Progress in techniques for improving the output performance of triboelectric nanogenerators

Abstract

As a new technology based on the combined effects of contact electrification and electrostatic induction, triboelectric nanogenerators (TENGs) are already widely studied for energy harvesting and novel sensor design. In recent years, with the progress of material synthesis and device technology, researchers have been developing various working mechanisms and designs to enhance the output performance of TENGs. In this review, the research advance in improving the output performance of TENGs through different strategies is comprehensively reviewed. The main contents of this review include three aspects: surface material modifications of TENGs, mechanical design, and power management. Firstly, the basic principles and working modes of TENGs are introduced. Secondly, the surface material selection and treatment methods of TENGs are classified. According to the different treatment methods, the modification methods of surface materials are divided into physical and chemical modifications. Meanwhile, the production of high-performance degradable materials is considered to enhance the performance of TENGs, including cellulose-based materials, artificial polymer materials and bio-based materials. Then, the methods for improving the output power of TENGs through mechanical design are classified, including multilayer structures, TENG networks and adjusting the mechanical motion frequency of TENGs. In addition, the existing methods for improving the power supply efficiency of TENGs by using switch management are summarized. Power management methods are classified according to the types of switches, including mechanical switches and electronic switches. Finally, based on the current progress, we discuss the critical problems and systematically suggest future research directions and challenges.

---

Broader context

As human society and technology continue to advance, global energy challenges become increasingly difficult. In addition, the number of mobile electronic devices has been surging rapidly in recent decades. Although the power consumption of commonly seen individual mobile electronic devices is usually at the watt level, the total energy consumption is incalculable owing to their quantity (in billions). It is urgent to seek new energy substitutes to relieve the pressure of energy demand. The triboelectrification effect, as one of the major energy transducing mechanisms, is first utilized to construct triboelectric nanogenerators (TENGs). The working principles of TENGs are triboelectricity and electrostatic induction. Due to their ability to collect environmental kinetic energy, TENGs have been widely used to solve the problem of distributed energy for Internet of Things devices. With the development of TENGs, they have been used in many applications, such as environmental energy collection, industrial production, environmental monitoring and biomedical equipment. With the development of Industry 4.0 and the Internet of Things, TENGs are expected to be widely used in automated production lines and industrial equipment. By collecting energy from mechanical motion, TENGs provide autonomous power for sensors and smart devices. Due to the limitation of material surface and low-frequency environment, the achievable output hinders the practical application of TENGs. In this paper, the methods for improving TENG power generation are comprehensively summarized from different angles. Toward future advancement in this field, pressing problems that need to be solved and onward research directions are also discussed.

Introduction

As human society and technology advance, global energy challenges become increasingly difficult, including the dramatic plunge in fossil fuel reserves and deterioration of the ecological environment.¹ According to recent studies, it is necessary to reduce global greenhouse gas emissions by 40–70% by 2050 to limit global temperature increases within 2 °C.² Therefore, it is urgent to seek new energy substitutes to relieve the pressure of energy demand. In addition, the number of mobile electronic devices has surged rapidly in recent decades. Although the power consumption of commonly seen individual mobile electronic devices is usually at the watt level, the total energy consumption is incalculable owing to their quantity (in billions). The emergence of self-powered technology effectively addresses the demand issues for the Internet of Things devices for distributed energy.

In 2012, the team led by Prof. Wang originally proposed the concept of triboelectric nanogenerators (TENGs), advancing human cognition and application based on the triboelectric effect to a new era. Such technology displays the capability to convert low-frequency mechanical energy into electrical energy via both electrostatic induction and contact electrification effects.^3–5 TENGs have the advantages of high efficiency, low cost, simple manufacturing and availability of various raw materials. In comparison to electromagnetic and piezoelectric devices, TENGs are able to generate a higher output voltage owing to their extremely low capacitance.^6–8 Meanwhile, TENG-based devices and sensors exhibit significant merits in flexibility and portability.⁹ In recent years, TENGs have been widely employed in various fields of energy supply, thus reducing the reliance of human beings on traditional energy sources,^10,11e.g., batteries. Researchers’ exploration of clean energy involves solar energy,^2,12,13 wind energy,^14–16 blue energy^17,18 human kinetic energy,^19,20etc. TENGs show extensive potential for distributed and portable energy technologies, including implantable medical devices, environmental monitoring, and human–computer interaction.^21–26 In the field of transportation, especially where the parts of cars and trains need more sensitive sensors because of many parts with special shapes, TENGs have shown broad application prospects.^27–29 However, due to the limitation of material surface and low-frequency environment, the practical application of TENGs is hindered by the achievable output. Therefore, enhancing the output of TENGs of mechano-electric transduction has been a heated subject, whether to intensify sensitivity or boost harvested power.

For the aforementioned purposes, various methods have been proposed, e.g., surface material modifications,^30–34 structural optimization,^35–39 and circuit management.^40–44 Xin et al.⁴⁵ conducted a series of experiments with polyimide (PI) intermediate layers with different thicknesses, and studied the influence of intermediate layers on the output of TENGs. Experimental results showed that the output voltage of the TENG with 0.001 mm thick PI interlayers was increased by 5 times. Shin et al.⁴⁶ changed the triboelectric properties of polymer surfaces through a series of atomic-level chemical functionalization employing halogen and amine. To induce triboelectricity on polyester surfaces, Shin et al. functionalized TENG interfaces with aryl silane-terminated electron-accepting element halogen, and obtained adjustable surface charge density. Zhang et al.⁴⁷ developed bionic-fin-structure assisted multilayer structured TENGs (BFM-TENGs) for harvesting energy from the seabed. The BFM-TENG employed a multilayer contact structure, resulting in a peak power density of 444 W m⁻³ under ideal testing conditions. Xu et al.⁴⁸ developed an optimized spherical TENG element-based coupling network for harvesting water wave energy. The charge output of the coupled cells was more than 10 times that of the uncoupled cells. Kara et al.⁴⁹ developed a parallel synchronous switching inductive harvesting integrated circuit (p-SsHi) for harvesting energy from mechanical motion within the frequency range of 1 Hz to 5 Hz using TENGs. The p-SsHi utilized a step-down converter and switched capacitor DC-DC converter to convert the mechanical motion into electric output, and yielded a power output of 722 μW to the load for 4 ms.

In this study, methods from different (material, mechanical and electric) areas to enhance TENG power generation are summarized, and the current challenges are then respectively analyzed related to this technology. The framework of this review is illustrated in Fig. 1. The second section presents the fundamental working principle of TENGs. The third section summarizes various methods for enhancing the performance of TENGs from the aspect of material surfaces, e.g., surface physical modifications, functional group introduction, ion injection, chemical doping, and degradable material consideration. The fourth section classifies the mechanical configuration methods of TENG structures in detail, including increasing the contact area inside a single TENG, increasing the mechanical frequency and forming TENG networks. The next section introduces the application of power management technology in TENGs, and mainly summarizes the research in recent years from the angle of circuit switches. Finally, we will provide advisory perspectives on future research directions and challenges, focusing on the fundamental theory, material properties, mechanical optimization, and switch management of TENGs.

Fig. 1 Schematic overview of classification and application for enhancing TENG device output.

The basic principle of TENGs

Triboelectric effect

The working principle of TENGs is contact electrification (CE) and electrostatic induction. CE occurs when two surfaces contact and separate to generate electrostatic charges. Over the years, people have utilized various technological methods to study and characterize the process of triboelectrification. Many hypotheses have been put forward to explain the charge transfer in the process of contact friction, including the electronic transfer mechanism,⁵⁰ ion transfer mechanism,⁵¹ and free radical stabilization effect model.⁵² However, the source particles of surface charge generated by friction are still in dispute.⁵³ Recently, Wang et al.⁵⁴ proposed an electron cloud overlap model to explain the widespread existence of the CE effect at solid–solid and solid–liquid interfaces (Fig. 2(a)). Before two materials come into contact, their respective electron clouds remain separated without overlapping. When two atoms belonging to two materials are in close contact with one another, electron clouds overlap between the two atoms to form bonds. When an external force is applied, the energy barrier between the two materials decreases due to the overlapping of strong electron clouds. Electrons are then transferred between the two atoms to produce CE.

Fig. 2 Triboelectric effect and four basic modes of TENGs. (a) Triboelectrification effect, electron cloud overlapping model and electronic state. ΔE is the potential barrier between the two atoms. E_A and E_B are the two atoms belonging to two materials A and B, respectively. (b) Model and equivalent circuit of VCTENG. (c) Model and equivalent circuit of LSTENG. (d) Model and equivalent circuit of SETENG. (e) Model and equivalent circuit of FTENG.

Maxwell's displacement current

Maxwell's equations (MEs) are partial differential equations proposed by James Clerk Maxwell to describe the relationship between electric field, magnetic field, charge density and current density. Among them, the second term of displacement current is the theoretical basis of TENGs. That is, the polarization field generated by the electrostatic charge on the surface of the medium causes the current.⁵⁵ The indispensable communication and microelectronic technologies in modern society are based on the principles and theoretical implementation of Maxwell's equation system. The specific expressions are as follows:⁵⁶

∇·D = ρ_f (Gauss's law) (1)

∇·B = 0 (Gauss's law for magnetism) (2)

where E is the electric field, B is the magnetic field, H is the magnetization field, ρ_f is the free charge density, J_f is the free current density, and D is the displacement field.

D = ε₀E + P (5)

where P is the polarization field density and ε₀ is the vacuum permittivity. For each isotropic medium, eqn (5) is defined as

D = εE (6)

where ε is the permittivity of the dielectrics.

The second term in eqn (4) is the Maxwell's displacement current, which is defined as⁵⁷

The research shows that the fundamental theoretical basis of TENGs is the second term in eqn (7).

In addition, for the general theoretical explanation of TENGs, Wang derived an extended set of Maxwell's equations. The specific expressions are as follows:⁵⁴

∇·D′ = ρ_f − ∇·P_S (8)

∇·B = 0 (9)

where D = D′ + P_S = ε₀E + P + P_S, with P_S the mechano-driving created polarization owing to the relative movement of the charged media surfaces by mechanical triggering.

Furthermore, Wang derived the expanded Maxwell’s equations, which can be employed to explain the general theory of TENGs.⁵⁸ It is not only applicable to moving charged solid and soft medium with acceleration, but also to charged fluid or liquid medium.^55,58,59

∇·D′ = ρ_f − ∇·P_S (12)

∇·B = 0 (13)

where v(t) is the translation velocity of the medium and v_r = v_r(r, t) is the rotation velocity of the medium. Note that eqn ((12)–(15)) are regarded as the general MEs for shape-deformable, mechano-driven, slow-moving media at an arbitrary velocity field. These full MEs for a mechano-driven media system describe the coupling among three fields: mechano-electricity-magnetism.

To sum up, based on Maxwell's displacement current theory, the output power of TENGs can be calculated and enhanced by different processing methods. Such a theory is instructive for the design and application of efficient TENGs.

Electromechanical coupling modeling of TENGs

The output process of TENGs is described as follows: when two materials come into contact and then separate, the friction generates a potential difference between them, causing one material to become positively charged and the other negatively charged. This charge separation generates alternating current by driving the charge flow between two electrodes.^4,52,60,61 According to the configuration and driving mode, Wang et al.^62,63 classified TENGs into the following four types: vertical contact-separation (VCTENG),^64,65 lateral sliding (LSTENG),^66,67 single-electrode (SETENG)^68,69 and free-standing (FTENG).^70,71

Fig. 2(b) illustrates that VCTENGs are comprised of two distinct dielectric films stacked together. When these two dielectric films contact each other, surface charges with opposite signs will be formed on the two surfaces. When displacement is applied, the surfaces of the two materials come into contact, resulting in charge transfer at the contact region of the surfaces. When the charged surfaces are separated, a potential difference is generated between the two electrodes in an open circuit. If two electrodes are connected by a load, electrons will flow from one electrode to the other through the load, forming an opposite potential difference to balance the electric field. When the two dielectric films contact again, the potential difference formed by friction charges disappears and electrons will flow back.⁷² In the equivalent circuit, every two nodes are connected by electric field lines, and will form an equivalent capacitance. However, since the surface of dielectric material 2 is assumed to be infinite relative to the thickness, node 2 will shield the electric field lines between node 1 and node 3. Therefore, only two capacitors exist in the whole electrostatic system in the figure. VCTENGs show the advantages of simplicity in design and high instantaneous output power. The inherent output performance of VCTENGs is deduced through the principles of electrodynamics:⁷³

where d₀ is the effective thickness constant, which is defined as the sum of the ratio of the thickness of all dielectric materials between two electrodes to their relative dielectric constants, S is the contact area, σ is the electric charge density, Sσ is the amount of electricity generated by triboelectrification and x is the distance between two triboelectrically charged layers.

LSTENGs are driven by a force parallel to the horizontal direction of the thin film (Fig. 2c). Since air gaps are no longer needed in the process of charge separation, LSTENGs are more beneficial to packaging than TENGs in other working modes. Such modes are particularly suitable for rotary sliding⁷⁴ and packaging structures.⁷⁵ In the equivalent circuit, nodes 1, 2 and nodes 5, 6 respectively constitute two constant capacitances C₁ and C₂. Nodes 2, 3 and nodes 4, 5 form two triboelectric layers. In addition, a variable capacitor C₃ is formed between nodes 3 and 4. Under the condition of incomplete separation, the intrinsic output performance of LSTENGs is expressed as⁷⁶

where ω is the width of the whole structure, l is the geometric dimensions of two dielectrics in the length direction, and x is the lateral separation distance.

SETENGs usually consist of an independent layer and a pair of stationary electrodes (Fig. 2(d)). One of the friction surfaces of SETENGs is not limited by wire connection, so TENGs in this mode have been widely applied to collect energy from different sources, such as airflow,⁷⁷ raindrops⁷⁸ and human walking.^79,80 Under the condition of open circuit, the total charges of nodes 1, 2 and 3 are −σωl, σωl and 0, respectively. Capacitors C₁, C₂ and C₃ between the two electrodes are connected in series. From the conservation of capacitance and charge of each dielectric, the intrinsic output performance of SETENG is expressed as⁸¹

FTENGs generate electricity by the triboelectric effect using a pair of symmetrical electrodes. This working mode is applied to devices with different structures, e.g., grating structure⁸² and ball structure.⁸³ In Fig. 2(e), since capacitors C₁, C₂ and C₃ are connected in series, the total capacitance between two electrodes (#1 and #4) can be expressed bywhere g is the interval between two electrodes. In the case of the minimum achievable charge reference state, the short-circuit transfer charge and the open-circuit voltage can be expressed byTherefore, the basic equation of FTENG can be expressed by⁸⁴To sum up, this paper introduces four theoretical models of TENG modes in detail to help beginners to understand TENGs further. At the same time, such theoretical models provide a theoretical basis for the performance enhancement technology of TENGs.

Material selection

Material selection is widely used to select suitable triboelectric materials to enhance the electrical output of TENGs and other electrostatic applications.⁸⁵ Different materials exhibit varying capabilities to acquire or lose electrons, so generating electricity through friction heavily depends on the choice of triboelectric materials. In 1757, John Carl Wilcke published the first triboelectric sequence table of electrostatic charge, which listed common triboelectric materials in the order of polarity strength.⁸⁶ As the distance between the two materials in the sequence is farther, the corresponding charge transfer amount is larger.^87,88 However, such a triboelectricity sequence shows limitations in quantitatively characterizing the triboelectric properties of materials. In 2019, Wang's team measured the triboelectric charge density of triboelectric materials according to the triboelectric interaction between different materials and liquid metal mercury in contact-separation mode.⁶⁰ Many common materials were characterized by standardized measurement, and a quantitative triboelectric sequence was established. Such a quantitative triboelectric sequence lays the foundation for the research of energy harvesting, sensing and human–machine interfacing of the measured materials. To improve the output performance of TENGs, researchers have proposed various strategies, e.g., surface treatment and fabrication of composite materials. According to the methods and principles of treatment, the modification process is divided into physical and chemical modifications.⁸⁹ In addition, under the background of sustainable development, with the progress of green energy technology, biodegradable friendly materials have been arousing widespread interest of researchers.⁹⁰

Physical modification

Physical modifications of TENG interfaces are common methods to improve output. For instance, the surface charge density of materials can be increased by increasing the contact area.^91,92 Some researchers have attempted to introduce micro/nano structures on the surfaces of triboelectric materials to enhance the surface charge density by increasing the effective contact area.^93–95

Microstructure formation through templating

Template-assisted microstructure formation is a typical approach for designing and fabricating the interface microstructure of TENGs. Such a method does not require surfactant coating and high-vacuum equipment, which makes it a simple and economic choice. Due to the above advantages, several researchers have been exploring the formation of the microstructure interface of TENGs. For example, Mule et al.⁹⁶ utilized a hot stamping method to create micro-patterns on the surface of PTFE with sandpaper as a template. This approach effectively enlarges the contact area of PTFE, leading to improved performance of the TENG. When the load resistance was 10 MΩ, the maximum power reached 7.6 W m⁻². To develop a cost-effective and eco-friendly backup power source for gas sensors, Zhang et al.⁹⁷ utilized sandpaper as a template to fabricate nanostructured gelatin films for TENGs. The resulting film not only significantly enhanced the performance of TENG devices but also demonstrated excellent water degradability. TENG devices using such thin films obtained a significant peak voltage of 400 V and output power of 6.16 mW. Zhao et al.⁹⁸ produced a TENG using aluminum foil and Kapton film, and linearly polished the surface of the triboelectric layer with sandpaper (Fig. 3(a)). They observed that polishing the triboelectric layer four times with 3000-grit sandpaper resulted in the best performance, increasing the output voltage by 3 times and the output power by 5 times.

Fig. 3 Schematic diagram of the template used to improve the performance of TENGs. (a) Schematic diagram of grinding the surface of the triboelectric layer with sandpaper and TENG output voltage of different surface morphologies. (b) The manufacturing method of the complementary pattern of sandpaper. (c) Output voltage of the TENG made of triboelectric materials with different particle sizes. (d) Replicating the schematic diagram of the micro-nano structure with silicon mold. (e) Output performance of the TENG with different micro-nano structures.

Because template methods offered significant advantages in the processing efficiency and diversity of TENGs, many strategies have been put forward aiming at breakthroughs in electric outputs. Zhang et al.⁹⁹ patterned the copper electrodes and PDMS films by using sandpaper template (Fig. 3(b)). As can be seen from Fig. 3(c), the V_oc of TENGs increased with the increase of sandpaper particle size. When the sandpaper templates used for the PDMS film and copper electrode possesses the same particle size of 10 000, the corresponding maximum output short-circuit current (I_sc) density was 3.89 mA m⁻², the open-circuit voltage (V_oc) was 200 V, the transferred charge amount was 76 nC, and the power density was 4.36 W m⁻². Zhang et al.⁹⁴ fabricated microstructures on a silicon substrate by a combination of lithography and wet etching (Fig. 3(d)). Then, micro-nano dual-scale structures were copied from the silicon mold to produce PDMS films. The influence of the roughness of triboelectric surfaces on the output performance of TENGs was studied. Experimental results showed that the open circuit voltage and the short circuit current of the PDMS film with micro-nano structures were higher by 61.4% and 118% than those of the flat PDMS film, respectively (Fig. 3(e)). Likewise, Seol et al.¹⁰⁰ fabricated micro/nano structures on the contact interface of TENGs with silicon molds. Through the direct deformation of the TENG interface structure, the relationship between pressure and voltage in TENGs was clarified. Tcho et al.¹⁰¹ studied the influence of the shape of surface nanostructures on the output of TENG. They fabricated ordered dome-shaped and columnar nanostructures with precisely controlled height and diameter on PDMS thin films. The results showed that TENGs with dome-shaped PDMS exhibited greater force sensitivity than TENGs with columnar PDMS.

The above work demonstrates that microstructure formation through templating is an effective approach to increase the surface charge density by increasing the actual contact area. However, some limitations remain, particularly with regard to the reduced material lifespan, e.g., the increase of wear degree of surface roughness with the increase in the working time of TENG devices.

Fabricating patterns through etching

Etching has been widely used to fabricate surface micro/nano structures due to its high manufacturing efficiency, low pollution and processing diversity.^92,102 Etching can be divided into dry etching and laser etching. The related research results within the last decade are shown in Fig. 4.

Fig. 4 Etching based TENGs. (a) PI films were modified by in situ two-step plasma irradiation. (b) Comparison of output performance of modified PI films and untreated PI films. (c) Schematic diagram of the laser etching method. (d) Output of different droplets when they hit the triboelectric layer with 40° inclination angle of the TENG device. (e) Photograph of the prefabricated TENG on the surface after laser direct writing. (f) Microscopic images at different scratching rates. (g) Voltage comparison of the TENG before and after optimization.

