A comprehensive review on the mechanism of contact electrification

Abstract

Contact electrification (CE) is one of the earliest sciences and has a wide range of applications in production sector and life. Some breakthroughs have been made in recent years regarding the mechanism of CE. In this review, we consider different contact interfaces as the entry point, answer some basic scientific questions about CE and explore the material selection, modulation methods, and influencing factors for CE at different interfaces. It is clarified that the dominant mechanism of CE is mainly electron transfer, and the electron transfer process in CE of all contact types is explained by the electron-cloud-potential-well model, which is the fundamental basis for the regulation of CE carried out at different contact interfaces. In addition, we categorize the related research results on the mechanism and modulation means of the tribovoltaic effect related to the CE of semiconductors. Finally, this review describes the opportunities and challenges that the CE field might encounter in the future, with an aim to provide research ideas for related fields.

---

Jia Tian

Jia Tian is now pursuing her PhD degree in Electronic Science and Technology at Xi'an Jiaotong University. Her research focuses on tribotronics, self-powered sensors, and flexible semiconductor devices.

Yue He

Yue He received his Master's degree in Integrated Circuit Engineering from Xi'an Jiaotong University. He is currently pursuing a PhD degree in Electronic Science and Technology. His works aim to the promote the application of contact electrification, tribovoltaic effect and triboelectric nanogenerator.

Fangpei Li

Dr Fangpei Li received her PhD degree in Electronic Science and Technology in Xi'an Jiaotong University in 2021. Then, she joined the Northwestern Polytechnical University as a Postdoc working on the growth of perovskite materials and radiation detectors. Currently, she is an Assistant Professor in the School of Microelectronics, Xi'an Jiaotong University, focusing on self-powered, flexible, and wearable sensors and integrated systems.

Wenbo Peng

Dr Wenbo Peng received his PhD degree in Electronic Science and Technology in Xi'an Jiaotong University in 2016. He is now an Associate Professor in the School of Microelectronics, Xi'an Jiaotong University. His research mainly focuses on piezoelectric semiconductor devices and physics, piezo-/pyro-/flexo-phototronics, nanogenerators, and self-powered sensing systems.

Yongning He

Dr Yongning He is now a Professor in School of Microelectronics, Xi'an Jiaotong University. Her main research directions are semiconductor sensors and integrated systems, wide bandgap semiconductor devices and their reliability issues, microwave devices and their reliability issues, cold plasma sources and their applications.

---

1. Introduction

Contact electrification (CE) (usually called triboelectrification, TE) is a ubiquitous natural phenomenon that occurs when two solids contact each other, resulting in the transfer of surface charges. Almost all types of materials can be charged by CE, including metals, semiconductors, inorganic materials, and polymers.¹ In daily life and industry, it is usually regarded as a negative effect. For example, due to static electricity, the clothes on our body easily wrinkle in the dry season, and computer screens often adsorb dust particles due to CE. In industrial production, the static electricity generated by CE can easily lead to fire, dust explosion, dielectric breakdown, etc.² However, some positive applications of the surface charge generated by CE have also been gradually developed, such as electrophotography,³ electrostatic spraying,⁴ electrostatic printing,⁵ and electrostatic separation of particles.⁶ In addition to CE of two solid surfaces, the contact separation of a solid and liquid can also cause them to develop significant contact charges, which has important production applications in the fields of interfacial electrochemistry, catalysis, etc.⁷ Therefore, it is necessary to understand the mechanism of CE, which is important for controlling the effects of CE in production and life. However, due to the lack of suitable research methods, the progress of mechanistic studies on CE was once very slow, and there was no accurate answer to even the most basic charge carrier identity. In 2012, Zhonglin Wang's team invented the triboelectric nanogenerator (TENG),⁸ which uses the principle of CE and electrostatic induction between materials to achieve efficient conversion of mechanical energy to electrical energy, with broad application prospects in the fields of energy harvesting,^9–12 sensing,^13,14 and electrocatalysis.^15,16 The invention of TENG revived the interest in the mechanism of CE. Since then, using different TENG devices, researchers have confirmed that the main carriers of CE at various contact interfaces are electrons, from both macroscopic and microscopic scales, and have established a relatively complete model to explain the mechanism of CE.¹⁷

This review provides a comprehensive overview of the CE principles at different contact interfaces. This paper is divided into the following sections: (1) mechanism and regulation of CE at solid–solid surfaces, (2) mechanism and regulation of CE at solid–liquid surfaces, (3) mechanisms of CE at other interfaces, (4) mechanism and regulation of the tribovoltaic effect and (5) conclusions and prospects. By categorizing the related research, we believe that this will be beneficial in reducing the adverse effects of CE and increasing the potential of CE in applications. In addition, we discuss the opportunities and challenges for future CE research, aiming to provide insights for the development of related fields.

2. Mechanism and regulation of CE at solid–solid surfaces

CE in solid–solid cases (S–S) was the first to be discovered and studied, and since then, there have been conflicting opinions about the mechanism of S–S CE. For a long time, researchers carried out studies on the mechanism of CE at the macroscopic scale and found that almost all materials can be charged by CE. However, due to factors such as the impossibility of precisely controlling the roughness of material surfaces at the macroscopic scale and the interference of material fragments generated when materials are in contact and rubbed together, experiments on contact electrification at the macroscopic scale are full of randomness. With the development of science and technology, the emergence of means to characterise materials at the atomic scale (e.g., X-ray photoelectron spectroscopy (XPS) and atomic force microscopy (AFM)) has led to a turnaround in the study of the CE mechanism, but there are still some controversies. The main conflicting points include the identity of the charge transfer between the S–S interfaces and the specific process of charge transfer.

2.1. The electron-cloud-potential-well model

In response to the identity of charge transfer, some researchers advocate that CE is dominated by electron transfer,¹⁸ while others argue that the transfer of ions also plays a huge role in the CE process,^1,19 and some researchers also support species transfer in materials.^20–22 For a long time, these views were not supported by sufficient direct evidence and there was no complete theoretical model to explain them all, where the main difficulty was how to conveniently characterise the transferred charges in CE. Tangible evidence of electron transfer was not given until 2018, when Cheng Xu used TENG to quantitatively observe real-time charge transfer in CE at different temperatures, as shown in Fig. 1(a).²³ They observed that the triboelectric charges on the solid surface decayed at high temperatures (Fig. 1(b)), which is consistent with the law of electron thermionic emission (Fig. 1(c)). If the identity of the dominant transferred charge in the CE process is ions or charged material fragments, then neither the positive charge nor the negative charge should show a clear trend with temperature. Therefore, this experiment strongly excludes the role of ion transfer or charged material fragments in the CE process at the metal–dielectric interface and proves that electron transfer is dominant in this process. After this, using the TENG device as an observational tool, increasing evidence for the dominance of electron transfer was successively obtained.^25,26 For example, Shiquan Lin et al. used UV light to irradiate the charged region on the surface of insulators (Fig. 1(d)) and found that the decay of triboelectric charges is strongly correlated with the wavelength and intensity of the incident light and the results are consistent with the photoelectron emission model (as shown in Fig. 1(e and f)), which provides strong evidence for the electron transfer mechanism in CE.²⁴

Fig. 1 Effects of thermal excitation and photoexcitation on contact electrification (CE): (a–c) thermal excitation experiment.²³ Reproduced from ref. 23 with permission from John Wiley and Sons, Copyright 2018. (a) Setup of the measurement platform for thermal excitation. (b) Q_SC evolution with time at high temperatures. (c) Plots of ln (J/A₀/T) against 1/T. (d–f) Photoexcitation experiment.²⁴ Reproduced from ref. 24 with permission from John Wiley and Sons, Copyright 2019. (d) Setup of the measurement platform for photoexcitation. (e) Variation in the charge density transferred on the surface of SiO₂ with time under different irradiation wavelengths. (f) Variation in transferred charge density on the surface of polyvinyl chloride (PVC) with time at different irradiation powers.

In the metal–dielectric case, the mechanism can be described by the surface state model, as shown in Fig. 2(a–c) (the situation for metal–semiconductor is similar). Specifically, the energy band diagram of the dielectric is divided into the conduction band (whose bottom is named E_{C, d}) and valence band (whose top is named E_{V, d}), and generally there are surface states due to the existence of surface defects, while there is a uniform Fermi energy level for the metal (E_{F, m}) as a whole (Fig. 2(a)). If the E_{F, m} is higher than E_{V, d}, the electrons on the surface of the metal that are below the E_{F, m} can easily be transferred to the surface of the dielectric when in contact (Fig. 2(b)). Undisturbed by external factors, these charges are expected to remain on the surface forever (Fig. 2(c)). The ability to gain/lose the electrons in the metal and dielectric or semiconductor can be regulated by the applied bias voltage, temperature difference, light, etc. As shown in Fig. 2(d–f), when a bias voltage is applied to a metal–dielectric, if a negative bias voltage is applied to the metal side, the surface-state energy level of the dielectric will become lower and it will receive more electrons from the metal side at the time of contact, and the quantity of transferred charge will increase (Fig. 2(e)) compared to the case at zero bias (Fig. 2(d)). Conversely, if a positive voltage is applied to the metal side, the surface state energy level of the dielectric will become higher (Fig. 2(f)), and it can even be higher than the Fermi energy level of the metal, thus making the metal surface negatively charged.²⁷

Fig. 2 Modified surface state model of CE in metal–dielectric case: (a–c) schematic energy band diagrams for the metal and dielectric materials (a) before contact, (b) in contact, and (c) after contact. (d–f) Schematic energy band diagrams for the metal and dielectric materials (d) in contact with no bias, and in equilibrium with (e) negative and (f) positive bias.

In the semiconductor–semiconductor case, the energy band diagram can still be used to describe its electronic structure, and thus the surface state model can still partially explain its CE mechanism,²⁸ as show in Fig. 3(a–c). Specifically, before contact, there is a difference in the work function between the p-type semiconductor and n-type semiconductor (that is, the Fermi energy level difference), as shown in Fig. 3(a); and when the two materials are gradually close to each other until they come into contact, the transfer of electrons occurs under the driving force of the Fermi energy level difference, which results in the formation of a depletion region on the contact surface of the p-type (n-type) semiconductor, as shown in Fig. 3(b); after contact, the transferred electrons captured on the semiconductor surface generate a transient current through an external circuit, while the depletion layer width decreases due to the presence of an air gap and an external circuit load until a new equilibrium is reached at the time of complete separation, as shown in Fig. 3(c). CE involving semiconductors has led to the discovery of some new phenomena since 2019, the mechanism of which is mainly explained later in the supplementary explanation.

Fig. 3 Modified surface state model of CE in semiconductor–semiconductor case and dielectric–dielectric case: (a–c) schematic energy band diagrams for semiconductor case (a) before contact, (b) in contact and (c) after contact. (d–f) Charge transfer (d) before contact, (e) in contact, and (f) after contact between two different dielectrics.

In the dielectric–dielectric case, where the surface states can be described in terms of energy band structures, CE between the two materials can still be represented as in Fig. 3(d–f),²⁸ where due to the different valence and conduction band structures of each material, the surface state of dielectric A is assumed to have a lower energy than that of dielectric B (Fig. 3(d)). Once the two materials come into contact, some of the electrons are transferred from the surface of dielectric B to the surface of dielectric A. Even if the two materials separate, these electrons are not transferred back to dielectric B, resulting in a positive electrostatic charge for B and a negative electrostatic charge for A (Fig. 3(e)). These charges are surface-state bound charges, and in general if the conductivity of the material is quite low (e.g., an insulator), then these charges are expected to remain on the surface forever without interference from external factors (Fig. 3(f)). However, surface state models are clearly inappropriate for polymers, rubbers and composites, etc., which cannot be described by an energy band structure (their electronic structure is either generally represented by molecular orbitals and chain arrangements, or simply cannot be described by an electronic structure). For this reason, it is necessary to propose a unified and universal model to describe the mechanism of the CE process.

Before looking for a model, we should firstly return to the definition of the word contact in CE. In our daily studies on the properties of TENG, we do not overemphasize the difference between CE and TE, but essentially there is a difference between the two phenomena. In this study, the general TE consists of two processes, where firstly, two friction materials need to come into contact with each other under the action of an external force, and secondly the process of relative motion starts. Neglecting the relatively complicated process of frictional motion, here we focus on the first step of contact. It is known that the surface of all materials is not absolutely smooth, and the microscopic picture between two contacting materials is necessarily rough, which is one of the reasons why external forces are usually applied in TENG to make the friction layers fit tightly together or to create surface nanostructures to increase the contact area of the surfaces as much as possible. Then the question arises, is it possible to generate charge transfer between the atoms that are not actually in contact? Shengming Li et al. used the KPFM technique to adequately answer this question based on the change in surface potential between the tip and the semiconductor at a certain distance.²⁹ As shown in Fig. 4(a–g), when the spacing between the tip and the material to be tested is less than a certain distance (usually the atomic spacing in the equilibrium state), there will be an obvious surface potential change due to electron transfer, accompanied by a slight change in the phase shift of the tip.

Fig. 4 Exploring the mechanisms of CE at the micro-scale: (a–g) experiment to explore the relationship between contact electrification and phase shift.²⁹ Reproduced from ref. 29 with permission from the American Chemical Society, Copyright 2016. (a–f) ΔV–A_sp and Δφ–A_sp curves under the conditions of A₀ = 100, 70, and 50 nm. (g) Schematic of the tip–sample interaction force of tapping scans with different scanning parameters (A₀ and A_sp). (h) Electron-cloud-potential-well model proposed for explaining CE and charge transfer and release between two materials that may not have the well specified energy band structure.²³ Reproduced from ref. 23 with permission from John Wiley and Sons, Copyright 2018.

The causes of this phenomenon can be well understood in conjunction with knowledge of interatomic interactions. From an atomic scale, CE can be viewed as a process occurring between two atoms. As shown in Fig. 4(h), before the contact in the macroscopic sense, due to the large distance between the two atoms, they are two isolated atoms, and the transfer of electrons cannot take place due to the limitation of their respective potential wells. When the two atoms are moved by an external force and the atomic distance is at the equilibrium distance, their electron clouds begin to overlap. The external force continues to act, and the atomic distance is less than the equilibrium distance, at which time the increase in repulsive forces between the atoms is accompanied by an increase in the degree of overlap of the electron cloud, leading to a change in this system from a single-potential-well state to an asymmetric double-potential-well state. At this point, the energy barrier between the two atoms decreases due to the strong overlapping state of the electron cloud, the probability of electron transfer increases,^30–32 and the CE process on the atomic scale can easily take place (the two atoms in this state can be considered to be weakly bonded to each other). When the two materials separate, it is difficult for the electrons that have been transferred to cross the potential barrier and return to their original orbits. Unless additional energy is provided to the electrons from outside (e.g., by raising the temperature), the electrons can jump out of the potential well and transfer back to their original position or escape into the air. This is the electron-cloud-potential-well model (also named Wang transition for CE).²³ The electron-cloud-potential-well model was first proposed in the metal–dielectric case, but it is applicable to almost all S–S cases even liquid–solid cases.

