A ternary-dielectric dual-domain triboelectric nanogenerator for high charge density and tail-charge suppression

Abstract

Residual-charge accumulation and limited interfacial charge excitation have emerged as inherent bottlenecks that constrain the charge density and energy-conversion efficiency of sliding triboelectric nanogenerators (TENGs). These limitations commonly originate from binary-dielectric architectures, in which deep-level tail charges and incomplete interfacial depletion hinder charge transfer regardless of materials or structural configurations. Here, we introduce a ternary-dielectric dual-domain triboelectric architecture that overcomes this intrinsic constraint by reconfiguring the dielectric field landscape and enabling cross-domain charge management. The upper discharge domain – constructed by a PA/PTFE/PU ternary triboelectric layer with adjacent electrodes – realizes efficient interfacial charge generation and active removal of tail-inherent charges via corona-assisted regulation. Simultaneously, a lower polarization-discharge domain formed by promoting discharge electrodes harnesses space-volume-induced charge migration within the PU foam, converting previously lost inherent charges into additional effective output via dielectric-polarization discharge. By coupling these two charge domains, the system establishes a charge-generation and charge-depletion pathway that systematically reshapes the excitation–decay dynamics of triboelectric charges. The optimized device achieves a considerable charge density of 1.2 mC m⁻², an average power density of 22.53 W m⁻², and an energy density of 3.42 J m⁻². This work introduces a generalizable dielectric-domain engineering strategy that addresses long-standing charge-retention and tail-charge bottlenecks in TENGs, offering a mechanistic foundation for next-generation high-efficiency triboelectric energy-conversion systems.

---

Broader context

Sliding-mode triboelectric nanogenerators (TENGs) are widely regarded as promising power sources for distributed sensors and self-sustained environmental monitoring. However, their practical deployment has been fundamentally limited for more than a decade by two unresolved bottlenecks: (i) restricted interfacial charge excitation and (ii) persistent accumulation of tail-end inherent charges. These issues originate from incomplete charge-governing mechanisms in conventional binary or porous dielectrics, where charge generation, migration, and depletion occur in isolated and poorly coordinated pathways. As a result, large fractions of triboelectric charges remain unutilized, while accumulated residual charges induce electrostatic shielding that progressively suppresses energy-conversion efficiency. This work introduces a ternary-dielectric dual-domain triboelectric nanogenerator (TDT) that establishes a unified charge-management paradigm based on cross-domain dielectric engineering. By synergistically coupling corona-assisted interfacial excitation, space–volume-driven charge migration, and dielectric-polarization discharge, the TDT simultaneously enhances charge generation, suppresses tail-charge buildup, and recovers inherently dissipative volumetric charges. This mechanism-level integration overcomes the long-standing limitations of sliding TENGs and unlocks a new route toward high-efficiency micro-energy harvesting. Beyond device performance, the proposed charge-governing concept provides a generalizable strategy for improving dielectric-based energy-conversion systems and enables scalable, self-powered sensing in next-generation environmental and IoT infrastructures.

Introduction

The rapid proliferation of distributed electronics, intelligent sensors, and autonomous monitoring systems has intensified the demand for sustainable micro-energy technologies capable of operating reliably under ambient mechanical excitations. Triboelectric nanogenerators (TENGs), which utilize contact electrification coupled with electrostatic induction to convert ubiquitous mechanical motions into electricity, have emerged as a promising platform for self-powered systems due to their material diversity, structural adaptability, and compatibility with large-area, low-frequency energy sources.^1–5 These attributes have enabled broad exploration of TENGs in Internet of Things (IoT) nodes, wearable electronics, and long-term environmental or structural monitoring.^6–10 Among the different operation modes, sliding-mode TENGs are particularly attractive for harvesting broadband, low-frequency mechanical stimuli such as human motion, machine vibrations, and wind-induced movements. However, despite their versatility, the overall energy-conversion efficiency of sliding-mode TENGs remains fundamentally limited by intrinsic charge-excitation bottlenecks and persistent interfacial charge-retention phenomena. These long-standing constraints originate from the interplay between dielectric properties, surface charge trapping, and incomplete charge neutralization, and they continue to impede further advancement of TENG-based micro-energy systems. This has motivated growing interest in material-, interface-, and mechanism-level innovations aimed at overcoming the inherent restrictions of conventional dielectric architectures.

Over the past decade, extensive efforts have been devoted to boosting the output of sliding-mode TENGs through structural optimization, dielectric engineering, interfacial modification, charge injection, and advanced power-management circuits.^11–25 A particularly influential direction involves the use of millimeter-scale electropositive interlayers, such as polyurethane (PU) foam, melamine-formaldehyde (MF) foam, and other porous dielectrics, to introduce volumetric charge-excitation pathways that couple triboelectrification with electrostatic induction, localized ionization, and dielectric polarization. Representative advances include PU-foam-based hybrid AC/DC generators,²⁶ corona-volume-triboelectric coupling strategies that significantly elevate charge excitation,²⁷ and three-terminal porous-dielectric systems capable of providing coordinated AC/DC output.²⁸ These studies have shed light on synergistic interactions among the triboelectric effect, corona discharge, and dielectric polarization, demonstrating that volumetric dielectrics can partially alleviate the limited charge-transfer capability inherent to planar interfaces and yield power densities on the order of tens of W m⁻² Hz⁻¹.²⁹ Concurrently, structural innovations such as slider-integrated adjacent electrodes have been developed to broaden vertical electric-field distributions and capture spatially dispersed charges.^30,31 In addition, ternary electrification systems and dynamic dielectric-polarization discharge mechanisms have created new opportunities for tailoring triboelectric polarity and enhancing charge-collection efficiency.^32,33

Despite these advances, current binary or porous-dielectric TENG architectures remain fundamentally constrained by their inability to govern charge generation and charge dissipation in a coordinated manner. Existing approaches typically enhance either interfacial charge excitation or discharge-assisted charge extraction, yet rarely unify these processes within a single charge-governing framework. As a result, two long-standing bottlenecks persist. (i) Spatially dispersed or weakly bound charges, especially those originating from deep or volumetric dielectric regions, cannot be efficiently migrated, released, or reutilized, leading to significant charge loss through non-productive pathways. (ii) Tail-end inherent charges inevitably accumulate at sliding interfaces, producing strong electrostatic shielding that progressively suppresses effective charge transfer during continuous motion. These limitations are not isolated phenomena but interdependent outcomes of incomplete charge-regulation mechanisms that lack cross-domain coordination. Addressing them requires a charge-management paradigm capable of simultaneously enhancing charge generation, dynamically modulating interfacial charge evolution, and extracting or neutralizing residual charges before they induce detrimental shielding. Such an integrated solution, which couples generation, migration, and depletion within a single dielectric system, has not yet been realized, leaving the fundamental challenges in coordinated charge-governing dynamics unresolved.

Herein, this study introduces a ternary-dielectric dual-domain triboelectric nanogenerator (TDT) that realizes a charge-governing mechanism integrating corona discharge, space-volume-induced migration, and dielectric-polarization discharge. The upper discharge domain, constructed from a PA/PTFE/PU ternary triboelectric layer with adjacent electrodes, not only strengthens interfacial charge excitation but also actively neutralizes tail-end inherent charges through corona-assisted regulation, thereby reshaping the interfacial potential landscape. In parallel, the lower discharge domain, formed by a pair of promoting-discharge electrodes (PDE), harvests and reutilizes migrated residual charges originating from the volumetric PU-foam matrix, enabling additional direct-current output via polarization-discharge pathways. Through the coupling of these two charge domains, the TDT achieves the simultaneous enhancement of interfacial charge generation, suppression of tail-charge accumulation, and collection of volumetrically migrated inherent charges within a single sliding-mode TENG architecture for the first time. Guided by mechanism analysis, numerical modeling, and systematic structural optimization, the TDT delivers a considerable power density of 22.53 W m⁻² and an energy density of 3.42 J m⁻², substantially surpassing previously reported sliding-mode designs. To demonstrate practical viability, a rotary TDT incorporating a coaxial counter-rotating planetary-gear module was further developed. The system can directly illuminate 6000 LEDs and, when integrated with an efficient power-management circuit, it is able to continuously drive twelve 2 W commercial bulbs. Under a simulated 15 m s⁻¹ wind field, the TDT further demonstrated stable operation by powering 36 distributed hygrothermographs, underscoring its suitability for large-scale environmental energy harvesting and self-powered IoT sensing. These results highlight the TDT as a robust and scalable platform that advances sliding-mode TENGs in terms of output performance, energy utilization efficiency, and system-level integration, offering a promising route toward next-generation high-efficiency micro-energy harvesting technologies.