Energy sources for dry etching mainly include inductively coupled plasma and neutral beams, which can achieve efficient and accurate surface treatment effects without heating the materials.³⁴ Dry etching improves the friction stability and wear resistance of materials, and has intrigued many researchers. Sun et al.¹⁰³ used in situ two-step Ar + O₂ reactive ion etching to etch the surface of PI films, to improve the electron affinity of PI films (Fig. 4(a)). Additionally, they compared the outputs of the PI-based TENG under two different irradiation sequences. As shown in Fig. 4(b), I_sc and V_oc of the TENG irradiated with the films were increased by 5.65 and 3.1 times, respectively. Similarly, to enhance the efficiency of flexible TENGs, Kim et al.¹⁰⁴ constructed a TENG by fluorocarbon plasma etching of PDMS-CNT. The physical effects of the etched triboelectric surface morphology were utilized to improve the output performance of the TENG. Finally, they proposed a TENG prepared using PDMS containing 4% carbon nanotubes and etched with fluorocarbon for 60 seconds. The output voltage of this TENG was 248.7% higher than that of its counterpart. Phan et al.¹⁰² used a q-switched pulsed laser to treat surfaces of aluminium thin films by direct writing. By inductively coupled plasma reactive ion etching, the wrinkled nanostructures were formed on the surface of PTFE films. Compared with the TENGs of the original aluminium thin-film device, V_oc and I_sc were improved by 97.33% and 41.17%, respectively.

Laser etching is a micro-machining method, which applies the principle of photochemical reaction to process the surface of a medium to obtain the required patterns. Such a method shows characteristics of high stability, no consumables and no pollution. Therefore, laser etching has been widely applied to fabricate the interface microstructure of TENGs. Li et al.¹⁰⁵ utilized laser etching to design a micro-pillar array on the triboelectric surface of the TENG, increasing the frictional surface area. After tests, the amplitude values of V_oc and I_sc reached 56.37 V and 1.008 μA, respectively. When an external load of 100 MΩ was applied, the output power reached 56.82 μW. Liang et al.¹⁰⁶ used a CO₂ laser to fabricate superhydrophobic microstructures on the surface of silicone rubber to achieve a superhydrophobic surface TENG (Fig. 4c). In addition, the authors used a pump to automatically control the raindrop speed to collect energy. As shown in Fig. 4(d), the output voltage of the device after laser etching driven by water droplets reached 30 V and the maximum output current reached 14 μA. Yang et al.¹⁰⁷ used a laser etching method to build micro-nano structures on triboelectric surfaces of the arched TENG (Fig. 4(e)). Fig. 4(f) presents microscopic images of the surface before and after treatment. During laser etching, the parameters of ultraviolet nanosecond laser etching were optimized. As can be seen from Fig. 4(g), the average V_oc of the arc-shaped CS-TENG fabricated at an optimized rate of 111 mm s⁻¹ reached more than 6 V, i.e., it was improved by 50%.

The above research shows that the charge density of triboelectric layers is significantly improved through the formation of patterns on TENG interfaces by etching. However, limitations have been reported in the application of TENGs at solid–solid interfaces. For example, etching will damage the triboelectric surface, resulting in unstable electric output. Therefore, to enhance the durability of TENG devices, it is necessary to select materials with appropriate modulus, strength and elasticity.

Nanowire/rod structure formation through electrospinning

Electrospinning exhibits the advantages of a large surface area and adjustable micro/nano fibre composition. Therefore, electrospinning has been widely used in the preparation of TENG interfaces for application in fields such as filtration¹⁰⁸ and biosensors.^109,110

Electrospinning is a versatile technique by which natural polymers, polymer alloys, inorganic nanoparticles, metals and ceramics can be spun into fibres. Therefore, some researchers have embedded different materials on the surface of friction layers to change the surface electrification and enhance the electrical properties of TENGs. For example, Huang et al.¹¹¹ embedded graphene oxide (GO) into a polyvinylidene fluoride (PVDF) matrix by electrospinning, and formed a book-like TENG with poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV) nanofibers. Experimental results showed that it was feasible to improve the performance of the TENG with composite materials. The maximum output voltage and current of the TENG were 340 V and 78 μA, respectively. Chen et al.¹¹² made a kind of SiO₂/poly(vinylidenefluoride-co-trifluoroethylene) P(VDF-TrFE) composite film with a layered micro-nano structure by electrospinning and applied it to a tensile TENG. Compared with the pure P(VDF-TrFE) nanofiber film, the surface roughness of the composite film was significantly improved. After tests, the maximum instantaneous output voltage produced by the assembled TENG was about 170 V, which was three times higher than that of the pure P(VDF-TrFE) film. Mathew et al.¹¹³ designed a surface-enhanced vibration energy harvester via pure nylon-66 electrospun nanofibers. As shown in Fig. 5(a), the prepared nylon-66 solution was injected into an electrostatic spinning device through a syringe, and electrospinning was carried out at room temperature. The nylon-66 with 15 wt% concentration showed the best effect after 8 hours of spinning, and I_sc reached 10–15 μA (Fig. 5(b)). Zhang et al.¹¹⁴ compounded PDMS and barium titanate nanoparticles (BT NPs) into coaxial nanofiber core layers by the electrospinning in situ curing method (Fig. 5(c)). To obtain better output performance, they improved the dispersion of BT NPs in the PDMS core layer. The optimized TENG showed an average peak voltage of 1020 V and an I_sc of 29 μA (Fig. 5(d)).

Fig. 5 Electrospinning high performance TENGs. (a) Schematic diagram of the fabrication process of electrospinning. (b) Properties of the TENG with different concentrations of nylon-66 solution. (c) Schematic image of the coaxial electrostatic spinning device. (d) Output current and voltage of two kinds of the synthesized TENG. (e) Schematic diagram of the fabrication process of electrostatic spinning. (f) Output voltages of PVDF nanofibers with different concentrations of MoS₂/CNT.

Electrospinning is capable of producing ultra-thin fibres with unique characteristics and high performance. Electrospun nanofibers show high specific surface area and porosity, which increases the working area. Due to the above advantages, electrostatic spinning has been widely employed in wearable TENGs.^115–117 Xing et al.¹¹⁸ prepared a multi-ply and stabilized triboelectric yarn with a frictional charge using silica aerogel introduced into the electrospinning process of PTFE nano-coating. This study provided another way to develop wearable power sources for high-temperature-resistant textile products. Sun et al.¹¹⁹ used electrospinning to produce a flexible lightweight TENG (PMC-TENG) with MoS₂/carbon nanotube (MC) doped PVDF as the friction substrate (Fig. 5(e)). By optimization of the electrospinning process, the output voltage of PMC-TENG reached 300 V (Fig. 5(f)). At the same time, PMC-TENG maintained normal charging capacity and ran stably for more than 3000 cycles. Su et al.¹²⁰ used electrospinning to fabricate a flexible composite substrate TENG (F-TENG) based on silk fibroin protein and carbon nanotubes (CNTs). When the matching resistance was 30 MΩ, the power density of F-TENG using a vibrator and a clapper was 140.99 μW cm⁻² and 317.4 μW cm⁻², respectively.

By the generation of long and continuous nanofibers on the surface of the triboelectric layer, electrospinning improves the triboelectric effect and generates more charges.^121–123 Notably, the parameter setting and environmental conditions of electrospinning nanofibers are critical to electric outputs of TENGs, e.g., solution temperature, air humidity and temperature in a spinning environment.

Fabrication of TENGs via 3D printing

As a rapid prototyping manufacturing method, 3D printing shows a distinct advantage in complex prototypes. Based on 3D printing, TENG devices can be moulded in one step, eliminating the steps of cutting and mould manufacturing.^124–126 The relevant representative work within the last decade is shown in Fig. 6 and 7.

Fig. 6 3D printing high-performance TENGs. (a) Schematic diagram of the 3D printing process. (b) Output charges of fluffy columns of 0, 2, 4 and 6: vertical direction mode and rotation direction mode. (c) Schematic drawing of the AP-TENG. (d) The output of the AP-TENG and the traditional TENG. (e) Manufacturing method and structure of self-repairing 3D printing TENGs. (f) Open circuit voltage of the TENG based on pure PDMS and PDMS/PTFE composite triboelectric layer.

Fig. 7 Schematic diagrams of 3D printing for TENGs. (a) 3D printing of rigid polymers based on the combination of DIW. (b) TENG manufacturing of polymer tube-assisted FDM based on biomaterials and PDMS. (c) and (d) Manufacturing of triboelectric layers based on the DIW method. (e) and (f) Screen printing. (g) Photopolymer resin assisted stereolithography.

3D printing is known for rapid prototyping and diversified printable materials, which has attracted extensive interest of researchers in TENG manufacturing with different shapes. Yoon et al.¹²⁷ used 3D printing to design a biomimetic TENG (BV-TENG) with an intestinal villi structure, as depicted in Fig. 6(a), which effectively increased the contact surface area. To ensure full contact between friction layers, the PTFE powder was utilized as the triboelectric material. With the increase in the number of fluffs in the BV-TENG, the triboelectric area and the generated triboelectric charges increased. As shown in Fig. 6(b), the charge amount of the 24-villus BV-TENG was about 5 times and 4 times higher than that of the 0-villus BV-TENG, respectively. Qian et al.¹²⁸ fabricated an all-printed TENG (AP-TENG) with nanocellulose as the raw material (Fig. 6(c)). The printed three-dimensional shape and freeze-dried aerogel structure contributed to increasing the contact area, surface roughness, and mechanical elasticity of the device, thereby improving the electrical response. Compared with the template methods, the voltage output was increased by nearly 175% (Fig. 6(d)). Zheng et al.¹²⁹ developed PTFE/PDMS composite elastomer ink, which was used in the 3D printing method of direct ink writing (Fig. 6(e)). As shown in Fig. 6(f), the V_oc of the TENG based on pure PDMS was 12.0 V, but after replacing PDMS layers with PTFE/PDMS composite layers, V_oc was increased to 47 V.

Fig. 7 summarizes the commonly used 3D printing technologies in preparing TENGs. At present, four kinds of 3D printing technologies are used in TENG fabrication and research: fused deposition modeling (FDM), direct ink writing (DIW), stereolithography appearance (SLA) and screen printing. FDM is applied widely to produce microstructures of filaments, films and arrays. Haque et al.¹³⁰ proposed a method to manufacture TENGs with FDM and DIW methods (Fig. 7(a)). Triboelectric layers were made of rigid polymer acrylonitrile butadiene styrene, and DIW was used to prepare triboelectric layers based on the flexible polymer structure. Yu et al.¹³¹ used polymer pipes (polyamide (PA) pipes and tubular polyethylene (PE)) as the carrier of powder based on the FDM method, and the triboelectric layer with nanoporous structure was formed by PDMS 3D printing (Fig. 7(b)). Similarly, Mohapatra et al.¹³² fabricated TENG devices using FDM 3D printing. The maximum output voltage of the TENG device was enhanced by adjustment of the layer thickness, filling ratio and surface pattern. Experimental results showed that when the thickness of triboelectric layers with a 20% filling rate was 0.1 mm, the highest V_oc of the TENG was 8.9 V.

DIW is a method of manufacturing three-dimensional parts by extrusion of ink layer by layer along a digitally defined path through a nozzle. It is reported that DIW has been used to manufacture portable and wearable flexible TENGs. Wang et al.¹³³ applied DIW 3D printing to fabricate fully flexible single-electrode triboelectric nanogenerators (FFTENGs) with complex shapes and 3D structures. The results showed that the highest V_oc and I_sc of the FFTENG with standard 30 × 30 mm² size and a complex 3D structure, with a matching resistance of 80 MΩ, reached 60 V and 0.23 μA, respectively. Li et al.¹³⁴ used silicone elastomer as viscoelastic ink via the DIW method, and directly extruded it onto aluminium foil through the nozzle tip to form a pattern with microstructure on the electrode (Fig. 7c). Experimental results showed that TENGs prepared by 3D printing exhibited the highest output performance when the thickness of triboelectric films was 150 μm. Ahmed et al.¹³⁵ used DIW to deposit electrodes and triboelectric layers and prepare a flexible and stretchable TENG (Fig. 7(d)). Based on the single-electrode mode, the instantaneous output power density and open circuit voltage of the TENG reached 0.2 mW m⁻² and 8 V, respectively.

The screen printing method can be used not only on hard objects, but also on soft and elastic films. At the same time, screen printing is not limited by the texture of printed pages. Based on the above characteristics, many articles have been published to study the production of thin film TENGs. Godard and his team¹³⁶ used a screen-printing method to print 10 layers of thin poly(vinylidene fluoride-trifluoroethylene) on a polymer substrate (Fig. 7(e)). The TENG made via thin poly films generated 0.97 mW of power at 33 Hz and with an area of only 2.4 cm². Hong et al.¹³⁷ prepared a honeycomb pattern spacer by ultraviolet curing screen printing (Fig. 7(f)). SLA utilizes an optical process to aggregate materials, which is usually suitable for processing flexible products with complex plane patterns and shapes. Yang et al.¹³⁸ used an SLA-based 3D printer to prepare TENG backplane support layers with a special pattern and used a photopolymer resin composed of (meth)acrylate monomers, oligomers and a photoinitiator (Fig. 7(g)). With an external load of 1 MΩ, the TENG generated a maximum power of 0.13 W at a low frequency of 3 Hz, which lighted at least 50 LEDs. This strategy improves the direct contact and grounding connection between metal and metal.

Table 1 provides an overview of the literature on physical modifications of TENG interfaces. Various methods, such as templating and electrospinning, have been used to prepare microstructures that increase the surface area and roughness of TENG interfaces, resulting in improved output performance. As shown in Table 1, when using surface-treated triboelectric materials to fabricate TENGs, most researchers choose to use the VCTENG working mode. The main reason is that most micro-nano structures are easy to deform during the continuous friction of TENGs, which is the bottleneck limiting their lifespan. Compared with the other three working modes, the VCTENG exerts less influence on triboelectric materials. In addition, how to determine the quantitative relationship between the microstructure of physical modification and the generated charge needs further study. At present, the research mainly focuses on experimental analysis, and few studies have been done on the theoretical mechanism of enhancing the electric output of TENGs. Future research should focus on the influence of the theoretical mechanism of TENGs on output performance. For example, accurately describe the theoretical mechanism and model of TENG output change in the experiment.

Table 1 Comparison of properties of TENGs after physical modifications

Surface treatment method	Triboelectric material 1	Triboelectric material 2	Electrical output	Size (contact area)	Working mode	Lifespan test	Ref.
Templating	Gelatine film	PI	400 V, 49 μA	5 cm × 5 cm	VCTENG	10 000 cycles	97
	Kapton	Al	1500 V, 75 μA	20 cm × 20 cm	VCTENG	1.2 × 10^5 cycles	98
	PDMS	Cu	200 V, 3.89 mA m^−2	3 cm × 3 cm	VCTENG	30 days	99
	PDMS	Al	465 V, 13.4 μA cm^−2	2 cm × 4 cm	VCTENG	—	94
Etching	PTFE	Al	148 V, 9.6 μA	5 cm × 5 cm	VCTENG	3 months	102
	Silicone rubber	Raindrop	30 V, 14 μA	14 × 14 square arrays	SETENG	—	106
	PTFE	Cu	6 V	1 cm × 1 cm	VCTENG	—	107
	PDMS/CNT	Al	77.8 V, 1.98 mW	2 cm × 3 cm	VCTENG	10 000 cycles	104
	PDMS/CB/MWCNTs	FTO	49 V, 1.008 μA	2 cm × 2 cm	VCTENG	5000 cycles	105
Electrospinning	P(VDF-TrFE)	Raindrop	36 V, 10 μA	1.5 cm × 1.5 cm	SETENG	—	112
	Nylon-66	Cu	350 V, 300 μA	4 cm × 4 cm	VCTENG	10 000 cycles	113
	PHBV nanofiber mat	PDMS/BT@PVDF nanofiber mat	1020 V, 29 μA	9 cm × 6 cm	VCTENG	—	114
	MoS2/CNTs	Nylon	300 V, 11.5 μA	6 cm × 6 cm	VCTENG	6 months	119
	CNT-silk mix	PET	276 V, 9.20 μA	1.5 cm × 3 cm	VCTENG	—	120
3D printing	PTFE	Acrylonitrile butadiene styrene	1.7 μC m^−2	Cylindrical structure (diameter: 42 mm height:38 mm)	VCTENG	10 000 cycles	127
	Cellulose nanofibers (CNFs)	PDMS	35 V, 0.3 μA	3 cm × 3 cm	VCTENG	—	128
	PDMS/PTFE	Al	150 V, 3.2 μA	Barrel-shaped structure	VCTENG	1000 cycles	129
	Silicone	NBR/skin	60 V, 0.23 μA	3 cm × 3 cm	SETENG	1000 cycles	133

---

Chemical modification

Different from surface physical modifications, surface chemical modifications change the potential distribution and even polarity of triboelectric layers, which help improve the output performance of TENGs. Some researchers have attempted to chemically modify the triboelectric surface and develop innovative surface chemical modification methods, including functional group grafting, dielectric performance engineering, ion injection and ultraviolet (UV) radiation.

Functional group grafting. In principle, the output performance of TENGs is related to the electronic affinity of atoms on the surface of triboelectric materials. For example, nitro groups are easy to be negatively charged due to their superior electron-withdrawing ability when contacted with other groups. In contrast, methyl groups exhibit a positive charge when in contact.¹³⁹ Functional group grafting enhances the ability to gain or lose electrons of triboelectric layers by introducing electron accepting groups and electron-donating groups on the surface of the triboelectric material.^139,140 Related research is shown in Fig. 8 on functional group grafting within the last decade.

Fig. 8 Chemical surface modifications by introducing functional groups. (a) Schematic diagram of triboelectrification tendency after functionalization. (b) Open circuit voltage and short circuit current of the TENG composed of PI and modified ITO. (c) Schematic diagram of the surface functionalization process. (d) Three-dimensional structures of UIO-66 and ligands with different functional groups. (e) Schematic diagram of surface functionalization treatment using halogen and aminated molecules. (f) Output voltages of the bare PET:aminated PET pair and the bare PET:halogenated PET pair.

A self-assembled monolayer (SAM) is known for its simplicity, effectiveness and low cost, and is widely used in the surface functionalization of TENG interface materials. Wang et al.¹⁴¹ functionalized the surface with a SAM to enhance the electric output of TENGs. Four functional groups of hydroxyls (–OH), ester (–COOCH₃), amine (–NH₂) and chlorine (–Cl) were formed on the surface of gold (Au) through thiol-based SAM functionalization. Among them, amine groups greatly increased the surface chemical potential, and thus greatly enhanced the tendency of charge donation. The output power of the corresponding TENG was increased by nearly 4 times. Byun et al.¹⁴² used a SAM system to prepare the surface of materials containing electron-donating groups, neutral groups and electron-accepting groups. The schematic diagram of the functionalized triboelectrification tendency is shown in Fig. 8(a). The first peak of V_oc and I_sc of bare indium tin oxide (ITO) was negative, indicating that PI was negatively charged and the bare ITO substrate was positively charged. On the other hand, CF₃-ITO was negatively charged, which induced additional charges of opposite polarity to NH₂- and bare ITO (Fig. 8(b)). Shin et al.¹⁴³ proposed a method to change the triboelectric charge sequence of materials (Fig. 8(c)). The surface of the PET film was functionalized with poly(l-lysine) solution or trichloro(1H,1H,2H,2H-perfluorooctyl) silane. Because of the opposite triboelectricity of the two triboelectric surfaces, the output performance of the TENG manufactured with these two functionalized surfaces was significantly improved, i.e., under the force of 0.5 MPa, the maximum V_oc was 330 V and the maximum short-circuit current density reached 270 mA m⁻². Compared with its counterpart, the output power of the treated TENG was increased by about 4 orders of magnitude.

Other researchers have attempted to use composite materials with various functional groups, providing another view of functional groups that influence the output performance of TENGs. Wen et al.¹⁴⁴ studied the influence of fillers with different functional groups on the output performance of TENGs based on the Zr-based metal–organic framework UiO-66-R (R = –H, –NH₂, –NO₂, –Br) family as composite materials. As shown in Fig. 8(d), UiO-66 was constructed from Zr₆O₄(OH)₄(CO₂)₁₂ clusters of secondary building units (SBUs) connected with the BDC (1,4-benzenedicarboxylic acid) connector. UiO-66-R was constructed by the utilization of ligands with different functional groups –R (R = H, NH₂, NO₂, Br). Among them, –NO₂, –Br and –NH₂ showed electron-losing and electron-donating abilities, respectively. The experimental results showed that the composite materials UiO-66-NO₂ and UiO-66-Br exhibited higher output performance, followed by UiO-66 and UiO-66-NH₂. Shin et al.⁴⁶ used a series of atomic-level chemical functionalizations of halogen and amine to change the triboelectric properties of polymer surfaces (Fig. 8(e)). The surface of PET was functionalized by aryl silane terminal electron acceptor halogen to induce electronegativity. For the triboelectric positive side, the surface was functionalized with aminated molecules. Fig. 8(f) shows the TENG output voltages of the aminated-PET:PET pair and halogenated-PET:PET pair. After measurement, the signal of the TENG with the aminated PET:PET pair showed positive polarity. In contrast, the signal of the TENG with the halogenated PET:PET pair showed negative polarity.