2.2. Material selection and surface modification for the modulation of solid–solid CE

Theoretically, CE is a phenomenon that exists between all contact interfaces, regardless of the type of material. According to the electron-cloud-potential-well model, the contact material realises the cross-interfacial jump of electrons by generating a strong electron cloud overlapping between atoms, and the size of the electron cloud is correlated with the ability to gain or lose electrons.³³ The ability of atoms to gain or lose electrons varies for each material, and this effect is inevitably reflected in the difficulty and quantity of charge transfer in the CE process. In the practical application of TENG, how to choose materials has become a critical issue.

The traditional triboelectric series is sorted by the polarity of the triboelectric charge between two materials, but given that the microscopic rough interface contact between solid materials is not close, and there is no quantitative examination of the quantity of transferred charge, it has certain limitations in the application of material selection for TENGs. To quantitatively characterise the ability of materials for CE, Zou Haiyang et al. designed a solid–liquid TENG (SL-TENG), as shown in Fig. 5(a), which excludes the disturbing factors such as external charge and environmental conditions, and obtained the triboelectric series shown in Fig. 5(b) by using the normalisation method, which provides a certain reference for the TENG material selection.³⁴ Based on this, Di Liu et al. used a contact-separation TENG (CS-TENG) as a tool (Fig. 5(c)) and investigated the effects of air pressure, temperature, and thickness of the dielectric layer on the triboelectric charge density (TECD) through a series of tests (Fig. 5(d–g)), which demonstrated that in a vacuum environment it is possible to avoid the effects of air breakdown, hot electron emission, etc. They presented a new triboelectric series (Fig. 5(h)), which can reflect the intrinsic properties of the material to some extent and help understand the maximum TECD and energy density of various materials.³⁵

Fig. 5 Measurement of the triboelectric series: (a and b) principle for measuring the triboelectric charge density by SL-TENG.³⁴ Reproduced from ref. 34 with permission from Springer Nature, Copyright 2019. (a) Simplified model of the measurement method. The tested material contacts the liquid metal of mercury, and then separates periodically. The positive electrode of the electric meter is connected to the mercury, and the negative is connected to the copper electrode. (b) Quantified triboelectric series. The error bar indicates the range within a standard deviation. (c–h) Principle for measuring the triboelectric charge density by CS-TENG.³⁵ Reproduced from ref. 34 with permission from Springer Nature, Copyright 2022. (c) Triboelectric material pairs with a metal or a dielectric as its counterpart. (d) Surface charge density of the CS-TENG at various atmosphere pressures when the thickness of the PTFE film is 200 μm, 400 μm, and 600 μm. (e) Surface charge density of the CS-TENG under vacuum and atmosphere condition when the thickness of the PTFE film is 200 μm, 400 μm, and 600 μm. (f) Surface charge density of the CS-TENG at different temperatures and pressure of 300 Pa when the thickness of the PTFE film is 200 μm, 400 μm, and 600 μm. (g) Surface charge density of the CS-TENG at various temperatures under high vacuum when the thickness of the PTFE film is 200 μm, 400 μm, and 600 μm. (h) TECD of various materials under vacuum condition.

Based on the triboelectric series, we generally tend to choose the two materials that are farther apart in the table because their ability to gain or lose electrons differs more. However, this is not absolute, and the surface state of a material can have a significant effect on its ability to gain or lose electrons. Kyung-Eun Byu et al. found that it is possible to control the polarity and quantity of triboelectric charges by controlling the surface dipole and the surface electronic state through the electron-donating and electron-absorbing functional groups. Also, they significantly changed the position of the triboelectric charge in the conventional triboelectric series by modifying the SiO₂ substrate using only a single layer of an SAM film with a thickness of less than 1 nm.³⁶ Therefore, even if the same friction pair materials are chosen, the performance of CE will vary with the surface structure of their materials. For example, for polymer elastomers, crystallisation due to strain and temperature changes leads to rearrangement of the surface functional groups, which causes changes in the surface electron cloud density, and ultimately changes the position of the material in the triboelectric series.³⁷ Therefore, modulating the surface state by different surface modifications of friction layer materials is very important to enhance the efficiency of CE. Currently, the main research on the surface engineering of friction materials is centred on two means, chemical modification and physical modification.

Chemical modification is mainly aimed at organic materials to adjust their electron gaining/losing ability by means of changing chemical functional groups,^38,39 ion injection,^40,41 elemental doping,^42,43etc.,⁴⁴ which can significantly increase the difference in the contact barrier between the dielectric layer and the electrode. For example, researchers have found that by adding electron-donating or electron-absorbing functional groups to the surface of oxide and polymer materials, it is possible to modulate the position of the material in the triboelectric series. Specifically, the addition of halide groups moves the material towards the negative side of the triboelectric series, whereas the addition of amide groups tends to move the material towards the positive side.^36,38,39 Zekun Li et al. chemically modified SiO₂ nanoparticles with different chain lengths of perfluorosilane coupling agents, which increased the transferred charge density of the composite PVDF film by 6 times (up to 166 μC m⁻²) and the power density by 39 times (up to 3.12 W m⁻²). This solved the problem that increasing the dielectric constant of a material would weaken its built-in electric field in the medium to a certain degree, thus decreasing the power output of the generator and contact separation process, where the wear of the material surface can lead to the failure of the surface modification engineering.⁴⁵ Surface modification of triboelectric materials can improve the limitation of surface charge density, but not all materials are suitable or capable of being modified with functional groups.

In contrast, physical modification is generally performed by surface patterning techniques such as soft lithography,^46–48 photolithography,^49–51 and laser patterning^52–54 to fabricate on the surface of materials a regular and uniform texture of multiple micro-/nano-patterned structures, resulting in an increase in their effective charge density by improving their contact efficiency. Varghese et al. greatly increased the output power of TENG (∼180 times) using PDMS with micropyramidal arrays. Specifically, the TENGs based on PDMS with micropyramidal arrays could produce an output voltage of 400 V, a short-circuit current of 3 mA m⁻² and a peak power density of 0.9 W m⁻².⁵⁵ Almost all materials can be physically modified in an attempt to improve their performance of CE, but the micro/nano structures on the surface are prone to wear during the contact separation process or the sliding process, resulting in a reduction of the CE effect. In addition, physical modification can also lead to changes in the surface roughness of a material, which can also have an effect on the process of charge transfer in CE (this will be described in more detail in Section 2.3).^56,57 When choosing a suitable surface engineering method, the appropriate technique should be chosen based on the physicochemical properties of the material, which requires more research to be done to explore the mechanism of improving its CE performance.

In conclusion, the performance of TENGs can be significantly enhanced by precise material selection and targeted physical and chemical modification of their surface as well as modulation of their working environment. These strategies not only increase the output power of TENGs but also prolong their lifetime, which is a crucial step to promote the commercialization and utilization of TENG technology. In the future, with the further development of materials science and surface engineering technology, it is foreseeable that TENG technology will show greater application potential in the field of energy harvesting and sensors.

2.3. Factors affecting solid–solid CE

2.3.1. Curvature effect on solid–solid CE. Using the electron-cloud-potential-well model, it is easy to accept that CE can occur between two materials with different chemical compositions. However, there have also been many studies showing that CE can also occur between materials of the same chemical composition, which seems counterintuitive. For centuries it has been assumed that the contact charges arise from spatially homogeneous material properties (along the surface of the material) and that in a given friction pair of materials, one is uniformly positively charged and the other negatively charged. However, Mario M. Apodaca et al. showed that CE can occur between two identical materials, as shown in Fig. 6(a), and the magnitude of charge Q that develops scales not with the contact area, but rather with its square root.⁵⁸ Based on these results and combined with theoretical calculations, they concluded that CE between identical materials is driven by the inherent, molecular-scale fluctuations in the surface composition or structure of the material, and this behaviour appears to be generic to non-elemental insulators, as shown in Fig. 6(b). Baytekin et al. also presented evidence of a random “mosaic” of nanoscale reverse charged regions on each surface.²² They found that these mosaics of surface charges have the same topological characteristics for different types of dielectrics and hold much more charge per unit area than previously thought. They attributed this phenomenon to the transfer of charged material fragments. In contrast, John C. Angus et al. attributed different surface roughness to the differences in transient excitation electron concentrations in localised parts of the surface.⁵⁷ In 2019, Cheng Xu et al. found that the CE between identical materials is related to the curvature effect by investigating the output characteristics of TENG of identical materials with different curvature surfaces.^59,60 They found that the convex surface or positive curvature of the material is more likely to get negative charge, while the concave surface or negative curvature is more likely to get positive charge. Additionally, in the case of two curved surfaces in contact, increasing the applied force tended to result in the reversal of the direction of charge transfer. The above-mentioned law still held true at low vacuum and high temperature. Accordingly, they proposed a charge transfer model for CE between identical materials, where the curvature effect leads to a change in the surface state of the material. Specifically, different curvatures may lead to different surface energies due to surface tension, and thus lead to a jump of electrons after the contact of the surfaces of the same chemical composition,⁶⁰ as shown in Fig. 6(c–f). This curvature effect model can not only explain the phenomenon of surface charging of materials with different curvatures in the bulk, but also explain the mosaic pattern of positive and negative charge bias distribution resulting from microscopic local curvature differences.

Fig. 6 CE of identical materials. (a and b) Typical raw data of the charge Q developed on two contacting PDMS pieces and the possible scenarios of contact electrification of identical materials.⁵⁸ Reproduced from ref. 58 with permission from John Wiley and Sons, Copyright 2010. (c–f) Mechanism of CE between identical materials.⁵⁹ Reproduced from ref. 59 with permission from the American Chemical Society, Copyright 2019.

2.3.2. Effects of environmental conditions on solid–solid CE. According to the research in recent years, the effect of temperature on CE is very obvious.^{23,25,26,61,62} During the CE process, electrons tend to transfer from the high-temperature side to the low-temperature side,²⁶ and when the temperature of the friction pair is higher, the charge decay becomes faster, and the CE phenomenon is less obvious,²³ and the law is consistent with the electron thermionic emission model, which reveals the temperature dependence of CE. Taking the metal–dielectric case as an example, the effect of electron thermionic emission on CE can be explained by the energy band diagram, as shown in Fig. 7(a), assuming that the Fermi energy level (E_f) of the metal is higher than that of the highest occupied surface state energy level (E₀) of the dielectric. When the temperature of the metal increases, the probability of electrons in the metal running to higher energy levels increases and tends to leave the surface of the metal more, and thus when the metal is in contact with a dielectric, it is more likely to lose electrons, allowing the dielectric to gain a negative charge. On the contrary, when the temperature of the metal decreases, the probability of the electrons in the metal running to the high energy level decreases, and the probability of losing electrons to become positively charged when the metal is in contact with the dielectric decreases. Also, when the temperature of the metal decreases to a certain degree, polarity reversal of the CE may occur. In the case where the Fermi energy level (E_f) of the metal is lower than the highest occupied surface state energy level (E₀) of the dielectric, as shown in Fig. 7(b), the metal tends to gain electrons to be negatively charged when the temperature difference is 0; however, as the temperature of the metal increases to a certain extent, the electrons in the metal are more inclined to be transferred to the dielectric under the action of temperature, and thus the dielectric gains negative charge and the polarity reversal of friction charge transfer occurs.

Fig. 7 Band-structure model of the temperature-difference-induced charge transfer when (a) E_f > E₀ and (b) E_f < E₀.²⁶ Reproduced from ref. 26 with permission from John Wiley and Sons, Copyright 2019.

There are fewer studies on the effect of photon excitation on CE, but there is clear evidence that photoexcitation also affects the CE process.²⁴ By varying the intensity and wavelength of UV light, the decay of triboelectric charge induced by the photoelectron excitation occurs on the dielectric surface in the absence of a significant temperature increase, as shown in Fig. 1(g–i). In addition, Linlin Sun et al. found that 200 nm UV light attenuates the electron transfer of the CE process on the polymer surface, and O₃ can further enhance the inhibitory effect of UV light on CE. Subsequently, they proposed a photoelectron excitation model to explain this phenomenon, where UV light can make polymers positively charged; simultaneously, the highest occupied energy level on the polymer surface can be shifted to a higher energy level under an O₃ atmosphere. Therefore, under UV irradiation, electrons are more likely to be excited to the outside of the polymer,⁶³ as shown in Fig. 8. In addition, environmental conditions such as humidity,^64,65 interfacial pH,⁶⁶ atmosphere,^63,67,68 vacuum,^35,69,70 electric field,^65,71 and magnetic field⁷² also have an effect on the e charge transfer process of CE. An increase in humidity leads to an increase, and then a decrease in the transferred charge in CE, which Kun Wang et al. explained as a change in the surface state due to the adsorbed water layer at the interface.⁶⁴ Yaoyao Liu et al. performed acid–base treatment on PDMS films as a way to investigate the effect of interfacial pH on CE between PDMS and SiO₂ (acid solution results in the stronger hydrolysis of PDMS and promotes PDMS to generate more free radicals, which enhances its ability to gain electrons).⁶⁶ Shiquan Lin et al. investigated the effects of O₂, N₂, and Ar on the CE between metals and dielectrics and found that the adsorption of O₂ on the surface of the material makes the dielectrics more likely to be negatively charged in the CE, and an explanation was given in terms of surface state theory.⁶⁷ The vacuum degree can affect the CE by changing the threshold of the breakdown voltage at the interface,⁶⁹ and the maximum energy density of dielectric materials under vacuum is significantly higher than the measured value limited by air breakdown.³⁵ There are fewer studies on the effects of electric field and magnetic field on the mechanism of CE, and the current study shows that the direction and quantity of charge transfer between two materials can be purposefully controlled by an applied electric field,⁷¹ while studies on the magnetic field in relation to CE have been focused on the motion of ferromagnetic triboelectric materials for the mode regulation.⁷²

Fig. 8 Effect of ultraviolet light and ozone on electrification performance of polymers: (a) CS-TENG for the measurement under UV. (b) Amount of transferred electron (Q) between polymers and Cu under different UV light intensities. (c) Q between polymers and Cu under UV light and O₃. (d and e) Amount of photoelectron excitation on the polymer surface under UV light. (f) Amount of photoelectron excitation on polymer surface under different incident light intensities. (g) Amount of photoelectron excitation under UV light and O₃ atmosphere. (h) Photoelectron excitation model involving photo-excitation and the change of highest occupied energy levels of electrons.⁶³ Reproduced from ref. 63 with permission from Elsevier, Copyright 2021.