Results and discussion

Ternary-dielectric layer and dual-domain discharge mechanisms of the TDT

Fig. 1a depicts a sliding TENG model with the ternary-dielectric layer and the coupling of dual-domain discharge based on the triple-synergistic enhancement mechanism of corona discharge, the space volume effect, and dielectric polarization discharge. The ternary-dielectric material system was designed based on the unique space volume effect of polyurethane (PU) foam. By systematically comparing the charge transfer performance of various common friction materials paired with PU foam, an optimal ternary-dielectric system was ultimately selected, as illustrated in Fig. S1 (SI). Herein, a ternary-dielectric dual-domain triboelectric nanogenerator (TDT) is proposed with polyamide (PA)/polytetrafluoroethylene (PTFE)/paired adjacent electrodes on the upper slider and polyurethane (PU)/paired adjacent electrodes on the stator. Specifically, PA and PTFE films are alternately arranged at specific intervals on the slider, along with a pair of adjacent electrodes positioned on both sides of the film. A millimeter-thick polyurethane (PU) foam is adhered to the stator acrylic plate, and another set of paired adjacent electrodes is placed below as the slider electrodes, which forms the whole slider together with the upper PA/PTFE. As depicted in Fig. 1a, this innovative structural design employs a ternary-triboelectric layer composed of PA/PTFE/PU foam, and adjacent electrodes are arranged on both sides of this layer to construct an upper discharge region, aiming to enhance interface charge generation efficiency and simultaneously clear residual charges at the triboelectric interface tail. Meanwhile, a pair of promoting discharge electrodes (PDEs) is integrated at the bottom of the PU foam to form a lower discharge domain, further collecting the residual charges from the upper discharge domain due to the space volume effect of the PU foam, and generating additional direct current during sliding via dielectric-polarization discharge.

Fig. 1 Structure and mechanism of the ternary-dielectric and dual-domain TENG (TDT). (a) 3D structure diagram of TDT. (b) Schematic of the dual-domain discharge mechanism, including (i) upper discharge domain and (ii) lower discharge domain. (c) The working mechanism of the TDT based on triple-synergistic enhancement of corona discharge, the space volume effect, and dielectric-polarization discharge. The output performance comparison of a conventional binary-dielectric TENG and a ternary-dielectric TENG without PDE, including (d) Q_sc and (e) I_sc. The output performance comparison of the ternary-dielectric TENG without PDE and TDT, including (f) Q_sc and (g) I_sc.

As depicted in Fig. 1b(i), for the upper discharge domain, when the TDT is activated, the slider moves rightward, breaking the initial electrostatic equilibrium. Owing to its electronegativity, the PTFE film accumulates negative charges at the right-side contact interface, while the surface of the PU foam with strong electropositivity retains positive charges after being slid over. Meanwhile, the PA film captures a portion of the positive charges from the PU surface through sliding. The accumulated negative charges at the PTFE interface induce positive charges on the right-hand electrode, whereas the positive charges at the PU and PA interfaces jointly induce negative charges on the left-hand electrode. These two electrodes thereby generate a rightward direct current through the external circuit. Subsequently, under the conditions of a high voltage and narrow gap, corona discharge occurs at the left side of the upper slider electrode, where electron transfer neutralizes a portion of the surface charges, completing the charge regulation process in the upper discharge domain, namely the electrostatic friction discharge domain. As depicted in Fig. 1b(ii), for the lower discharge domain, upon completion of the charge regulation process in the upper discharge domain, a portion of isolated residual charges detach from the interface of the upper discharge domain and migrate to the upper surface of the acrylic substrate by leveraging the space volume effect of the PU foam. Simultaneously, the acrylic substrate generates a transient high voltage at its bottom owing to dielectric-polarization discharge, which induces opposite charges at the bottom-placed PDE, thereby forming a leftward direct current through the external circuit. Ultimately, corona discharge occurring in the bottom region of the PDE effectively removes residual charges, completing the charge regulation process in the lower discharge domain, namely the bottom-electrode coupled discharge domain. It is noteworthy that the charge removal in the lower discharge domain operates through a coupled dielectric-polarization and corona discharge mechanism. Owing to the high electrical insulation of the acrylic substrate, residual charges cannot migrate directly through the bulk material; however, the associated electric field penetrates the dielectric and induces charges of opposite polarity on the underlying PDE. Once the localized electric field on the PDE exceeds the air breakdown threshold, corona discharge is initiated, producing ion pairs. Under the influence of the strong electric field, the counter-ions migrate toward and adhere to the back surface of the substrate, thereby establishing electrostatic coupling with the residual charges on the top surface and accomplishing effective charge clearance. This mechanism successfully overcomes the problem of inherent charge accumulation at the tail caused by incomplete charge removal in the conventional binary-dielectric structure.

Notably, although corona discharge involves non-productive energy dissipation through air molecule ionization, it is regarded as a strategic energy investment in the TDT architecture. By actively introducing a discharge process, the system breaks through the charge-density ceiling inherent in traditional sliding TENGs. While this mechanism introduces additional microscale electrical losses, it significantly enhances the effective induced charge after electromechanical conversion, thereby achieving substantially greater energy gain compared to non-discharge operation modes. Fig. 1c systematically demonstrates the working mechanism of the TDT based on corona discharge, the space volume effect, and dielectric-polarization discharge. To distinguish the charge-migration pathways, electrical response tests were conducted on different materials under a high DC electric field. The results demonstrate that porous dielectrics such as PU foam exhibit a stable direct current, confirming the existence of bulk ion migration channels due to the space volume effect. In contrast, dense dielectrics like acrylic generate only transient polarization pulses, reflecting their characteristic dielectric-polarization behavior. This contrast effectively separates space-volume-driven migration from dielectric polarization as two independent physical processes, as depicted in Fig. S2 (SI). Based on the above principle, the charge collection characteristics of each electrode in the TDT are shown in Fig. S3 (SI). The charge collection processes of electrode pairs within the same discharge domain exhibit symmetry. The physical picture of the TDT is shown in Fig. S4 (SI).

To verify the working mechanism of the TDT, the electric potential distributions of the TDT during sliding were obtained within the COMSOL Multiphysics environment as depicted in Fig. S5 (SI); as the rightward displacement of the slider increased, the electric potential increased significantly, which is highly consistent with the aforementioned theoretical analysis. Furthermore, the numerical simulations of the corona discharge region in the TDT were performed, obtaining the electric field distribution around the upper and lower discharge domains. As depicted in Fig. S6 (SI), all four discharge regions developed sufficient electric field strength to initiate corona discharge, with the discharge space distribution distinctly biased toward the left side of the electrodes. Based on the reciprocating linear sliding platform driven by a linear motor, further verifications were conducted on the output performance enhancement of the sliding TENG by the ternary-dielectric layer and dual-domain discharge. When the lower discharge domain was not coupled, the ternary-dielectric TENG already demonstrated significant performance advantages. Compared with the conventional binary-dielectric TENG, the ternary material system effectively mitigates the charge density reduction caused by decreased electrode spacing, resulting in an increase of approximately 155% in transferred charge (Q_sc) and approximately 138% in short-circuit current (I_sc), respectively, as depicted in Fig. 1d and e. With the introduction of the lower discharge domain, the bottom-placed promoting discharge electrode (PDE) facilitates corona discharge by reducing the discharge gap. This not only creates conditions for collecting residual charges from the upper discharge domain but also generates additional DC output in the lower discharge domain. As a result, compared with the conventional ternary-dielectric TENG without PDE, the proposed TDT achieved an increase of approximately 68% in Q_sc and I_sc simultaneously, as depicted in Fig. 1f and g. Overall, the corona discharge serves as the dominant mechanism of TDT, contributing approximately 60% in Q_sc, responsible for generating charge and resetting charge in the output channel. The space volume effect and dielectric polarization act as key supporting mechanisms, successfully enhancing the overall output performance by 39.2% through the secondary recycling of residual charges and the deep cleaning of the interface. The tight synergy of these three mechanisms is the core driving force for the TDT to achieve high-performance output.

The synergistic mechanism of dual-domain discharge and the ternary-dielectric layer

This work proposes a performance enhancement mechanism for a sliding TENG based on the synergistic coupling between a ternary-dielectric layer and a dual-domain discharge structure. The mechanism operates through two complementary pathways. On the one hand, the lower discharge domain utilizes the space volume effect of PU foam to capture isolated charges generated by corona discharge in the upper discharge domain. This process not only creates an additional DC output but also effectively removes isolated residual charges from the output channel of the upper discharge domain, thereby significantly increasing its output charge density. On the other hand, compared to a conventional binary-dielectric structure, the ternary-dielectric structure leverages its triboelectric series differences to generate transfer charges during slider movement that effectively neutralize inherent charges at the slider tail edge. This enables deep cleaning of residual charges at the triboelectric interface while effectively avoiding the negative impact of tail inherent charges on the output channel performance.