High dielectric particle doping. The dielectric constant of triboelectric layers significantly affects the electric output of TENGs.¹⁴⁵ Doping nanoparticles with a higher dielectric constant inside the triboelectric layer raise the dielectric constant of the triboelectric material, thus improving the output performance of TENGs.¹⁴⁶ For example, Lai et al.¹⁴⁷ developed a new triboelectric material, NaNbO₃PDMS, which modified the electronegativity of PDMS by varying the content of NaNbO₃. When the mass ratio of NaNbO₃ was 7 wt%, the outputs of the NaNbO₃PDMS-based TENG reached the highest with a V_oc of 550 V, an I_sc of 16 μA, and an effective power density of 5.5 W m⁻². Additionally, molecules with different electron repulsion and attraction abilities were introduced into triboelectric materials to enhance electrical performance. Yu et al.¹⁴⁸ reported a method of AlO_x doping by sequential infiltration synthesis of polymers. The output voltage of the device composed of PDMS doped with AlO_x reached 2.3 V.

Some researchers have explored doping methods to improve the binding force and charge-holding capacity of triboelectric materials. Chen et al.¹⁴⁹ modified triboelectric materials of TENGs by filling and forming pores with high dielectric constant nanoparticles (SiO₂, TiO₂, BaTiO₃ and SrTiO₃) (Fig. 9(a)). Compared with the TENG based on a pure PDMS film, the power of the TENG based on the modified triboelectric material was increased by more than 5 times at 2.5 Hz. Seung's team¹⁵⁰ reported a high-performance nanocomposite, which consisted of high-dielectric ceramic materials such as barium titanate and P(VDF-TrFE). As shown in Fig. 9(b), the charge absorption performance of P(VDF-TrFE)/BTO was improved by 18 times. The experimental results showed that the V_oc of the TENG with P(VDF-TrFE)/BTO was 1130 V under the thrust of 6 kilogram-force at 5 Hz. After using graphene oxide solution (GO) to modify PDMS, Harnchana et al.¹⁵¹ added sodium dodecyl sulfate (SDS) as an activator to form the PDMS@GO@SDS composite material with a porous structure (Fig. 9(c)). The introduction of oxygen functional groups of the GO molecule and anionic head groups of the SDS molecule enhanced the negative charge of PDMS. The output voltage and current of the porous PDMS@GO@SDS composite TENG were 3 times higher than that of the planar PDMS. Song et al.¹⁵² used tetramethylpiperidinooxy-oxidized cellulose nanofibers (TOCN) and calcium copper titanate nanoparticles (CaCu₃Ti₄O₁₂, CCTO) to prepare composite aerogel films (Fig. 9(d)). The dielectric constant of the TOCN composite film was significantly improved by doping CCTO (a material with a high dielectric constant). Experimental results showed that the composite film-based TENG with 20% CCTO content performed best under various experimental conditions. The V_oc of the TENG based on the TOCN/CCTO composite film was 3.37 times higher than that of pure TOCN-based TENG devices (Fig. 9(e)).

Fig. 9 Improving the output performance of TENGs by doping high dielectric particles. (a) Schematic diagram of the fabrication process of sponge PDMS. (b) Schematic diagram of the TENG based on ferroelectric composite materials. (c) Schematic diagram of the porous PDMS@GO@SDS composite TENG and its output voltage. (d) Fabrication process of the TOCN/CCTO aerogel film. (e) The open circuit voltage of TOCN-based TENG devices.

Ion injection and radiation. Ion injection increases the triboelectric surface charge and significantly improves the electrical output performance of TENGs.¹⁵³ The methods of ion injection include anti-static gun injection and high voltage corona discharge polarization.¹⁵⁴ Oxygen and argon are two kinds of ions commonly used in ion injection due to their multiple charge ions.¹⁵⁵ Wu et al.¹⁵⁶ proposed a surface charge injection technique to realize directional charge accumulation by excitation circuit. Experimental results showed that the output charge density on PI films reached 880 μC m⁻² after charge injection. Lee et al.¹⁵⁷ fabricated a high-performance TENG based on PI via ion injection. Compared with TENGs based in bare PI films, the output power was increased by 7 times (Fig. 10(a) and (b)). Seo et al.¹⁵⁸ used polydimethylsiloxane-carbonyl iron (PDMS-Fe) to prepare a single-electrode mode TENG (C-TENG) with ciliated microstructures (Fig. 10(c)). On this basis, the authors implanted ions into PDMS-Fe 10 wt% composite films with an anti-static gun. After ion injection, the closed-circuit voltage output of C-TENG was doubled to 59.9 V (Fig. 10(d)). Zhou et al.¹⁵⁹ fabricated a new electret film TENG (E-TENG) by corona charging. The effects of different corona charging voltages, charging distances, and charging times were comprehensively studied on the output performance of the E-TENG. Electrons and negative ions were injected into the surface of the film by corona charging. After corona charging, both the current and voltage of the E-TENG were increased by 7 times.

Fig. 10 Ion injection and ultraviolet radiation methods. (a) Schematic diagram of the fabrication process for the PI-based TENGs. (b) V_oc of TENGs with different polymers. (c) Working mechanism of the fabricated single-electrode mode TENG. (d) Output voltage of the TENG based on PDMS-Fe 10 wt% with/without ion injection. (e) Three-dimensional diagram of ultraviolet radiation. (f) I_sc and V_oc of the DTENG under dark and ultraviolet irradiation.

Ultraviolet radiation is widely used to modify the surface properties of carbon-based and oxide materials, e.g., wettability, adhesion and biocompatibility.¹⁶⁰ Such a method involves exposing the material to UV radiation, which results in the formation of functional groups on the surface and the emergence of new bonds. Liu et al.¹⁶¹ achieved the continuous and rapid fabrication of high-performance TENGs through an ultrafast UV radiation strategy, and combined it with multiple electrode arrays to collect low-energy sources. Ren et al.¹⁶² proposed a p–n junction-based direct-current triboelectric nanogenerator (DTENG) with ultraviolet light modulation (Fig. 10(e)). Triboelectric layers of the TENG were composed of two materials, p-Si and n-GaN. Owing to the coupling effect of the triboelectric effect and photovoltaic effect, ultraviolet light enhanced the carrier concentration and increased the output current and output voltage. Experimental results showed that I_sc and V_oc under ultraviolet irradiation were about 13 times and 4 times higher than that of the comparison group (Fig. 10(f)). Luo et al.¹⁶³ prepared a TENG (FC-TENG) by application of the fluorocarbon polymeric coating with jagged topographies as a triboelectric layer. They templated the surface microstructure of liquid crystal polymer networks by using 365 nm ultraviolet light. Under the force of 10 N and the frequency of 4 Hz, the V_oc and I_sc of the FC-TENG were higher than those of the TENG without surface microstructure, which increased from 106.7 V and 0.69 μA to 194.9 V and 1.28 μA, respectively.

As illustrated in Table 2, this section summarizes chemical treatment methods to improve the electrical output of TENGs, including functional group grafting, high dielectric particle doping, ion injection, etc. Similar to the physical surface modification methods, most researchers chose the working mode of VCTENG to improve the lifespan of TENGs. In addition, compared with several chemical surface modification methods, the lifespan of high dielectric particle doping and ion injection is longer than that of other chemical surface modification methods. However, many significant challenges remain in these chemical modification methods. For example, lifespan and performance are a difficult trade-off for surface-modified TENGs. When the fabricated TENGs are working, treated functional groups may lose their function due to polishing or abrasion of the triboelectric surfaces. Material selection is a simple solution. Flexible materials show a low frictional force. Diamond-like carbon film has a low frictional coefficient and high hardness. Interfacial liquid lubrication is another strategy to reduce material wear. For doped high dielectric nanoparticles, the influence of their specific gravity on the composites needs to be further explored to optimize the doping dose of such nanoparticles. TENGs treated by ion injection usually present high stability, but the cost is high and the operation process is relatively more complicated. In future research, more attention should be given to flexible and economic ion injection methods.

Table 2 Comparison of properties of TENGs after chemical modifications

Surface treatment method	Triboelectric material 1	Triboelectric material 2	Electrical output	Size (contact area)	Working mode	Lifespan test	Ref.
Functional group grafting	FEP	Au	240 V, 1.75mA m^−2	4.5 cm × 4.5 cm	FTENG	—	141
	PFA, PTFE, PDMS, etc.	ITO·SiO2, Al2O3	20 V, 2.2 μA	0.785 cm^2	VCTENG	—	142
	PET, F-PET	FOTS, P-PET, ITO	330 V, 270 mA m^−2	2 cm × 2 cm	VCTENG	7200 cycles	143
	PDMS@UiO66-NO2	Cu	191 V, 17.3 μA	2 cm × 2 cm	VCTENG	30 days	144
	PET: Cl	PEI(b)-PET	520 V, 110 mA m^−2	2 cm × 2 cm	VCTENG	500 cycles	46
High dielectric particle doping	NaNbO3PDMS	Cu	550 V, 16 μA	2 cm × 2 cm	VCTENG	—	147
	AlOx-doped PMMA	ITO	3.1 V, 5.5 μA m^−2	1 cm × 1 cm	VCTENG	—	148
	CS-PDMS film	Cu	338 V, 9.06 μA m^−2	2 cm × 2 cm	VCTENG	15 000 cycles	149
	P(VDF-TrFE)	Al	1130 V, 1.5 mA	3 cm × 3 cm	VCTENG	—	150
	PDMS@GO@SDS	PEN	438 V, 11 μA cm^−2	2 cm × 2.5 cm	SETENG	—	151
	TOCN/CCTO	PVDF	152 V, 33.8 μA	2 cm × 2 cm	VCTENG	50 000 cycles	152
Ion injection	PI	PDMS	860 μC m^−2	2 cm × 2 cm	VCTENG	10 800 cycles	157
	PDMS-Fe 10 wt%	Wind	70 V, 250 nA	3 cm × 3 cm	SETENG	1600 cycles	158
	PI, PTFE	Acrylic, Cu	9.04 W m^−2	2.5 cm × 2.5 cm	VCTENG	42 000 cycles	156
	PTFE	Al	2450 V, 108 μA	7 cm × 7 cm	VCTENG	150 000 cycles	159
UV radiation	N-GaN	P-Si	3.35 V, 154 nA	1 cm × 1 cm	LSTENG	—	162
	MAPDMS-F	Raindrop	249 μC m^−2	5 cm × 5 cm	SETENG	—	161
	PDMS	Fluorocarbon Polymer	194.9 V, 1.28 mA	7 cm × 7 cm	VCTENG	10 000 cycles	163

---

Degradable material consideration

In addition to elucidating the physical and chemical modifications of materials, we hereby present state-of-the-art considerations of the development and application of degradable triboelectric materials in TENGs.

Fabrication of biodegradable TENGs with high performance displays great significance for reducing the environmental pollution caused by fossil fuels. The extensive application of heavy metals and non-degradable materials causes waste of resources and environmental pollution.¹⁶⁴ For example, since the 1980s, the amount of plastic garbage drifting into the ocean has been increasing yearly, attracting the attention of marine scientists and environmentalists all over the world.¹⁶⁵ So far, marine debris has become a global problem, with small islands composed of garbage visible in almost all ocean currents.¹⁶⁶ TENGs show great potential to be used in distributed sensing and blue energy harvesting. To achieve sustainable development and mitigate the adverse effects on the ecosystem, adopting renewable and easily prepared biodegradable materials for TENG applications is of critical importance. In recent years, natural fibre materials, artificial biodegradable materials, and biobased materials have been increasingly considered for the development of degradable TENGs (DB-TENGs), which offer high performance and environmental friendliness.^167–169

Cellulose-based degradable material. As a renewable and degradable natural molecular material, cellulose can be obtained from a wide range of sources and is friendly to the environment.¹⁷⁰ Because cellulose contains a large number of oxygen atoms, it has a high tendency to lose electrons. TENGs are fabricated with commonly used electronegative materials to obtain better output performance. In 2013, Zhong et al.¹⁷¹ fabricated a paper-based nanogenerator (pNG). The peak power density of the pNG was 90.6 μW cm⁻², and the maximum voltage reached 110 V. Such a study showed a pioneering role of the paper-based self-power supply system in energy collection and sensing. In 2016, Yao et al.¹⁷² proposed a TENG using fibreboard made of recycled cardboard fibres to collect energy from human gait. When the TENG was impacted by normal gait, it produced 30 V and 90 μA of electric outputs. In 2017, Peng et al.¹⁷³ utilized a new PDMS/cellulose nanocrystal flake composite as triboelectric layers of TENGs. Compared with its counterpart, the electrical performance of the TENG based on such composite was improved by 10 times, and the instantaneous output power density reached up to 0.76 mW cm⁻².

Each glucose unit of the cellulose molecule has three hydroxyl groups, which makes cellulose possess certain chemical reactivity. It is an effective way to raise the output performance of TENGs by introducing strong electrophilic or electron-repelling functional groups with rich hydroxyl groups on the surface of CNFs. Therefore, many studies have reported that CNF materials are modified by simple manufacturing methods to strengthen the electrical output of TENGs. For example, Yang et al.¹⁶⁸ proposed a single-electrode mode TENG based on CNFs. As shown in Fig. 11(a), the TENG comprised PVDF-coated cellulose filter paper (CFP) as a triboelectric layer and conductive Cu-coated CFP as an electrode layer. Due to the microstructure on the surface of PVDF-coated CFP, such a TENG gave significant electrical outputs, i.e., the maximum output voltage of the TENG was 192 V (Fig. 11(b)). Nie et al.¹⁷⁴ utilized chemical functionalization to enhance the surface polarization and hydrophobicity of CNFs. The authors used triethoxy-1H,1H,2H,2H-tridecafluoro-n-octylsilane (PFOTES) to modify CNF functional groups (Fig. 11(c)). The polarity of CNFs modified by PFOTES was significantly improved. As shown in Fig. 11(d), the I_sc of the TENG based on PFOTES-CNF was about 2 times higher than that of its counterpart. Then, Roy et al.¹⁷⁵ developed a high-performance TENG based on cellulose. Allicin was grafted onto CNF by a “thiol–ene” chemical reaction (Fig. 11(e)). The CNF film modified by allicin presented desirable mechanical and thermal stability. As shown in Fig. 11(f), the peak voltage and current of the TENG with the modified membrane reached 7.9 V and 5.13 μA, respectively, i.e., about 6.5 times higher than that of its counterpart (1.23 V, 0.80 μA).

Fig. 11 Manufacturing and properties of TENGs based on cellulose. (a) The fabrication process of Cu-coated CFP and PVDF-coated CFP. (b) Output voltage of TENGs with different amounts of PVDF solution. (c) Fabrication process of the PFOTES-CNF film. (d) I_sc of the PFOTES-CNF film-based TENG. (e) Synthesis mechanism of nano-cellulose fibres grafted with allicin. (f) Output current of the TENG based on different cellulose materials.

Artificial degradable materials. Artificial degradable materials are known for their excellent plasticity and processability, and they decompose into harmless substances after invalidation.⁹⁰ Based on the above characteristics, such degradable materials are increasingly being used for TENGs. At present, several artificial degradable materials are commonly used in TENGs: polylactic acid (PLA), polylactic acid co-glycolic acid (PLGA), polyvinyl alcohol (PVA), polycaprolactone (PCL) and poly-3-hydroxybutyrate (PHB/V). In 2016, Zheng et al.¹⁷⁶ proposed a TENG composed of biodegradable polymers (BDPs). The TENG showed desirable biocompatibility and biodegradability. Under the simulated motion with a frequency of 1 Hz, the V_oc and I_sc of the TENG reached 40 V and 1 μA, respectively. Zhang et al.¹⁷⁷ utilized a rechargeable carbon black (CB)/TPU composite material to fabricate a stretchable TENG (CT-TENG). The CB/TPU composite material, serving as a triboelectric layer of the TENG, presents acceptable stretchability. After corona charging, the V_oc of the CT-TENG reached 41 V.

Because of its favourable biocompatibility and biodegradability, BDP has intrigued more researchers in the field of implantable medical devices and wearable devices. In Fig. 12(a), Yang et al.¹⁷⁸ fabricated a TENG based on a polyvinyl alcohol/phytic acid (PVA/PA) hydrogel (PH-TENG). Owing to the desired mechanical and electrical properties of such material, the PH-TENG worked normally in the range of stretching of 200% and bending angles of 30 to 90 (Fig. 12(b)). To improve effective communication between doctors and patients, a multi-channel self-powered medical nursing human–machine interaction system was developed. Liu et al.¹⁷⁹ reported a humidity-resisting TENG (HR-TENG) with polyvinyl alcohol/lithium chloride (PVA/LiCl) as triboelectric layers (Fig. 12(c)). The electrical output of HR-TENG increased significantly with the increase of relative humidity (from 20% to 90%), and the maximum output was increased by 5 times. When the humidity was 90%, the maximum output of HR-TENG reached 244 μC m⁻² (Fig. 12(d)). Kamboj et al.¹⁸⁰ fabricated a laser-irradiated graphene (LIG) film with a melting structure by chemical deposition and laser irradiation (Fig. 12(e)). A composite matrix with higher conductivity was prepared by coating polymers such as polyvinyl alcohol and polyethylene oxide (PEO) on the molten structure of LIG thin films. The TENG yielded a V_oc of 10 V and an I_sc of 10 μA, which provided a unique idea for the development of a multifunctional wearable power supply (Fig. 12(f)).

Fig. 12 Fabrication and properties of artificial degradable materials and bio-based degradable materials. (a) Structure diagram of the PH-TENG. (b) Output voltage of the PH-TENG under different strains and the voltage signal of the PH-TENG under different angle gestures. (c) Fabrication process and output voltage of the HR-TENG composed of PVA and PVDF. (d) Output voltage and current of the HR-TENG under different humidity conditions. (e) The surface structure and output current of the TENG composed of LIG/PEO and PVA/LIG. (f) V_oc and I_sc of the proposed TENG under continuous finger tapping stress.

Biobased degradable materials. Biobased degradable materials such as silk fibroin (SF),^181–184 gelatine and protein^185–187 have been attempted to be applied in energy collection electronic devices. In addition to biocompatibility, high tensile strength, and renewability, SF also adds many benefits to hybrid energy systems, such as tunability and multifunctionality, making SF one of the most all-encompassing materials for hybrid devices.¹⁸⁸ When combined with conductive nano-materials, the silk materials are functionalized on the mesoscopic scale.¹²⁰ Based on the above characteristics, SF has been widely used to fabricate high-performance TENGs. Candido et al.¹⁸⁹ mixed the SF into polyvinyl alcohol films to fabricate flexible TENGs. The addition of SF directly affected the dielectric properties of the devices by changing their surface morphology. V_oc and I_sc were 172 V and 8.5 μA, respectively. Gong et al.¹⁹⁰ functionalized the SF by adding glycerol and polyurethane, which improved the mechanical toughness of a silk protein film. In addition, a biodegradable TENG with high stretchability was fabricated with the demonstration of lighting more than 25 commercial LED bulbs per square centimetre.

In addition, the connective tissues of some animals are treated to form gels, which present good biocompatibility and biodegradability and are widely used in the medical field.¹⁹¹ For example, Xue et al.¹⁹² proposed an efficient protein-based composite film with hydroxypropyl methylcellulose as a filler (Fig. 13(a)). The film overcame the problems of weak mechanical flexibility and electrical conductivity of protein. The experimental results showed that the TENG based on the composite film and PTFE obtained higher output signals. Finally, the TENG reached an average V_oc of 120 V and an I_sc of 12 μA (Fig. 13(b)). Han et al.¹⁹³ fabricated a fish gelatine (FG) film from fish scales and combined it with a PTFE/PDMS composite film to form a TENG (FG-TENG) (Fig. 13(c)). The FG-TENG showed satisfactory output performance because of its strong electron-donating ability, i.e., when the size was 5 × 5 cm², both V_oc and I_sc were the highest, which reached 130 V and 0.35 μA, respectively (Fig. 13(d)). He et al.¹⁸² fabricated a conductive hydrogel based on the SF and polyacrylamide (PAM) (Fig. 13(e)). The hydrogel presented desirable biocompatibility, elasticity, and mechanical tensile properties. Experimental results showed that the TENG based on the hydrogel provided energy for lighting 20 commercial LED lamps. In addition, a sensor based on the hydrogel was fabricated, which was used to identify and monitor human motion (Fig. 13(f)).

Fig. 13 Application of biobased degradable materials in TENGs. (a) The production process of an egg white substrate. (b) Output performance of the TENG with cellulose/egg white protein composite films under different humidity conditions. (c) Fabrication process and output voltage of the TENG based on fish glue. (d) Output voltage and current of the TENG with different sizes. (e) The fabrication process and output performance of the TENG based on silk fibroin. (f) Output performance of the TENG based on the PSGP hydrogel.

Summarily, the emergence of DB-TENGs paves an effective path for new environmental protection and material-degradable energy conversion. DB-TENGs show their potential in energy collection, signal sensing and implantable medical care. At present, biodegradable materials used in TENGs can be divided into three categories: cellulose, animal-based materials and artificial materials (Table 3). Among them, cellulose exhibits satisfactory biocompatibility and biodegradability, but its polarity is weaker than that of the other two kinds of degradable materials. Similar to common triboelectric materials, biodegradable triboelectric materials also improve the output performance of DB-TENGs by surface modification. When preparing DB-TENGs, most researchers also adopt the working mode of the VCTENG to obtain a long lifespan. In addition, some scholars have studied the degradation rate of DB-TENGs and achieved good results. Although many achievements were made in this field, there is still much space for improving the TENG performance based on cellulose, e.g., chemical modifications and physical doping of high cellulose triboelectric materials. Animal-based biodegradable materials, such as silk fibroin and gelatine, are expected to be used in implantable medical devices because of their favourable biocompatibility.