2.4. Simulation on CE

In addition to theoretical calculations and experimental characterisation, simulation is also an important tool commonly used to study the mechanism of CE, which can save research time, effort and cost, and corroborate with theoretical and experimental results. However, the progress in the application of simulation in CE research is relatively slow. In the study of CE by simulation, multiphysics modelling and finite element method (FEM) simulation are the most commonly used methods. Between them, FEM is easier to discretize the partial differential equations, more suitable for dealing with the complex region, and its calculation accuracy is more reliable, but it cannot directly simulate the phenomenon of charge transfer, while it can only simulate the working process and output characteristics of TENGs. However, by combining the coupled multiphysics field analysis, the operating state of TENG can be fully reflected.⁷³ Peljo et al. used FEM to simulate the CE of a bimetallic system by solving the Laplace equation, ∇²ϕ = 0, for the electrostatic potential in a two-dimensional axisymmetric vacuum.⁷⁴Fig. 9 shows the simulated potential distribution and charge distribution of Janus Ag–Au nanoparticles with a radius of 1.2 nm, where the distribution of positive and negative charges depends on the ratio of Ag and Au. Seongsu Kim et al. also observed the potential distribution of graphene-based friction electric nanogenerators (GTNGs) using the FEM-based COMSOL Multiphysics field software.⁷⁵ However, most of the current CE-related research using FEM has focused on the simulation of electrostatic induction with the aim of optimising the overall TENG output,^76,77 and therefore the contribution to the CE mechanism is far less than what it can be.

Fig. 9 FEM simulation of CE at a bimetallic contact system: (a–e) 2D FEM simulation of potential distributions of spherical Ag–Au Janus particles with different Ag compositions. (f) Surface charge density vs. arc length as shown in (c).⁷⁴ Reproduced from ref. 74 with permission from the American Chemical Society, Copyright 2016.

Of course, besides FEM, there are other simulation tools that can be used in CE-related studies. For example, the discrete element method (DEM), which considers the interaction force between particles, can simulate the trajectory of the particles, and is also applicable to particles of different shapes and sizes. Therefore, DEM can be used for the CE process of particles under different conditions, such as the effect of gas velocity on the CE of particles,^78,79 the relationship between the CE properties of particles and the surface energy or other properties,⁸⁰ and the process of triboelectric charge transfer and distribution during the collision between particles and walls.^81,82

3. Mechanism and regulation of CE at solid–liquid surfaces

The interface between a liquid and solid (L–S) is the most important surface science in electrochemistry,^83–85 catalysis,^86–88 energy^89,90 and even biology.^91,92 In the history of CE research, most studies focused on the S–S case, while targeting CE between L–S can be traced as far back as the work of EI-Kazzaz and Rose-Innes et al. in the 1980s.⁹³ Their work was carried out mainly on the CE between liquid metals and solid dielectrics, which can form better contact between interfaces due to their better mobility compared to solid metals, but this work was essentially an exploration of the mechanisms of S–S CE. In most cases, the liquids of interest that will come into contact with solids are generally referred to as solutions.

Truly, the first work on the mechanism of CE between solid–liquid solutions was the study of CE between a water droplet and a dielectric surface by Matsui and Yatsuzuka et al.^94–96 Their experimental results showed that when a droplet falls on PTFE and slides on its surface, it always becomes positively charged, and a simple explanation of this principle was presented using the electric double layer (EDL) model. However, regarding the mechanism of L–S CE, Touchard et al. suggested that physicochemical reactions are the physical basis of flow electrification,⁹⁷ whereas Lewis used a molecular approach to describe the charge transfer mechanism in water/polymers.⁹⁸ Furthermore, Zimmer-mann et al. suggested that the CE between a simple electrolyte solution and a polymer CE between simple electrolyte solutions and polymers is caused by ion adsorption and transfer.⁹⁹ Some studies have shown that a water film adsorbed on the surface of a solid is the basic cause of L–S charge transfer,¹ whereas others hold a different view, suggesting that water is not required in charge transfer.¹⁰⁰ In addition to considering ionic transfer, some studies have shown that the charge transfer between a liquid and dielectric is dominated by electron transfer, such as the electrification between oils and polymers.¹⁰¹ However, these works were focused on the study of the distribution of ions and the structure of water molecules on the liquid side, and very few studies have been done on the carrier properties of the surface on the solid side.

3.1. A hybrid EDL model and the “two-step” formation

At the L–S interface, the carrier identity is usually assumed to be ionic in L–S CE due to the involvement of the solution.^94–96,102 In recent years, with the emergence of L–S TENG,^103–105 there has been renewed interest in the mechanistic study of L–S CE. Some researchers used KPFM to observe the effect of temperature on the attenuation of the solid surface charge after L–S CE and the effect of solute, pH of the aqueous solution and hydrophilicity of the solid surface on the liquid–solid-CE at the microscopic scale.¹⁰⁶ The results showed that electron transfer and ion transfer occur in L–S CE because the surface charge decayed at high temperature, but there was always a residual surface charge, as shown in Fig. 10(a–c). The solutes in aqueous solution, such as Na⁺ and Cl⁻, reduced the electron transfer between the aqueous solution and solid, while the ion transfer was significantly affected by the pH of the solution. In addition, the CE between hydrophilic surfaces and aqueous solutions may be dominated by ion transfer, and the CE between hydrophobic surfaces and aqueous solutions is more likely to be dominated by electron transfer. Other related studies have also confirmed the dominance of electron transfer in L–S CE from different perspectives. The transferred charge of CE of deionised water droplets with PTFE exceeds the theoretical transferred charge in the presence of ionic transfer only, and there is also charge transfer in the CE of oil with PTFE,¹⁰⁷ as shown in Fig. 10(d–f). Also, there is an accumulated charge in CE of water droplets with PTFE by a asymptotic saturation process,¹⁰⁸etc., as shown in Fig. 10(g–i).

Fig. 10 Evidence for the existence of electron transfer in liquid–solid CE: (a–c) temperature effect on the CE between the DI water and the SiO₂.¹⁰⁶ Reproduced from ref. 106 with permission from Springer Nature, Copyright 2020. (a) Setup of the contact charge experiment. (b) Decay of the CE charge on the SiO₂ surface at different substrate temperatures. (c) Charge density on the SiO₂ surface in the charging and heating cycle tests. (d–f) Squeezing system for studying the liquid–solid contact electrification.¹⁰⁷ Reproduced from ref. 107 with permission from John Wiley and Sons, Copyright 2019. (d) Schematic of experimental method. (e and f) Amount of charges on deionized water (50 μL). (e) After contacting with PTFE membrane and (f) theoretical calculation based on ion transfer model. (g–i) Sliding-type TENG for studying the continuous electrification process between liquid and solid surfaces.¹⁰⁸ Reproduced from ref. 108 with permission from the American Chemical Society, Copyright 2020. (g) Sliding motion of the droplet on the PTFE surface. (h) Charge saturation process on the PTFE surface and the insert is the detailed charge transfer process at the beginning state and saturated state. (i) Soaking experiments, where the PTFE surface is previously immersed in different liquids before the electrification test.

The conventional EDL model does not consider the electron transfer at the interface, only that the ionisation reaction at the interface between the electrode and the electrolyte induces an increase in the surface charge. This leads to the appearance of a Stern layer and a diffusion layer of compensating charge at the L–S interface, both of which are composed of ions. The Stern layer is formed by ions strongly adsorbed on the charged electrode (usually hydrated ions), whereas the diffusion layer is composed of ions in the electrolyte of opposite polarity to the electrode, and the ion concentration decreases with an increase in distance from the electrode surface (Fig. 11(a and b)). Considering the presence of electron transfer at the L–S interface, Wang et al. modified the EDL model by combining the electron-cloud-potential-well model and illustrated the “two-step” process of hybrid EDL model formation by considering both electron transfer and ion adsorption.¹⁰⁹

Fig. 11 EDL model for liquid–solid CE.¹⁰⁹ Reproduced from ref. 109 with permission from the American Chemical Society, Copyright 2022: (a and b) traditional model for EDL. (a) EDL on the electrode surface and (b) EDL on the ionized solid surface. The ion distribution near the surface can be illustrated as the Stern layer and diffuse layer. (c and d) Wang's hybrid EDL model and the “two-step” processes about its formation. (c) In the first step, the molecules and ions in the liquid impact the solid surface due to the thermal motion and the pressure from the liquid leads to electron transfer between them; meanwhile, the ions may also attach to the solid surface. (d) In the second step, the free ions in the liquid would be attracted to the electrified surface due to electrostatic interactions, forming an EDL.

In Section 2.1, we discussed the electron-cloud-potential-well model. This model shows the charge transfer mechanism of CE at the atomic scale, which can explain CE in any case, including of course the L–S case. At the L–S interface, liquid molecules will undergo collisions with solid atoms at the L–S interface due to the presence of liquid pressure, leading to electron transfer due to overlapping electron clouds. Liquid molecules/atoms that have gained or lost electrons from the solid surface are pushed away from the L–S interface due to the flow of the liquid, while the transferred charge remains on the solid surface. This is the first step of the hybrid EDL model, as shown in Fig. 11(c). In the second step, the free counterions in the liquid are attracted to the charged surface due to electrostatic interactions, forming an EDL, as shown in Fig. 11(d). In the second step, the ionisation reaction on the solid surface occurs in parallel, generating additional surface charge and free-migrating ions, which are also involved in the formation of the EDL. This is the hybrid EDL model (abbreviated as Wang's hybrid EDL model).

3.2. Material selection and interface regulation of liquid–solid CE

Based on the mechanism of L–S CE, it is clear that the formation of EDL is influenced by the ability of the solid material to gain or lose electrons. Therefore, similar to the S–S case, the solid material should be chosen to be far away from the solution in the triboelectric series. In 2016, Burgo et al. quantified the position of water in the triboelectric series and found that water is located between air and glass.¹¹⁰ However, nothing more has been reported about the position of water in the triboelectric series in recent years. When choosing a solid material, it can be based on the triboelectric series summarised by Di Liu et al.³⁵ In general, the ability of a solid dielectric to gain or lose electrons during CE is closely related to its own electronegativity, which depends on the electron affinity of the atoms in its chemical structure. Therefore, Shaoxin Li et al. combined the standard electrode potentials of electrically active materials in liquids and proposed the concept of unifying the work function, electronegativity of the triboelectric series, and standard electrode potentials. This can also be used as a method for screening contact materials.¹¹¹ In addition, when selecting solid triboelectric materials, it is also necessary to pay attention to the type of material. When a semiconductor or a metal forms a friction pair with an aqueous solution, the aqueous solution can be considered as a liquid semiconductor,¹¹² and thus there is a tribovoltaic effect at the liquid–semiconductor/metal interface, which has a different mechanism from the liquid–dielectric CE (the exact mechanism will be described in Section 5), but is still consistent with the hybrid EDL model and the two-step formation process.

In the case of pure polymers, chemical modification is often required to improve their triboelectric properties due to the adverse effects of high crystallinity and low dielectric constant. The modification methods and principles are similar to that mentioned in Section 2.2, but it should be noted that when the contact interface is an L–S interface, the contact angle difference due to the solid surface morphology and the hydrophilic/hydrophobic properties due to the different surface work functions also play an important role in L–S CE because the liquid is mobile.^113–115 For example, Xiaojuan Li et al. utilised temperature-sensitive polycaprolactone (PCL) as a probe reflecting the friction and lubrication state of the L–S interface. As shown in Fig. 12, they constructed a PCL-based L–S TENG, where the temperature controls the conformation of PCL, and thus the substrate has variable surface compositions, flexibly controlling the L–S CE. The short-circuit current and the open-circuit voltage of the PCL-based TENG were reduced by more than 40 times when the temperature was increased from 20 °C to 40 °C. At a temperature of 20 °C, the power output can be increased again to the initial level.¹¹³

Fig. 12 Reversible temperature-sensitive L–S triboelectrification with polycaprolactone material.¹¹³ Reproduced from ref. 107 with permission from John Wiley and Sons, Copyright 2019. (a) Design and fabrication of the PCL-F-3D AAO-based liquid–solid TENG. (b) Comparison of I_SC and V_OC at different temperatures. (c) Static CA and SA of water at different temperatures. (d) Electrical signal changes in situ as the temperature increases. (e–h) Mechanism of thermal response of PCL. Conformation of PCL chains at air–polymer interface and water–polymer interface at (e) 20 °C and (f) 40 °C. (g and h) PCL reorganization graphically illustrated by specific groups at air–polymer interface and water–polymer interface at (g) 20 °C and (h) 40 °C.

3.3. Effects of environmental conditions on liquid–solid CE

The hybrid EDL model assumes that electron transfer proceeds in parallel with ion transfer, and that the transferred electrons are usually trapped in the surface state, whereas the additional charge from the ionisation reaction is trapped in the atomic orbitals of the atom. Given that the electrons in the surface state are mobile and relatively more unstable, it can be assumed that the potential barrier of the surface state is lower than that of the atomic orbitals, and that the surface charge generated by electron transfer is more susceptible to excitation and escape. This is confirmed by the results of the heating treatment of the L–S contact interface, as shown in Fig. 10(a–c). Ion adsorption plays a more significant role in L–S CE than in S–S contact. Thus different ionic species, ionic concentration, and pH of the liquid can lead to the generation of different contact voltages and effective contact charges.^{106,107,116,117} In addition, controlling external stimuli has an important effect on the friction charge at the L–S interface.¹¹⁸ Similar to S–S CE, electron thermionic emission due to temperature accelerates the dissipation of charge from the solid surface,¹⁰⁶ and the temperature difference between the liquid and solid affects the CE process.¹¹³ Proton excitation is also commonly used as a control condition to improve the charge transport at the L–S interface. For example, Xinglin Tao et al. suggested that water molecules exposed to 365 nm UV light can be excited to higher energy states, promoting the solid surface in a saturated state to further capture electrons from water molecules and increase the saturation charge density of the L–S CE,¹¹⁹ as shown in Fig. 13(a–d). In addition, Mengli Zheng et al. observed the CE between a ferrimagnetic solid and different O₂-containing liquids under a magnetic field, and also investigated the contribution of dissolved O₂ molecules to L–S CE, and based on the experimental results, a spin-select electron transfer model based on the radical pair mechanism was proposed,¹²⁰ which provided a strategy to control the EDL through a magnetic field (Fig. 13(e–h)).