Fig. 2a depicts the residual charge formation process in the upper discharge region during the corona discharge, further systematically illustrating that the residual charges inhibit the upper discharge output channel during corona discharge. First, a portion of spatially distributed residual charges form at the triboelectric interface after corona discharge (Fig. 2a(i)). Second, these residual charges have an electrostatic shielding effect in subsequent operational cycles, hindering the charge separation in the triboelectric interface and significantly reducing the interface charge transfer efficiency (Fig. 2a(ii)). Last, due to insufficient charge separation, the induced charge density on the electrode surface significantly attenuates, ultimately leading to degraded electrical output performance (Fig. 2a(iii)). Fig. 2b reveals the intrinsic mechanism by which different substrate materials (wood, acrylic, and acrylic + PDE) regulate the isolated charges in the upper discharge domain. Owing to the unique space volume effect of the PU foam, isolated charges generated after discharge in the upper domain can migrate to the upper surface of the bottom substrate and are subsequently collected and utilized during the discharge process of the lower discharge domain. The collection efficiency primarily depends on the discharge support capability of the substrate material, with quantitative classification and numerical simulations of relevant materials, as depicted in Fig. S7 (SI). When the substrate lacks discharge support capability (Fig. 2b(i) and (ii)), all or part of the isolated charges transform into residual charges on the triboelectric interface, inhibiting the charge separation process on the right side of the triboelectric interface in subsequent cycles. In contrast, the introduction of a paired PDE enhances corona discharge in the bottom discharge domain, significantly improving the clearance and collection efficiency of isolated charges from the upper discharge domain (Fig. 2c(iii)). According to IEEE Std 539-2020,³⁴ the corona inception voltage based on Peek's law can be expressed using eqn (1) and (2).

where V_c is the corona inception voltage, m is the roughness coefficient of the electrode, g_v is the air breakdown field strength, r is the radius of the electrode, d is the distance between the electrode and acrylic, δ is the relative air density, E₀ is the air dielectric strength (21.1 kV cm⁻¹), and K is the constant (0.308) determined by the applied voltage characteristics.

Fig. 2 The synergistic mechanism of dual-domain discharge and ternary-dielectric layer. (a) Schematic of the residual charges inhibiting the upper discharge output channel during corona discharge. (b) Schematic of the comparison of the collection and utilization of residual charges in the upper discharge domain when using different substrates (wood, acrylic, and acrylic + PDE). (c) Schematic of the inherent charge formation in the triboelectric interface tail of the binary-dielectric structure. (d) Schematic of the mechanism of inherent charge elimination at the tail of the triboelectric interface in ternary dielectric structures, revealed by comparison with binary dielectric structures. The output performance comparison of the ternary-dielectric TENG with different substrates, including (e) Q_sc and (f) I_sc. The output performance comparison of ternary- and binary-dielectric TENGs with PDE, including (g) Q_sc and (h) I_sc.

It can be analyzed that the corona inception voltage V_c is positively correlated with the discharge gap d, thereby reducing the discharge gap facilitates the occurrence of corona discharge. By inducing opposite charges at the bottom, the layout of a paired PDE effectively shortens the discharge gap, thereby enhancing the charge collection capability of the lower discharge domain.

Conventional binary-dielectric sliding TENGs exhibit notable limitations in charge management, exhibiting distinctly asymmetric charge removal behavior during operation, as illustrated in Fig. 2c. During the rightward sliding discharge phase, the negatively isolated charges accumulated at the interface tail of the binary TENG cannot be effectively eliminated. This phenomenon stems from inherent characteristics of the binary-dielectric architecture. As positive charges surrounding the triboelectric interface tail are cleared during the corona discharge (Fig. 2c(i)), the region becomes prone to excessive accumulation of negative charges (Fig. 2c(ii)). Specifically, these negatively isolated charges remain confined to the triboelectric interface and can only rely on the three-dimensional volume effect of the PU foam for limited downward migration (Fig. 2c(iii)), ultimately resulting in persistent accumulation of inherent charges as corona discharge continues to progress during sliding. Fig. 2d illustrates the mechanism by which the ternary-dielectric structure clears the inherent charges at the triboelectric interface tail compared to the binary-dielectric structure. In the traditional binary-dielectric TENG, the discharge process mainly occurs in the area on the right side of the electrode (Fig. 2d(i)). Due to the close space position, the inherent charges accumulated at the tail of this structure exert a significant electrostatic shielding effect on the charge output channel, interfering with the normal electrostatic induction process and leading to a significant reduction in the charge density of the output channel. The TDT structure completely eliminates the inherent charges at the triboelectric interface tail through the synergistic effect of the optimization of the ternary-dielectric structure and PDE. In the ternary-dielectric structure, only a tiny amount of surface charges generated by normal contact electrification exist at the PTFE tail, and the magnitude of these charges is much lower than that of the inherent charges at the triboelectric interface in the binary-dielectric structure (Fig. 2d(ii)). Meanwhile, TDT can transfer the effective discharge area to the left side of the electrode, avoiding the interference of the tail charges on electrostatic induction and significantly increasing the output charge density. To directly visualize the suppression of tail-charge accumulation, we performed Kelvin probe force microscopy (KPFM) on the PTFE film in the upper discharge domain after 100 sliding cycles. As depicted in Fig. S8 (SI), the conventional binary-dielectric slider exhibits a concentrated negative potential at the tail region, confirming tail-charge retention. In contrast, the TDT shows a smooth potential distribution near the background level, demonstrating real-time clearance of tail charges via corona discharge and dielectric-polarization discharge.

Further comparison verifications were conducted on the output performance enhancement of the sliding TENG with different substrate materials and ternary-/binary-dielectric materials. As depicted in Fig. 2e and f, the PDE attached to acrylic had the optimal output performance, followed by the acrylic substrate and wood substrate, indicating that the proposed PDE has optimal charge collection capability for the upper discharge domain, achieving an increase of approximately 178%/53.2% in Q_sc and 111%/49.8% in I_sc, respectively, compared with the wood substrate and acrylic substrate. The lower discharge domain is designed to capture and reuse residual charges migrating from the upper domain. Independent electrical measurements confirm that this domain consistently delivers a stable direct-current output across different sliding speeds, as depicted in Fig. S9 (SI). This sustained output provides direct evidence that residual charges are effectively collected and utilized. As depicted in Fig. 2g and h, the ternary-dielectric dual-domain TENG (TDT) had superior output performance compared to the binary-dielectric dual-domain TENG (BDT), achieving an increase of approximately 155% in Q_sc and 59% in I_sc, respectively, further demonstrating that the unique working characteristics of the ternary-dielectric structure provide an efficient suppression approach for the removal of inherent charges at the triboelectric interface tail. To ensure a rigorous comparison, all geometric parameters of the TDT and binary-dielectric counterparts were strictly aligned to maintain identical effective contact areas. We further compared the TDT with two binary-dielectric structures, including spaced arrangement and closely packed arrangement. As depicted in Fig. S10 (SI), the TDT exhibits significantly higher Q_sc than any binary configuration, even under the same total contact area. The closely packed binary-dielectric structure easily suffers from inter-electrode breakdown, while both binary configurations remain limited by inherent tail-charge accumulation. In contrast, the integrated ternary-dielectric and dual-domain design of the TDT not only prevents breakdown but also actively suppresses interfacial tail charges, thereby achieving superior charge density and space energy efficiency without increasing the contact area. Finally, the influence of the sliding direction of the TDT on its output characteristics is analyzed. Specifically, the Q_sc of the upper discharge domain increased rapidly when the slider moved toward the electrode side, exhibiting excellent charge transfer efficiency. In contrast, when sliding in the reverse direction, the Q_sc was significantly suppressed due to the obstruction of the charge transfer channel by inherent charges at the triboelectric layer tail. As depicted in Fig. S11 (SI), I_sc of the upper discharge domain in TDT further validated this phenomenon. When the slider moved in the forward direction, the peak I_sc was significantly higher than that during the reverse sliding, and the waveform of the I_sc exhibited greater stability during the decay phase.