Table 3 Performance comparison of different types of DB-TENGs

Degradable material	Degradable triboelectric material	Surface treatment method	Electrical output	Size (contact area)	Working mode	Lifespan test	Ref.
Cellulose	Cardboard	Nanowire structure	30 V, 90 μA	40 cm^2	VCTENG	5 × 10^5 cycles	172
	PDMS/CNCF composite film	High dielectric particle doping	0.76 mW cm^−2	1.5 cm × 1.5 cm	VCTENG	—	173
	Cross-linking of PEI and CNF	Nanowire structure	13.3 W m^−2	1 cm × 2 cm	VCTENG	2 months	194
	CNF-PEI-Ag	High dielectric particle doping	286V, 0.43 W m^−2	4 cm × 1.5 cm	VCTENG	10 000 cycles	195
	PPy-coated cellulose paper	—	60V, 0.83 W m^−2	1 cm × 2 cm	VCTENG	10 000 cycles	196
Artificial material	PVA	Templating	1.47 V, 3.9 nA	10 cm^2	VCTENG	—	197
	PVA	High dielectric particle doping	270 V, 27 μA	4 cm × 4 cm	VCTENG	30 000 cycles	198
	PLGA	Near infrared laser	28 V, 220 nA	1.2 cm × 1.2 cm	VCTENG	28 days	199
	PLGA	Nanowire structure	90 V, 1.5 μA	—	SETENG	50 000 cycles	200
	CB/TPU composite material	Ion injection	41 V	2 cm × 4 cm	VCTENG	4500 cycles	177
Animal-based	Silk/silk fibroin	Spray-coating	213.9 V, 0.34 μA	4 cm × 6 cm	VCTENG	7 days	201
	Silk/silk fibroin	Electrospinning	28.13 V, 2.71 μA	5 cm × 3 cm	VCTENG	—	202
	Gelatin	High dielectric particle doping	130 V/0.35 μA	5 cm × 5 cm	VCTENG	10 000 cycles	193
	Silk/silk fibroin	Templating	3 μA, 50V	2 cm × 2 cm	VCTENG	1000 bending cycles	203
	Chitosan-silk fibroin-airlaid paper composite (CSA) film	Templating	268.8 mW m^−2	4 cm × 4 cm	VCTENG	7 days	24

---

Mechanical design

The design of mechanical structures is crucial for the sensing performance and transduction efficiency of TENGs. Wireless sensors and self-powered sensors need to work for long periods, so the reliability and stability of TENGs are quite important in practical applications. Through reasonable mechanical design, the space utilization ratio and structural stability of TENG devices can be optimized. Additionally, the improvement of the mechanical structure also expands the applications of TENGs. By the design of mechanical structures, different types of energy conversion and applications are reached, e.g., pressure sensors, vibration sensors, and gas flow sensors. In recent years, many researchers have tried to propose various structural design methods of TENGs, including multilayer/multipoint in a single TENG, TENG networks and mechanical adjustment of low-frequency excitation.

Multilayer and multipoint in a single TENG

To improve the utilization ratio of space, many scholars have explored the structural design methods of TENGs.^204–206 The research directions of improving the space utilization rate mainly focus on two main aspects: multilayer and multipoint.^207,208 Related research within the last decade is shown in Fig. 14 and 15.

Fig. 14 Multilayer structures in a single TENG to improve output performance. (a) Structural schematic diagram of the BFM-TENG and a comparison of output performance of different layers. (b) Schematic diagram and output voltage of the PC-TENG. (c) Schematic diagram of the multilayer structure and output performance of the MLS-TENG. (d) Structural schematic diagram and output performance of the VSE-TENG. (e) Structural schematic diagram of the MH-TENG.

Fig. 15 Structure and output performance of TENGs based on multipoint contact. (a) Schematic diagram of the NDM-FTENG. (b) Output performance of the NDM-FTENG. (c) Structural schematic diagram of the T-TENG. (d) Changes in the electrical output of the T-TENG as the number of particles increases. (e) The schematic diagram of the TENG with multilayer superposition. (f) The changing trend of I_SC and Q_SC with the number of units.

Multilayer structure. Multilayer structures enhance the electrical output of TENGs by increasing the contact area and contact frequency. The application of TENGs with multilayer structures has intrigued more researchers to collect energy in low-frequency environments. To increase the contact area and space utilization rate of TENGs, some researchers have carried out many related studies on multilayer TENGs in recent years. Zhang et al.⁴⁷ proposed a bionic-fin-structure assisted with a multilayer-structured TENG (BFM-TENG), which was used to collect seabed energy (Fig. 14(a)). Such a TENG was composed of multilayer structures and ultra-thin dielectric materials. The BFM-TENG reached a peak power density of 444 W m⁻³ under ideal test conditions, which was 1–2 orders of magnitude higher than that of its counterpart. Inspired by the unique stacked multi-cell structure of fish gills, Yin et al.²⁰⁹ designed a parallel-cell TENG (PCTENG). As shown in Fig. 14(b), compared with a zigzag TENG (Z-TENG) of the same size, the overall volume power density of the PCTENG reached 143.7 W m⁻³, which was 9 times as high as that of the Z-TENG. Zhou et al.²¹⁰ proposed a multilayer stacked TENG (MLS-TENG) based on the rotation-translation mechanism (Fig. 14(c)). In the mechanism, rotary motions induced by fluid were transformed into linear reciprocating motions. Owing to the design of multilayer stacked structures, V_oc was increased from 860 V to 2410 V, and the efficient energy collection rate was 2 mJ min⁻¹ in a river. Li et al.²¹¹ proposed a V-shaped stacked TENG (V-TENG), which effectively increased the effective contact area and the space utilization rate (Fig. 14(d)). The effects of different parameters and stacking units were systematically studied on a V-TENG with self-charge excitation (VSE-TENG). The peak power of the VSE-TENG reached 3.96 mW, which was 22 times as high as that of the V-TENG. Liu et al.²¹² fabricated a multilayered helical spherical TENG (MH-TENG) composed of two spring-like multilayer spiral units, and its space utilization rate reached 92.5% (Fig. 14(e)). At a trigger frequency of 1.0 Hz, the output current and power of the MH-TENG reached 200.3 μA and 16.2 mW, respectively.

Multipoint contact structure. Multipoint contact is another strategy to enhance the space utilization rate. The key point of multipoint contact structures is that the movements of internal rolling balls are guided by the movements of TENGs. Jing et al.²¹³ designed a fully enclosed TENG (FE-TENG), which combined multipoint contact and bionic fish-like structures. The loss of friction energy between the disorderly movements of rolling balls was avoided, and the output voltage reached 150 V in an actual river environment. Wang et al.³⁹ proposed an annular TENG (A-TENG), which consisted of an annular shell and inner balls. Experimental results showed that with the increase of the diameter and number of balls, the friction area between nylon balls and the inner wall of the shell was increased, thus increasing the total charges on the surface of nylon balls. Liu et al.²¹⁴ proposed a nodding duck structure multi-track freestanding TENG (NDM-FTENG) for wave energy collection in low-frequency ranges (Fig. 15(a)). By sliding nylon balls on the arc-shaped dielectric composite film, the tracks were connected with each other to make nylon balls move synchronously. After tests, a single NDM-FTENG module block illuminated 32 LEDs at the same time, and the instantaneous power density reached 4 W m⁻³ (Fig. 15(b)). Xu et al.²¹⁵ designed a high-output TENG with a tower structure (T-TENG), which was used to collect water wave energy in any direction (Fig. 15(c)). The TENG was a multilayer structure composed of PTFE balls and nylon films. From Fig. 15(d), it is found that the voltage of the T-TENG increased with the increase in the number of parallel units. Besides, by the addition of 1 to 10 units in a module, it was found that the power density was increased linearly from 1.03 W m⁻³ to 10.6 W m⁻³. Wang et al.²¹⁶ modified multilayer structures on the basis of a stackable structure to make full use of the space (Fig. 15(e)). Experimental results showed that with the increase in the number of units, I_SC increased from 52.24 to 261.79 μA, and Q_SC increased from 3.5 to 14.72 μC (Fig. 15(f)). Finally, the peak power density of the TENG reached 4 W m⁻³, which was about 29% higher than that of their previous work.

In summary, the single TENG with multilayer or multipoint contact improves the space utilization rate of the device. The frequency of mechanical movement is raised by the efficient movement of multiple triboelectric layers or rolling balls inside TENGs, thus enhancing the output power of TENGs. At present, the single TENG with multilayer or multipoint contact has been widely used in energy harvesting in low-frequency environments and has produced satisfactory results.

TENG networks

In a large amplitude excitation environment, the energy generated per unit is limited. To achieve large-scale energy harvesting for such environments, some researchers have suggested using conductive cables to connect more devices to generate more electrical energy. In the background of collecting blue energy, Prof. Wang suggested using a TENG network composed of multiple TENG units to collect wave energy on a large scale.²¹⁷ Some researchers integrated several single devices into networks to increase the energy collection scale of TENGs.^218–220 Common network structures can be divided into one-dimensional chain connections and two-dimensional network connections. The relevant representative work within the last decade is illustrated in Fig. 16 and 17.

Fig. 16 TENG networks with one-dimensional chain structures. (a) The structure of the SS-TENG array and the output of different segments. (b) Digital photo of the four-HW-NG network, output performance of the single HW-NG and HW-NG network. (c) The structure of T-TENG networks and output voltage of the T-TENG network in a tank. (d) Schematic diagram and output power of a V-shaped network composed of multiple duck units.

Fig. 17 Two-dimensional connected TENG networks. (a) The 6 × 6 array of TS-TENG, and the output voltage of arrays with different sizes. (b) Schematic diagrams of the hexagonal TENG network, and the output voltage at different frequencies. (c) TENG networks composed of multiple spherical TENGs, and the output currents of TENG networks with different numbers. (d) Schematic diagrams of the kelp TENG network, and the output voltage at different frequencies. (e) Digital photos of the TENG array and output performance under different loads. (f) Ocean emergency self-powered wireless SOS system composed of solid–liquid TENGs and the trend of output current of the TENG network changing with the number of TENGs.

One-dimensional chain connection. In various TENG network structures, one-dimensional chain connection structures are commonly utilized to collect unidirectional energy sources. For example, Yuan et al.²²¹ designed a serpentine spiral multilayer TENG, which improved the output performance and solved the problem of low space utilization rate. Zhang et al.²²² designed a sea snake TENG (SS-TENG) with segments of TENG units linked by springs (Fig. 16(a)). Every part of the SS-TENG was easy to bend, which improved the moving speed of rolling balls and the output power. Owing to air gap structures, the dielectric shielding effect of water on the device performance was solved, and the highest output lighted 152 LEDs. Ren et al.²²³ proposed a hybrid nanogenerator (HW-NG), and a network was formed by four HW-NGs (Fig. 16(b)). In the network, all TENG units were connected in parallel, while EMGs were connected in series. Experimental results showed that when the charging time of the network was 60 s, the voltage of a 0.47 mF capacitor reached 4.3 V. As shown in Fig. 16(c), based on high-power TENGs (T-TENGs) with tower structures, Xu et al.²¹⁵ connected multiple T-TENGs in parallel to form a T-TENG network. Due to the unique design and mechanism of T-TENGs, the power density of the network was increased proportionally with the number of parallel units. Under the wave excitation, the induced peak current I_SC was 5.8 μA. Ahmed et al.²²⁴ introduced a fully enclosed duck-shaped TENG (Fig. 16(d)). The authors connected a series of TENG units to two floating structures and found that the peak power increased with the increase of the unit number, and the highest power reached 1.366 W m⁻².

Two-dimensional network. Compared with one-dimensional TENG networks, two-dimensional TENG networks show higher energy output and flexibility. On the one hand, redundant space and interaction between TENG units are avoided. On the other hand, the energy conversion efficiency and stability of the devices are enhanced. Liu et al.²²⁵ used multiple torus structure TENGs (TS-TENGs) composed of a torus shell and inner balls to manufacture a TENG network (Fig. 17(a)). The current amplitude in the TENG network was almost proportional to the unit quantity. The average current amplitude of 2 cells was 0.57 μA, and that of 16 cells was 2.60 μA. In addition, they also studied the charging performance of the TENG network under the same excitation conditions. Experiments showed that the modified TENG network charged the 1 μF capacitor to 13.38 V in one minute. Liang et al.²²⁶ proposed a hexagonal TENG network based on spherical TENGs with spring-assisted multilayer structures (Fig. 17(b)). The influence of water wave frequency, amplitude and water wave type on the TENG network was studied. Under the action of a longitudinal shock wave, when the frequency of the water wave was 1.0 Hz and the amplitude was 2.5 V, V_OC reached the maximum value, about 354 V. Xue et al.⁴⁸ demonstrated the energy collection of TENGs using treated silicone rubber as a triboelectric material in water waves (Fig. 17(c)). The aggregation performance of multiple TENG units was explored, and the coupling behaviour of each unit in a TENG network was proposed. Through reasonable design, the points and outputs of the modified TENG network were 10 times more than that of the uncoupled elements. Inspired by kelp, Wang et al.²²⁷ prepared a ribbon TENG with flexible materials (Fig. 17(d)). With the independent swing of every single strip, two adjacent single strips come into contact. The maximum I_sc and V_oc of a single TENG unit reached 10 μA and 260 V, respectively. In practical application, higher output performance was obtained through an integrated network of multiple kelp-inspired biomimetic TENGs.

Solid–liquid interface-based sensors show promising research prospects,^228–230e.g., marine environmental monitoring and raindrop monitoring systems. Meanwhile, solid–liquid TENG networks have demonstrated potential for large-scale applications in the ocean. For example, Wei et al.²³¹ presented a droplet-based TENG, which effectively obtained wave energy from the ocean (Fig. 17(e)). An integrated array composed of 24 TENGs successfully demonstrated the ability to collect actual rainwater and seawater energy. Furthermore, under the action of mimicked ocean waves, the amplified TENG showed satisfactory output performance. The output peak power reached 23.3 μW when the load resistance was 500 MΩ. As shown in Fig. 17(f), a kind of solid–liquid contact TENG was prepared by Li et al.²³² The energy output was 48.7 times as high as that of a solid–solid TENG with the same area. Besides, a TENG network was formed by the use of multiple TENGs. The electric output exhibited an accurate linear relationship with the number of TENG units, which supplied power to hundreds of LEDs.

Connections between units within the TENG network. External mechanical connection modes present a significant influence on the performance of TENG networks, especially on the energy conversion efficiency of units in networks.²³³ For example, rigid connections provide a strong constraint for TENG units in networks, which makes TENG units move synchronously. TENG units in flexible networks show higher flexibility. Therefore, researchers have studied the influence of mechanical connections on the stability and output of TENG networks. Xu et al.⁴⁸ reported a coupling network based on optimized spherical TENG elements. The authors compared the performance of rigid connections with that of flexible connections. In Fig. 18(a), the output of both flexible networks was higher than that of rigid networks. Yang et al.²³⁴ proposed a TENG network that realized self-assembly. Such TENG units were self-assembled through adaptive magnetic joints (Fig. 18(b)). In addition, the average power of the TENG network composed of 18 TENG units reached 9.89 mW. Bai et al.²³⁵ demonstrated a high-performance tandem disk TENG (TD-TENG) to collect water wave energy (Fig. 18(c)). The influence of connection point position on the overall output performance was studied. Experimental results showed that the symmetrical connection of shells made it difficult to be pushed by waves, so the bottom connection was much better than the middle connection. Through surface modification and optimal design, the peak power and average power of the TD-TENG were increased to 45.0 mW and 7.5 mW, which were about 35 times and 24 times that of the typical spherical shell structure device, respectively.

Fig. 18 Connection modes of TENG networks. (a) Comparison of the spherical TENG networks connected in different ways. (b) The schematic diagram of the self-assembled TENG network based on adaptive magnetic joints. (c) The connection mode of the TD-TENG in water and the output current. (d) Diagram of the TENG array before sealing and waterproofing.

When TENGs are connected to a large-scale grid, it is especially critical to ensure the structural strength of networks and the efficiency of power transmission. Some researchers have provided theoretical and general methods for the internal cable connection of large TENG networks. For example, to build large-scale TENG networks, Liu et al.²³⁶ proposed a rigid power cable for TENG networks. The plane structure based on a steel belt enabled TENG units to deflect and move vertically without interference. When the cable swung with waves, TENGs generated power while keeping the rhombic array unchanged. As shown in Fig. 18(d), the cable consisted of two spring steel bars and three polymer film layers. The outermost layer was a PTFE film, which assembled another TENG with seawater, and the maximum V_oc was 24 V. Liu et al.²³⁷ proposed four basic electrical networking topologies for large TENG networks. The influence of cable resistance and the asynchronous output phase of each unit on the output of networks was systematically studied. With the reduction of cable length in a TENG network, the production cost and complexity of the network were decreased, leading to a reduction of the energy loss in the circuit.

Table 4 presents the performance comparison of different types of TENG networks. As the number of units increases, the array performance rapidly improves. Ref. 225 shows that the performance improved over 40 times when 16 TENGs are networked. Meanwhile, different connection strategies exhibit different energy capture efficiencies. The movement of the rigidly connected array is stiff resulting in a decrease in overall energy collection efficiency. On the other hand, a flexible connection is better than a rigid connection in increasing internal freedom.⁴⁸ Therefore, flexible connections are more favored by researchers.^221,222 Moreover, the damping of cables cannot be ignored when a large-scale network array is constructed. A reasonable layout of the array requires careful consideration based on the number of units.

Table 4 Performance comparison of different types of TENG networks

TENG networks	Triboelectric materials	Voltage (single TENG)	Current (single TENG)	Output (array)	Size (array)	Connection modes	Ref.
One-dimensional network	Nylon-PTFE	55 V	—	3 W m^−3	1 × 3	Flexible	222
	Nylon-PTFE	105 V	1.3 μA	10.6 W m^−3	1 × 10	Flexible	215
	Nylon-Kapton	325 V	65.5 μA	1.36 W m^−2	1 × 3	Flexible	224
	FEP-conductive fiber	300 V	0.3 mA	347 W m^−3	1 × 3	Flexible	221
Two-dimensional network	Nylon-FEP	83.4 V	1 μA	41.6 μA	4 × 4	Flexible	225
	Al-FEP	—	—	3.33 W m^−3	1 × 7	Flexible	226
	Silicone rubber-Ag	1780 V	1.8 μA	4.47 W m^−3	4 × 4	Rigid	48
	FEP-water	58 V	40 nA	23.3 μW	4 × 6	Rigid	231
	PTFE-water	400 V	40 μA	290 μA	3 × 6	—	232

---

To sum up, the current research on TENG networks is divided into one-dimensional structures and two-dimensional networks. The one-dimensional structure is simple and easy to produce and maintain. The two-dimensional connection strategy is complex but shows a higher power density. Some studies envisaged the construction of three-dimensional networks, but further experimental verification is still required. In addition, other researchers also studied the mechanical connection mode of TENG networks. However, these studies cannot cope with large-scale harsh environments, e.g., strong winds and large waves. Overall, flexible connections work better in the marine environment than rigid connections. Therefore, it is recommended that future research focuses on a more comprehensive study of the effect of connection mode on array efficiency.

Mechanical regulation towards low-frequency excitation. Low-frequency vibrations significantly reduce the effective working time of TENGs, thereby reducing energy conversion efficiency. Typically, the output voltage and current of TENGs under low-frequency excitation are smaller than those under high-frequency excitation.^238–240 Increasing the working frequency of TENGs via mechanical design under low-frequency excitation is an efficient strategy for enhancing the electric output. Raising the output cycles of TENGs is another method to enhance the electric output.^241,242Fig. 19 summarizes the relevant research on adjusting the output frequency within the last decade.

Fig. 19 TENGs with adjustable motion frequency based on gear structures. (a) Structure diagram and output performance of the GUA-TENG. (b) Schematic diagram of the motion principle and output performance of the HH-TENG. (c) Principle and output comparison of the LR-TENG based on one-way bearing.

Some researchers have conducted research from the perspective of increasing the output cycles. By adding transmission structures to TENGs, the multiple output cycles were obtained by one excitation.²⁴³ Zhai et al.²⁴⁴ proposed a gear-driven unidirectional acceleration TENG to collect disordered low-frequency water wave energy (Fig. 19(a)). With the design of special gears, the TENG converted the disordered energy into high-frequency water flow kinetic energy, and the rotational speed was increased by 25 times. Experimental results showed that the peak voltage and peak current at 120 rpm reached 7.1 V and 1.5 mA, respectively. Choi et al.²⁴⁵ developed a handheld TENG (HH-TENG), which used a simple transmission unit to collect the energy generated by finger movements (Fig. 19(b)). Through the transmission units, linear reciprocating motions were converted into high-speed unidirectional rotating motions with long displacement. Such a TENG not only generated reliable DC voltage, but also yielded an open circuit voltage larger than 2.5 kV. Tcho et al.²⁴⁶ fabricated a linear-to-rotational TENG (LR-TENG) based on the sliding mode of the grating (Fig. 19(c)). The LR-TENG converted linear mechanical energy into electrical power through a gear system with a transmission ratio of 5.25. The authors used the same materials as the LR-TENG to fabricate a vertical contact mode TENG (V-TENG) as the comparison object. The higher transmission ratio increased the rotating speed of the LR-TENG. At the same level of applied force and mechanical energy, the surface charge density of the LR-TENG was much higher than that of the V-TENG, i.e., the power generated by the LR-TENG was 100 times that of the V-TENG. Yun et al.²⁴⁷ designed an exo-shoe TENG (ES-TENG) with a two-way gearbox to collect energy from daily walking. In the process of reciprocating low-frequency natural stepping motion (1 Hz), the output performance of the ES-TENG was enhanced by 10 times that of the comparison group, and the specific power reached 13 μW g⁻¹.