Fig. 13 Effects of environmental conditions on L–S CE: (a–d) effect of UV on CE at the water–PTFE interface.¹¹⁹ Reproduced from ref. 119 with permission from the American Chemical Society, Copyright 2021. (a) Schematic of the experimental setup. (b) Snapshots of a water droplet falling, contacting, and separating from PTFE. Transferred charge (c) and current (d) of water-PTFE TENG, in which only the aluminium electrode is connected to a grounded electrometer. (e–h) Spin-selected electron transfer in liquid–solid contact electrification.¹²⁰ Reproduced from ref. 120 with permission from Springer Nature, Copyright 2022. (e) Experimental setup and magnetization of the O₂ molecules and ferrimagnetic samples. (f) Effect of the O₂ concentration in the DI water on the charge transfer between the Fe₃O₄ and CoFe₂O₄ samples and the DI water. (g) Spin-selected electron transfer at the O₂-containing liquid and ferrimagnet interface without a magnetic field and with a magnetic field. (h) Magnetic field-induced T-S spin conversion of the [HO₂··e⁻] radical pair and the vector representation of the T₀-S conversion of the [HO₂··e⁻] radical pair.

4. Mechanisms of CE at other interfaces

In addition to the S–S and L–S interfaces, in recent years, researchers have also begun to pay attention to the CE at other interfaces. For example, due to the fluidity of liquid, it can be expected that the liquid–liquid (L–L) contact interface will be very tight, and it is easy to obtain a large effective contact area.¹²¹ In their study on the mechanism of L–L CE, Ruotong Zhang et al. explored the charge, pH, and ζ-potential variations under different L–L systems such as HFE–water, hexadecane (Hex)–water, and oleic acid (OA)–water and investigated their respective CE mechanisms. As shown in Fig. 14(a–c), in the Hex–water system, the predominant mechanism of L–L CE is preferential ion adsorption, while an electron transfer mechanism also plays a role. The OA–water system is dominated by functional group dissociation, while preferential ion adsorption is present. Similarly, in the HFE–water system, the simultaneous preferential adsorption of ions, in addition to the dominant electron transfer also occurs.¹²²

Fig. 14 CE principle at other interfaces. (a–c) CE at liquid–liquid interface.¹²² Reproduced from ref. 122 with permission from John Wiley and Sons, Copyright 2022. (a) Conceptual illustration of the usage of the L–L TENG system. (b) Mechanism of CE between liquid and liquid. (c) Summary of dominant mechanisms and co-existing mechanisms in HFE–water, Hex–water, and OA–water systems, respectively. (d) Gas collision model involving the initial charge on the solid surface at gas–solid interface.¹²³ Reproduced from ref. 123 with permission from MDPI, Copyright 2023. (e–g) Working principle of the GL-TENG.¹²⁴ Reproduced from ref. 124 with permission from American Association for the Advancement of Science, Copyright 2022. (e) Schematic of GL-TENG and a detailed discharge effect mechanism. Schematic of the working principle of (f) initial charge accumulation process and (g) steady operation process.

Linlin Sun et al. investigated gas–solid charge transfer using a single-electrode TENG and found that polar molecules are attracted to the initial charge on the surface of solid particles. This attraction enhances the gas–solid collision and increases the surface area of the solid, causing the travelling distance and the initial charge to increase the transferred charge in the gas–solid CE.¹²³ Accordingly, they proposed a model of gas–solid contact charging involving the initial charge of the solid and gas collisions, as shown in Fig. 14(d), which provides a mechanism different from the general understanding of classical gas–solid interfacial chemistry.

In addition, gas–liquid two-phase flow can effectively improve the problems of small L–S CE contact area and slow contact separation because of the rheological properties of the fluid.^124,125 As shown in Fig. 14(e–g), initially CE does not occur when the liquid flows through the saturated-charged PTFE, and with the generation of gas–liquid two-phase flow at the outlet, the gas–liquid two-phase flow rapidly replaces the air as the interstitial medium. Given that the breakdown voltage of the gas–liquid two-phase flow is lower than that of pure air, the accumulated charge on the surface of PTFE forms an electric field, which breaks through the gas–liquid two-phase flow and produces a huge discharge.¹²⁴

5. Tribovoltaic effect

Conventional TENGs often use metals and polymers as their power-generating layers, inducing an alternating current (AC), low energy density (nA–μA cm⁻²), and high resistivity (MΩ–GΩ). However, most semiconductor-based sensors at present require a direct current (DC) to operate normally. Thus, the output of conventional TENGs needs to be rectified to meet the current requirement of semiconductor devices. However, rectifiers usually take up too much valuable area of the device, as well as require additional energy consumption. If semiconductor materials are used as the friction layers, i.e., two semiconductors of different doping types slide over each other, a DC can be obtained without rectification. In 2019, this phenomenon was summarized by Zhong Lin Wang as a new semiconductor-based contact electrification effect, which is also known as the tribovoltaic effect.²⁸ The emergence of the tribovoltaic effect has revolutionized the field of nanogenerators and attracted a great deal of interest from researchers. In recent years, there has been a proliferation of literature addressing its basic theories and potential applications.

5.1. Mechanism of tribovoltaic effect

The tribovoltaic effect is an epoch-making discovery that has attracted interest from scholars worldwide, devoting their energy to research in this field and eager to rigorously elaborate its mechanism. Shortly after the concept of the tribovoltaic effect was introduced, Ran Xu et al. used N-type Si sliding on a P-type Si surface and found that between the two electrodes, the direct current flowed in the same direction of the built-in electric field formed in the dynamic PN junction near the contacting surfaces, which was not dependent on the direction of sliding.¹²⁶ By 2020, Shiquan Lin et al. used deionized water to slide on the surface of an Si wafer. They found that the electron–hole pairs on the Si surface were excited by the energy released at the interface during the sliding process and further separated by the built-in electric field, similar to the photovoltaic effect.²⁸ Subsequently, Mingli Zheng et al. detected the flow of DC between the two semiconductors using an N-type diamond tip sliding on the surface of P-type Si, experimentally verifying the existence of the solid–solid tribovoltaic effect for the first time.¹²⁷ Based on this, the electron transfer process during the tribovoltaic effect between the two semiconductors was first described microscopically by the energy band theory (Fig. 15). They proposed that the basic principle of the tribovoltaic effect to generate direct current was that after the contact between the two oppositely doped semiconductors, due to the difference in their work function, the electron will transfer from one semiconductor to the other at the contact interface. In this process, energy is released, exciting electron–hole pairs. Then, the pairs are separated by the built-in electric field, and thus a direct current is detected in the external circuit.

Fig. 15 Energy band diagram to illustrate the tribovoltaic effect at a PN heterojunction.¹²⁸ Reproduced from ref. 128 with permission from John Wiley and Sons, Copyright 2020: (a–c) energy band diagrams of the p-type semiconductor and n-type semiconductor at (a) separation state, (b) contact state, and (c) sliding state. (d) Movement of electrons and holes at the contact interface when the tribovoltaic effect occurs.

It is undeniable that this discovery is a breakthrough in the field of the tribovoltaic effect. Nevertheless, a more rigorous explanation is demanded regarding the mechanism of energy generation and release during the sliding process. Also, the excitation process of the electron–hole pairs still needs a more rational and rigorous explanation. Won Hyung Lee et al. used the Hall effect to quantify the changes in electron density accumulated inside semiconductors,¹²⁹ not only laying a theoretical foundation for solving the above-mentioned problems but also providing a powerful analytical method at the same time. Soon after, the concept of “bindington” was introduced by Zhong Lin Wang. When the tribovoltaic effect occurs between two materials, chemical bonding occurs at their contact interface, and this process releases a large amount of energy, resulting in the existence of a bindington (Fig. 16(a–c)). After that, the bindington excites electron–hole pairs, and the electrons and holes undergo directional separation by the built-in electric field, which produces a direct current.¹¹⁹ The section on interfacial bonding theory was confirmed by Youchao Huang et al. by using density-functional theory (DFT). The results of DFT calculations showed that during the tribovoltaic effect of liquid–solid, the chemical bond formation and energy release would occur at the contact interface (Fig. 16(d and e)), which proved the existence and validity of bindington.¹³¹

Fig. 16 Bindington theory and its verification by using density functional theory (DFT): (a and b) energy band diagrams when the deionized (DI) water (a) is separate from and (b) slides with N-type Si. (c) Bindington at the contact interface.¹³⁰ Reproduced from ref. 130 with permission from John Wiley and Sons, Copyright 2021. (d and e) Process of chemical bond formation between the surface of H₂O and IGZO under the calculation of DFT.¹³¹ Reproduced from ref. 131 with permission from John Wiley and Sons, Copyright 2022. It is clear to find (d) dangling bond at the surface of IGZO before contact with H₂O and (e) chemical bond formation at the sliding state.

However, whether the built-in electric field or the interfacial electric field dominates the direct current generation process remains controversial. The direct current output determined by the built-in electric field can be illustrated as follows. Two different materials have different work functions. After their contact, the transfer of electrons occurs, causing the surfaces of the two materials to be oppositely charged, which in turn generates a built-in electric field, leading to the directional separation of friction-generated, i.e., bindington-excited, electron–hole pairs, and the detection of a direct current in the external circuit^132–134 (Fig. 17(a–c)). However, by applying positive and negative bias voltages to the GaN/Si sliding heterojunction, Yunkang Chen et al. found that the output current was regulated by the external electric field, and the direction of the output current did not always coincide with the direction of the built-in electric field. The original theory that the built-in electric field determines the current direction can no longer be used to explain this phenomenon. They proposed that an electric field exists at the contact interface, which is in the opposite direction of the built-in electric field, i.e., the interfacial electric field, and believed that it was the interfacial electric field, not the built-in electric field that dominated the directional drift motion of the carriers¹³⁵ (Fig. 17(d)). Zhaozheng Wang's group and Zhi Zhang et al. also summarized the same point based on their experimental results^136,137 (Fig. 17(e and f)). By 2023, it has also been suggested that there is competition between the built-in field and the interfacial electric field. Which electric field dominates the direct current generation process will be influenced by the state of the contact interface and external environmental factors, and there is a competitive mechanism between the two electric fields, which cannot make sweeping generalizations¹³⁸ (Fig. 17(g)). More accurate methods and more rigorous theories are needed to explain the current generation mechanism of the tribovoltaic effect more completely.

Fig. 17 Debate on the domination of built-in electric field and interfacial electric field on the mechanism of current generation and direction when the tribovoltaic effect occurs. (a–c) Built-in electric field dominates the current direction.^132–134 Reproduced from ref. 132 with permission from Elsevier, Copyright 2022. Reproduced from ref. 133 with permission from the American Chemical Society, Copyright 2022. Reproduced from ref. 134 with permission from Elsevier, Copyright 2023. (d–f) Interfacial electric field dominates the current and direction.^135–137 Reproduced from ref. 135 with permission from the American Chemical Society, Copyright 2022. Reproduced from ref. 136 with permission from the Royal Society of Chemistry, Copyright 2022. Reproduced from ref. 137 with permission from John Wiley and Sons, Copyright 2022. (g) Built-in electric field and the interfacial electric field both control the current direction; however, there would be a competition mechanism with each other depending on the experimental factors.¹³⁸ Reproduced from ref. 138 with permission from John Wiley and Sons, Copyright 2023.

5.2. Tribovoltaic effect at different contact interfaces

5.2.1. Metal–semiconductor interface. The existence of the tribovoltaic effect at the metal–semiconductor interface was first discovered by Jun Liu et al. They detected a direct current with a maximum density of 10⁶ A m⁻² between the two materials by sliding a metal probe over a multilayer film composed of the semiconductor material MoS₂.¹³⁹ Soon after, by sliding Fe-made probes on Si wafers, Shisheng Lin et al. found a similar phenomenon¹⁴⁰ (Fig. 18(a)). In 2019, Zhi Zhang et al. used metal sliding on Si wafers with different doping types and doping concentrations and used an energy band model to summarize, for the first time in a relatively complete way, the mechanism of Si-based metal–semiconductor tribovoltaic effect.¹⁴⁴ In subsequent reports, semiconductors such as asymmetric graphene oxide (aGO),¹⁴⁵ PEDOT,^146–148 WO₃,¹⁴⁸ IGZO,^131,149 GaN,^150,151 4H–SiC,^133,152 and CsPbBr₃ (ref. 141) (Fig. 18(b)) were also used to study the tribovoltaic effect at the metal–semiconductor interface with great results.

Fig. 18 Tribovoltaic effect at different interfaces and structures. (a and b) Tribovoltaic effect at metal–semiconductor (MS) interface.^140,141 Reproduced from ref. 140 with permission from John Wiley and Sons, Copyright 2019. Reproduced from ref. 141 with permission from John Wiley and Sons, Copyright 2022. (c and d) Tribovoltaic effect at semiconductor–semiconductor (SS) interface.^126,142 Reproduced from ref. 126 with permission from Elsevier, Copyright 2019. Reproduced from ref. 142 with permission from John Wiley and Sons, Copyright 2022. (e and f) Tribovoltaic effect at liquid–semiconductor (LS) interface.^127,130 Reproduced from ref. 127 with permission from Elsevier, Copyright 2020. Reproduced from ref. 130 with permission from John Wiley and Sons, Copyright 2021. (g–i) Tribovoltaic effect at some “sandwich” structures: (g) metal–insulator–semiconductor (MIS) structure,¹⁴³ (h) semiconductor–insulator–semiconductor (SIS) structure,¹⁴³ and (i) liquid–insulator–semiconductor (LIS) structure.¹²⁹ Reproduced from ref. 143 with permission from the American Chemical Society, Copyright 2019. Reproduced from ref. 129 with permission from John Wiley and Sons, Copyright 2021.