Output performance optimization of the upper discharge domain

To comprehensively explore the influence of the structural dimensions on the electrical output performance of the upper discharge domain in TDT, a reciprocating sliding platform driven by a linear motor was built, as depicted in Fig. 3a. This section systematically investigates the optimization of key structural parameters in the upper discharge domain of the TDT. The considered parameters include the width ratio R₁ between the triboelectric dielectric and the electrode (defined as L₁ : L₂), the width ratio R₂ between the two triboelectric dielectrics PA and PTFE (L₁ : L₃), the thickness T₁ of the upper triboelectric layer comprising PA and PTFE, the thickness T₂ of the lower triboelectric layer made of PU foam, and the thickness T₃ of the acrylic substrate, as well as the slider width W₁ and the applied pressure F on the slider. During experimental testing, the linear motor drives the slider to perform bidirectional motion, with the discharge behavior demonstrating distinct directional dependence. When the slider moves toward the electrode side, it effectively avoids the influence of inherent charge at the triboelectric layer tail. In contrast, it exhibits a discharge pattern similar to binary-dielectric structure during the reverse sliding, as depicted in Fig. S12 (SI). It should be noted that such a directional dependence only exists in a single TDT unit. In the integrated rotary TDT structure, the upper discharge domain can eliminate the adverse effect of the motion direction on output performance through maintaining a continuous contact between the back-end of the electrode and the triboelectric dielectric. This design maintains stable charge density during bidirectional motion, thereby effectively overcoming the directional limitations of the conventional structures. To systematically validate the inherent charge theory and ensure the integrity of experimental data, this work simultaneously recorded the performance data under the optimal working conditions where the slider slides towards the electrode side and under the non-optimal working conditions where the slider slides towards the dielectric side. This experimental approach led to significant asymmetric characteristics in the data. Unlike the symmetric outputs commonly reported in previous studies, the asymmetric results obtained in this experiment provide a more comprehensive and reliable experimental basis for evaluating the structure optimization and output performance of the TDT.

Fig. 3 Output performance optimization of the upper discharge domain. (a) Schematic of the reciprocating sliding platform and the structural dimensions of the upper discharge domain. (b) The output waveforms of the upper discharge domain fabricated with different R₁, including (i) Q_sc and (ii) I_sc. (c) The output waveforms of the upper discharge domain fabricated with different R₂, including (i) Q_sc and (ii) I_sc. (d) The peak Q_sc of the upper discharge domain fabricated with different dielectric thicknesses. (e) The peak Q_sc of the upper discharge domain fabricated with different slider widths. (f) The peak Q_sc of the upper discharge domain under different external applied pressures and sliding speeds. (g) Durability and stability test of the upper discharge domain in the TDT.

The parameters that can be optimized in the upper discharge domain of the TDT structure include the width ratio of the triboelectric dielectric to the electrode and the thickness of the triboelectric dielectric to the substrate. Under the condition of maintaining constant width of the slider and equal widths of PA and PTFE, the effect of the width ratio R₁ between the PA and the electrode (1 : 1, 1.5 : 1, 2.5 : 1, 4 : 1 and 9 : 1) on the output performance of the upper discharge domain in TDT was systematically investigated. As depicted in Fig. 3b, with the increase of R₁, the Q_sc and I_sc of the upper discharge domain in the TDT exhibited a trend of increasing first and then decreasing, where the TDT with R₁ = 4 : 1 achieved optimal output performance. The triboelectric generation ability can be adversely impacted by both excessively wide electrodes and excessively narrow triboelectric dielectric layers. Conversely, excessively narrow electrodes can lead to a diminished discharge space within the upper discharge region, resulting in insufficient charge transfer. In the optimization of the dielectric width ratio R₂ (1 : 5, 1 : 2, 1 : 1, 2 : 1 and 5 : 1), with R₁ fixed at 4 : 1, the output performance of the upper discharge domain also exhibited a significant variation with changes in R₂, as depicted in Fig. 3c. The triboelectric generation ability can be adversely impacted by both excessively narrow triboelectric dielectric layers. Specifically, when the PA width proportion is too small, its ability to capture positive charges becomes insufficient, leading to degraded output performance. Conversely, a reduction in the PTFE width proportion significantly weakens the triboelectric charging capability of the TDT, resulting in a substantial decrease in output performance. Through systematic parameter optimization, it is determined that the upper discharge domain in the TDT with R₂ = 1 : 1 achieved optimal performance in both charge transfer efficiency and current output. In the optimization of the dielectric thickness (Fig. 3c), the effects of three key thickness parameters were systematically analyzed using the single-variable control method, including the thickness of the upper triboelectric layer T₁ (corresponding to the PA and PTFE films), the thickness of the lower triboelectric layer T₂ (corresponding to the PU foam), and the thickness of the acrylic substrate T₃. As depicted in Fig. 3d, the thickness of each layer significantly affects the output performance of the upper discharge domain in the TDT. The observed thickness-dependent trends arise from distinct physical roles of each layer. T₁ governs surface charge density and electrostatic induction efficiency, and too thin a film stores insufficient charge, while excessive thickness weakens induction. T₂ modulates space charge migration through the porous foam, and an insufficient thickness limits the migration volume, whereas overly thick foam extends the transfer path and increases loss. T₃ acts as a capacitive medium for polarization discharge, an extremely thin substrate results in a significant shrinkage of the effective sensing area due to the easily deformable structure, while a thick one weakens the electrostatic induction due to the decreased cross-substrate electric field. Therefore, deviations from the optimal thickness in any layer degrade charge generation, migration, or recovery, underscoring the need for balanced structural design. The space volume effect is also sensitive to the porosity of the porous medium. As depicted in Fig. S13 (SI), porosity gradient experiments show that higher porosity enhances charge migration through an interconnected pore network, provided mechanical resilience is maintained. Excessively high porosity, however, reduces the structural strength and triboelectric contact area. As depicted in Fig. S2 (SI), cross material comparisons further confirm that both chemically distinct PU and melamine foams exhibit stable bulk conduction under high electric field, indicating that the effect relies on the porous microstructure rather than the material composition.

Through parameter optimization, the optimal thickness combination was determined as T₁ = 0.1 mm, T₂ = 2 mm, and T₃ = 1 mm. Under this configuration, the upper discharge domain achieved a peak Q_sc of 2.64 µC, fully demonstrating the critical importance of thickness optimization in enhancing the output performance of the TDT. Based on the above optimized dielectric width and thickness, the influence of the overall slider width W₁ on the output performance was further investigated. As depicted in Fig. 3e, the Q_sc increased nonlinearly with slider width, primarily constrained by the space utilization efficiency, where the upper discharge domain achieved optimal space utilization efficiency at a slider width of 40 mm, corresponding to a peak Q_sc of 2.64 µC. In the optimization of the applied load and the sliding speed of the upper discharge domain, as depicted in Fig. 3f, the peak Q_sc of the upper discharge domain under a sliding speed of 0.5 m s⁻¹ exhibited a significant positive correlation with the applied pressure F. In this work, the applied pressure was controlled by adjusting the weight of the mass block on the TDT. When the pressure increased from 10 N to 50 N, the Q_sc increased by 0.93 µC, which is mainly attributed to the increase of the actual interface contact area and the reduction of the discharge gap in the upper discharge domain, resulting in more triboelectric charge generation and easier occurrence of discharge. As the pressure further increased, the growth rate of Q_sc decreased significantly, conforming to the saturation trend predicted by the Hertz contact theory. According to the working principle of the TENG, the output performance of the TENG is proportional to the triboelectric contact area as follows.

A = π × a² (3)

where A is the triboelectric contact area, a is the actual contact radius, R is the curvature radius of the slider, and E is the equivalent elastic modulus. Thereby, the transferred charge theoretically satisfies as follows

Q_sc ∝ F^2/3 (5)

This trend aligns with the experimentally observed saturation of the Q_sc with increasing applied pressure. Considering the mechanical strength limits of the upper discharge domain in the TDT, further increasing the weight of the mass block to enhance the output performance is not recommended, thereby the mass block with a weight of 50 N was finally determined as the optimal parameter. Furthermore, under all of the above optimized parameters, the peak Q_sc of the upper discharge domain in the TDT under the applied pressure of 30 N exhibited an increase of only 2.2% upon varying the sliding speed within the range of 0.1 m s⁻¹ to 1 m s⁻¹, as depicted in Fig. 3f, indicating that sliding velocity has limited influence on charge transfer efficiency. This phenomenon can be explained by the quasi-static capacitive model of sliding-mode TENGs. The short-circuit transferred charge Q_sc is determined primarily by geometric displacement and charge density, rather than by the sliding speed. In the TDT, corona discharge continuously sustains charge-density saturation through the space volume effect, thereby keeping Q_sc stable over a wide speed range. This displacement-dominated characteristic ensures that the TDT delivers consistent energy output even under variable-speed operating conditions. The detailed theoretical derivation is provided in Supplementary Note S1 (SI). After output performance optimization of the upper discharge domain, to evaluate the long-term operational stability, a continuous operation test of 36 000 cycles was conducted. As depicted in Fig. 3g, the upper discharge domain in the TDT exhibited an upward trend in the output at the initial stage and then entered a stable operating state. After 100 cycles, the output performance tended to be stable and maintained excellent consistency throughout the testing period. After 36 000 cycles, the performance degradation is only 4.6%, fully demonstrating excellent long-term operational reliability and durability of the optimized upper discharge domain in the TDT.