To sum up, making full use of charge excitation is one of the key factors to improve the output performance of TENGs. TENGs usually use optimized structures and surface texture to maximize the friction surface area, thus enhancing the charge excitation effect. In mechanical design, it is an effective means to improve the charge excitation ability by increasing the effective friction area. Elastic materials and soft contact designs are usually used to increase the effective friction area, such as rubber, animal fur, etc. In addition, the energy conversion efficiency is improved by adjusting the movement frequency of the mechanical system to match the environmental movement frequency. Similarly, the working frequency of TENGs in low-frequency environments, such as ocean and human walking, is improved by adding gears. The combination of mechanical design and surface modification methods of other materials further improves the performance and application potential of TENGs.

Other types of configurations

To cope with various specific situations, several researchers have proposed many unique designs. Besides the mechanical design methods mentioned above, to improve the output performance of TENGs, the mechanical design of triboelectric layers has also attracted the attention of scholars. For example, some researchers pre-deform the triboelectric layers so that the two triboelectric surfaces fit better.^248,249 Lu et al.²⁵⁰ developed a bidirectional gear transmission TENG (BGT-TENG). As shown in Fig. 20(a), FEP films and copper electrodes were attached to a flywheel and the inner wall of a holder, respectively. Rotating the flywheel caused friction between the two material surfaces and yielded electricity outputs. The authors cut FEP films into different lengths and inserted them on the flywheel at different angles, so that the FEP films were deformed in advance. The maximum output energy was generated when the mounting angle of FEP films was 30 and the length was 45 mm. Yang et al.²⁵¹ fabricated a travel switch integrated mechanical regulation TENG (TSMR-TENG) that converted linear motion into rotary motion. The TSMR-TENG stored the rotational energy through a coil spring, and rotated with an inertia wheel when the travel switch was turned on. As shown in Fig. 20(b), FEP films on the inertia wheel rubbed against the copper electrode to generate electric energy. Among them, the FEP film obtained higher output energy when the installation angle was 60 and the film length was 60 mm, i.e., under a mimic random vibration, V_oc and I_sc were 400 V and 17 μA, respectively. Yang et al.²⁵² used a flywheel structure and pre-deformed FEP to design a TENG for collecting wave energy. The peak power of this TENG reached 11 mW, and it maintained a stable output after 100 000 cycles. Fu et al.²⁵³ designed a soft–soft contact TENG (SSTENG) using ultra-soft PU foam and a spring steel sheet. The TENG maintained 100% initial output after 1 million cycles. More importantly, the characteristics of PU foam greatly improved the output performance of the SSTENG, and the charge density reached 144 μC m⁻².

Fig. 20 Other strategies for improving TENG output. (a) Structure diagram and output performance of the BGT-TENG. (b) Structure diagram and output performance of the TSMR-TENG. (c) Experimental platform and output performance of the MS-TENG.

Animal fur is also a common material for soft contact design. In addition, the introduction of soft fur also gives TENGs excellent stability and durability.^254,255 Based on the above characteristics, scholars have studied the application of animal fur in TENGs. He et al.²⁵⁶ reported a soft contact fiber structure of TENGs to increase the effective friction area and improve the output. They used degradable fiber rabbit hair, paper and PVDF to construct a soft-contact triboelectric layer for the TENG. Compared with the direct contact TENG, the output of the soft contact TENG under high-speed operation was improved by at least 350%. Han et al.²⁵⁷ made an improved segmented soft contact rotating TENG to collect wind energy. They used rabbit hair as a typical triboelectric material, and controlled the contact state of the TENG by adjusting the gap between brushes and electrodes. After 480 000 cycles of operation, the transferred charge showed no obvious attenuation, and the efficiency of mechanical conversion to electric energy reached 15.4%. Chen et al.²⁵⁸ used animal fur as a triboelectric material to prepare an ultra-durable, low-wear TENG for efficient energy collection. Compared with the traditional copper electrode, the output of TENGs with brushes can be increased by more than ten times. In addition, after 300 000 cycles of continuous operation, the attenuation of the transferred charge was 5.6%.

In addition, some scholars exert extra pressure on the triboelectric layer through mechanical design to increase the effective contact area of the triboelectric interface. To study the influence of pressure on sliding TENGs, Zhang et al.²⁵⁹ introduced a flexible magnet layer into sliding TENGs (MS-TENGs). In Fig. 20c, the influence of factors such as magnetic preload and load resistance on the MS-TENG was analysed from theoretical aspects of mechanical characteristics and electromechanical coupling. With the increase of magnetic preload, triboelectric layers of TENGs contacted better, thus obtaining higher output performance. Experimental results showed that V_oc with different magnetic forces reached 12.73 V, 16.67 V, and 19.62 V at the frequency of 2 Hz, respectively. Huang et al.²⁶⁰ used a magnetic response layer to fabricate a new type of magnetically assisted contactless TENG. In an arch structure, PDMS/Fe–Co–Ni powder layers play the role of a magnetic response layer. By adjusting the parameters of magnetic force, the effective contact area of the TENG interface was raised and the output performance of TENG was enhanced.

Power management

Since the appearance of TENGs, the surface treatment of materials and mechanical design have paved the way for the sustainable development of TENGs. Various miniature energy harvesting or sensing devices are widely used. However, due to the high output impedance of TENGs, the output voltage stays high and the output current is limited.^261,262 In addition, the working environment of TENGs is characterized by randomness, which makes the output of electrical signals unstable. These factors limit TENGs from supplying power directly to sensors and energy storage devices. Therefore, it is crucial to enhance charge transfer and minimize output impedance via rational circuits to achieve efficient energy supply and storage. This section summarizes the research progress from the perspective of switch types in circuits. The switch types in circuits can be divided into mechanical and electronic switches.

Power management with mechanical switches

Travel switches. As one of the mechanical switches, travel switches are triggered by the periodic movement of TENGs.²⁵¹ One of their key features is the ability to convert the continuous release of electric energy from the TENG into an instantaneous release.^263–265 Moreover, the operation frequency of the travel switch is directly related to the output frequency of TENGs, i.e., the working frequency of the travel switch is generally 1–2 times as high as the output frequency of electrical signals.²⁶⁶ For example, Zi et al.²⁶⁷ added charging cycles and motion trigger switches to a TENG system, which enhanced the maximum energy storage efficiency by 50% and increased the saturation voltage by at least 2 times. This study presented another strategy for effectively collecting energy from the environment. Wu et al.²⁶⁸ developed an opposite-charge-enhanced transistor-like TENG with two electrodes and a slider, the turn-off current of which was similar to that of a common independent sliding TENG. Its turn-off working principle was similar to the source-drain conductance caused by the gate trigger of a transistor. Furthermore, a method of using travel switches and anti-charge enhancements was proposed to raise the output performance of the built TENG. Cheng et al.²⁶⁹ designed a travel switch in the electrode contact of TENGs, which triggered the electrical switch connection through the movement of TENGs. The instantaneous output current and power peak value were greatly improved, which were 0.53 A and 142 W respectively when the load was 500 Ω.

Additionally, considering that travel switches use the collision of moving parts to make their contacts move to connect or disconnect control circuits, researchers have carried out research from the perspective of mechanical design. For example, Le et al.²⁷⁰ developed a discontinuous conductive rotating solid–liquid TENG (DCR-TENG) by using the strategy of activating the switch with periodic motions. As shown in Fig. 21(a), the device mainly included a rotating disc TENG and a stationary electric contact (SEC). The DCR-TENG generated electricity by the contact between negatively charged PVDF films and positively charged deionized (DI) water. As shown in Fig. 21(b), when the DI water and PVDF films completely overlapped on electrode E1, the switch was activated, so that the two electrodes and the SEC were merged into one channel. As the disc continues to rotate, a cycle completed four stages, which also represented a complete cycle of the current generated by the device. To compare the effect of the travel switch, a continuous conductive TENG (CCR-TENG) was fabricated. As shown in Fig. 21(c), when the resistance was 20 MΩ, the output voltage of the CCR-TENG was only 0.54 V. In contrast, the voltage of the DCR-TENG was around 1.54 V. Chung et al.²⁷¹ developed a dielectric liquid-based self-operating switch TENG (DLSS-TENG) to overcome current limitations caused by air breakdown in triboelectric materials. The DLSS-TENG controlled the field emission on the electrode surface through the movement of the dielectric liquid. During the back-and-forth motion of PTFE films on the electrode, the dielectric liquid functioned as a travel switch (Fig. 21(d)). When the PTFE contacted the electrode, field emission occurred, and electrons flowed directly to the electrode. Through this series of processes, the output of the DLSS-TENG was amplified. As shown in Fig. 21(e), when the PTFE plate moved at a low frequency of 0.5 Hz, the DLSS-TENG reached a maximum peak voltage of 140 V and an amplification current output of 30 mA.

Fig. 21 TENG structures with travel switches. (a) Radial symmetric structure of the DCR-TENG. (b) Schematic diagram of the DCR-TENG in one movement period. (c) Comparison of output performance of the DCR-TENG and CCR-TENG. (d) The working principle of the DLSS-TENG in one cycle. (e) Output voltage of the DLSS-TENG.

Voltage-triggered switches. According to Paschen's law, when the voltage output of TENGs is high enough, it will generate an air discharge. Mechanical switches triggered by the air discharge generated from TENGs are called voltage-triggered switches. In recent years, some researchers have developed various types of voltage-triggered switches, thereby raising the overall performance and efficiency of TENG systems.^272–274

As a kind of voltage-triggered switch, spark switches mainly adopt gas discharge to generate an arc under an electric field, so as to realize circuit switching. Compared with ordinary mechanical switches, spark switches present higher reliability and a longer service lifespan due to their characteristic of non-mechanical contact. Researchers have developed various circuits based on spark switches to enhance the energy conversion efficiency of TENGs. Li et al.²⁷⁵ proposed a method of using a spark switch to realize energy accumulation and rapid release for developing simpler and more robust equipment. He et al.²⁷⁶ prepared a double-tip-assisted TENG, and the generated charges quickly accumulated at a tip, providing a conduction path for the accumulated charges through air discharge. Wang et al.²⁷⁷ reported an automatic spark switch that realized energy accumulation and rapid release. As shown in Fig. 22(a), the maximum energy was released through the spark switch in a three-stage working process. The capacitor located at the back stored the charge when the voltage generated by the TENG was lower than the breakdown voltage between the A-board and the B-board (V_AB). When the TENG moved to a sufficient separation distance, and the voltage was higher than the V_AB, the spark switch was triggered and released the stored charges in the capacitor. As shown in Fig. 22(b), a stable high voltage was generated via high-energy output by a half-wave rectification, i.e., the spark switch withstood a huge voltage, and the starting voltage of 7540 V was realized (Fig. 22(c)). Cheng et al.²⁷⁸ used a tungsten electrode and an adjacent stainless steel plate electrode to form a TENG triggered by an air discharge switch. When the voltage was sufficiently high and the two electrodes of the switch were relatively close, the switch was triggered by an air arc discharge, resulting in electrical output. When the two electrodes were far apart, only corona discharge occurred (Fig. 22(d)). Additionally, the discharge energy varied with the distance between electrodes. When d = 0.4 mm, the maximum output energy per cycle was 172 μJ, which was 45 times of the minimum output energy of 3.8 μJ when d = 20 mm (Fig. 22(e)). When the load was 2 MΩ, the peak power and output energy of the TENG with a spark switch were 1600 times and 30 times higher than that of the TENG without switches, respectively.

Fig. 22 Voltage-triggered switches in TENG systems. (a) The three-stage process of releasing energy through automatic spark switches. (b) The circuit diagram of TENGs driving external load. (c) Voltage and output charge when the switch is on and off. (d) The schematic diagram of spark switches. (e) The output energy of TENGs during the whole movement period. (f) Overlooking optical image of the MEMS-TENG. (g) The performance of the MEMS-TENG and the process of switching operation (the performance of the MEMS on the left axis and the switching response on the right axis).

The micro-electromechanical system (MEMS) switch is also a kind of voltage-triggered switch.^279,280 It usually consists of a movable metal contact that is suspended from a fixed metal contact by a thin flexible arm. When an electrical voltage is introduced to the arm, the arm undergoes deformation. It causes the mobile contact to make contact with the stationary contact, thereby allowing the circuit to become connected. The MEMS shows the advantages of small size, low power consumption, superior performance and mass production.^281–283 He et al.²⁸⁴ developed a flexible textile-based TENG, which used a diode switch integrated instantaneous discharge to amplify the current. TENG textiles exhibit the characteristics of softness, flexibility and lightness, which enable them to maintain moderate output under various operations, even if they are kneaded randomly. The flexible TENG managed by the diode increased the short-circuit current by 25 times. Mousavi et al.²⁸⁵ used the CMOS micromachining process to produce a micro-sized TENG. They integrated the TENG with MSMS sensors and drivers to convert environmental vibration into voltage generated between a conductive aluminum layer and a dielectric layer (Fig. 22(f)). The micro-TENG (MEMS-TENG) used a microcantilever, two fixed measuring points and a fixed center electrode to create a MEMS switch. The high voltage generated between the electrodes operated a low-frequency electrostatic MEMS switch. When the amplitude of vibration frequency exceeded a predetermined threshold, the MEMS switch was activated. Experiments showed that the vibration sensor operated with V_DC at 1.7 V where resonance dynamic pull-in happens at 10.6 kHz (Fig. 22(g)). The results showed that the MEMS-TENG provided good signal-to-noise ratio, sensitivity and quietness for MEMS switches in frequency-sensitive and acceleration-sensitive modes. Lin et al.²⁸⁶ fabricated a self-powered and autonomous vibration wake-up system through the integration of MEMS switches and signal TENGs (S-TENGs). The TENG was used as a self-powered accelerometer to control MEMS switches. When the acceleration increased, the MEMS switch was turned on by the S-TENG. Meanwhile, the energy generated by the S-TENG was used to send an alarm signal through a wireless transmitter.

In summary, mechanical switches require mechanical structures to control switches. Mechanical switches are mainly divided into travel switches and voltage-triggered switches. Friction and resistance in mechanical components will consume a certain amount of energy, leading to energy losses. In the future, it may be necessary to study new mechanical structures based on the MEMS and micromechanical structures to realize mechanical switches with high precision, speed, and reliability. Additionally, research into high-strength, wear-resistant, and high-temperature-resistant materials helps solve energy loss and wear problems in mechanical switches.

Power management with electronic switches

Electronic switches in TENG systems are primarily employed to manage the flow of current between TENGs and external loads, which is typically realized by means of semiconductor devices.^287,288 Such switches utilize various types of electronic components to rectify and adjust the output voltage of the rectifier, e.g., diodes, transistors and thyristors.

Transistor switches are employed to adjust the output voltage and current to meet the load requirements. In TENG circuit systems, transistor switches act as voltage regulators to keep the output voltage of the TENG-based equipment stable, and also serve as current regulators to restrict the current flowing through the load. Niu et al.²⁸⁹ designed two switches controlled by logic circuits to periodically extract energy from capacitors. When the voltage of the capacitor reached the optimal value, the electronic switch was closed, and the energy was transferred to a post-storage capacitor through electromagnetic conversion, which enables alternating current to be converted into direct current with 60% efficiency. Harmon et al.²⁹⁰ designed a circuit management method based on a silicon rectifier. The switching circuit reduced the matching impedance of the VCTENG from 15 MΩ to 2 MΩ, and maintained 84.3% of the peak power under the load of 15 MΩ. Fang et al.²⁹¹ developed a self-powered ferroelectric transistor memory integrated module based on an arched TENG. The memory device was activated by tapping on the TENG with fingers. Xi et al.²⁹² proposed a novel method of combining the transistor with TENGs, which promoted the development of bipolar junction transistors. Furthermore, a highly sensitive and stable frequency monitoring sensor was designed via such innovations. Harmon et al.²⁹⁰ employed semiconductor devices to create a power management system tailored for TENGs, as presented in the complete circuit diagram depicted in Fig. 23(a). The energy transfer in this system was partitioned into four stages (Fig. 23(b)). The first stage involved the energy generated by the TENG being rectified and stored in the capacitor C_in. In the second stage, when the voltage reached the threshold, the energy flowed from the C_in capacitor to L, C_out, and R. In the third stage, the diode D₆ was conducted, locking the voltage in C_in, and no current flowed. Finally, the current flowed from C_out to the resistor R, and all the energy was consumed by R. As shown in Fig. 23(c), the output voltage was proportional to the frequency. As illustrated in Fig. 23(d), owing to the suitable turn-on time and ultra-low energy loss of the thyristor, this switch circuit was capable of reducing the matching impedance of the TENG output from 150 MΩ to 2 MΩ. Moreover, the TENG maintained 84.3% of the AC peak power under a 150 MΩ load.

Fig. 23 TENG systems with electronic switches. (a) Ideal buck converter topology for a power management system. (b) The energy transfer from TENGs to loads is divided into four stages. (c) Relationship between the output performance of TENGs and frequency. (d) Relationship between the output power and load resistance. (e) Detailed circuit diagram of the CEC and TENG integration. (f) Output power of the TENG integrated with the CEC at different water wave frequencies.

Compared with other types of transistors, metal-oxide-semiconductor field-effect transistors (MOSFETs) present the advantages of higher input impedance, lower power consumption, and faster switching. In 2016, Peng et al.²⁹³ studied the performance and physical mechanism of triboelectric MOSFET gates through calculations and finite element analysis. This work provided guidance for developing MOSFETs in TENG circuit managements. Song et al.²⁹⁴ designed a flexible PCB power management module, and realized the maximum energy transfer efficiency of the circuit by accurately controlling the opening and closing time. Experiments showed that the DC power of the energy collection circuit with power management reached 69.3% of the maximum AC power under the load of 10 kΩ and 47 μF. Kim et al.²⁹⁵ reported a system of MOSFETs gated by a triboelectric potential in two working modes. Such a system exhibited the ability of continuous working and environmental immunity. Moreover, the durability and stability of the proposed system were significantly enhanced by annealing the MOSFET. Liang et al.²⁹⁶ developed a charge excitation circuit (CEC) to collect water wave energy. As shown in Fig. 23(e), to realize automatic circuit switching, the system adopted two N-type MOSFETs and one P-type MOSFET. The CEC was a capacitor group consisting of two identical capacitors C₁ and C₂, which were switched from parallel to series independently. When the two triboelectric layers of the TENG were separated, C₁ and C₂ were charged in parallel. When the TENG was contacted, C₁ and C₂ were no longer charged and switched in series, which doubled the voltage of the capacitor bank. When the voltage of each capacitor reached a steady state, two capacitors in series charged C_E, and the charge transfer increased the output current. Through the integration of the CEC, the current was enhanced by 208 times, and the maximum power reached 15.8 mW (Fig. 23(f)).

In summary, electronic switches usually use semiconductor devices, e.g., transistors and diodes. However, electronic devices are susceptible to external environmental factors such as temperature, humidity and electromagnetic fields, which affect the stability of switches. In addition, the aging and failure of electronic equipment will lead to the failure of switches or a decline in performance. It is suggested to develop high-performance devices with low power consumption and high reliability in the future to solve the problems of energy loss and stability of electronic switches.

Recent advances and outlook

In recent years, researchers have improved the properties of TENG through physical, chemical, and mechanical design, such as electrospinning, functional group grafting, and the TENG network. However, as an emerging technology, some obstacles also exist, such as short lifespans, elusive mechanisms of charge transfer, and low energy conversion efficiency. In this section, the studies of the past five years are discussed statistically. Then we summarize the existing research and achievements. Finally, an outlook is provided, including research focus and key application scenarios to provide a reference for the research community.

Development statistics

We make a comprehensive analysis of TENG-related publications by searching for related keywords in the scientific citation index database Web of Science. A total of 1530 articles on improving the output of TENGs are found from 2018 to 2022. Furthermore, we classify the related research: including 459 articles on “methods of improving performance based on material selection”, 719 articles on “methods of improving performance based on mechanical design (MD)”, and 352 articles on “methods of improving performance based on circuit switch (CS)”. Among them, 459 articles on TENG improvement methods based on material selection are divided into three categories, including 209 articles on “methods of TENGs based on surface physical modifications (PM)”, 181 articles on “methods based on surface chemical modifications (CM)” and 69 articles on “the selection of biodegradable materials (DB)”. Relevant statistics are shown in Fig. 24.

Fig. 24 Statistical data of related research. (a) Proportion of the related documents in each year. (b) Keywords and quantity of the related literature in 2018.