5.2.2. Semiconductor–semiconductor interface. After the discovery of the tribovoltaic effect at the metal–semiconductor interface, Ran Xu et al. found that the tribovoltaic effect also occurred at the semiconductor–semiconductor interface. By sliding an N-type Si slider on a P-type Si wafer, a direct current was also detected at the electrode¹²⁶ (Fig. 18(c)). This report pioneered the study of the tribovoltaic effect only between semiconductors. Mingli Zheng et al. investigated the effect of normal pressure on the output current by using a diamond probe sliding on an Si surface in 2020,¹²⁸ which extended the tribovoltaic effect at the semiconductor–semiconductor interface to more semiconductors besides Si. Numerous reports concentrated on easily accessible semiconductor materials such as perovskite,¹⁵³ GaN,^135,136,154 and MXene¹⁴² (Fig. 18(d)). Meanwhile, Bi₂Te₃,¹³⁷ Cu₂O,¹⁵⁵ and WS₂ (ref. 156) were also adopted. In 2024, Morten Willatzen et al.¹⁵⁷ proposed a quantum mechanical model in which electron–hole pairs were excited when the bindington energy released by bonding was higher than the forbidden bandwidth at the interface.³² This theory gives a general expression for the rate of generation at the semiconductor–semiconductor interface and microscopically shows the process and mechanism of direct current generation in the case of the tribovoltaic effect between a P-type semiconductor and an N-type semiconductor.

5.2.3. Liquid–semiconductor interface. In addition to a solid, a liquid can also be involved in the formation of direct currents as a tribovoltaic layer. Shiquan Lin's team dropped deionized (DI) water or NaCl solution on the surface of an Si wafer, and by applying an external force to cause the Si wafer to move laterally, thus sliding against the liquid, they detected a direct current of about 100 nA in the external circuit.¹²⁷ This discovery inspired researchers to devote their attention to the tribovoltaic effect at the liquid–semiconductor interface. Subsequently, studies on the tribovoltaic effect between solutions and semiconductors gradually emerged,^{119,130,158,159} which are shown in Fig. 18(e and f). When studying the tribovoltaic effect that occurs between an aqueous solution and a semiconductor, the aqueous solution is usually treated as a liquid semiconductor material.^160–162 When an aqueous solution is in contact with a solid semiconductor material, the transfer of electrons and the establishment of a built-in electric field also occur. When the relative sliding between two materials occurs, molecules or ions in the aqueous solution will also form chemical bonds with the dangling bonds on the surface of the solid semiconductor, thus releasing energy to produce bindingtons and excite the electron–hole pairs. Electrons and holes will be separated by the built-in electric field to produce a direct current.

5.2.4. Metal–insulator–semiconductor, semiconductor–insulator-semiconductor, and liquid–insulator–semiconductor structures. Besides the above-mentioned three types of contact interfaces consisting of only two layers of materials, insulator materials are sometimes added between a metal and a semiconductor or between two semiconductors to form a “sandwich” structure, which is used to regulate the output of the tribovoltaic effect to adapt to a wider range of application scenarios. For instance, after investigating the tribovoltaic effect at the metal–semiconductor interface composed of a metal probe and MoS₂,¹³⁹ Jun Liu et al. added TiO₂ with thicknesses of 5, 10, 30, 60, 100 and 200 nm as a dielectric layer between the metal and MoS₂, constituting a metal–insulator–semiconductor three-layer structure (Fig. 18(g)). They found that the voltage and current could be artificially controlled by adjusting the thickness of the insulator, and the output open-circuit voltage and short-circuit current followed the principle of increasing, and then decreasing when the thickness of TiO₂ increased. In particular, when the thickness of TiO₂ was 60 nm, the output open-circuit voltage and short-circuit current had the maximum values of 0.7 V and 2 μA, which were significantly improved compared with the structure of only Al and MoS₂ (0.3 V and 0.6 μA), respectively. This is because the TiO₂ films obtained by electron beam evaporation usually contain a large number of defects, grain boundaries, etc., as conductive paths. A suitable film thickness allows frictionally excited carriers to conduct through these paths, resulting in the generation of current. At the same time, they also added SiO₂ between MoS₂ and Si to form a semiconductor–insulator–semiconductor structure (Fig. 18(h)). It was found that the mechanism of the tribovoltaic current, in this case, was due to the frictionally excited carriers crossing the dielectric layer to reach the electrode through tunnelling. Thus, a current was detected in the external circuit.¹⁴³ In 2021, Won Hyung Lee et al. discussed the tribovoltaic effect on the sandwich structure of NaCl solution–HfO₂–Si (Fig. 18(i)), which is a significant theoretical basis for the application of semiconductor-based liquid sensor systems.¹²⁹

5.3. Interface lubrication techniques to enhance the tribovoltaic effect performance

The prerequisite for the tribovoltaic effect to occur is the relative sliding of two materials. As the number of slides increases, wear of the material surfaces inevitably occurs. This problem is particularly acute when two solids rub against each other. These abrasions deplete the tribovoltaic layer that is supposed to generate the electric current and have a great influence on the output of the tribovoltaic effect.¹⁶³ Therefore, it is thought that the wear process of the two tribovoltaic layers can be slowed down and the output of the tribovoltaic effect can be enhanced by interfacial lubrication. Given that graphene oxide (GO) solution has good lubrication properties, and simultaneously conducts electricity very well, it can also increase surface carriers while sliding, and thus it is an ideal lubricant for sliding interfaces. In 2022, Wenyan Qiao et al. first used GO solution to lubricate the contact interface of Cu/Si and found that the tribovoltaic effect between Cu and P-type Si could still maintain a high current performance after 30 000 cycles.¹⁶⁴ Subsequently, in 2023, Di Yang et al. found that by sliding a stainless-steel ball over an Si wafer, although greater normal pressure and higher sliding frequency increased the output, they also increased the wear on the Si surface, and the output was not stable in the long run. It was discovered that the drop of polyalphaolefin SpectraSyn 4 (PAO 4) as a lubricant at the contact interface of the two materials could significantly reduce the coefficient of friction (CoF) at the interface from 0.76 to 0.16 at a pressure of 5 N and sliding frequency of 5 Hz. At the same time, the output of the device did not decrease significantly after about 600 continuous sliding cycles.¹⁶⁵ Soon after, they also discovered that an MXene solution could achieve a similar effect.¹⁶⁶ In 2024, Kong-Qiang Wei et al. investigated the influence of various liquid lubricants at the sliding interface of N-type WS₂ and P-type Si on the output of the tribovoltaic effect, such as ethanol, deionized water, and NaCl solution. They also analysed the mechanism of the influence on the output when polar liquids, non-polar liquids, and ionic solutions were used as lubricants.¹⁵⁶ In addition to liquid lubricants, solid lubricants can sometimes reduce the roughness of the surface of a material and increase its durability. Among them, diamond-like carbon (DLC) film is a solid lubricant with good wear resistance, and its CoF remains low (CoF < 0.01) under high normal force and high friction frequency for long periods of sliding.¹⁶⁷

In summary, the interface lubrication technique plays an important role in enhancing the performance of the tribovoltaic effect, i.e., the output of tribovoltaic nanogenerators (TVNGs). The operating principle of TVNGs is based on contact electrification, tribovoltaic effect and electrostatic induction, and their output mainly depends on the frictional behaviour and charge transfer efficiency between interfaces. However, conventional frictional interfaces may limit the energy conversion efficiency and reduce the device lifetime due to their high frictional resistance or surface wear. By introducing a lubricant (e.g., a liquid lubricant or a solid lubrication layer) at the interface, the friction characteristics of the contact interface can be significantly improved by reducing the surface roughness and friction resistance, thereby increasing the uniformity and effectiveness of the interfacial contact. On the one hand, lubrication can reduce the surface wear and prolong the service life of TVNGs. On the other hand, optimised lubrication conditions can enhance the charge separation and transfer process, thereby boosting the charge density and output voltage generated by TVNGs. In addition, interfacial lubrication is believed to be adapted to a variety of complex environments, enabling TVNGs to maintain a stable performance under conditions such as humidity and temperature changes. In conclusion, the interfacial lubrication technology provides an effective method to enhance the output and application reliability of TVNGs, which can effectively expand their application potential in the field of energy harvesting.

5.4. Surface modification techniques to modulate the tribovoltaic effect performance

Modifying the surface of materials by using physical or chemical methods can also modulate the performance of the tribovoltaic effect. Physical methods include adding a TiO₂ layer between two sliding friction layers to regulate the output voltage and output current,^143,168 magnetron sputtering to obtain wear-resistant material layers for increased durability,¹⁴⁸ and epitaxy on a regular Si wafer to modulate the interfacial electric field and inhibit the compounding of electron–hole pairs, causing more carriers to be available for conduction.¹⁶⁹ Even more interesting, Ruizhe Yang et al. discovered that by heating VO₂ to undergo a phase change from an insulating state to a metallic state, the output current could be increased by more than 20 times without affecting the output voltage.¹⁷⁰ Zhihao Zhao and group improved the contact efficiency through optimizing the graphic design of the electrode microstructure.¹⁷¹ Kai Xiao et al. also found an enhancement in output by varying the doping type and doping concentration of GaN.¹⁵¹ Other novel structures such as adding water, (CH₃OH)₂, C₂H₅OH, and C₆H₁₄ as a tribovoltaic layer in the middle of P-type Si and N-type Si, where the liquid has a relative sliding between both the upper and lower layers of Si, can generate bindington-excited carriers at two contact interfaces, which can lead to a larger current output.¹⁷² Chemical methods include the Cu-catalysed azide–alkyne cycloaddition (CuAAC) reaction to obtain the 1,2,3-triazole moiety on the surface of Si, which significantly reduces the work function of the silicon-carbon-bound organic monolayer, thereby maximizing the output current.¹⁷³ Alternatively, the bonding of F-containing function groups (FFGs) on fluorinated graphite paper (GP) by using CF4 plasma treatment resulted in a significant change in the work function of the FGPs, enhancing the output.¹⁵⁸ You-Sun Lee et al. used ChCl to passivate the internal defects in the CsFAMA perovskite. The results showed a significant increase in the tribovoltaic charge density, mobility, and built-in electric field, consequently improving the performance.¹³⁴

In this way, the surface modification technique is also one of the significant means to enhance the performance of TVNG. By physically modifying, chemically modifying or microstructuring the surface of the tribovoltaic material layer, its surface charge trapping ability, effective contact area and charge density can be dramatically enhanced. Specifically, surface modification can improve the frictional properties of the material by introducing functionalized molecules or specific physical material layers, for instance, to reinforce its electrophilic properties. In addition, by building microstructures, such as roughness optimisation and surface design, it is possible to increase the actual contact area, thus improving the energy conversion efficiency of TVNG. At the same time, the surface modification technique can also improve the durability and stability of materials to avoid performance degradation because of prolonged use. Taking all the above-mentioned aspects together, surface modification provides diverse solutions for achieving efficient and stable TVNGs.

The above-mentioned typical content on the techniques to enhance and modulate the output of the tribovoltaic effect is displayed in Fig. 19.

Fig. 19 Different methods to enhance the performance of tribovoltaic effect. (a–c) Using the lubrication technique at the sliding interface of two materials, such as liquid lubricant (a) PAO 4 (ref. 165) and (b) MXene solution,¹⁶⁶ and solid lubricant (c) diamond-like carbon (DLC).¹⁶⁷ Reproduced from ref. 165 with permission from John Wiley and Sons, Copyright 2022. Reproduced from ref. 166 with permission from Springer Nature, Copyright 2023. Reproduced from ref. 167 with permission from Elsevier, Copyright 2022. (d–g) Physical^171,172 and chemical^134,173 methods to enhance the output by modifying the surface of the layer of materials. Reproduced from ref. 171 with permission from the Royal Society of Chemistry, Copyright 2023. Reproduced from ref. 172 with permission from American Association for the Advancement of Science, Copyright 2021. Reproduced from ref. 134 with permission from Elsevier, Copyright 2022. Reproduced from ref. 173 with permission from Elsevier, Copyright 2023.

5.5. Factors affecting the tribovoltaic effect

Similar to the contact electrification effect, the tribovoltaic effect can be affected by the operating conditions and environmental factors, as shown in Fig. 20. The tribovoltaic effect is based on the relative sliding of two materials, and thus control of the parameters during sliding will directly affect its output. In 2021, Jia Meng et al. investigated that when an Al foil slid on a flexible substrate consisting of a mixture of PEDOT:PSS and cotton fibres, both the output voltage and output current increased as the normal pressure increased. The output current tended to increase as the sliding speed increased, but there was no significant change in the output voltage.¹⁴⁶ However, Jinchao Xia et al. found that at the metal–semiconductor interface consisting of Cu and 4H–SiC, an increase in the sliding velocity brought about a simultaneous improvement in both the open-circuit voltage and short-circuit current, but an increase in the normal pressure only led to an increase in the short-circuit current, with essentially no effect on the open-circuit voltage.¹³³ Subsequently, Andris Šutka et al. found that an increase in normal pressure would lead to a larger output current when the tribovoltaic effect occurred between W and a WO₃ film.¹⁴⁸ Ruizhe Yang et al. discovered that the increase in sliding velocity could enhance the output of the tribovoltaic effect occurring between Al and PEDOT.¹⁴⁷ Xiyan Xu et al. reported that as the contact area increased, the output of the tribovoltaic effect would increase simultaneously.¹⁶⁹ Nevertheless, if the number of sliding circles increases, the output of the tribovoltaic effect may decrease under some conditions. For instance, Shuo Deng et al. attributed the decrease to the fact that as the number of sliding times increased, water molecules may be adsorbed at the interface to form a layer of dense oxide film with an increasing thickness, which prevented the directional movement of electrons and holes controlled by the built-in electric field, leading to a decrease in the amplitude of the output current.¹³² Shisheng Lin et al. and Zhihao Zhao et al. studied the effect of surface roughness on the material surface in addition to the sliding parameter.^140,171 Meiqi Wang et al. employed a different approach to study the effect of the energy level position on the tribovoltaic effect employing the modulation technique of the doping type and doping concentration of the material, etc.^150,151

Fig. 20 Influence of operating conditions and environmental factors on tribovoltaic effect: (a) sliding speed and the normal force.¹⁴⁶ Reproduced from ref. 146 with permission from the American Chemical Society, Copyright 2021. (b) Time for the sliding process.¹³² Reproduced from ref. 132 with permission from Elsevier, Copyright 2022. (c) Doping type and concentration.¹⁵¹ Reproduced from ref. 151 with permission from AIP Publishing, Copyright 2022. (d) Temperature and humidity.¹³³ Reproduced from ref. 133 with permission from the American Chemical Society, Copyright 2022. (e and f) Coupling effect of the photovoltaic effect and the tribovoltaic effect.^141,155 Reproduced from ref. 141 with permission from John Wiley and Sons, Copyright 2022. Reproduced from ref. 155 with permission from John Wiley and Sons, Copyright 2023. (g) Coupling effect of the thermoelectric effect and the tribovoltaic effect.¹⁷⁴ Reproduced from ref. 174 with permission from Elsevier, Copyright 2021.