Output performance optimization of the lower discharge domain

This section systematically investigates the optimization of key structural parameters in the lower discharge domain of the TDT. As depicted in Fig. 4a, the considered parameters include the width of PDE (W₂), spatial position of PDE (I, II and III), and the discharge gap (G_ap). The spatial positions correspond to the PDE being placed beneath the PTFE and PA films in left (I), central (II) and right (III), respectively. As depicted in Fig. 4b, the lower discharge domain in the TDT exhibited an output performance trend of rising first and then tending to be stable with the increase of W₂, and the PDE with right-side positioning demonstrated superior output performance. Conversely, the PDE left-sided positioning may even lead to the reversal of output polarity. Theoretically, as depicted in Fig. 4c, the spatial position of the PDE determines the polarity source of its induced charges. When the PDE is located on the right side, it is mainly dominated by the positive charge region in the PU formation, resulting in the output of induced negative charges in the PDE owing to the space volume effect of the PU foam and dielectric-polarization of the acrylic substrate. When it moves to the left side, it is then dominated by the negative charge region, leading to the output of induced positive charges. This switching of the dominant induction region caused by the position change is the primary reason for the reversal of the output polarity. Therefore, when the width of the PDE continuously increases and exceeds the critical value, even though the area of the induced charge region will be increased to some extent due to the increase in the electrode area, the decrease in the distance between the two electrodes in a pair of PDEs will lead to an enhanced interference between regions of opposite polarities. Ultimately, this causes the output of the lower discharge region to stop increasing continuously. Based on the comprehensive consideration of device stability and energy conversion efficiency, the PDE with right-sided positioning and a width of 4 mm was finally determined as the optimal parameter combination, with the peak Q_sc reaching 0.24 µC. Furthermore, to visually verify the effect of the PDE position, color-changing silica gel that turns pink in a wet state was employed for visualization. After 300 discharge cycles, the silica gel above differently positioned PDEs showed significant differences. As depicted in Fig. 4d, when the PDE was located on the right side, the discharge activity in the lower domain was most active, causing that the silica gel completely turned blue due to sufficient dehydration. This observation strongly corroborates the theoretical analysis that right-sided PDE exhibits superior electrical output characteristics. The corona inception voltage of the PDE under different G_ap was systematically measured. As depicted in Fig. 4e, the corona inception voltage decreased significantly with the reduction of the discharge gap. As depicted in Fig. 4f, further tests indicated that when the discharge gap increases from approximately 0 mm to 5 mm, the peak Q_sc of the lower discharge domain decreased by about 58.3%, fully demonstrating the significant influence of the discharge gap on the output performance. Theoretically, the discharge channel of the lower discharge region exhibits similarities to those described by the classical Paschen's law for parallel-plate electrodes. In a uniform electric field between parallel plates, the breakdown voltage V_b can be expressed as follows.where p is the gas pressure, d is the spacing of the discharge gap, and A and B are the empirical constants of the primary and secondary Townsend ionization coefficients, respectively. γ_se is the secondary electron emission coefficient of the electrode material affected by the surface state.

Fig. 4 Output performance optimization of the lower discharge domain. (a) Schematic of the structural dimensions of the lower discharge domain. (b) The peak Q_sc of the lower discharge domain fabricated with different PDE widths and spatial positions. (c) Theoretical illustration of output polarity reversal caused by the change of the dominant region when the position of the PDE changes. (d) Verification of the effect of the PDE position using color-changing silica gel. (e) Verification of Paschen curves under different discharge gaps. (f) The Q_sc of the lower discharge domain under a discharge gap of 0–5 mm. (g) The peak Q_sc of the lower discharge domain under different external applied pressures and sliding speeds. (h) Durability and stability test of the lower discharge domain in the TDT.

Considering the non-uniform electric field generated by the PDE structure, it is necessary to introduce the field enhancement factor β for correction. Thereby, the maximum electric field strength E_max ≈ βV/d, and the corrected expression of V_b can be expressed as follows.

The discharge characteristics of the lower discharge region are highly consistent with the theoretical predictions of Paschen's law, confirming the applicability of this theory in the discharge process of the lower discharge domain. This discovery provides an important theoretical basis for optimizing the performance of the lower discharge region. Based on the comprehensive consideration of the output performance of the device and the feasibility of its structural design, the PDE with a discharge gap of 0.1 mm was finally determined as the optimal parameter, with the peak Q_sc reaching 0.24 µC. In the optimization of the applied load and the sliding speed of the lower discharge domain, similar to the upper discharge domain, as depicted in Fig. 4g, the peak Q_sc of the lower discharge domain under a sliding speed of 0.5 m s⁻¹ exhibited a significant positive correlation with the applied pressure within the range of 10 N to 50 N, and the peak Q_sc of the lower discharge domain under the mass block of 30 N showed a stable trend with the sliding speed within the range of 0.1 m s⁻¹ to 1 m s⁻¹. Based on the comprehensive consideration of output performance and friction loss, the mass block with a weight of 50 N is also appropriate for the lower discharge domain, with the peak Q_sc reaching 0.24 µC under the sliding speed of 0.5 m s⁻¹. Under the same external applied pressure and sliding speed, the output of the upper discharge domain is greater than that of the lower discharge domain, with the peak Q_sc reaching 2.64 µC and 0.24 µC under the mass block of 50 N and sliding speed of 0.5 m s⁻¹. Following the output performance optimization of the upper discharge domain, a continuous operation test of 36 000 cycles was conducted to evaluate the long-term operational stability of the lower discharge domain. As depicted in Fig. 4h, the lower discharge domain exhibited an initial running-in period after which its output performance stabilized. The results demonstrated a performance degradation of only 3.4% after the complete cycling test, indicating long-term operational reliability comparable to that of the upper discharge domain and further validating the durability of the overall TDT structural design. Although the TDT maintains high output stability after 36 000 cycles, the material aging induced by sustained corona discharge requires attention. As depicted in Fig. S14 (SI), scanning electron microscopy (SEM) characterization reveals that the copper electrode surface undergoes microscopic erosion and oxidation due to high-energy ion bombardment, resulting in increased surface roughness. Localized strong electric fields cause polymer molecular chain breakage and the formation of micro-cracks. The polyurethane foam structure exhibits partial collapse, which may weaken the space volume effect. Additionally, strong oxidizing substances such as ozone generated during discharge alter the surface chemistry of the dielectric materials, leading to the rearrangement of polar functional groups and thereby affecting long-term output stability. These combined physical and chemical aging mechanisms may contribute to the performance degradation of the device under prolonged operation.

Output performance of the TDT after optimization

Based on the collaborative optimization of the upper and lower discharge domains, the comprehensive output performance of the TDT was systematically measured and evaluated. First, to ensure reliable operation in real-world environments, we systematically evaluated the effects of humidity and temperature on the TDT's output performance. As depicted in Fig. 5a, humidity is the dominant environmental factor. With temperature fixed at 25 °C, the Q_sc exhibits a clear nonlinear decrease as relative humidity rises from 10% to 90%, primarily due to charge leakage through microscopic surface water films and an increased corona-discharge threshold under humid conditions. In contrast, as depicted in Fig. 5b, temperature has a much weaker influence, within the typical operating range of 0–40 °C at 50% RH, and Q_sc varies by only about 12.4%. For practical deployment in humid environments, we recommend adopting fully sealed packaging combined with hydrophobic surface treatments to suppress moisture ingress and maintain stable output. Subsequently, the correlation between the sliding distance and electrical output characteristics was investigated. As depicted in Fig. 5c, the peak Q_sc of the TDT exhibited an approximately linear growth trend with increasing sliding distance, while the I_sc also shows a corresponding increase, demonstrating the continuous charge accumulation and efficient release of the TDT during the sliding process, as well as the optimized energy conversion efficiency. Subsequently, as depicted in Fig. 5d, the charging performance of the optimized TDT for a series of standard capacitors from 4.7 µF to 220 µF at a rated voltage of 25 V were systematically measured. As depicted in Fig. 5e, the peak V_oc reached up to 4.8 kV under varying resistances, with a peak power of 54.07 mW achieved at a matched resistance of 300 MΩ, corresponding to the impedance-matching point of the TENG. Beyond this point, the current drops steeply while the voltage approaches saturation. The slight voltage decrease observed at very high resistances results from internal charge dissipation, a characteristic behavior clearly reflected in the logarithmic-scale plot. The corresponding single-cycle energy output reaches 8.2 mJ, as depicted in Fig. 5f. As depicted in Fig. 5g–i, a comparative study with other high-performance sliding TENGs demonstrates that the TDT developed in this work achieved a power density of 22.53 W m⁻² and an energy density of 3.42 J m⁻², exhibiting significant competitive advantages and the breakthrough in energy harvesting efficiency achieved by the TDT, as depicted in Table S1 (SI).

