the system relies solely on the limited interfacial charge and slower ion redistribution within the nano-confined ionogel. Thus, the material-dependent behavior is governed by the interplay between work function-induced electron transfer, EDL asymmetry stability, and electrochemical activity, which collectively modulate both the amplitude and persistence of the steady-state signal. And other detailed tests and structural characterizations of the ionogel are demonstrated in the Supplementary Material (Fig. S4-Fig. S8). Varying the content of ionic liquid in the TPU/[EMIM][TFSI] membrane will also greatly affect the amplitude and stability of electrical output. Six ionogels with different doping concentrations (including 5 wt%, 10 wt%, 15 wt%, 20 wt%, 25 wt% and 30 wt%) were prepared, and the experiments were conducted using the same closed-loop electrical connection with a fixed contact position at 4 cm from the corner electrode, a fixed load resistance of 10 MΩ, and a fixed contact pressure of 3200 Pa. As shown in Fig. 3a and Fig. S9, the electrical response gradually increases as the ionic liquid content increases from 5% to 20%, which may be due to the shortening of the Debye length and Debye time. By further increasing the ion content, the voltage and current response peaks tend to stabilize, indicating that the number of ions involved in EDL generation has reached saturation. On the other hand, the stability of the electrical signal output shows a decreasing trend with further increase of the ionic liquid content. In particular, when the doping concentration is further increased to 20–30%, a significant overshoot phenomenon has been observed, which could be attributed to the rapid formation of EDL. The increased free ions are rapidly and densely adsorbed on the electrode surface, followed by partial ion reflux or re-equilibration. For accurate sensing applications, low ionic liquid content is preferable. Factors affecting the output performance of iontronic devices. a Effect of ionic liquid doping concentration on voltage output. b Effect of contact pressure at the interface between the movable electrode and the ionogel on the voltage output performance. c Effect of contact pressure at the interface between the movable electrode and the ionogel on the current output performance. d Effect of external resistance on voltage output performance. e Effect of external resistance on current output performance. f Effect of contact area on voltage output performance. g Effect of ambient temperature on the voltage output The effect of contact pressure on the voltage and current response was also investigated and the results are shown in Fig. 3b, c. As the contact pressure gradually increases (0.5 kPa, 0.8 kPa, 1.6 kPa, 2.4 kPa, 3.2 kPa, 4.8 kPa, 6.4 kPa, 9.6 kPa, 16 kPa, 22.4 kPa, 32 kPa), the voltage and current output accordingly show a slowly increasing trend. The growth rate of output electrical signal is much slower than that of the pressure amplitude since the device responds to the contact behavior itself rather than the pressure. The slow increase in the electrical signal caused by pressure is attributed to the more effective contact and the decrease in ion transmission resistance due to gel deformation. Besides, the magnitude of the external load resistance also greatly affects the electrical output of the device. As shown in Fig. 3d, e, 0 Ω, 100 Ω, 1 kΩ, 10 kΩ, 100 kΩ, 1 MΩ, and 10 MΩ are selected as external resistors, respectively. In the 0–1 MΩ range, the current output of the device (7–8 nA) is essentially independent of the magnitude of the load resistance. As the resistance reaches 10 MΩ, the current output decays to approximately 6 nA. This is because in the entire closed-loop electrical system, although increasing the load resistance weakens the transmission of electrons, the charge transfer rate of the entire electrical system mainly depends on the slower moving ions. The voltage signal gradually increases from 0.05 mV to 206 mV as the load resistance changes. Consequently, in order to improve the sensitivity of the device, the load resistance was set to 1 MΩ and 10 MΩ respectively when testing the current and voltage in subsequent application demonstrations. In addition, the effect of the movable metal electrode size on the electrical output performance was investigated (Fig. 3f). The movable electrode sizes were set to 0.5 cm2, 1 cm2, 2 cm2, 3 cm2, 5 cm2 and other test conditions were identical. The results show that the electrical signal output increases with increasing contact area, which is attributed to the formation of more EDL at the interface. Moreover, the effect of ambient temperature was also investigated, as shown in Fig. 3g, which demonstrates the contact response performance of the device at 20 °C, 30 °C, 40 °C, 60 °C and 80 °C. The higher the temperature, the larger the electrical output amplitude, which is attributed to the high temperature promoting ion migration and rapid EDL formation. To investigate the effect of ambient humidity on output performance, the response of device to finger contact was tested under different ambient humidity conditions. As shown in Fig. S10, the device’s response to finger contact exhibits a clear increasing trend with rising ambient humidity. This behavior can be attributed primarily to the hydration state of the fingertip skin. As humidity increases, the enhanced hydration promotes greater charge generation at the finger–ionogel interface, thereby amplifying the contact electrification effect. Moreover, elevated ambient humidity facilitates the formation of a thin moisture layer on the microscale surface structures of the fingertip, which increases the effective contact area and improves interfacial charge transfer efficiency during touch. This combined effect—enhanced interfacial charge density and improved conformal contact—leads to a higher electrical output. Relevant tests were also performed for contact response stability as well, as shown in Fig. S11a–c. At the same touch position, more than 200 contact-separation cycle operations were performed. The current output values do not show any obvious fluctuation or signal degradation. Besides, to verify that the device is not affected by water evaporation (i.e. its long-term storage stability), the ionogel was exposed to air for a period of time before undergoing contact response testing. As shown in Fig. S 12–13, the ionogel maintained its contact response output performance without degradation after exposure to air for 1, 3, and 7 days. This also demonstrates the good stability and durability of the device. Analysis of ion-electron positive feedback coupled transport mechanism In order to explore the intrinsic mechanism of the unique output characteristics of this iontronic device, the phenomena of solid-solid triboelectricity, chemical reaction and ionic piezoelectricity that may exist inside the device were firstly investigated. As shown in Fig. S14, the ionogel film was replaced with a pure TPU polymer film, and other test conditions remained unchanged. No voltage or current response was observed from the test data, indicating that the electrical signal output cannot be obtained by relying solely on solid-solid friction. As mentioned earlier, when the electrodes involve active metals, we speculate that chemical reactions make partial contributions in the device. In order to investigate the contribution of pure ionic triboelectricity and exclude possible chemical reactions, both the corner electrode and the movable electrode were replaced with Au sheets, as shown in Fig. 4a. The voltage and current responses of point touch mode are shown in Fig. 4b, c. The trend is consistent with the previous test results: voltage and current amplitudes decrease or increase as the contact distance becomes farther or closer. This suggests that a purely ionic triboelectric phenomenon is still suitable for constructing an ionic touch panel, but the duration of the electric signal will be shortened. Despite the choice of gold for both the movable and corner electrodes, considering the relatively slow rate of ion diffusion in ionogel, the initial density of EDLM remains much lower than that of EDLC. Thus, the asymmetric EDL is still formed and provides power for charge transfer in closed-loop electrical system. In addition, ionic piezoelectricity may be introduced during contact process due to the presence of pressure. In order to investigate the role of ionic piezoelectricity, a typical sandwich-type device structure consisting of an upper electrode (Al), an ionogel film (middle layer), and a lower electrode (Cu) was used, as shown in Fig. S15. In the 600 s long-term contact response test, ion piezoelectricity makes no contribution since there is no force gradient across the device structure. And both the contact electrical signal amplitude and stability do not match the typical transient response characteristics of ionic piezoelectricity. As shown in Fig. 4d, the fixed corner electrode is copper, while the metal type of the movable electrode is changed. These voltage and current output amplitudes are ranked in order: Cu