Environmental factors can also have an impact on the output of the tribovoltaic effect. Jun Chen's team first investigated the enhancement by ambient humidity on the tribovoltaic effect of Al and asymmetric graphene oxide (aGO).¹⁴⁵ Subsequently, Jinchao Xia et al. focused on the effect of an increase in temperature and humidity on the output properties of the tribovoltaic effect at the interface between 4H–SiC and Cu.¹³³ Similarly, a few works about the effects of temperature or humidity on the tribovoltaic effect have been reported.^152,175,176 When light is applied to the surface of a semiconductor material, the photovoltaic effect occurs. Also, the photovoltaic effect can be coupled with the tribovoltaic effect, in which light and friction simultaneously excite electron–hole pairs, jointly regulating the output performance of the device.^{141,154,155,159,177} A group found that the heat generated during friction would cause a thermoelectric effect on the surface of Si wafers, resulting in a significant increase in the number of majority carriers on the surface, which was much higher than that in the interior of Si. The concentration gradient resulted in a diffusion process of carriers from higher to lower temperatures, which in turn caused the formation of a thermoelectric current. The total current would contain both the tribovoltaic current and the thermoelectric current to co-generate outputs.¹⁷⁴

5.6. Difference and relation between the tribovoltaic effect and the contact electrification effect

At some point, the tribovoltaic effect will be brought up together with the contact electrification effect simultaneously. The two effects are both different and related. The differences can be summarized in three points as follows. Firstly, they have different operation modes. The contact electrification effect operates in the contact-separation mode, whereas the sliding mode has been used in most of the literature studying the tribovoltaic effect thus far, except for a very few reports.¹⁷⁸ Secondly, they have different internal mechanisms. The mechanism of the contact electrification effect originates from the difference in work functions or chemical potentials of the two materials,¹⁷⁹ which leads to the electron transfer at the contact interface (Fig. 4(h)). However, the mechanism of the tribovoltaic effect is that during sliding, chemical bonding occurs at the interface of the two materials in contact. A large amount of energy will be released in this process to produce “bindington”. Subsequently, the bindington excites many electron–hole pairs, which are directionally separated by a built-in and/or interfacial electric field to produce an electric current (Fig. 17). Finally, the triboelectric nanogenerator (TENG) based on the contact electrification effect usually produces an alternating current, while the tribovoltaic nanogenerator (TVNG) based on the tribovoltaic effect tends to generate a direct current. However, the contact electrification effect will also influence the tribovoltaic effect. When two materials are brought into contact, the contact electrification effect will introduce net charges at the interface to generate an electric field and establish a potential barrier at the interface, thus affecting the direction of the direct current induced by the tribovoltaic effect.¹⁸⁰

In general, the tribovoltaic effect is not exactly a sub-section of CE, but rather an extension of the research on CE on semiconductor materials. We emphasize that CE is a phenomenon that charge transfer (in most cases electron transfer) occurs at the contact interface of two materials, and thus forms oppositely charged surfaces, and relative sliding is not necessary in this process. The concept of tribovoltaic effect is that in the presence of a semiconductor material, the introduced energy from contact or sliding leads to the excitation of electron–hole pairs between the interfaces, which subsequently separate in a directional manner, driven by either the built-in electric field or interfacial electric field. Specifically, in the vast majority of cases, the CE phenomenon is generated at the contact interface of two materials, and the contact-mode tribovoltaic effect is actually a special case of CE, whereas the process of generating the sliding-mode tribovoltaic effect requires that the material surfaces first come into contact, i.e., CE is a prerequisite for sliding-mode tribovoltaic effect.

6. Conclusions and prospects

Herein, we reviewed the mechanisms of CE at solid–solid, solid–liquid, liquid–liquid, solid–gas, and liquid–gas interfaces and introduced the tribovoltaic effect, which is associated with semiconductor-based CE. The fundamental mechanism of CE occurrence is the electron transfer due to the overlapping of electron clouds at the atomic scale. Whether at solid–solid, solid–liquid, solid–gas, or liquid–gas interfaces, the charge transfer due to CE involves electron transfer, and the electron transfer dominates the CE process in the vast majority of cases, as shown in Fig. 21. Therefore, CE should be defined as a quantum-mechanical electron transfer process that occurs in any material, in any state (solid, liquid, and gas), and in any application environment, which is universal and unique in nature. Furthermore, the modulation of the CE process by the range of material selection, surface modification methods, and environmental factors was discussed in detail, and several simulation methods commonly used to study the CE process were briefly outlined. Besides, the mechanism of the tribovoltaic effect associated with semiconductor-based CE was also introduced in detail. The basic principle of the tribovoltaic effect can be expressed as follows. The energy released from the process of contact and sliding will excite electron–hole pairs. Then, they will be separated by the built-in electric field or the interfacial electric field, depending on which one is the dominant, to generate a direct current. It is also found that tribovoltaic effect can show a distinct performance in the cases of different interfaces such as metal–semiconductor, semiconductor–semiconductor, liquid–semiconductor, and “sandwich” structure of metal–insulator–semiconductor, semiconductor–insulator-semiconductor, and liquid–insulator–semiconductor. Meanwhile, we also reviewed the techniques of interface lubrication and surface modification for enhancing the output and some factors that can have a significant influence on tribovoltaic effect. Lastly, the difference and relation between tribovoltaic and CE was discussed concisely.

Fig. 21 Summary of research advances in CE mechanisms.^{23,109,122–124} Reproduced from ref. 23 with permission from John Wiley and Sons, Copyright 2018. Reproduced from ref. 109 with permission from the American Chemical Society, Copyright 2021. Reproduced from ref. 122 with permission from John Wiley and Sons, Copyright 2022. Reproduced from ref. 123 with permission from MDPI, Copyright 2023. Reproduced from ref. 124 with permission from the American Association for the Advancement of Science, Copyright 2022.

We believe that an in-depth and comprehensive review of the mechanisms, regulation, and influencing factors of CE and the tribovoltaic effect is crucial for understanding the working principles of TENGs and TVNGs. As shown in Fig. 22, by analysing these mechanisms in depth, researchers can design more efficient energy harvesting devices and self-powered sensors, which will propel the advancement of micro/nano energy, blue energy and self-powered sensing technologies, among others. In addition to the above-mentioned fields, contact-electro-catalysis, as emerging technology, can utilize CE to improve the efficiency and selectivity of chemical reactions, which is of great significance in fields such as the chemical industry and environmental treatment. Moreover, TVNGs, in addition to their roles as energy harvesters and self-powered sensors, also exhibit potential applications in the field of data storage (for instance, if magnetic semiconductor materials are used as the sliding counterparts, the triboelectric current may be influenced by the magnetization directions of the two sliding materials).

Fig. 22 Perspectives on application areas of CE, including TENG- or TVNG-based energy harvesting, self-powered sensors, contact-electro-catalysis and data storage.^{133,176,180–183} Reproduced from ref. 133 with permission from the American Chemical Society, Copyright 2022. Reproduced from ref. 176 with permission from John Wiley and Sons, Copyright 2024. Reproduced from ref. 180 with permission from Elsevier, Copyright 2023. Reproduced from ref. 181 with permission from the Royal Society of Chemistry, Copyright 2020. Reproduced from ref. 182 with permission from John Wiley and Sons, Copyright 2024. Reproduced from ref. 183 with permission from Springer Nature, Copyright 2022.

Despite the considerable progress in recent years in the mechanistic studies of CE, there are still many pressing issues that need to be addressed. Future research in the field of CE should work towards the following directions. For example, there is no clear explanation for the quantification of transfer and the tunnelling mechanism in the electron transfer process, which makes it very difficult for us to control the electron transfer process through the interface effect, and we should explore the influence of the interface effect on CE by designing a new method, which is combined with theoretical calculations and simulations to explore the nature of electron transfer in CE. The study of the mechanism of the tribovoltaic effect is still in its infancy and many of its intrinsic mechanisms are still unclear (e.g., it is still controversial whether the built-in electric field or the interfacial electric field dominates the DC current generation process), and it can be considered to combine the quantum mechanical theory and the carrier transport theory to quantitatively study the tribovoltaic effect. How can TENG-based CE reduce the influence of environmental factors and the wear of materials over a long time scale, while maintaining a stable and continuous charging and discharging state? It also necessary for researchers to continuously explore the CE process itself because it is unstable. In addition, most of the current studies on CE at solid–gas, liquid–liquid and liquid–gas interfaces are biased towards the application level, and more in-depth and detailed studies on the specific mechanisms are needed in the future to improve its application in production and life. By reviewing the previous work, we summarise the future directions of CE research, which we hope will promote the development of CE and related application fields.

Data availability

No primary research results, software or code have been included and no new data were generated or analyzed as part of this review.

Author contributions

Conceptualization and methodology: Jia Tian and Yue He; investigation: Yue He; visualization: Jia. Tian; writing – original draft: Jia Tian and Yue He; writing – review & editing: Jia Tian, Yue He, and Fangpei Li; supervision: Wenbo Peng and Yongning He. All authors discussed the results and reviewed the manuscript.

Conflicts of interest

The authors declare no conflicts of interest.

Acknowledgements

This research was supported by the National Natural Science Foundation of China (Grant No. 62174131 and 61704135), the China Postdoctoral Science Foundation (Grant No. 2018T111055 and 2017M613138) and the Postdoctoral Research Project of Shaanxi Province (Grant No. 2017BSHEDZZ30).