Fig. 5 Output performance of the optimized TDT. (a) The peak Q_sc of the TDT within the relative humidity range of 10% to 90%. (b) The peak Q_sc of the TDT at environmental temperatures ranging from 0 °C to 40 °C. (c) The peak Q_sc and I_sc of the TDT with increasing sliding distance. (d) The charging performance of the TDT for a series of standard capacitors. (e) The load characteristics of the TDT. (f) The V–Q curve of the TDT under matched resistance. Comparison of the output performance with that of reported high-output sliding TENGs, including (g) power density, (h) energy density and (i) M-E efficiency.

To verify the reproducibility of the TDT's performance, ten structurally identical TDT units were independently fabricated and tested. Statistical analysis shows that the average transferred charge Q_sc is 2.58 µC, with a relative standard deviation of only 3.1%. The average power density reaches 22.12 W m⁻², and 95% of the devices yield outputs stably distributed within the interval of [21.8, 22.5] W m⁻². The exceptionally low performance variation confirms the high reliability of the TDT output and further reflects the good robustness of the ternary-dielectric dual-domain architecture against minor fabrication variations. Additionally, by integrating a miniature tension-compression sensor into the linear sliding platform, the mechanical-to-electrical conversion efficiency of the TDT was quantitatively evaluated. Detailed calculation is provided in Note S2 (SI). Under the optimal matched load, the M–E efficiency reaches 33.8%, which notably exceeds those of conventional binary-dielectric structures (typically 10–20%) and other high-performance TENGs reported in the literature, as depicted in Table S2 (SI†). These results confirm that the TDT's integrated charge-generation, migration, and neutralization processes achieve high energy utilization efficiency, highlighting the advantage of the ternary-dielectric dual-domain architecture. It is noteworthy that the ternary-dielectric dual-domain synergy mechanism proposed in this work exhibits notable generality. Its underlying charge-cycling principle can be extended to diverse material systems and geometric configurations. By employing ternary material combinations with varying electrical gradients, novel porous migration media, and highly blocking dielectric layers, the mechanism can be adapted to specific application scenarios requiring full transparency, flexibility, or high-temperature tolerance. Furthermore, as the mechanism relies on charge cycling induced by relative displacement, its applicability extends from linear motion to rotary motion, curved-surface sliding, and even complex human motion, thereby enabling the design of diverse, high-performance energy harvesting systems.

Application demonstrations of the TDT

To demonstrate the practical application of the TDT in environmental energy harvesting, this work developed an innovative high-output rotary TDT that integrates a coaxial counter-rotating planetary gear structure with a pair of symmetric TDTs, as depicted in Fig. 6a(i). This model operates through the collaboration of three core subsystems, including the wind cups that convert wind energy into mechanical energy, a coaxial counter-rotating planetary gear that enables motion transmission and direction conversion, and a pair of symmetrical TDTs that accomplish electromechanical conversion. The overall and component pictures of the high-output rotary TDT are shown in Fig. S15 (SI). Specifically, the wind cups drive the planetary gear, causing relative rotation between the central gear and the outer ring gear. This drives the synchronous rotation of the PA/PTFE layer and the PDE layer, relative rotation of the PA/PTFE and PU layer, and relative rotation of the PDE layer and the PU layer, achieving efficient dual-layer triboelectric generation based on the symmetrical TDT structure, as depicted in Fig. S16 (SI). Meanwhile, the torque coupling generated by the counter-rotation can effectively suppress system resonance, reduce mechanical stress, and enhance operational stability. In addition, the planetary gear set has a torque amplification function, which can increase the output torque while maintaining a low cut-in wind speed, as depicted in Fig. S17 (SI). In summary, the coaxial counter-rotating planetary gear enhances the output performance of the TDT while ensuring operational stability and low-wind speed startup ability through a triple synergistic mechanism that multiplies the relative speed of the triboelectric layers, suppresses resonance, and amplifies torque. Additionally, as depicted in Fig. 6a(ii), to further validate the environmental adaptability and practical potential of the rotary TDT, on the one hand, a water-wheel-driven rotary TDT is developed for harvesting kinetic energy from water flow environments. On the other hand, a wind-water co-rotary TDT with a gear-free mechanical structure is developed for simultaneously capturing wind energy through an upper cup-type rotor and hydrodynamic energy through a lower impeller, which is mainly attributed to a single axis with differential counter-rotation in the wind-water co-rotary TDT. Actually, the TDT structure proposed in this work represents an abstract high-performance mechanism module designed to enhance the charge density limit in sliding-mode systems, rather than a single device constrained to a specific form. To address bidirectional or irregular mechanical excitation in real-world scenarios, it should be integrated into application-specific energy harvesting systems. The coaxial counter-rotating motion mode suppresses detrimental mechanical resonance, enabling the rotary TDT to efficiently convert irregular external excitations into regular mechanical forces applied to its internal units. This approach mitigates the performance degradation of individual units under non-regular motion, fully leverages the synergistic advantages of TDT in charge management, and enables omnidirectional high-efficiency electrical output.

Fig. 6 Application demonstrations of the TDT. (a) A schematic of a high-output rotary TDT in terms of the coaxial counter-rotating planetary gear structure, water-wheel-driven structure and wind-water co-rotary structure. (b) The Q_sc of the high-output rotary TDT with triboelectric dielectric staggered-phase and same-phase structures. (c) The I_sc of the high-output rotary TDT with different numbers of TDT units. (d) The power supply circuit for the LED. (e) 6000 LEDs are lighted by high-output rotary TDT. (f) The power management circuit. (g) 12 × 2 W commercial bulbs are powered by high-output rotary TDT with PMC. (h) 36 hygrothermographs are powered by high-output rotary TDT with PMC. (i) Voltage–time curve of driving 36 hygrothermographs by the high-output rotary TDT at a wind speed of 15 m s⁻¹.

For the rotary TDT with a coaxial counter-rotating planetary gear structure, a stepper motor was used to simulate wind-induced rotational excitation, and the output characteristics of a high-performance rotary TDT were systematically evaluated and demonstrated. Specifically, the influence of the dielectric phase and the number of TDT units on the output performance were mainly investigated. As depicted in Fig. 6b, with 4 TDT units on one side of the fixed disc, the dielectric staggered-phase structure exhibited a superior output performance. This design effectively suppressed the charge cancellation effect between a pair of PDEs caused by the polarization of the acrylic substrate during the same phase structure. The Q_sc values of staggered-phase and same-phase TDTs with different thicknesses of the acrylic substrate were measured. As depicted in Fig. S18 (SI), different from the same-phase structure, the thickness of the acrylic substrate in the staggered-phase structure has no significant impact on the electrical output characteristics, and the Q_sc of the TDT in the staggered-phase was significantly greater than that of the same-phase structure under the same thickness of the acrylic substrate. Based on a staggered-phase triboelectric dielectric arrangement, a fixed disc area and the thickness of the acrylic substrate, the influence of the number of TDT units on the output performance was investigated. As depicted in Fig. 6c and Fig. S19 (SI), both the Q_sc and I_sc of the high-output rotary TDT showed a rapid increase, followed by a gradual decrease. The optimal performance was achieved with 6 TDT units, after which a slight attenuation in the output was observed as the number of units was increased to 8. Fig. S20 (SI) shows the influence of the inner diameters of the coated materials where the PA/PTFE layer and PDEs are located on the electrical output of the high-output rotary TDT. In both cases, as the inner diameter decreased, the Q_sc of the high-output rotary TDT first increased and then decreased. This is mainly because an excessively large inner diameter leads to a reduction in the frictional interface area, while an excessively small inner diameter brings the adjacent PDEs closer, making it easier to cause breakdown and resulting in a decrease in the charge density. Therefore, the inner diameters of the disks where the PA/PTFE layer and PDEs are located are both selected as 50 mm.