Fig. 24(a) shows that the proportion of theme types in each year has been relatively stable in the past five years. Fig. 24(b) shows the retrieval situation in 2018, and we searched the related literature in five years according to the above keywords. By comparison, the number of related articles in the past five years shows an apparent upward trend year by year. The most significant proportion is the design of mechanical aspect and circuit switches, which is mainly due to many studies on the combination of mechanical structures and circuits in practical application. The power extraction efficiency of TENGs can be enhanced by circuit management. In addition, using the corresponding mechanical structures for different application scenarios can better adapt to the actual application scenarios. Subsequently, with the combination of innovative designs and new materials, various studies based on TENGs have been reported. According to the different application scenarios, TENGs composed of different structures and materials are developed. For example, the structure of vertical contact–separation mode is adopted when collecting vibration energy, and hydrophobic materials are more inclined to be used when collecting wave energy.

Conclusion

In this review, we thoroughly discuss and summarize the research in recent years on improving the output performance of TENGs, including methods in material processing, mechanical design, and circuit management. Firstly, we summarize and discuss the surface treatment of the TENG interface and the application of biodegradable materials. According to different treatment methods, the research on improving the performance of TENGs by surface treatment of materials is summarized into two categories: physical modifications and chemical modifications. Such methods mainly enhance the output of TENGs by adjusting the surface charge density of triboelectric layers. In addition, it is considered to improve the output performance of TENGs by the preparation of high-performance biodegradable materials. Secondly, we summarize the research progress of improving the output performance based on mechanical design, including multilayer structures, TENG networks, and gear structures. Among them, the design of multilayer structures and gear structures is mainly manifested in improving the output performance of TENGs by increasing the motion frequency of devices. TENG networks improve the total output power by increasing the number of TENG units. Finally, we present state-of-the-art research in circuit management from the perspective of circuit switches, including mechanical and electronic switches. The working principle of mechanical switches is generally to accumulate and then release energy, which improves energy extraction efficiency. Electronic switches usually store the energy output from TENGs into the front-end capacitor based on optimal capacitance theory. These switches are usually used to collect mechanical energy in the environment with irregular frequency and improve the energy conversion efficiency of TENG devices.

With the increasing demand for renewable energy and self-powered technology, improving the performance of TENGs will promote their wide use in industrial applications. This involves the integration with the Internet of Things, smart cities and renewable energy systems. TENG-based intelligent sensing technology provides a wide range of material selection and diverse structural design, which has great potential in the construction of intelligent applications related to the Internet of Things in the 5G era.^297,298 In the intelligent transportation field, the power consumption of microelectronic chips like sensors is decreasing and the power of TENGs is increasing, which makes it possible for all sensors in the transportation field to realize self-driving.^299,300 In addition, in terms of smart wearable devices, these methods of improving performance contribute to the development of new wearable devices. Through the continuous efforts of researchers, the idea of collecting human mechanical energy to provide power for biomedical systems has been achieved.^301,302 The improved TENGs with biocompatibility realize continuous monitoring of physiological signals to support timely disease diagnosis, and promote revolutionary changes in medical technology.

Obstacles and prospects

In summary, researchers have raised the output performance of TENGs from various perspectives, and achieved a series of exciting results, laying the foundation for future energy collection. However, their application in various environments still has some limitations. The following suggestions are put forward to promote the development of TENGs in the fields of energy collection and intelligent sensing:

(1) The position of materials in the triboelectric series determines the output performance of TENGs during operation. Physical modifications of material surfaces can improve their triboelectric properties and even change their position in the triboelectric series. However, researchers’ understanding of the friction-induced charging mechanism remains in its infancy. The research community is devoted to understanding and establishing a precise quantitative relationship between the output charge and the physically modified surface, which is groundbreaking and challenging research. Additionally, optimizing the charge output of TENGs can be accomplished by altering the contact area and introducing micro/nano structures on the material surface. Unfortunately, the TENG working mechanism necessitates the mutual contact and friction of two triboelectric layers, which can cause uneven surfaces to wear and tear, impacting the output stability of TENGs. For instance, micro/nano structures may deform or detach during friction, and sanding may cause TENG devices to function erratically. Therefore, to improve the lifespan and output stability of TENGs, developing materials with appropriate modulus, elasticity and excellent wear resistance is a path worth considering. In addition, TENG devices can also be affected by optics and heat when they collect mechanical energy. Therefore, a physical modification method based on coupling multiple physical fields enhances the adaptability and output performance of TENGs in various environments, which is an inspiring strategy.

(2) Introducing chemical functional groups or performing chemical modifications on the surface of materials enhances the electrical performance and chemical stability of TENGs. However, such materials also inevitably suffer from wear and tear due to repeated friction. Different from physical modifications, some self-healing chemical bonds or self-healing polymer materials can repair damaged TENG surfaces, which significantly improves the lifespan of TENGs. Furthermore, in certain specific environments, the surface may be affected by pollution or deposits and even corroded, which impacts its performance and output stability. Surface self-cleaning technology can be used to minimize the adhesion of pollutants and extend its service life. Additionally, by introducing chemical functional groups on the materials, the hydrophilicity or hydrophobicity of the surface is improved and the corrosion can be slowed or avoided.

(3) TENGs based on degradable materials effectively reduce the environmental pollution caused by heavy metals and non-degradable plastics and promote the development of green energy technology. Although degradable materials can be naturally degraded in the environment, TENG devices need to exhibit certain long-term reliability and availability. For some implantable medical TENGs, it is necessary to use materials that can degrade inside the body of patients. In addition, the surface microstructure remains a crucial factor affecting the electrical properties of degradable materials. To ensure the stability and efficiency of modified devices, the performance of degradable materials must meet certain requirements, e.g., wear resistance, conductivity, transparency, and flexibility. Meanwhile, DB-TENGs have been widely studied, but high-performance DB-TENGs are still rare. Therefore, it is suggested to improve the output performance of DB-TENGs by combining various methods to expand their application prospects.

(4) In terms of mechanical design, this review mainly introduces the related research of multilayer structures, adjusting mechanical frequency, and TENG networks in recent years. It is an effective means to improve the output charge density through mechanical design. It has been found that the cells of a stacked structure interfere with each other, and the energy output efficiency of cells can be improved by changing the arrangement of TENGs. TENG networks have been widely applied in the field of large-scale energy collection. It is suggested that the layout and arrangement of TENG networks may be improved to ensure the structural stability and reliability of TENG networks under strong excitation. The mechanisms and regularities of charge transfer and potential distribution in TENG networks are still unclear. It is recommended to increase the studies combining theoretical calculations and experimental analysis in future explorations. At the same time, the output characteristics of TENG networks are closely related to circuit matching, so it may be necessary to conduct in-depth research on their output characteristics to achieve maximum energy conversion efficiency.

(5) The research on power management has made great progress in recent years. But power management based on switches still has plenty of space for improvement. For example, the long-term use of mechanical switches will lead to mechanical wear and looseness, which will result in a decrease in energy conversion efficiency. In future work, it is suggested to enhance the mechanical durability of mechanical switches by improving the wear resistance of mechanical materials and enhancing the reliability of the structural design. The response speed of the TENG electronic switches directly affects their practicality. Hence, it is suggested to improve the response speed of the electronic switch by optimizing circuit structure, material selection and control algorithm. In addition, the circuit topology plays a key role in system performance. For different application scenarios, it is suggested to design a more adaptable circuit topology. For example, in the case of high-voltage output, the difficulty of a high-voltage excitation circuit can be solved by choosing a suitable circuit structure.

(6) With the further development of the Internet of Things, the improved TENGs have been widely applied to convenient facilities in smart homes, robots in industry, and detection systems in smart cities by combining wireless sensors.^303–306 TENGs are moving toward intelligent self-powered sensors and display huge development prospects. For example, TENG-based self-powered sensors are used for human identity recognition and home safety monitoring. In view of the research hotspots of intelligent industries such as robots and mechanical arms, TENGs still present significant potential for development in sensing systems such as electronic skin. In the traffic of smart cities, the self-powered sensors developed by energy collection technology combined with signal processing provide an important technical basis for remote control and range monitoring.

Author contributions

C. C. wrote the original draft, F. S., Q. Z., C. H., Y. G., H. G., Z. L., and Z. W. revised the manuscript, and Y. P. and Z. L. provided funding.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was funded by the National Natural Science Foundation of China (No.: 62001281; No.: 62225308) and the Shanghai Science and Technology Committee (22dz1204300).