References

1. L. S. McCarty and G. M. Whitesides, Electrostatic charging due to separation of ions at interfaces: contact electrification of ionic electrets, Angew. Chem., Int. Ed., 2008, 47(12), 2188–2207.
2. G. Lüttgens, S. Lüttgens and W. Schubert. Static Electricity: Understanding, Controlling, Applying, Wiley-VCH, Germany, 2017.
3. D. M. Pai and B. E. Springett, Physics of electrophotography, Rev. Mod. Phys., 1993, 65(1), 163.
4. S. E. Law, Agricultural electrostatic spray application: a review of significant research and development during the 20th century, J. Electrost., 2001, 51, 25–42.
5. P. Galliker, J. Schneider and H. Eghlidi, et al., Direct printing of nanostructures by electrostatic autofocussing of ink nanodroplets, Nat. Commun., 2012, 3(1), 890.
6. Y. Higashiyama and K. Asano, Recent progress in electrostatic separation technology, Part. Sci. Technol., 1998, 16(1), 77–90.
7. E. Gileadi, Problems in interfacial electrochemistry that have been swept under the carpet, J. Solid State Electrochem., 2011, 15, 1359–1371.
8. F. R. Fan, Z. Q. Tian and Z. L. Wang, Flexible triboelectric generator, Nano Energy, 2012, 1(2), 328–334.
9. Z. L. Wang, J. Chen and L. Lin, Progress in triboelectric nanogenerators as a new energy technology and self-powered sensors, Energy Environ. Sci., 2015, 8(8), 2250–2282.
10. J. Bae, J. Lee and S. M. Kim, et al., Flutter-driven triboelectrification for harvesting wind energy, Nat. Commun., 2014, 5(1), 4929.
11. Z. L. Wang, T. Jiang and L. Xu, Toward the blue energy dream by triboelectric nanogenerator networks, Nano Energy, 2017, 39, 9–23.
12. Y. Yang, H. Zhang and Z. H. Lin, et al., Human skin based triboelectric nanogenerators for harvesting biomechanical energy and as self-powered active tactile sensor system, ACS Nano, 2013, 7(10), 9213–9222.
13. G. Yao, L. Xu and X. Cheng, et al., Bioinspired triboelectric nanogenerators as self-powered electronic skin for robotic tactile sensing, Adv. Funct. Mater., 2020, 30(6), 1907312.
14. C. Wu, T. W. Kim and J. H. Park, et al., Self-powered tactile sensor with learning and memory, ACS Nano, 2019, 14(2), 1390–1398.
15. K. Han, J. Luo and Y. Feng, et al., Self-powered electrocatalytic ammonia synthesis directly from air as driven by dual triboelectric nanogenerators, Energy Environ. Sci., 2020, 13(8), 2450–2458.
16. X. F. Li, A. Berbille and T. Y. Wang, et al., Defect Passivation Toward Designing High-Performance Fluorinated Polymers for Liquid–Solid Contact-Electrification and Contact-Electro-Catalysis, Adv. Funct. Mater., 2024, 34(25), 2315817.
17. X. Xu, Z. L. Wang and Z. Yang, Triboelectric junction: a model for dynamic metal-semiconductor contacts, Energy Environ. Sci., 2024, 17(1), 149–157.
18. C. Liu and A. J. Bard, Electrons on dielectrics and contact electrification, Chem. Phys. Lett., 2009, 480(4–6), 145–156.
19. A. F. Diaz, D. Wollmann and D. Dreblow, Contact electrification: ion transfer to metals and polymers, Chem. Mater., 1991, 3(6), 997–999.
20. W. R. Salaneck, A. Paton and D. T. Clark, Double mass transfer during polymer-polymer contacts, J. Appl. Phys., 1976, 47(1), 144–147.
21. J. Lowell, The role of material transfer in contact electrification, J. Phys. D: Appl. Phys., 1977, 10(17), L233.
22. H. T. Baytekin, A. Z. Patashinski and M. Branicki, et al., The mosaic of surface charge in contact electrification, Science, 2011, 333(6040), 308–312.
23. C. Xu, Y. Zi and A. C. Wang, et al., On the electron-transfer mechanism in the contact-electrification effect, Adv. Mater., 2018, 30(15), 1706790.
24. S. Lin, L. Xu and L. Zhu, et al., Electron transfer in nanoscale contact electrification: photon excitation effect, Adv. Mater., 2019, 31(27), 1901418.
25. C. Xu, B. Zhang and A. C. Wang, et al., Effects of metal work function and contact potential difference on electron thermionic emission in contact electrification, Adv. Funct. Mater., 2019, 29(29), 1903142.
26. S. Lin, L. Xu and C. Xu, et al., Electron transfer in nanoscale contact electrification: effect of temperature in the metal–dielectric case, Adv. Mater., 2019, 31(17), 1808197.
27. Y. S. Zhou, Y. Liu and G. Zhu, et al., In situ quantitative study of nanoscale triboelectrification and patterning, Nano Lett., 2013, 13(6), 2771–2776.
28. Z. L. Wang and A. C. Wang, On the origin of contact-electrification, Mater. Today, 2019, 30, 34–51.
29. S. Li, Y. Zhou and Y. Zi, et al., Excluding contact electrification in surface potential measurement using kelvin probe force microscopy, ACS Nano, 2016, 10(2), 2528–2535.
30. Y. Zhang and T. Shao, Effect of contact deformation on contact electrification: a first-principles calculation, J. Phys. D: Appl. Phys., 2013, 46(23), 235304.
31. C. B. Duke and T. J. Fabish, Contact electrification of polymers: A quantitative model, J. Appl. Phys., 1978, 49(1), 315–321.
32. X. Shen, A. E. Wang and R. M. Sankaran, et al., First-principles calculation of contact electrification and validation by experiment, J. Electrost., 2016, 82, 11–16.
33. S. Li, J. Nie and Y. Shi, et al., Contributions of different functional groups to contact electrification of polymers, Adv. Mater., 2020, 32(25), 2001307.
34. H. Zou, Y. Zhang and L. Guo, et al., Quantifying the triboelectric series, Nat. Commun., 2019, 10(1), 1427.
35. D. Liu, L. Zhou and S. Cui, et al., Standardized measurement of dielectric materials' intrinsic triboelectric charge density through the suppression of air breakdown, Nat. Commun., 2022, 13(1), 6019.
36. K. E. Byun, Y. Cho and M. Seol, et al., Control of triboelectrification by engineering surface dipole and surface electronic state, ACS Appl. Mater. Interfaces, 2016, 8(28), 18519–18525.
37. Z. Liu, S. Li and S. Lin, et al., Crystallization-induced shift in a triboelectric series and even polarity reversal for elastic triboelectric materials, Nano Lett., 2022, 22(10), 4074–4082.
38. S. H. Shin, Y. E. Bae and H. K. Moon, et al., Formation of triboelectric series via atomic-level surface functionalization for triboelectric energy harvesting, ACS Nano, 2017, 11(6), 6131–6138.
39. K. W. Biegaj, M. G. Rowland and T. M. Lukas, et al., Surface chemistry and humidity in powder electrostatics: A comparative study between tribocharging and corona discharge, ACS Omega, 2017, 2(4), 1576–1582.
40. S. Wang, Y. Xie and S. Niu, et al., Maximum surface charge density for triboelectric nanogenerators achieved by ionized-air injection: methodology and theoretical understanding, Adv. Mater., 2014, 26(39), 6720–6728.
41. S. Li, Y. Fan and H. Chen, et al., Manipulating the triboelectric surface charge density of polymers by low-energy helium ion irradiation/implantation, Energy Environ. Sci., 2020, 13(3), 896–907.
42. Q. Z. Guo, L. C. Yang and R. C. Wang, et al., Tunable work function of Mg_xZn_1–xO as a viable friction material for a triboelectric nanogenerator, ACS Appl. Mater. Interfaces, 2018, 11(1), 1420–1425.
43. S. N. Chen, C. H. Chen and Z. H. Lin, et al., On enhancing capability of tribocharge transfer of ZnO nanorod arrays by Sb doping for anomalous output performance improvement of triboelectric nanogenerators, Nano Energy, 2018, 45, 311–318.
44. J. Wang, S. Xu and G. Li, et al., Regulation of deep and shallow hole/electron trap states and charge conducting behaviors of dielectric tribo-materials for maximizing retained charges, Energy Environ. Sci., 2024, 17(19), 7382–7393.
45. Z. Li, L. Zhang and L. Guo, et al., Manipulating functional groups between polyvinylidene difluoride and nanoparticles for high-performance triboelectric nanogenerator, Nano Res., 2023, 16(9), 11855–11861.
46. T. Huang, M. Lu and H. Yu, et al., Enhanced power output of a triboelectric nanogenerator composed of electrospun nanofiber mats doped with graphene oxide, Sci. Rep., 2015, 5(1), 13942.
47. W. Li, Y. Zhang and L. Liu, et al., A high energy output nanogenerator based on reduced graphene oxide, Nanoscale, 2015, 7(43), 18147–18151.
48. S. J. Park, M. L. Seol and D. Kim, et al., Triboelectric nanogenerator with nanostructured metal surface using water-assisted oxidation, Nano Energy, 2016, 21, 258–264.
49. J. H. Lee, I. Yu and S. Hyun, et al., Remarkable increase in triboelectrification by enhancing the conformable contact and adhesion energy with a film-covered pillar structure, Nano Energy, 2017, 34, 233–241.
50. L. Zhao, Q. Zheng and H. Ouyang, et al., A size-unlimited surface microstructure modification method for achieving high performance triboelectric nanogenerator, Nano Energy, 2016, 28, 172–178.
51. H. Zhang, Y. Yang and Y. Su, et al., Triboelectric nanogenerator for harvesting vibration energy in full space and as self-powered acceleration sensor, Adv. Funct. Mater., 2014, 24(10), 1401–1407.
52. M. Muthu, R. Pandey and X. Wang, et al., Enhancement of triboelectric nanogenerator output performance by laser 3D-Surface pattern method for energy harvesting application, Nano Energy, 2020, 78, 105205.
53. S. Zhu, Y. Xia and Y. Zhu, et al., High-performance triboelectric nanogenerator powered flexible electroluminescence devices based on patterned laser-induced copper electrodes for visualized information interaction, Nano Energy, 2022, 96, 107116.
54. H. Chen, W. Yang and P. Huang, et al., A multiple laser-induced hybrid electrode for flexible triboelectric nanogenerators, Sustainable Energy Fuels, 2021, 5(14), 3737–3743.
55. H. Varghese, H. M. A. Hakkeem and K. Chauhan, et al., A high-performance flexible triboelectric nanogenerator based on cellulose acetate nanofibers and micropatterned PDMS films as mechanical energy harvester and self-powered vibrational sensor, Nano Energy, 2022, 98, 107339.
56. A. E. Wang, I. Greber and J. C. Angus, Contact charge transfer between inorganic dielectric solids of different surface roughness, J. Electrost., 2019, 101, 103359.
57. J. C. Angus and I. Greber, Tribo-electric charging of dielectric solids of identical composition, J. Appl. Phys., 2018, 123(17), 174102.
58. M. M. Apodaca, P. J. Wesson and K. J. M. Bishop, et al., Contact electrification between identical materials, Angew. Chem., Int. Ed., 2010, 5(49), 946–949.
59. C. Xu, B. Zhang and A. C. Wang, et al., Contact-electrification between two identical materials: curvature effect, ACS Nano, 2019, 13(2), 2034–2041.
60. G. Makov and A. Nitzan, Electronic properties of finite metallic systems, Phys. Rev. B, 1993, 47(4), 2301.
61. A. C. Wang, B. Zhang and C. Xu, et al., Unraveling temperature-dependent contact electrification between sliding-mode triboelectric pairs, Adv. Funct. Mater., 2020, 30(12), 1909384.
62. X. Xia, H. Wang and H. Guo, et al., On the material-dependent charge transfer mechanism of the contact electrification, Nano Energy, 2020, 78, 105343.
63. L. L. Sun, C. Y. Li and X. Chen, et al., Effect of ultraviolet light and ozone on electrification performance of polymers, Nano Energy, 2021, 90, 106587.
64. K. Wang, Z. Qiu and J. Wang, et al., Effect of relative humidity on the enhancement of the triboelectrification efficiency utilizing water bridges between triboelectric materials, Nano Energy, 2022, 93, 106880.
65. Y. Zhang, T. Pähtz and Y. Liu, et al., Electric field and humidity trigger contact electrification, Phys. Rev. X, 2015, 5(1), 011002.
66. Y. Liu, G. Liu, T. Bu and C. Zhang, Effects of interfacial acid–base on the performance of contact–separation mode triboelectric nanogenerator, Mater. Today Energy, 2021, 20, 100686.
67. S. Lin, L. Xu and W. Tang, et al., Electron transfer in nano-scale contact electrification: Atmosphere effect on the surface states of dielectrics, Nano Energy, 2019, 65, 103956.
68. K. Han, W. Tang and J. Chen, et al., Effects of environmental atmosphere on the performance of contact–separation mode TENG, Adv. Mater. Technol., 2019, 4(2), 1800569.
69. N. Luo, D. Yang and M. Feng, et al., Vacuum discharge triboelectric nanogenerator with ultrahigh current density, Cell Rep. Phys. Sci., 2023, 4(3), 101320.
70. J. Wang, C. Wu and Y. Dai, et al., Achieving ultrahigh triboelectric charge density for efficient energy harvesting, Nat. Commun., 2017, 8(1), 88.
71. Y. S. Zhou, S. Wang and Y. Yang, et al., Manipulating nanoscale contact electrification by an applied electric field, Nano Lett., 2014, 14(3), 1567–1572.
72. S. Du, S. Fu and W. He, et al., High Durable Rotary Triboelectric Nanogenerator Enabled by Ferromagnetic Metal Particles as a Friction Material, Adv. Funct. Mater., 2023, 33(48), 2306491.
73. Z. Zhang, N. Yin and Z. Wu, et al., Research methods of contact electrification: Theoretical simulation and experiment, Nano Energy, 2021, 79, 105501.
74. P. Peljo, J. A. Manzanares and H. H. Girault, Contact potentials, fermi level equilibration, and surface charging, Langmuir, 2016, 32(23), 5765–5775.
75. S. Kim, M. K. Gupta and K. Y. Lee, et al., Transparent flexible graphene triboelectric nanogenerators, Adv. Mater., 2014, 26(23), 3918–3925.
76. L. Lin, Y. Xie and S. Wang, et al., Triboelectric active sensor array for self-powered static and dynamic pressure detection and tactile imaging, ACS Nano, 2013, 7(9), 8266–8274.
77. T. Jiang, W. Tang and X. Chen, et al., Figures-of-merit for rolling-friction-based triboelectric nanogenerators, Adv. Mater. Technol., 2016, 1(1), 1600017.
78. C. Pei, C. Y. Wu and M. Adams, DEM-CFD analysis of contact electrification and electrostatic interactions during fluidization, Powder Technol., 2016, 304, 208–217.
79. W. Wei and Z. Gu, Electrification of particulate entrained fluid flows—mechanisms, applications, and numerical methodology, Phys. Rep., 2015, 600, 1–53.
80. S. Naik, B. Hancock and Y. Abramov, et al., Quantification of tribocharging of pharmaceutical powders in V-blenders: experiments, multiscale modeling, and simulations, J. Pharm. Sci., 2016, 105(4), 1467–1477.
81. C. Pei, C. Y. Wu and M. Adams, Numerical analysis of contact electrification of non-spherical particles in a rotating drum, Powder Technol., 2015, 285, 110–122.
82. S. Naik, S. Sarkar and V. Gupta, et al., A combined experimental and numerical approach to explore tribocharging of pharmaceutical excipients in a hopper chute assembly, Int. J. Pharm., 2015, 491(1–2), 58–68.
83. F. Zaera, Probing liquid/solid interfaces at the molecular level, Chem. Rev., 2012, 112(5), 2920–2986.
84. O. M. Magnussen and A. Groß, Toward an atomic-scale understanding of electrochemical interface structure and dynamics, J. Am. Chem. Soc., 2019, 141(12), 4777–4790.
85. S. M. Oja, M. Wood and B. Zhang, Nanoscale electrochemistry, Anal. Chem., 2013, 85(2), 473–486.
86. A. Kolmakov and M. Moskovits, Chemical sensing and catalysis by one-dimensional metal-oxide nanostructures, Annu. Rev. Mater. Res., 2004, 34, 151–180.