In the motor-driven test, a power supply circuit for the LED was built, as depicted in Fig. 6d. Video S1 (SI), and Fig. 6e shows that the optimized system can light up 6000 LEDs simultaneously. Cooperating with the designed power management circuit (PMC), as depicted in Fig. 6f, which includes Zener diodes, current-limiting inductors, bidirectional thyristor, protection diodes, etc., the system can stably power 12 commercial bulbs with 2 W, as depicted in Fig. 6g and Video S2 (SI). The designed PMC conditions the high-voltage, high-impedance output of the TDT into a stable DC source. Its operation includes three stages, including energy buffering via a storage capacitor, threshold-controlled switching using Zener-SCR components for under-voltage lockout, and impedance matching through an LC network to enable efficient energy transfer to low-impedance loads. Benefiting from a passive-trigger design that minimizes switching losses, the PMC achieves an energy conversion efficiency of 87.2%. Detailed calculations of the energy efficiency are provided in Note S3 (SI). Subsequently, the I_sc of the high-output rotary TDT on varied wind speeds was measured, exhibiting a relatively stable output under different wind speeds, as depicted in Fig. S21 (SI). Furthermore, the high-output rotary TDT with PMC successfully achieved a continuous power supply for 36 hygrothermographs at a wind speed of 15 m s⁻¹, as depicted in Fig. 6h and Video S3 (SI). The voltage–time curve of the energy storage capacitor continuously accumulated during the energy storage phase stably maintained a periodic fluctuation around 1.5 V during the working phase, and gradually decayed only after power-off, as depicted in Fig. 6i. At a wind speed of 15 m s⁻¹, the charging rates of the high-output rotary TDT for capacitors with capacitance values from 1.1 mF to 6.8 mF demonstrated efficient charging performance, as depicted in Fig. S22 (SI).

To further validate the environmental adaptability and practical potential of the rotary TDT, a water-wheel-driven rotary TDT was further deployed in a natural river to demonstrate its reliable operation under realistic environmental or mechanical conditions. As demonstrated in Fig. S23 (SI) and Video S4 (SI), the system successfully captures flow kinetic energy and drives a commercial Bluetooth sensor via the integrated PMC. This demonstration confirms the stable operation and energy autonomy of the system under realistic, unsteady hydrodynamic conditions. Additionally, to demonstrate system-level IoT functionality, a wind-water co-rotary TDT was used to power a Bluetooth-enabled environmental monitoring node integrating light, temperature, and humidity sensors. As demonstrated in Fig. S24 (SI) and Video S5 (SI), the node operates stably under combined wind-water excitation. Real-time data transmission confirms reliable sensing, and manually shading the light sensor results in an immediate drop in recorded light intensity, verifying responsive operation fully powered by the TDT. This transition from basic power demonstration to an intelligent self-powered sensing node illustrates the new application space enabled by the TDT for multi-energy autonomous monitoring systems. The aforementioned results robustly verified the efficient electrical output performance and energy supply capacity of the high-output rotary TDT in practical applications. To evaluate the energy effectiveness of the rotary TDT in practical applications, its net energy balance was systematically calculated in this study. The overall energy balance of the system adheres to the law of energy conservation, where the total external mechanical work input (W_mech) equals the sum of the net available electrical energy (E_available), mechanical friction losses (Q_friction), and circuit conversion losses (E_loss(PMC)). Experimental results show that the system's electromechanical conversion efficiency is approximately 33.8%, and the power management circuit efficiency is approximately 87.2%, yielding an overall net energy balance rate of approximately 29.5%. This confirms the high efficiency of the system in capturing and converting ambient mechanical energy. The detailed energy distribution is summarized in Table S3 (SI), and the calculation process is provided in Note S4 (SI).

Conclusions

In summary, we have developed a ternary-dielectric dual-domain triboelectric nanogenerator (TDT) that provides an integrated and systematically validated approach to addressing the persistent challenges in inefficient interfacial charge migration and tail-charge accumulation in sliding-mode TENGs. By coupling corona discharge, space-volume-driven charge migration, and dielectric-polarization discharge within an integrated dual-domain architecture, the TDT simultaneously enhances interfacial charge generation, suppresses inherent tail charges, and recovers spatially dispersed residual charges. This multi-domain synergy establishes a novel charge-management pathway that enables highly efficient electromechanical energy conversion. Leveraging this mechanism-level innovation, the optimized TDT achieves a charge density of 1.2 mC m⁻², a power density of 22.53 W m⁻², and an energy density of 3.42 J m⁻², outperforming those of all previously reported sliding TENGs. Beyond device-level enhancements, we further demonstrate a high-output rotary TDT integrated with a coaxial counter-rotating planetary-gear module, capable of directly illuminating 6000 LEDs, stably powering twelve 2 W commercial lamps via an efficient power-management circuit, and continuously driving 36 environmental sensors under a simulated 15 m s⁻¹ wind field. These results collectively validate the scalability, robustness, and practical applicability of the proposed architecture. Overall, this work advances sliding-mode TENGs by establishing a charge-generation and charge-regulation strategy that effectively improves charge utilization and energy-conversion efficiency. The TDT architecture not only provides a promising direction for high-output micro-energy harvesting but also serves as a versatile and scalable platform for next-generation self-powered IoT devices, distributed sensor networks, and environmental monitoring systems.

Experimental

Fabrication of the TDT

The TDT is mainly composed of a slider and a stator. The slider adopts a layered design with an upper and a lower part. The upper slider uses an acrylic plate with dimensions of 80 mm × 40 mm × 2 mm as the substrate. On the acrylic surface, a PTFE film with dimensions of 60 mm × 16 mm × 0.1 mm, a PA film with dimensions of 60 mm × 4 mm × 50 µm, and two copper foil electrodes arranged at intervals are pasted in parallel and in sequence. The lower slider uses an acrylic substrate of the same size, and a pair of copper foils used as promoting discharge electrodes (PDE) are fixed right below the corresponding positions of the upper slider electrodes. The upper and lower sliders are precisely aligned and mechanically fastened by acrylic locating pins at the four corners. The stator part uses an acrylic plate with dimensions of 150 mm × 60 mm × 1 mm as the substrate, and its upper surface is covered and adhered with a PU foam with dimensions of 150 mm × 60 mm × 2 mm.

Fabrication of the high-output rotary TDT

The structure of the high-output rotary TDT is mainly composed of a central layer, two PDE layers, two PU layers and a planetary gear. The central layer utilizes an annular acrylic substrate with an outer diameter of 300 mm, an inner diameter of 10 mm, and a thickness of 2 mm. PTFE films, PA films, and copper electrodes are symmetrically attached to both sides of the substrate. Specifically, four PTFE films and four PA films with a 72° central angle, along with eight copper electrodes with an 18° central angle, are arranged in a staggered symmetrical configuration on both surfaces. The PDE section consists of upper and lower annular acrylic plates, each with an outer diameter of 300 mm, an inner diameter of 10 mm, and a thickness of 2 mm. Both plates are equipped with eight copper electrodes, each corresponding to an 18° central angle and aligned with the rotational direction of the central layer films. The PU layers are constructed using disc acrylic substrates with an outer diameter of 300 mm and an inner diameter of 10 mm, each coated with a 2 mm thick PU foam. The final assembly is achieved by axially stacking, from top to bottom, the planetary gear, the upper PDE layer, the upper PU layer, the central layer, the lower PU layer, and the lower PDE layer. The central gear of the planetary gear, the central layer, and the PDE layers are rigidly connected to the wind cups, whereas the PU layers are fixed to the outer ring of the planetary gear set, establishing a counter-rotating motion relative to the central and PDE layers.

Fabrication of the water-wheel-driven rotary TDT

The water-wheel-driven TDT consists of a water turbine, a sealed waterproof chamber, and three functional layers, including the central layer, PDE layer, and PU layer. The central layer is an annular acrylic substrate with an outer diameter of 190 mm and an inner diameter of 10 mm, patterned with four PTFE films and four PA films, each spanning 72°, along with eight 18° copper electrodes in a staggered arrangement. The PDE layer comprises eight copper electrodes mounted on the inner wall of a cylindrical acrylic chamber with an outer diameter of 200 mm and an inner diameter of 190 mm. The PU layer is formed from a disc acrylic substrate coated with 2 mm thick PU foam. These layers are assembled radially inside the sealed chamber, with the PDE layer bonded to the chamber wall. The chamber, central layer, and PDE layer rotate with the water turbine, while the PU layer is fixed to a central shaft, enabling relative motion for energy conversion.

Fabrication of the wind-water co-rotary TDT

The wind-water co-rotary TDT integrates wind cups and a water turbine with a sealed chamber housing three stacked layers: the central, PDE, and PU layers. The central layer is an annular acrylic substrate with an outer diameter of 280 mm, patterned with alternating PTFE and PA films (each 72°) and eight 18° copper electrodes. The PDE layer consists of eight copper electrodes attached to the inner surface of an acrylic chamber with an outer diameter of 300 mm. The PU layer is a disc substrate coated with 2-mm-thick PU foam. These components are axially assembled within the chamber. The chamber, central layer, and PDE layer are fixed to the wind cups, while the PU layer is connected to the water turbine, establishing counter-rotational motion for dual-energy harvesting.