Notes and references

1. F. Johnsson and J. Kjarstad, Tsinghua Univ, Inst Thermal Engn, Beijing, Peoples R China, 2015.
2. M. M. Farid, A. M. Khudhair, S. A. K. Razack and S. Al-Hallaj, Energy Convers. Manage., 2004, 45, 1597–1615.
3. F. R. Fan, Z. Q. Tian and Z. L. Wang, Nano Energy, 2012, 1, 328–334.
4. S. M. Niu, S. H. Wang, L. Lin, Y. Liu, Y. S. Zhou, Y. F. Hu and Z. L. Wang, Energy Environ. Sci., 2013, 6, 3576–3583.
5. S. H. Wang, L. Lin and Z. L. Wang, Nano Lett., 2012, 12, 6339–6346.
6. B. Mahapatra, K. K. Patel, Vidya and P. K. Patel, Mater. Today-Proc., 2021, 46, 5523–5529.
7. H. Kulah and K. Najafi, IEEE, Maastricht, Netherlands, 2004.
8. O. Langhamer, K. Haikonen and J. Sundberg, Renewable Sustainable Energy Rev., 2010, 14, 1329–1335.
9. P. P. Wang, M. M. Hu, H. Wang, Z. Chen, Y. P. Feng, J. Q. Wang, W. Ling and Y. Huang, Adv. Sci., 2020, 7, 29.
10. F. R. Fan, W. Tang and Z. L. Wang, Adv. Mater., 2016, 28, 4283–4305.
11. J. Chen and Z. L. Wang, Joule, 2017, 1, 480–521.
12. Y. Ji, K. W. Zhang and Y. Yang, Adv. Sci., 2018, 5, 8.
13. K. W. Zhang, S. H. Wang and Y. Yang, Adv. Energy Mater., 2017, 7, 8.
14. B. Chen, Y. Yang and Z. L. Wang, Adv. Energy Mater., 2018, 8, 13.
15. Y. X. Bian, T. Jiang, T. X. Xiao, W. P. Gong, X. Cao, Z. N. Wang and Z. L. Wang, Adv. Mater. Technol., 2018, 3, 7.
16. L. X. He, C. G. Zhang, B. F. Zhang, O. U. Yang, W. Yuan, L. L. Zhou, Z. H. Zhao, Z. Y. Wu, J. Wang and Z. L. Wang, ACS Nano, 2022, 16, 6244–6254.
17. F. Shen, Z. J. Li, H. Y. Guo, Z. B. Yang, H. Wu, M. Wang, J. Luo, S. R. Xie, Y. Peng and H. Y. Pu, Adv. Electron. Mater., 2021, 7, 26.
18. Y. T. Hao, X. M. Li, B. D. Chen and Z. Y. Zhu, Front. Mar. Sci., 2022, 9, 6.
19. Q. Zhang, C. F. Xin, F. Shen, Y. Gong, Y. L. Zi, H. Y. Guo, Z. J. Li, Y. Peng, Q. Zhang and Z. L. Wang, Energy Environ. Sci., 2022, 15, 3688–3721.
20. Z. J. Li, J. Luo, S. R. Xie, L. M. Xin, H. Y. Guo, H. Y. Pu, P. L. Yin, Z. B. Xu, D. Zhang, Y. Peng, Z. B. Yang and H. Naguib, Nano Energy, 2021, 81, 9.
21. C. Chen, Z. Wen, J. H. Shi, X. H. Jian, P. Y. Li, J. T. W. Yeow and X. H. Sun, Nat. Commun., 2020, 11, 9.
22. J. B. Yu, L. Chen, X. J. Hou, J. L. Mu, J. He, W. P. Geng, X. J. Qiao and X. J. Chou, J. Materiomics, 2022, 8, 247–256.
23. Q. Zheng, H. Zhang, B. J. Shi, X. Xue, Z. Liu, Y. M. Jin, Y. Ma, Y. Zou, X. X. Wang, Z. An, W. Tang, W. Zhang, F. Yang, Y. Liu, X. L. Lang, Z. Y. Xu, Z. Li and Z. L. Wang, ACS Nano, 2016, 10, 6510–6518.
24. H. Chen, Q. X. Lu, X. Cao, N. Wang and Z. L. Wang, Nano Res., 2022, 15, 2505–2511.
25. P. S. Rui, W. Zhang and P. H. Wang, ACS Nano, 2021, 15, 6949–6960.
26. R. Wang, M. M. Ji, X. Jin, Q. R. Zhang and T. F. Jiao, Nano Futures, 2022, 6, 16.
27. T. Guo, G. X. Liu, Y. K. Pang, B. Wu, F. B. Xi, J. Q. Zhao, T. Z. Bu, X. P. Fu, X. J. Li, C. Zhang and Z. L. Wang, Extreme Mech. Lett., 2018, 18, 1–8.
28. L. Lin, Y. F. Hu, C. Xu, Y. Zhang, R. Zhang, X. N. Wen and Z. L. Wang, Nano Energy, 2013, 2, 75–81.
29. C. G. Zhang, Y. B. Liu, B. F. Zhang, O. Yang, W. Yuan, L. X. He, X. L. Wei, J. Wang and Z. L. Wang, ACS Energy Lett., 2021, 6, 1490–1499.
30. X. Y. Wei, G. Zhu and Z. L. Wang, Nano Energy, 2014, 10, 83–89.
31. X. Cui, Y. M. Zhang, G. W. Hu, L. Zhang and Y. Zhang, Nano Energy, 2020, 70, 7.
32. S. Kim, J. Ha and J. B. Kim, Integr. Ferroelectr., 2016, 176, 283–290.
33. A. Sutka, J. Ruza, M. Jarvekulg, A. Linarts, K. Malnieks, V. Jurkans, I. Gornevs, J. Blums, K. Rubenis and M. Knite, J. Electrost., 2018, 92, 1–5.
34. W. Kim, T. Okada, H. W. Park, J. Kim, S. Kim, S. W. Kim, S. Samukawa and D. Choi, J. Mater. Chem. A, 2019, 7, 25066–25077.
35. T. Jiang, L. M. Zhang, X. Y. Chen, C. B. Han, W. Tang, C. Zhang, L. Xu and Z. L. Wang, ACS Nano, 2015, 9, 12562–12572.
36. V. Palaniappan, X. Zhang, D. Maddipatla, B. B. Narakathu, B. J. Bazuin and M. Z. Atashbar, IEEE, Electr Network, 2021.
37. N. Y. Cui, L. Gu, Y. M. Lei, J. M. Liu, Y. Qin, X. H. Ma, Y. Hao and Z. L. Wang, ACS Nano, 2016, 10, 6131–6138.
38. I. Kim, T. Y. Ahn and D. Kim, Micromachines, 2021, 12, 14.
39. C. J. Wang, F. Meng, Q. Fu, C. H. Fan and L. Cui, Micromachines, 2022, 13, 11.
40. S. M. Niu, Y. S. Zhou, S. H. Wang, Y. Liu, L. Lin, Y. Bando and Z. L. Wang, Nano Energy, 2014, 8, 150–156.
41. A. Kawaguchi, H. Uchiyama, M. Matsunaga and Y. Ohno, Appl. Phys. Express, 2021, 14, 5.
42. X. L. Cheng, L. M. Miao, Y. Song, Z. M. Su, H. T. Chen, X. X. Chen, J. X. Zhang and H. X. Zhang, Nano Energy, 2017, 38, 448–456.
43. F. B. Xi, Y. K. Pang, W. Li, T. Jiang, L. M. Zhang, T. Guo, G. X. Liu, C. Zhang and Z. L. Wang, Nano Energy, 2017, 37, 168–176.
44. J. H. Yang, J. Q. Cao, J. Han, Y. Xiong, L. Luo, X. Z. Dan, Y. J. Yang, L. L. Li, J. Sun and Q. J. Sun, Nano Energy, 2022, 101, 10.
45. C. F. Xin, Z. J. Li, Q. Zhang, Y. Peng, H. Y. Guo and S. R. Xie, Nano Energy, 2022, 100, 12.
46. S. H. Shin, Y. E. Bae, H. K. Moon, J. Kim, S. H. Choi, Y. Kim, H. J. Yoon, M. H. Lee and J. Nah, ACS Nano, 2017, 11, 6131–6138.
47. C. G. Zhang, Z. H. Zhao, O. Yang, W. Yuan, L. L. Zhou, X. Yin, L. Liu, Y. H. Li, Z. L. Wang and J. Wang, Adv. Mater. Technol., 2020, 5, 7.
48. L. Xu, T. Jiang, P. Lin, J. J. Shao, C. He, W. Zhong, X. Y. Chen and Z. L. Wang, ACS Nano, 2018, 12, 1849–1858.
49. I. Kara, M. Becermis, M. A. A. Kamar, M. Aktan, H. Dogan and S. Mutlu, IEEE Trans. Circuits Syst. I-Regul. Pap., 2021, 68, 210–223.
50. C. Xu, B. B. Zhang, A. C. Wang, W. Z. Cai, Y. L. Zi, P. Z. Feng and Z. L. Wang, Adv. Funct. Mater., 2019, 29, 8.
51. L. S. McCarty and G. M. Whitesides, Angew. Chem., Int. Ed., 2008, 47, 2188–2207.
52. Z. L. Wang and A. C. Wang, Mater. Today, 2019, 30, 34–51.
53. Y. J. Sun, X. Huang and S. Soh, Angew. Chem., Int. Ed., 2016, 55, 9956–9960.
54. Z. L. Wang, Rep. Prog. Phys., 2021, 84, 68.
55. Z. L. Wang, Mater. Today, 2022, 52, 348–363.
56. Z. L. Wang, Mater. Today, 2017, 20, 74–82.
57. Z. L. Wang, Nano Energy, 2020, 68, 10.
58. Z. L. Wang, J. Phys. Commun., 2022, 6, 085013.
59. Z. L. Wang, Int. J. Mod. Phys. B, 2023, 37, 32.
60. H. Y. Zou, Y. Zhang, L. T. Guo, P. H. Wang, X. He, G. Z. Dai, H. W. Zheng, C. Y. Chen, A. C. Wang, C. Xu and Z. L. Wang, Nat. Commun., 2019, 10, 9.
61. Y. L. Zi, S. M. Niu, J. Wang, Z. Wen, W. Tang and Z. L. Wang, Nat. Commun., 2015, 6, 8.
62. Y. Yang, H. Zhang, R. Liu, X. Wen, T. C. Hou and Z. L. Wang, Adv. Energy Mater., 2013, 3, 1563–1568.
63. J. Tian, X. Chen and Z. L. Wang, Nanotechnology, 2020, 31, 242001.
64. G. Khandelwal, A. Chandrasekhar, N. P. Maria Joseph Raj and S. J. Kim, Adv. Energy Mater., 2019, 9, 1803581.
65. J. Wang, Z. Wen, Y. Zi, P. Zhou, J. Lin, H. Guo, Y. Xu and Z. L. Wang, Adv. Funct. Mater., 2016, 26, 1070–1076.
66. A. C. Wang, B. Zhang, C. Xu, H. Zou, Z. Lin and Z. L. Wang, Adv. Funct. Mater., 2020, 30, 1909384.
67. J. Wu, X. Wang, J. He, Z. Li and L. Li, J. Mater. Chem. A, 2021, 9, 6583–6590.
68. J. Park, A. Y. Choi, C. J. Lee, D. Kim and Y. T. Kim, RSC Adv., 2017, 7, 54829–54834.
69. Y. Lin, S. Duan, D. Zhu, Y. Li, B. Wang and J. Wu, Adv. Intell. Syst., 2021, 2100120.
70. S. Wang, Y. Xie, S. Niu, L. Lin and Z. L. Wang, Adv. Mater., 2014, 26, 2818–2824.
71. W. Paosangthong, R. Torah and S. Beeby, J. Phys.: Conf. Ser., 2019, 1407, 012124.
72. G. Zhu, C. F. Pan, W. X. Guo, C. Y. Chen, Y. S. Zhou, R. M. Yu and Z. L. Wang, Nano Lett., 2012, 12, 4960–4965.
73. S. Niu and Z. L. Wang, Nano Energy, 2015, 14, 161–192.
74. L. Lin, S. Wang, Y. Xie, Q. Jing, S. Niu, Y. Hu and Z. L. Wang, Nano Lett., 2013, 13, 2916–2923.
75. P. Bai, G. Zhu, Y. Liu, J. Chen, Q. Jing, W. Yang, J. Ma, G. Zhang and Z. L. Wang, ACS Nano, 2013, 7, 6361–6366.
76. S. M. Niu, Y. Liu, S. H. Wang, L. Lin, Y. S. Zhou, Y. F. Hu and Z. L. Wang, Adv. Mater., 2013, 25, 6184–6193.
77. J. Chen, Y. Huang, N. Zhang, H. Zou, R. Liu, C. Tao, X. Fan and Z. L. Wang, Nat. Energy, 2016, 1, 1–8.
78. S. Liu, W. Xu, J. Yang, S. Liu, Y. Meng, L. Jia, G. Chen, Y. Qin and X. Li, ACS Appl. Electron. Mater., 2022, 4, 2851–2858.
79. W. Yang, J. Chen, X. Wen, Q. Jing, J. Yang, Y. Su, G. Zhu, W. Wu and Z. L. Wang, ACS Appl. Mater. Interfaces, 2014, 6, 7479–7484.
80. F. Yi, L. Lin, S. Niu, P. K. Yang, Z. Wang, J. Chen, Y. Zhou, Y. Zi, J. Wang and Q. Liao, Adv. Funct. Mater., 2015, 25, 3688–3696.
81. S. M. Niu, Y. Liu, S. H. Wang, L. Lin, Y. S. Zhou, Y. F. Hu and Z. L. Wang, Adv. Funct. Mater., 2014, 24, 3332–3340.
82. Y. Xie, S. Wang, S. Niu, L. Lin, Q. Jing, J. Yang, Z. Wu and Z. L. Wang, Adv. Mater., 2014, 26, 6599–6607.
83. P. Cheng, H. Guo, Z. Wen, C. Zhang, X. Yin, X. Li, D. Liu, W. Song, X. Sun and J. Wang, Nano Energy, 2019, 57, 432–439.
84. S. M. Niu, Y. Liu, X. Y. Chen, S. H. Wang, Y. S. Zhou, L. Lin, Y. N. Xie and Z. L. Wang, Nano Energy, 2015, 12, 760–774.
85. D. Liu, Y. Gao, L. Zhou, J. Wang and Z. L. Wang, Nano Res., 2023, 1–20.
86. https://en.wikipedia.org/wiki/Van_de_Graaff_generator .
87. A. F. Diaz and R. M. Felix-Navarro, J. Electrost., 2004, 62, 277–290.
88. R. Y. Zhang and H. Olin, EcoMat, 2020, 2, 13.
89. S. A. Lone, K. C. Lim, K. Kaswan, S. Chatterjee, K. P. Fan, D. Choi, S. Lee, H. L. Zhang, J. Cheng and Z. H. Lin, Nano Energy, 2022, 99, 33.
90. S. Y. Chao, H. Ouyang, D. J. Jiang, Y. B. Fan and Z. Li, EcoMat, 2021, 3, 19.
91. R. Dharmasena and S. R. P. Silva, Nano Energy, 2019, 62, 530–549.
92. W. X. Yang, X. L. Wang, H. Q. Li, J. Wu, Y. Q. Hu, Z. H. Li and H. Liu, Nano Energy, 2019, 57, 41–47.
93. F. R. Fan, L. Lin, G. Zhu, W. Z. Wu, R. Zhang and Z. L. Wang, Nano Lett., 2012, 12, 3109–3114.
94. X. S. Zhang, M. D. Han, R. X. Wang, F. Y. Zhu, Z. H. Li, W. Wang and H. X. Zhang, Nano Lett., 2013, 13, 1168–1172.
95. C. K. Jeong, K. M. Baek, S. M. Niu, T. W. Nam, Y. H. Hur, D. Y. Park, G. T. Hwang, M. Byun, Z. L. Wang, Y. S. Jung and K. J. Lee, Nano Lett., 2014, 14, 7031–7038.
96. A. R. Mule, B. Dudem and J. S. Yu, Energy, 2018, 165, 677–684.
97. D. Z. Zhang, Y. Yang, Z. Y. Xu, D. Y. Wang and C. Du, J. Mater. Chem. A, 2022, 10, 10935–10949.
98. L. M. Zhao, Q. Zheng, H. Ouyang, H. Li, L. Yan, B. J. Shi and Z. Li, Nano Energy, 2016, 28, 172–178.
99. X.-W. Zhang, G.-Z. Li, G.-G. Wang, J.-L. Tian, Y.-L. Liu, D.-M. Ye, Z. Liu, H.-Y. Zhang and J.-C. Han, ACS Sustainable Chem. Eng., 2018, 6, 2283–2291.
100. M. L. Seol, S. H. Lee, J. W. Han, D. Kim, G. H. Cho and Y. K. Choi, Nano Energy, 2015, 17, 63–71.
101. I.-W. Tcho, W.-G. Kim, S.-B. Jeon, S.-J. Park, B. J. Lee, H.-K. Bae, D. Kim and Y.-K. Choi, Nano Energy, 2017, 42, 34–42.
102. H. Phan, P. N. Hoa, H. A. Tam and P. D. Thang, J. Sci., 2021, 6, 84–91.
103. X. Sun, Y. J. Liu, N. Luo, Y. Liu, Y. E. Feng, S. G. Chen and D. A. Wang, Nano Energy, 2022, 102, 16.
104. M. K. Kim, M. S. Kim, H. B. Kwon, S. E. Jo and Y. J. Kim, RSC Adv., 2017, 7, 48368–48373.
105. T. Li, P. Pan, Z. C. Yang and X. P. Yang, J. Mater. Sci., 2022, 57, 6723–6733.
106. C. Liang, L. W. Zhang, N. Zhao, W. Zhou, X. Deng, Y. G. Zhou, Y. X. Hou and W. Q. Zhang, Front. Energy Res., 2022, 10, 10.
107. Y. Y. Yang, J. H. Qian, Y. H. Dong, M. A. Nikitina, U. Zafari and X. Q. Xiao, Results Eng., 2022, 14, 5.
108. F. E. Ahmed, B. S. Lalia and R. Hashaikeh, Desalination, 2015, 356, 15–30.
109. F. Chen, Y. Wu, Z. Ding, X. Xia, S. Li, H. Zheng, C. Diao, G. Yue and Y. Zi, Nano Energy, 2019, 56, 241–251.
110. M. Zhang, X. Zhao, G. Zhang, G. Wei and Z. Su, J. Mater. Chem. B, 2017, 5, 1699–1711.
111. T. Huang, M. X. Lu, H. Yu, Q. H. Zhang, H. Z. Wang and M. F. Zhu, Sci. Rep., 2015, 5, 8.
112. X. Chen, J. Xiong, K. Parida, M. Guo, C. Wang, C. Wang, X. Li, J. Shao and P. S. Lee, Nano Energy, 2019, 64, 103904.
113. D. T. Mathew, K. V. Vijoy, N. V. Nayar, N. Manoj, K. J. Saji, S. C. Pillai and H. John, ACS Sustainable Chem. Eng., 2022, 10, 14126–14135.
114. X. Zhang, S. S. Lv, X. C. Lu, H. Yu, T. Huang, Q. H. Zhang and M. F. Zhu, Nano Energy, 2020, 75, 10.
115. J. C. Zhao, F. J. Li, Z. L. Wang, P. Dong, G. T. Xia and K. Wang, J. Mater. Sci.: Mater. Electron., 2021, 32, 14715–14727.
116. Y. H. Wu, J. K. Qu, W. A. Daoud, L. Y. Wang and T. Qi, J. Mater. Chem. A, 2019, 7, 13347–13355.
117. J. Y. Xia, Z. P. Zheng and Y. P. Guo, Composites, Part A, 2022, 157, 7.
118. F. J. Xing, Z. Q. Ou, X. B. Gao, B. D. Chen and Z. L. Wang, Adv. Funct. Mater., 2022, 32, 2205275.
119. C. Y. Sun, G. Q. Zu, Y. Wei, X. L. Song and X. J. Yang, Langmuir, 2022, 38, 1479–1487.
120. M. Su and B. Kim, ACS Appl. Nano Mater., 2020, 3, 9759–9770.
121. M. Ranjbar-Mohammadi, S. Kargozar, S. H. Bahrami and M. Joghataei, J. Ind. Text., 2017, 46, 1170–1192.
122. Y. Kim, S. Jang, B. J. Kang and J. H. Oh, Appl. Phys. Lett., 2017, 111, 073502.
123. Y.-P. Chen, H.-Y. Liu, Y.-W. Liu, T.-Y. Lee and S.-J. Liu, J. Nanomater., 2019, 2019, 2626085.
124. X. R. Zhou, K. Parida, O. Halevi, Y. Z. Liu, J. Q. Xiong, S. Magdassi and P. S. Lee, Nano Energy, 2020, 72, 10.
125. N. Hu, D. J. Chen, D. Wang, S. C. Huang, I. Trase, H. M. Grover, X. J. Yu, J. X. J. Zhang and Z. Chen, Phys. Rev. Appl., 2018, 9, 6.
126. S. Bose, S. Vahabzadeh and A. Bandyopadhyay, Mater. Today, 2013, 16, 496–504.
127. H.-J. Yoon, D.-H. Kim, W. Seung, U. Khan, T. Y. Kim, T. Kim and S.-W. Kim, Nano Energy, 2019, 63, 103857.
128. C. Qian, L. Li, M. Gao, H. Yang, Z. Cai, B. Chen, Z. Xiang, Z. Zhang and Y. Song, Nano Energy, 2019, 63, 103885.
129. R. H. Zheng, Y. X. Chen, H. Chi, H. Qiu, H. Xue and H. Bai, ACS Appl. Mater. Interfaces, 2020, 12, 57441–57449.
130. R. I. Haque, O. Chandran, S. Lani and D. Briand, Nano Energy, 2018, 52, 54–62.
131. Z. H. Yu, Y. M. Wang, J. Q. Zheng, Y. Xiang, P. Zhao, J. Q. Cui, H. M. Zhou and D. Q. Li, Nano Energy, 2020, 68, 7.
132. A. Mohapatra, N. Divakaran, Y. Alex, P. V. A. Kumar and S. Mohanty, Mater. Today Nano, 2023, 22, 9.
133. Z. W. Wang, C. C. Luan, Y. B. Zhu, G. X. Liao, J. P. Liu, X. J. Li, X. H. Yao and J. Z. Fu, Nano Energy, 2021, 90, 11.
134. H. Li, R. H. Li, X. T. Fang, H. W. Jiang, X. R. Ding, B. Tang, G. F. Zhou, R. Zhou and Y. Tang, Nano Energy, 2019, 58, 447–454.
135. A. Ahmed, I. Hassan, I. M. Mosa, E. Elsanadidy, G. S. Phadke, M. F. El-Kady, J. F. Rusling, P. R. Selvaganapathy and R. B. Kaner, Nano Energy, 2019, 60, 17–25.
136. N. Godard, L. Allirol, A. Latour, S. Glinsek, M. Gerard, J. Polesel, F. D. Dos Santos and E. Defay, Cell Rep. Phys. Sci., 2020, 1, 13.
137. D. Hong, Y. M. Choi, Y. Jang and J. Jeong, Int. J. Energy Res., 2018, 42, 3688–3695.
138. U. J. Yang, J. W. Lee, J. P. Lee and J. M. Baik, Nano Energy, 2019, 57, 293–299.
139. C. Yao, X. Yin, Y. Yu, Z. Cai and X. Wang, Adv. Funct. Mater., 2017, 27, 1700794.
140. J. W. Lee, S. Jung, J. Jo, G. H. Han, D. M. Lee, J. Oh, H. J. Hwang, D. Choi, S. W. Kim, J. H. Lee, C. Yang and J. M. Baik, Energy Environ. Sci., 2021, 14, 1004–1015.
141. S. H. Wang, Y. L. Zi, Y. S. Zhou, S. M. Li, F. R. Fan, L. Lin and Z. L. Wang, J. Mater. Chem. A, 2016, 4, 3728–3734.
142. K. E. Byun, Y. Cho, M. Seol, S. Kim, S. W. Kim, H. J. Shin, S. Park and S. Hwang, ACS Appl. Mater. Interfaces, 2016, 8, 18519–18525.
143. S. H. Shin, Y. H. Kwon, Y. H. Kim, J. Y. Jung, M. H. Lee and J. Nah, ACS Nano, 2015, 9, 4621–4627.
144. R. M. Wen, B. Zhao, L. M. Fan, J. M. Guo and J. Y. Zhai, Adv. Mater. Technol., 2023, 8, 2201330.
145. M. P. Kim, C. W. Ahn, Y. Lee, K. Kim, J. Park and H. Ko, Nano Energy, 2021, 82, 11.
146. X. Kang, C. Pan, Y. Chen and X. Pu, RSC Adv., 2020, 10, 17752–17759.
147. M. H. Lai, L. Cheng, Y. Xi, Y. H. Wu, C. G. Hu, H. Y. Guo, B. L. Du, G. L. Liu, Q. P. Liu and R. C. Liu, J. Phys. D: Appl. Phys., 2018, 51, 10.
148. Y. Yu, Z. Li, Y. Wang, S. Gong and X. Wang, Adv. Mater., 2015, 27, 4938–4944.
149. J. Chen, H. Y. Guo, X. M. He, G. L. Liu, Y. Xi, H. F. Shi and C. G. Hu, ACS Appl. Mater. Interfaces, 2016, 8, 736–744.
150. W. Seung, H. J. Yoon, T. Y. Kim, H. Ryu, J. Kim, J. H. Lee, J. H. Lee, S. Kim, Y. K. Park, Y. J. Park and S. W. Kim, Adv. Energy Mater., 2017, 7, 8.
151. V. Harnchana, H. V. Ngoc, W. He, A. Rasheed, H. Park, V. Amornkitbamrung and D. J. Kang, ACS Appl. Mater. Interfaces, 2018, 10, 25263–25272.
152. Y. H. Song, M. Liu, J. K. Bao, Y. Hu, M. H. Xu, Z. B. Yang, Q. L. Yang, H. P. Cai, C. X. Xiong and Z. Q. Shi, Carbohydr. Polym., 2022, 298, 10.
153. S. Li, Y. Fan, H. Chen, J. Nie, Y. Liang, X. Tao, J. Zhang, X. Chen, E. Fu and Z. L. Wang, Energy Environ. Sci., 2020, 13, 896–907.
154. J. N. Dong, S. Y. Huang, J. J. Luo, J. W. Zhao, F. R. Fan and Z. Q. Tian, Nano Energy, 2022, 95, 7.
155. R. K. Vatsa and D. Mathur, J. Phys. Chem. Lett., 2020, 11, 9842–9845.
156. H. Y. Wu, S. K. Fu, W. C. He, C. C. Shan, J. Wang, Y. Du, S. H. Du, B. X. Li and C. G. Hu, Adv. Funct. Mater., 2022, 32, 8.
157. J. W. Lee, S. Jung, T. W. Lee, J. Jo, H. Y. Chae, K. Choi, J. J. Kim, J. H. Lee, C. Yang and J. M. Baik, Adv. Energy Mater., 2019, 9, 11.
158. J. Seo, S. Hajra, M. Sahu and H. J. Kim, Mater. Lett., 2021, 304, 4.
159. T. Zhou, L. M. Zhang, F. Xue, W. Tang, C. Zhang and Z. L. Wang, Nano Res., 2016, 9, 1442–1451.
160. J. E. Kim, K. Panda, J. I. J. Choi and J. Y. Park, Nanoscale, 2019, 11, 6120–6128.
161. Y. B. Liu, C. G. Zhang, B. F. Zhang, W. Yuan, O. Yang, Y. X. Hu, L. L. Zhou, Z. H. Zhao, J. F. Xiao, Z. L. Wang and J. Wang, J. Mater. Chem. A, 2022, 10, 16547–16554.