Measurement and characterization

A comprehensive experimental platform was established to systematically evaluate the performance characteristics of the TDT. A linear actuator (H80BN) combined with a stepper motor driver (DM542) was used for the output performance testing and optimization of the linear-mode TDT. A stepper motor (86BYG250D) equipped with a frequency converter was employed for the precise speed regulation and output characteristic research of the high-output rotary TDT. A blower with a frequency converter was utilized to simulate the actual wind field environment and verify the energy supply capacity of the high-output rotary TDT. The short-circuit current and transferred charge were measured using a Keithley 6514 electrometer, and the open-circuit voltage was recorded using a Trek model 370 electrostatic voltmeter. Finite element simulations of the potential distribution were carried out based on the COMSOL Multiphysics 6.0.

Author contributions

S. G.: conceptualization, methodology visualization, formal analysis, funding, writing – original draft and writing – review and editing. Y. S.: investigation, data curation, methodology, visualization, software development and writing – review and editing. L. L.: formal analysis and visualization. Y. L.: investigation and visualization. J. W.: methodology, supervision and writing – review and editing.

Conflicts of interest

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

Data availability

The datasets supporting the findings of this study, including raw electrical measurements, COMSOL simulation files, and processed analysis data, are available from the corresponding author upon reasonable request.

Additional experimental videos and supplementary data are provided in the supplementary information (SI). Supplementary information includes figures (Fig. S1–S24), tables (Tables S1 and S2), notes (Note S1–S4) and video (S1–S5). See DOI: https://doi.org/10.1039/d5ee07296d.

Acknowledgements

This work was supported by the Scientific and Technological Research Program of Chongqing Municipal Education Commission of China (KJQN202000628), and Beijing Key Laboratory of High-Entropy Energy Materials and Devices, Beijing Institute of Nanoenergy and Nanosystems (No. GS2025MS027).

Notes and references

1. J. Wang, W. Ding, L. Pan, C. Wu, H. Yu, L. Yang, R. Liao and Z. L. Wang, ACS Nano, 2018, 12, 3954–3963.
2. H. Wu, Z. Wang, B. Zhu, H. Wang, C. Lu, M. Kang, S. Kang, W. Ding, L. Yang, R. Liao, J. Wang and Z. L. Wang, Adv. Energy Mater., 2023, 13, 2300051.
3. S. Gao, H. Wei, J. Wang, X. Luo, R. Wang, Y. Chen, M. Xiang, X. Chen, H. Xie and S. Feng, Nano Energy, 2024, 122, 109323.
4. L. Wang, Y. Zhang, X. Zhang, X. Cheng, S. Zhai, X. Bi, H. Li, Z. Wang and T. Cheng, Adv. Energy Mater., 2025, 15, 2403318.
5. S. Yong, J. Wang, L. Yang, H. Wang, H. Luo, R. Liao and Z. L. Wang, Adv. Energy Mater., 2021, 11, 2101194.
6. K. Dong, X. Peng and Z. L. Wang, Adv. Mater., 2020, 32, 1902549.
7. J. Yang, K. Hong, Y. Hao, X. Zhu, Y. Qin, W. Su, H. Zhang, C. Zhang, Z. L. Wang and X. Li, Adv. Mater. Technol., 2025, 10, 2400554.
8. Y. Jiang, X. Liang, T. Jiang and Z. L. Wang, Engineering, 2024, 33, 204–224.
9. S. Li, Y. Hu, P. He, J. Qu, J. Zhao, J. Mo and Z. Fan, Nano Energy, 2025, 133, 110436.
10. A. Baburaj, S. Jayadevan, A. K. Aliyana, N. K. SK and G. K. Stylios, Adv. Sci., 2025, 12, 2417414.
11. X. Li, Z. Zhao, Y. Hu, Y. Gao, L. He, W. Qiao, B. Zhang, Y. Xu, Z. L. Wang and J. Wang, Energy Environ. Sci., 2024, 17, 1244–1254.
12. C. Cao, Z. Li, F. Shen, Q. Zhang, Y. Gong, H. Guo, Y. Peng and Z. L. Wang, Energy Environ. Sci., 2024, 17, 885–924.
13. G. Li, J. Wang, Y. He, S. Xu, S. Fu, C. Shan, H. Wu, S. An, K. Li, W. Li, P. Wang and C. Hu, Energy Environ. Sci., 2024, 17, 2651–2661.
14. W. Guo, Y. Lei, X. Zhao, R. Li, Q. Lai, S. Liu, H. Chen, J. Fan, Y. Xu, X. Tang, Q. Sun and Q. Sun, Nano Energy, 2024, 131, 110324.
15. J. Han, J. Li, X. Zhang, L. Zhao and C. Wang, Chem. Eng. J., 2024, 489, 151493.
16. Y. Wu, X. Wang, Y. Wang, Y. Nan, H. Xu, H. Zhou, M. Ren, J. Duan, Y. Huang and B. Hou, Adv. Eng. Mater., 2023, 25, 2201442.
17. K. Xiao, W. Wang, K. Wang, H. Zhang, S. Dong and J. Li, Adv. Funct. Mater., 2024, 34, 2404744.
18. Y. Dong, N. Wang, D. Yang, J. Wang, W. Lu and D. Wang, Adv. Funct. Mater., 2023, 33, 2300764.
19. H. Liu, X. Li, Z. Li, H. Xiang, Z. Bai, H. Wu, G. Liu and H. Zhou, Nano Energy, 2024, 124, 109464.
20. H. Wu, S. Fu, W. He, C. Shan, J. Wang, Y. Du, S. Du, B. Li and C. Hu, Adv. Funct. Mater., 2022, 32, 2203884.
21. L. Tan, Q. Zeng, F. Xu, Q. Zhao, A. Chen, T. Wang, X. Tao, Y. Yang and X. Wang, Adv. Mater., 2024, 36, 2313878.
22. Q. Zhao, H. Wu, J. Wang, S. Xu, W. He, C. Shan, S. Fu, G. Li, K. Li and C. Hu, Adv. Energy Mater., 2023, 13, 2302099.
23. X. Ji, D. Zhang, L. Zhou, Z. Xu, Y. Wu and C. Yang, Chem. Eng. J., 2024, 485, 149753.
24. Z. X. Wang, Y. H. Wu, W. B. Jiang, Q. Y. Liu, X. B. Wang, J. W. Zhang, Z. Y. Zhou, H. W. Zheng, Z. F. Wang and Z. L. Wang, Adv. Funct. Mater., 2021, 31, 2103081.
25. J. Oh, M. Kang, J. Park, T. Y. Kim, K. Park and J. Seo, Adv. Electron. Mater., 2024, 10, 2400013.
26. Q. Zeng, A. Chen, X. Zhang, Y. Luo, L. Tan and X. Wang, Adv. Mater., 2023, 35, 2208139.
27. S. Gao, H. Wei, R. Wang, X. Luo, Y. Liu, C. Huang, Y. Lei and X. Deng, Nano Energy, 2024, 129, 110060.
28. S. Gao, R. Wang, H. Wei, X. Luo, J. Zhang and X. Chen, Nano Energy, 2024, 129, 110070.
29. S. Fu, H. Wu, W. He, Q. Li, C. Shan, J. Wang, Y. Du, S. Du, Z. Huang and C. Hu, Adv. Mater., 2023, 35, 2302954.
30. W. He, C. Shan, H. Wu, S. Fu, Q. Li, G. Li, X. Zhang, Y. Du, J. Wang, X. Wang and C. Hu, Adv. Energy Mater., 2022, 12, 2201454.
31. C. Shan, W. He, H. Wu, S. Fu, Q. Tang, Z. Wang, Y. Du, J. Wang, H. Guo and C. Hu, Adv. Energy Mater., 2022, 12, 2200963.
32. C. Shan, K. Li, H. Wu, S. Fu, A. Liu, J. Wang, X. Zhang, B. Liu, X. Wang and C. Hu, Adv. Funct. Mater., 2024, 34, 2310332.
33. K. Li, S. Gong, S. Fu, H. Guo, C. Shan, H. Wu, J. Wang, S. Xu, G. Li, Q. Zhao, X. Wang and C. Hu, Energy Environ. Sci., 2024, 17, 8942–8953.
34. IEEE Std 539-2020. Definitions of Terms Relating to Corona and Field Effects of Overhead Power Lines. IEEE: New York, USA, 2020.

---

Footnote

† S. Gao and Y. Sun contributed equally to this work.

---