162. L. L. Ren, A. F. Yu, W. Wang, D. Guo, M. M. Jia, P. W. Guo, Y. F. Zhang, Z. L. Wang and J. Y. Zhai, Nano Lett., 2021, 21, 10099–10106.
163. Q. Luo, J. J. Gao, S. Y. Lin, D. L. Xiong, H. N. Sun and J. B. Guo, J. Mater. Chem. C, 2022, 10, 1808–1815.
164. J. Wang, Z. Wen, Y. L. Zi, P. F. Zhou, J. Lin, H. Y. Guo, Y. L. Xu and Z. L. Wang, Adv. Funct. Mater., 2016, 26, 1070–1076.
165. A. A. Budnikov, P. V. Zharkov and Y. D. Chashechkin, Moscow Univ. Phys. Bull., 2012, 67, 403–408.
166. M. Connan, V. Perold, B. Dilley, C. Barbraud, Y. Cherel and P. G. Ryan, Mar. Pollut. Bull., 2021, 169, 11.
167. P. Bai, G. Zhu, Z.-H. Lin, Q. Jing, J. Chen, G. Zhang, J. Ma and Z. L. Wang, ACS Nano, 2013, 7, 3713–3719.
168. M. Y. Yang, X. Tian and T. Hua, Chem. Eng. J., 2022, 442, 9.
169. J. Kaur, H. Singh, R. S. Sawhney, T. Sui and M. A. R. Loja, Mater. Chem. Phys., 2022, 291, 13.
170. M. L. Foresti, A. Vazquez and B. Boury, Carbohydr. Polym., 2017, 157, 447–467.
171. Q. Z. Zhong, J. W. Zhong, B. Hu, Q. Y. Hu, J. Zhou and Z. L. Wang, Energy Environ. Sci., 2013, 6, 1779–1784.
172. C. H. Yao, A. Hernandez, Y. H. Yu, Z. Y. Cai and X. D. Wang, Nano Energy, 2016, 30, 103–108.
173. J. Peng, H. Zhang, Q. Zheng, C. M. Clemons, R. C. Sabo, S. Gong, Z. Ma and L. S. Turng, Nanoscale, 2017, 9, 1428–1433.
174. S. X. Nie, Q. Fu, X. J. Lin, C. Y. Zhang, Y. X. Lu and S. F. Wang, Chem. Eng. J., 2021, 404, 9.
175. S. Roy, H. U. Ko, P. K. Maji, L. V. Hai and J. Kim, Chem. Eng. J., 2020, 385, 10.
176. Q. Zheng, Y. Zou, Y. L. Zhang, Z. Liu, B. J. Shi, X. X. Wang, Y. M. Jin, H. Ouyang, Z. Li and Z. L. Wang, Sci. Adv., 2016, 2, 9.
177. W. Y. Zhang, Q. Liu, S. Y. Chao, R. P. Liu, X. Cui, Y. Sun, H. Ouyang and Z. Li, ACS Appl. Mater. Interfaces, 2021, 13, 42966–42976.
178. J. Yang, J. An, Y. S. Sun, J. J. Zhang, L. L. Zu, H. Li, T. Jiang, B. D. Chen and Z. L. Wang, Nano Energy, 2022, 97, 11.
179. D. Liu, J. M. Liu, M. S. Yang, N. Y. Cui, H. Y. Wang, L. Gu, L. F. Wang and Y. Qin, Nano Energy, 2021, 88, 10.
180. N. Kamboj, K. Athira and R. S. Dey, Pandit Deendayal Petr Univ, Raysan, India, 2021.
181. S. C. Narayanan, K. Karpagam and A. Bhattacharyya, Fibers Polym., 2015, 16, 1269–1275.
182. F. He, X. You, H. Gong, Y. Yang, T. Bai, W. Wang, W. Guo, X. Liu and M. Ye, ACS Appl. Mater. Interfaces, 2020, 12, 6442–6450.
183. W. Lu, P. Yu, M. Jian, H. Wang, H. Wang, X. Liang and Y. Zhang, ACS Appl. Mater. Interfaces, 2020, 12, 11825–11832.
184. R. Ranjana, N. Parushuram, K. Harisha, S. Asha, B. Narayana, M. Mahendra and Y. Sangappa, J. Mater. Sci.: Mater. Electron., 2020, 31, 249–264.
185. C. M. Jiang, Q. Zhang, C. X. He, C. Zhang, X. H. Feng, X. J. Li, Q. Zhao, Y. B. Ying and J. F. Ping, Fundamental Res., 2022, 2, 974–984.
186. M. Lee, J. Shin, S. Kim and S. Gandla, Sensors, 2022, 22, 11.
187. C. M. Jiang, C. X. He, R. Z. Lin, X. J. Li, Q. Zhao, Y. B. Ying, J. Z. Song and J. F. Ping, Nano Energy, 2022, 101, 10.
188. S. Strassburg, S. Zainuddin and T. Scheibel, Adv. Energy Mater., 2021, 11, 19.
189. I. C. M. Candido, G. D. Oliveira, S. J. L. Ribeiro, M. Cavicchioli, H. S. Barud, L. G. Silva and H. P. de Oliveira, Nano Energy, 2023, 105, 9.
190. H. Gong, Z. J. Xu, Y. Yang, Q. C. Xu, X. Y. Li, X. Cheng, Y. R. Huang, F. Zhang, J. Z. Zhao, S. Y. Li, X. Y. Liu, Q. L. Huang and W. X. Guo, Biosens. Bioelectron., 2020, 169, 9.
191. S. K. Ghosh, M. P. Kim, S. Na, Y. Lee, J. Park, S. Cho, J. Cho, J. J. Kim and H. Ko, Nano Energy, 2022, 100, 12.
192. J. H. Xue, H. J. Xiang, Y. R. Zhang, J. Yang, X. Cao and Z. L. Wang, Energy Environ. Mater., 2023, e12492,  DOI:10.1002/eem2.12492.
193. Y. J. Han, Y. F. Han, X. P. Zhang, L. Li, C. W. Zhang, J. H. Liu, G. Lu, H. D. Yu and W. Huang, ACS Appl. Mater. Interfaces, 2020, 12, 16442–16450.
194. H. Y. Mi, X. Jing, Q. F. Zheng, L. M. Fang, H. X. Huang, L. S. Turng and S. Q. Gong, Nano Energy, 2018, 48, 327–336.
195. C. Y. Zhang, X. J. Lin, N. Zhang, Y. X. Lu, Z. M. Wu, G. L. Liu and S. X. Nie, Nano Energy, 2019, 66, 11.
196. X. X. Shi, S. Chen, H. L. Zhang, J. X. Jiang, Z. Q. Ma and S. Q. Gong, ACS Sustainable Chem. Eng., 2019, 7, 18657–18666.
197. Q. J. Liang, Q. Zhang, X. Q. Yan, X. Q. Liao, L. H. Han, F. Yi, M. Y. Ma and Y. Zhang, Adv. Mater., 2017, 29, 7.
198. H. Ryu, J. H. Lee, T. Y. Kim, U. Khan, J. H. Lee, S. S. Kwak, H. J. Yoon and S. W. Kim, Adv. Energy Mater., 2017, 7, 6.
199. Z. Li, H. Q. Feng, Q. Zheng, H. Li, C. C. Zhao, H. Ouyang, S. Noreen, M. Yu, F. Su, R. P. Liu, L. L. Li, Z. L. Wang and Z. Li, Nano Energy, 2018, 54, 390–399.
200. X. Peng, K. Dong, C. Y. Ye, Y. Jiang, S. Y. Zhai, R. W. Cheng, D. Liu, X. P. Gao, J. Wang and Z. L. Wang, Sci. Adv., 2020, 6, 10.
201. C. R. Liu, J. Q. Li, L. F. Che, S. Q. Chen, Z. K. Wang and X. F. Zhou, Nano Energy, 2017, 41, 359–366.
202. A. Y. Choi, C. J. Lee, J. Park, D. Kim and Y. T. Kim, Sci. Rep., 2017, 7, 6.
203. Z. J. Xu, W. Qiu, X. W. Fan, Y. T. Shi, H. Gong, J. N. Huang, A. Patil, X. Y. Li, S. T. Wang, H. B. Lin, C. Hou, J. Z. Zhao, X. Guo, Y. Yang, H. Z. Lin, L. F. Huang, X. Y. Liu and W. X. Guo, ACS Nano, 2021, 15, 12429–12437.
204. C. Wu, H. Huang, S. Yang and G. J. Wen, IEEE Sens. J., 2020, 20, 13999–14006.
205. W. Tang, B. Meng and H. X. Zhang, Nano Energy, 2013, 2, 1164–1171.
206. G. L. Liu, H. Y. Guo, L. Chen, X. Wang, D. P. Wei and C. G. Hu, Nano Res., 2016, 9, 3355–3363.
207. C. Song, X. Zhu, M. Wang, P. Yang, L. Chen, L. Hong and W. Cui, Sustainable Energy Technol. Assess., 2022, 53, 102767.
208. L. M. Zhang, C. B. Han, T. Jiang, T. Zhou, X. H. Li, C. Zhang and Z. L. Wang, Nano Energy, 2016, 22, 87–94.
209. P. L. Yin, K. C. Aw, X. M. Jiang, C. F. Xin, H. Y. Guo, L. H. Tang, Y. Peng and Z. J. Li, Nano Energy, 2022, 95, 11.
210. H. L. Zhou, X. L. Wei, B. C. Wang, E. Y. Zhang, Z. Y. Wu and Z. L. Wang, Adv. Funct. Mater., 2023, 33, 2210920.
211. G. Li, S. K. Fu, C. Y. Luo, P. Wang, Y. Du, Y. T. Tang, Z. Wang, W. C. He, W. L. Liu, H. Y. Guo, J. Chen and C. G. Hu, Nano Energy, 2022, 96, 10.
212. S. J. Liu, X. Liang, P. F. Chen, H. R. Long, T. Jiang and Z. L. Wang, Small Methods, 2023, 7, 2201392.
213. Z. Jing, J. Zhang, J. Wang, M. Zhu, X. Wang, T. Cheng, J. Zhu and Z. L. Wang, Nano Res., 2022, 1–7.
214. L. Liu, X. Yang, L. Zhao, H. Hong, H. Cui, J. Duan, Q. Yang and Q. Tang, ACS Nano, 2021, 15, 9412–9421.
215. M. Xu, T. Zhao, C. Wang, S. L. Zhang, Z. Li, X. Pan and Z. L. Wang, ACS Nano, 2019, 13, 1932–1939.
216. H. Wang, C. Q. Zhu, W. C. Wang, R. J. Xu, P. F. Chen, T. L. Du, T. X. Xue, Z. Y. Wang and M. Y. Xu, Nanomaterials, 2022, 12, 12.
217. Z. L. Wang, Nature, 2017, 542, 159–160.
218. H. B. Lin, M. H. He, Q. S. Jing, W. F. Yang, S. T. Wang, Y. Liu, Y. L. Zhang, J. Li, N. Li, Y. W. Ma, L. H. Wang and Y. N. Xie, Nano Energy, 2019, 56, 269–276.
219. S. S. Lv, X. Zhang, T. Huang, H. Yu and M. F. Zhu, Nano Energy, 2021, 89, 9.
220. C. Y. Ye, D. Liu, P. F. Chen, L. N. Y. Cao, X. J. Li, T. Jiang and Z. L. Wang, Adv. Mater., 2023, 35, 2209713.
221. W. Yuan, B. F. Zhang, C. G. Zhang, O. Yang, Y. B. Liu, L. X. He, L. L. Zhou, Z. H. Zhao, J. Wang and Z. L. Wang, One Earth, 2022, 5, 1055–1063.
222. S. L. Zhang, M. Y. Xu, C. L. Zhang, Y. C. Wang, H. Y. Zou, X. He, Z. J. Wang and Z. L. Wang, Nano Energy, 2018, 48, 421–429.
223. Z. W. Ren, X. Liang, D. Liu, X. J. Li, J. F. Ping, Z. M. Wang and Z. L. Wang, Adv. Energy Mater., 2021, 11, 10.
224. A. Ahmed, Z. Saadatnia, I. Hassan, Y. L. Zi, Y. Xi, X. He, J. Zu and Z. L. Wang, Adv. Energy Mater., 2017, 7, 10.
225. W. Liu, L. Xu, T. Bu, H. Yang, G. Liu, W. Li, Y. Pang, C. Hu, C. Zhang and T. Cheng, Nano Energy, 2019, 58, 499–507.
226. X. Liang, T. Jiang, G. Liu, T. Xiao, L. Xu, W. Li, F. Xi, C. Zhang and Z. L. Wang, Adv. Funct. Mater., 2019, 29, 1807241.
227. N. Wang, J. D. Zou, Y. X. Yang, X. Y. Li, Y. L. Guo, C. Jiang, X. T. Jia and X. Cao, Nano Energy, 2019, 55, 541–547.
228. H. Y. J. Jung, H. Ouro-Koura, A. Salalila, M. Salalila and Z. D. Deng, Nano Energy, 2022, 99, 8.
229. J. You, J. J. Shao, Y. H. He, F. F. Yun, K. W. See, Z. L. Wang and X. L. Wang, ACS Nano, 2021, 15, 8706–8714.
230. B. Im, S. K. Lee, G. Kang, J. Moon, D. Byun and D. H. Cho, Nano Energy, 2022, 95, 9.
231. X. Wei, Z. Zhao, C. Zhang, W. Yuan, Z. Wu, J. Wang and Z. L. Wang, ACS Nano, 2021, 15, 13200–13208.
232. X. Li, J. Tao, X. Wang, J. Zhu, C. Pan and Z. L. Wang, Adv. Energy Mater., 2018, 8, 1800705.
233. G. F. He, Y. J. Luo, Y. Zhai, Y. Wu, J. You, R. Lu, S. K. Zeng and Z. L. Wang, Nano Energy, 2021, 87, 9.
234. X. Yang, L. Xu, P. Lin, W. Zhong, Y. Bai, J. Luo, J. Chen and Z. L. Wang, Nano Energy, 2019, 60, 404–412.
235. Y. Bai, L. Xu, C. He, L. P. Zhu, X. D. Yang, T. Jiang, J. H. Nie, W. Zhong and Z. L. Wang, Nano Energy, 2019, 66, 8.
236. G. L. Liu, L. F. Xiao, C. Y. Chen, W. L. Liu, X. J. Pu, Z. Y. Wu, C. G. Hu and Z. L. Wang, Nano Energy, 2020, 75, 8.
237. W. Liu, L. Xu, G. Liu, H. Yang, T. Bu, X. Fu, S. Xu, C. Fang and C. Zhang, iScience, 2020, 23, 101848.
238. F. Xing, Y. Jie, X. Cao, T. Li and N. Wang, Nano Energy, 2017, 42, 138–142.
239. C. Rodrigues, A. Gomes, A. Ghosh, A. Pereira and J. Ventura, Nano Energy, 2019, 62, 660–666.
240. Y. Yuan, H. L. Zhang, J. Wang, Y. H. Xie, S. A. Khan, L. Jin, Z. C. Yan, L. Huang, T. S. Pan, W. Q. Yang and Y. Lin, Mater. Res. Express, 2018, 5, 9.
241. J. Yoo, D. Yoo, S. Lee, J. Y. Sim, W. Hwang, D. Choi and D. S. Kim, Nano Energy, 2019, 56, 851–858.
242. K. W. Han, J. N. Kim, A. Rajabi-Abhari, V. T. Bui, J. S. Kim, D. Choi and I. K. Oh, Adv. Energy Mater., 2021, 11, 9.
243. W. Kim, D. Bhatia, S. Jeong and D. Choi, Nano Energy, 2019, 56, 307–321.
244. N. N. Zhai, Z. Wen, X. P. Chen, A. M. Wei, M. Sha, J. J. Fu, Y. N. Liu, J. Zhong and X. H. Sun, Adv. Energy Mater., 2020, 10, 9.
245. S. Choi, S. Cho, Y. Yun, S. Jang, J. H. Choi, Y. Ra, M. La, S. J. Park and D. Choi, Adv. Mater. Technol., 2020, 5, 8.
246. I. W. Tcho, S. B. Jeon, S. J. Park, W. G. Kim, I. K. Jin, J. K. Han, D. Kim and Y. K. Choi, Nano Energy, 2018, 50, 489–496.
247. Y. Yun, S. Jang, S. Cho, S. H. Lee, H. J. Hwang and D. Choi, Nano Energy, 2021, 80, 105525.
248. M. F. Yin, X. H. Lu, G. D. Qiao, Y. H. Xu, Y. Q. Wang, T. H. Cheng and Z. L. Wang, Adv. Energy Mater., 2020, 10, 8.
249. W. X. Yang, Y. Q. Wang, Y. K. Li, J. L. Wang, T. H. Cheng and Z. L. Wang, Nano Energy, 2019, 66, 8.
250. X. H. Lu, Y. H. Xu, G. D. Qiao, Q. Gao, X. S. Zhang, T. H. Cheng and Z. L. Wang, Nano Energy, 2020, 72, 7.
251. W. X. Yang, Q. Gao, X. Xia, X. S. Zhang, X. H. Lu, S. T. Yang, T. H. Cheng and Z. L. Wang, Extreme Mech. Lett., 2020, 37, 7.
252. Y. F. Yang, X. Yu, L. X. Meng, X. Li, Y. H. Xu, T. H. Cheng, S. M. Liu and Z. L. Wang, Extreme Mech. Lett., 2021, 46, 8.
253. S. K. Fu, H. Y. Wu, C. C. Shan, K. X. Li, W. C. He, Q. Y. Li, X. K. Yu, S. H. Du, G. Li and C. G. Hu, Nano Energy, 2023, 116, 9.
254. B. B. Zhang, Y. H. Gu, H. Y. Guo, Y. Y. Gao, Y. L. Zou, R. Zhang, F. Chun, Z. M. Lin, T. Yang, W. M. Peng, Y. Fang, X. E. Yu and W. Q. Yang, Nano Energy, 2022, 104, 8.
255. A. Ali, V. Baheti and J. Militky, Mater. Chem. Phys., 2019, 231, 33–40.
256. J. W. He, X. H. Guo, C. F. Pan, G. Cheng, M. L. Zheng, Y. L. Zi, H. Z. Cui and X. Y. Li, Nanotechnology, 2023, 34, 12.
257. J. J. Han, Y. W. Feng, P. F. Chen, X. Liang, H. Pang, T. Jiang and Z. L. Wang, Adv. Funct. Mater., 2022, 32, 8.
258. P. F. Chen, J. An, S. Shu, R. W. Cheng, J. H. Nie, T. Jiang and Z. L. Wang, Adv. Energy Mater., 2021, 11, 10.
259. F. S. Qin Zhang, C. Cao, Y. Gong, B. Wang, Z. Li, Y. Peng, Q. Zhang, C. Hu and H. Guo, Adv. Funct. Mater., 2023, 2301655.
260. L. B. Huang, G. X. Bai, M. C. Wong, Z. B. Yang, W. Xu and J. H. Hao, Adv. Mater., 2016, 28, 2744–2751.
261. S. Lu, W. Q. Lei, Q. Wang, W. Q. Liu, K. C. Li, P. F. Yuan and H. Yu, Nano Energy, 2022, 103, 10.
262. X. S. He, H. Y. Zhang, J. J. Jiang and X. Liu, Nanotechnology, 2023, 34, 15.
263. W. Xu, J. Y. Yang, S. S. Liu, Y. Meng, D. Feng, L. J. Jia, S. D. Liu, B. L. Wang and X. H. Li, Nano Energy, 2021, 86, 7.
264. C. D. Le, T. H. Nguyen, D. L. Vu, C. P. Vo and K. K. Ahn, Mater. Today Sustainability, 2022, 19, 8.
265. S. M. Liu, X. Li, Y. Q. Wang, Y. F. Yang, L. X. Meng, T. H. Cheng and Z. L. Wang, Nano Energy, 2021, 83, 8.
266. Y. J. Luo, P. F. Chen, L. N. Y. Cao, Z. J. Xu, Y. Wu, G. F. He, T. Jiang and Z. L. Wang, Adv. Funct. Mater., 2022, 32, 10.
267. Y. L. Zi, J. Wang, S. H. Wang, S. M. Li, Z. Wen, H. Y. Guo and Z. L. Wang, Nat. Commun., 2016, 7, 8.
268. H. Wu, S. Wang, Z. K. Wang and Y. L. Zi, Nat. Commun., 2021, 12, 8.
269. G. Cheng, Z. H. Lin, L. Lin, Z. L. Du and Z. L. Wang, ACS Nano, 2013, 7, 7383–7391.
270. C. D. Le, C. P. Vo, T. H. Nguyen, D. L. Vu and K. K. Ahn, Nano Energy, 2021, 80, 7.
271. J. Chung, S. H. Chung, Z. H. Lin, Y. Jin, J. Hong and S. Lee, Nano Energy, 2021, 88, 8.
272. G. Q. Xu, J. J. Fu, C. Y. Li, C. H. Li, H. Y. Wang and Y. L. Zi, ACS Appl. Mater. Interfaces, 2022, 4(39), 44398–44404.
273. Z. Wang, Q. Tang, C. C. Shan, Y. Du, W. C. He, S. K. Fu, G. Li, A. P. Liu, W. L. Liu and C. G. Hu, Energy Environ. Sci., 2021, 14, 6627–6637.
274. J. J. Shao, W. Tang, T. Jiang, X. Y. Chen, L. Xu, B. D. Chen, T. Zhou, C. R. Deng and Z. L. Wang, Nanoscale, 2017, 9, 9668–9675.
275. A. Y. Li, Y. L. Zi, H. Y. Guo, Z. L. Wang and F. M. Fernandez, Nat. Nanotechnol., 2017, 12, 481–487.
276. S. S. He, Z. X. Zhang, C. Zhai, X. B. Xue, Z. Y. Yang, Y. Q. Zheng, D. F. Cui and C. Y. Xue, Energy Technol., 2020, 8, 10.
277. Z. Wang, W. L. Liu, W. C. He, H. Y. Guo, L. Long, Y. Xi, X. Wang, A. P. Liu and C. G. Hu, Joule, 2021, 5, 441–455.
278. G. Cheng, H. W. Zheng, F. Yang, L. Zhao, M. L. Zheng, J. J. Yang, H. F. Qin, Z. L. Dua and Z. L. Wang, Nano Energy, 2018, 44, 208–216.
279. Y. X. Lin, J. H. Nie, Y. Bai, S. Y. Li, L. Xu, F. Wang, Y. F. Ding, J. W. Tian, Y. F. Li, X. Y. Chen and H. L. Shen, Nano Energy, 2020, 73, 9.
280. J. Chen, H. Y. Guo, Z. Y. Wu, G. Q. Xu, Y. L. Zi, C. G. Hu and Z. L. Wang, Nano Energy, 2019, 64, 9.
281. A. S. Algamili, M. H. M. Khir, J. O. Dennis, A. Y. Ahmed, S. S. Alabsi, S. S. B. Hashwan and M. M. Junaid, Nanoscale Res. Lett., 2021, 16, 21.
282. X. Z. Wang, B. Yang, J. Q. Liu, Y. B. Zhu, C. S. Yang and Q. He, Sci. Rep., 2016, 6, 10.
283. Z. H. Wang, J. Q. Zhao, L. K. Gong, Z. Wang, Y. H. Qin, Z. Z. Wang, B. B. Fan, J. H. Zeng, J. Cao and C. Zhang, ACS Appl. Mater. Interfaces, 2023, 15, 23328–23336.
284. T. Y. Y. He, H. Wang, J. H. Wang, X. Tian, F. Wen, Q. F. Shi, J. S. Ho and C. K. Lee, Adv. Sci., 2019, 6, 15.
285. M. Mousavi, M. Alzgool, B. Davaji and S. Towfighian, Mech. Syst. Signal Process., 2023, 187, 14.
286. Y. Lin, Y. C. Qi, J. Q. Wang, G. X. Liu, Z. Z. Wang, J. Q. Zhao, Y. Lv, Z. Zhang, N. Tian, M. B. Wang, Y. F. Chen and C. Zhang, Sensors, 2022, 22, 11.
287. S. Potu, M. Navaneeth, R. K. Rajaboina, B. Gollapelli, J. Vallamkondu, S. Mishra, H. Divi, A. Babu, K. U. Kumar and P. Kodali, ACS Appl. Energy Mater., 2022, 5, 13702–13713.
288. X. X. Wu, J. J. Zhang, C. H. Lee and M. F. Lin, Nanoscale, 2023, 15, 3823–3831.
289. S. M. Niu, X. F. Wang, F. Yi, Y. S. Zhou and Z. L. Wang, Nat. Commun., 2015, 6, 8.
290. W. Harmon, D. Bamgboje, H. Y. Guo, T. S. Hu and Z. L. Wang, Nano Energy, 2020, 71, 8.
291. H. J. Fang, Q. Li, W. H. He, J. Li, Q. T. Xue, C. Xu, L. J. Zhang, T. L. Ren, G. F. Dong, H. L. W. Chan, J. Y. Dai and Q. F. Yan, Nanoscale, 2015, 7, 17306–17311.
292. F. B. Xi, Y. K. Pang, W. J. Li, T. Z. Bu, J. Q. Zhao, G. X. Liu, T. Guo, W. B. Liu and C. Zhang, Microsyst. Nanoeng., 2018, 4, 8.
293. W. B. Peng, R. M. Yu, Y. N. He and Z. L. Wang, ACS Nano, 2016, 10, 4395–4402.
294. Y. Song, H. B. Wang, X. L. Cheng, G. K. Li, X. X. Chen, H. T. Chen, L. M. Miao, X. S. Zhang and H. X. Zhang, Nano Energy, 2019, 55, 29–36.
295. W. G. Kim, J. K. Han, I. W. Tcho, J. Y. Park, J. M. Yu and Y. K. Choi, Nano Energy, 2020, 76, 8.
296. X. Liang, T. Jiang, Y. W. Feng, P. J. Lu, J. An and Z. L. Wang, Adv. Energy Mater., 2020, 10, 8.
297. Q. Zhang, T. Jin, J. G. Cai, L. Xu, T. Y. Y. He, T. H. Wang, Y. Z. Tian, L. Li, Y. Peng and C. K. Lee, Adv. Sci., 2022, 9, 13.
298. L. Liu, X. G. Guo and C. Lee, Nano Energy, 2021, 88, 36.
299. J. Qian, D. S. Kim and D. W. Lee, Nano Energy, 2018, 49, 126–136.
300. H. Askari, Z. Saadatnia, A. Khajepour, M. B. Khamesee and J. A. Zu, Adv. Eng. Mater., 2017, 19, 9.
301. Z. R. Liu, J. H. Nie, B. Miao, J. D. Li, Y. B. Cui, S. Wang, X. D. Zhang, G. R. Zhao, Y. B. Deng, Y. H. Wu, Z. Li, L. L. Li and Z. L. Wang, Adv. Mater., 2019, 31, 8.
302. X. P. Chen, X. K. Xie, Y. N. Liu, C. Zhao, M. Wen and Z. Wen, Adv. Funct. Mater., 2020, 30, 19.
303. L. Liu, X. Guo, W. Liu and C. Lee, Nanomaterials, 2021, 11, 40.
304. Q. Zhang, L. Li, T. H. Wang, Y. C. Jiang, Y. Z. Tian, T. Jin, T. Yue and C. K. Lee, Nano Energy, 2021, 90, 11.
305. Y. Fang, T. Y. Tang, Y. F. Li, C. Hou, F. Wen, Z. Yang, T. Chen, L. N. Sun, H. C. Liu and C. K. Lee, iScience, 2021, 24, 24.
306. Q. W. Zheng, T. Y. Sun, L. Fang, L. Zheng and H. X. Li, Nano Energy, 2022, 102, 107706.

---

Footnote

† These authors contributed equally to this work.

---