87. I. I. Nedrygailov, C. Lee and S. Y. Moon, et al., Hot electrons at solid–liquid interfaces: A large chemoelectric effect during the catalytic decomposition of hydrogen peroxide, Angew. Chem., Int. Ed., 2016, 128(36), 11017–11020.
88. S. H. Lee, I. I. Nedrygailov and S. Oh, et al., Hot electron flux at solid–liquid interfaces probed with Pt/Si catalytic nanodiodes: Effects of pH during decomposition of hydrogen peroxide, Catal. Today, 2018, 303, 282–288.
89. T. Krupenkin and J. A. Taylor, Reverse electrowetting as a new approach to high-power energy harvesting, Nat. Commun., 2011, 2(1), 448.
90. Y. Wang, S. Gao and W. Xu, et al., Nanogenerators with superwetting surfaces for harvesting water/liquid energy, Adv. Funct. Mater., 2020, 30(26), 1908252.
91. N. S. Lewis, Developing a scalable artificial photosynthesis technology through nanomaterials by design, Nat. Nanotechnol., 2016, 11(12), 1010–1019.
92. H. Yuan, J. Zhang and S. Rencus-Lazar, et al., The engineering of molecular packing in amino acid crystals for the enhanced triboelectric effect, Nano Energy, 2023, 110, 108375.
93. A. El-Kazzaz and A. C. Rose-Innes, Contact charging of insulators by liquid metals, J. Electrost., 1985, 16, 157–163.
94. M. Matsui, N. Murasaki and K. Fujibayashi, et al., Electrification of pure water flowing down a trough set up with a resin sheet, J. Electrost., 1993, 31, 1–10.
95. K. Yatsuzuka, Y. Mizuno and K. Asano, Electrification phenomena of pure water droplets dripping and sliding on a polymer surface, J. Electrost., 1994, 32(2), 157–171.
96. K. Yatsuzuka, Y. Higashiyama and K. Asano, Electrification of polymer surface caused by sliding ultrapure water, IEEE Trans. Ind. Appl., 1996, 32(4), 825–831.
97. G. G. Touchard, T. W. Patzek and C. J. Radke, A physicochemical explanation for flow electrification in low-conductivity liquids in contact with a corroding wall, IEEE Trans. Ind. Appl., 1996, 32(5), 1051–1057.
98. T. J. Lewis, Charge transport, charge injection and breakdown in polymeric insulators, J. Phys. D: Appl. Phys., 1990, 23(12), 1469.
99. R. Zimmermann, S. Dukhin and C. Werner, Electrokinetic measurements reveal interfacial charge at polymer films caused by simple electrolyte ions, J. Phys. Chem. B, 2001, 105(36), 8544–8549.
100. H. T. Baytekin, B. Baytekin and S. Soh, et al., Is water necessary for contact electrification?, Angew. Chem., Int. Ed., 2011, 123(30), 6898–6902.
101. H. Kitabayashi, K. Tsuji and K. Itoh, A streaming electrification model based on differences of work function between solid materials and insulating oil, J. Electrost., 2005, 63(6–10), 735–741.
102. H. W. Cheng, P. Stock and B. Moeremans, et al., Characterizing the influence of water on charging and layering at electrified ionic-liquid/solid interfaces, Adv. Mater. Interfaces, 2015, 2(12), 1500159.
103. Z. H. Lin, G. Cheng and S. Lee, et al., Harvesting water drop energy by a sequential contact-electrification and electrostatic-induction process, Adv. Mater., 2014, 26(27), 4690–4696.
104. G. Cheng, Z. H. Lin and Z. Du, et al., Simultaneously harvesting electrostatic and mechanical energies from flowing water by a hybridized triboelectric nanogenerator, ACS Nano, 2014, 8(2), 1932–1939.
105. G. Zhu, Y. Su and P. Bai, et al., Harvesting water wave energy by asymmetric screening of electrostatic charges on a nanostructured hydrophobic thin-film surface, ACS Nano, 2014, 8(6), 6031–6037.
106. S. Lin, L. Xu and A. Chi Wang, et al., Quantifying electron-transfer in liquid-solid contact electrification and the formation of electric double-layer, Nat. Commun., 2020, 11(1), 399.
107. J. Nie, Z. Ren and L. Xu, et al., Probing contact-electrification-induced electron and ion transfers at a liquid–solid interface, Adv. Mater., 2020, 32(2), 1905696.
108. F. Zhan, A. C. Wang and L. Xu, et al., Electron transfer as a liquid droplet contacting a polymer surface, ACS Nano, 2020, 14(12), 17565–17573.
109. S. Lin, X. Chen and Z. L. Wang, Contact electrification at the liquid–solid interface, Chem. Rev., 2021, 122(5), 5209–5232.
110. T. A. L. Burgo, F. Galembeck and G. H. Pollack, Where is water in the triboelectric series?, J. Electrost., 2016, 80, 30–33.
111. S. Li, Z. Zhang and P. Peng, et al., A green approach to induce and steer chemical reactions using inert solid dielectrics, Nano Energy, 2024, 122, 109286.
112. F. Williams, S. P. Varma and S. Hillenius, Liquid water as a lone-pair amorphous semiconductor, J. Chem. Phys., 1976, 64(4), 1549–1554.
113. X. Li, L. Zhang and Y. Feng, et al., Reversible temperature-sensitive liquid–solid triboelectrification with polycaprolactone material for wetting monitoring and temperature sensing, Adv. Funct. Mater., 2021, 31(17), 2010220.
114. H. Cho, J. Chung and G. Shin, et al., Toward sustainable output generation of liquid–solid contact triboelectric nanogenerators: The role of hierarchical structures, Nano Energy, 2019, 56, 56–64.
115. S. Lin, M. Zheng and J. Luo, et al., Effects of surface functional groups on electron transfer at liquid–solid interfacial contact electrification, ACS Nano, 2020, 14(8), 10733–10741.
116. B. D. Chen, W. Tang and C. He, et al., Ultrafine capillary-tube triboelectric nanogenerator as active sensor for microliquid biological and chemical sensing, Adv. Mater. Technol., 2018, 3(1), 1700229.
117. L. Zhang, X. Li and Y. Zhang, et al., Regulation and influence factors of triboelectricity at the solid-liquid interface, Nano Energy, 2020, 78, 105370.
118. X. Tao, J. Nie and S. Li, et al., Effect of photo-excitation on contact electrification at liquid–solid interface, ACS Nano, 2021, 15(6), 10609–10617.
119. M. Zheng, S. Lin and Z. Tang, et al., Photovoltaic effect and tribovoltaic effect at liquid-semiconductor interface, Nano Energy, 2021, 83, 105810.
120. S. Lin, L. Zhu and Z. Tang, et al., Spin-selected electron transfer in liquid–solid contact electrification, Nat. Commun., 2022, 13(1), 5230.
121. P. Wang, S. Zhang and L. Zhang, et al., Non-contact and liquid–liquid interfacing triboelectric nanogenerator for self-powered water/liquid level sensing, Nano Energy, 2020, 72, 104703.
122. R. Zhang, H. Lin and Y. Pan, et al., Liquid–liquid triboelectric nanogenerator for harvesting distributed energy, Adv. Funct. Mater., 2022, 32(51), 2208393.
123. L. Sun, Z. Wang and C. Li, et al., Probing contact electrification between gas and solid surface, Nanoenergy Adv., 2023, 3(1), 1–11.
124. Y. Dong, S. Xu and C. Zhang, et al., Gas-liquid two-phase flow-based triboelectric nanogenerator with ultrahigh output power, Sci. Adv., 2022, 8(48), eadd0464.
125. Y. Dong, M. Feng and J. Cheng, et al., Ring-shaped single-electrode triboelectric nanogenerator (RSE-TENG) for energy harvesting and liquid flow rate monitoring of gas-liquid two-phase flow, Nano Energy, 2024, 119, 109083.
126. R. Xu, Q. Zhang and J. Y. Wang, et al., Direct current triboelectric cell by sliding an n-type semiconductor on a p-type semiconductor, Nano Energy, 2019, 66, 104185.
127. S. Lin, X. Chen and Z. L. Wang, The tribovoltaic effect and electron transfer at a liquid-semiconductor interface, Nano Energy, 2020, 76, 105070.
128. M. Zheng, S. Lin and L. Xu, et al., Scanning Probing of the Tribovoltaic Effect at the Sliding Interface of Two Semiconductors, Adv. Mater., 2020, 32(21), e2000928.
129. W. H. Lee, S. G. Yoon and H. Jin, et al., Electron Density-Change in Semiconductor by Ion-Adsorption at Solid-Liquid Interface, Adv. Mater., 2021, 33(10), e2007581.
130. M. Zheng, S. Lin and L. Zhu, et al., Effects of Temperature on the Tribovoltaic Effect at Liquid-Solid Interfaces, Adv. Mater. Interfaces, 2021, 9(3), 2101757.
131. Y. Huang, D. Liu and X. Gao, et al., Flexible Liquid-Based Continuous Direct-Current Tribovoltaic Generators Enable Self-Powered Multi-Modal Sensing, Adv. Funct. Mater., 2022, 33(1), 2209484.
132. S. Deng, R. Xu and W. Seh, et al., Current degradation mechanism of tip contact metal-silicon Schottky nanogenerator, Nano Energy, 2022, 94, 106888.
133. J. Xia, X. Luo and J. Li, et al., Wear-Resisting and Stable 4H-SiC/Cu-Based Tribovoltaic Nanogenerators for Self-Powered Sensing in a Harsh Environment, ACS Appl. Mater. Interfaces, 2022, 14(49), 55192–55200.
134. Y.-S. Lee, S. Jeon and D. Kim, et al., High performance direct current-generating triboelectric nanogenerators based on tribovoltaic p–n junction with ChCl-passivated CsFAMA perovskite, Nano Energy, 2023, 106, 108066.
135. Y. Chen, Z. Zhang and Z. Wang, et al., Friction-Dominated Carrier Excitation and Transport Mechanism for GaN-Based Direct-Current Triboelectric Nanogenerators, ACS Appl. Mater. Interfaces, 2022, 14(20), 24020–24027.
136. Z. Wang, Z. Zhang and Y. Chen, et al., Achieving an ultrahigh direct-current voltage of 130 V by semiconductor heterojunction power generation based on the tribovoltaic effect, Energy Environ. Sci., 2022, 15(6), 2366–2373.
137. Z. Zhang, Z. Wang and Y. Chen, et al., Semiconductor Contact-Electrification-Dominated Tribovoltaic Effect for Ultrahigh Power Generation, Adv. Mater., 2022, 34(20), e2200146.
138. L. Gong, Z. Wang and R. Luan, et al., Competitive Mechanism between Interfacial Electric Field and Built-In Electric Field for Silicon-Based Tribovoltaic Effect, Adv. Funct. Mater., 2024, 34(10), 2310703.
139. J. Liu, A. Goswami and K. Jiang, et al., Direct-current triboelectricity generation by a sliding Schottky nanocontact on MoS₂ multilayers, Nat. Nanotechnol., 2018, 13(2), 112–116.
140. S. Lin, R. Shen and T. Yao, et al., Surface States Enhanced Dynamic Schottky Diode Generator with Extremely High Power Density Over 1000 W m^-2, Advanced Science, 2019, 6(24), 1901925.
141. H. Yuan, Z. Xiao and J. Wan, et al., A Rolling-Mode Al/CsPbBr₃ Schottky Junction Direct-Current Triboelectric Nanogenerator for Harvesting Mechanical and Solar Energy, Adv. Energy Mater., 2022, 12(25), 2200550.
142. X. Luo, L. Liu and Y. C. Wang, et al., Tribovoltaic Nanogenerators Based on MXene-Silicon Heterojunctions for Highly Stable Self-Powered Speed, Displacement, Tension, Oscillation Angle, and Vibration Sensors, Adv. Funct. Mater., 2022, 32(23), 2113149.
143. J. Liu, F. Liu and R. Bao, et al., Scaled-up Direct-Current Generation in MoS(2) Multilayer-Based Moving Heterojunctions, ACS Appl. Mater. Interfaces, 2019, 11(38), 35404–35409.
144. Z. Zhang, D. Jiang and J. Zhao, et al., Tribovoltaic Effect on Metal-Semiconductor Interface for Direct-Current Low-Impedance Triboelectric Nanogenerators, Adv. Energy Mater., 2020, 10(9), 1903713.
145. J. Chen, P. He and T. Huang, et al., Boosting carrier transfer at flexible Schottky junctions with moisture: A strategy for high-performance wearable direct-current nanogenerators, Nano Energy, 2021, 90, 106593.
146. J. Meng, Z. H. Guo and C. Pan, et al., Flexible Textile Direct-Current Generator Based on the Tribovoltaic Effect at Dynamic Metal-Semiconducting Polymer Interfaces, ACS Energy Lett., 2021, 6(7), 2442–2450.
147. R. Yang, M. Benner and Z. Guo, et al., High-Performance Flexible Schottky DC Generator via Metal/Conducting Polymer Sliding Contacts, Adv. Funct. Mater., 2021, 31(43), 2103132.
148. A. Šutka, M. Zubkins and A. Linarts, et al., Tribovoltaic Device Based on the W/WO₃ Schottky Junction Operating through Hot Carrier Extraction, J. Phys. Chem. C, 2021, 125(26), 14212–14220.
149. Y. Huang, D. Liu and T. Zhu, et al., A mechanically tunable electromagnetic wave harvester and dual-modal detector based on quasi-static van der Waals heterojunction, Nano Energy, 2022, 99, 107399.
150. M. Wang, J. Yang and S. Liu, et al., Metal-Gallium Arsenide Based Tribovoltaic Nanogenerators and its Application for High-Precision Self-Powered Displacement Sensors, Adv. Mater. Technol., 2022, 8(3), 2200677.
151. K. Xiao, Y. Feng and W. Wei, et al., A Cu/P-type GaN triboelectric nanogenerator with power density over 100 W/m², Appl. Phys. Lett., 2022, 120(9), 093901.
152. J. Xia, A. Berbille and X. Luo, et al., Reversal in Output Current Direction of 4H-SiC/Cu Tribovoltaic Nanogenerator as Controlled by Relative Humidity, Small, 2023, 20(2), 2305303.
153. C. Ma, B. Kim, S.-W. Kim and N.-G. Park, Dynamic halide perovskite heterojunction generates direct current, Energy Environ. Sci., 2021, 14(1), 374–381.
154. L. Ren, A. Yu and W. Wang, et al., p–n Junction Based Direct-Current Triboelectric Nanogenerator by Conjunction of Tribovoltaic Effect and Photovoltaic Effect, Nano Lett., 2021, 21(23), 10099–10106.
155. S. Sriphan, S. Worathat and U. Pharino, et al., Highly Flexible Tribovoltaic Nanogenerator Based-on P-N Junction Interface: Comparative Study on Output Dependency Dominated by Photovoltaic Effect in Freestanding-Mode, Adv. Funct. Mater., 2023, 33(43), 2305106.
156. K.-Q. Wei, D.-J. Sun and M.-N. Liu, et al., Direct Current Nanogenerator Based on Tribovoltaic Effect at WS₂ Semiconductor Interface, ACS Appl. Nano Mater., 2024, 7(2), 1748–1756.
157. M. Willatzen, Z. Zhang and Z. L. Wang, Theory of Tribovoltaics: Direct Current Generation at a p–n Semiconductor Interface, PRX Energy, 2024, 3(1), 013009.
158. H.-H. Huang, R.-C. Wang and Y.-J. Chen, Fluorinated graphite paper used for self-powered water speed sensors by immersion-type tribovoltaic effect-dominated triboelectric nanogenerators, Nano Energy, 2022, 93, 106887.
159. M. Yalcin, D. K. Takci and A. F. Demirel, Investigation of the photovoltaic effect and tribovoltaic effect at the GaAs semiconductor-liquid interface, J. Electrost., 2024, 128, 103896.
160. J. Liu, M. I. Cheikh and R. Bao, et al., Tribo-Tunneling DC Generator with Carbon Aerogel/Silicon Multi-Nanocontacts, Adv. Electron. Mater., 2019, 5(12), 1900464.
161. Y. Yu and X. Wang, Chemical modification of polymer surfaces for advanced triboelectric nanogenerator development, Extreme Mech. Lett., 2016, 9, 514–530.
162. A. Iqbal, M. S. Hossain and K. H. Bevan, The role of relative rate constants in determining surface state phenomena at semiconductor–liquid interfaces, Phys. Chem. Chem. Phys., 2016, 18(42), 29466–29477.
163. X. Zhang, D. Ren and H. Wu, et al., Ultrahigh voltage direct current quasi-tribovoltaic nanogenerator by switchable tribo-bias induction and deposited charge extraction, Energy Environ. Sci., 2024, 17(12), 4175–4186.
164. W. Qiao, Z. Zhao and L. Zhou, et al., Simultaneously Enhancing Direct-Current Density and Lifetime of Tribovotaic Nanogenerator via Interface Lubrication, Adv. Funct. Mater., 2022, 32(46), 2208544.
165. D. Yang, L. Zhang and N. Luo, et al., Tribological-behaviour-controlled direct-current triboelectric nanogenerator based on the tribovoltaic effect under high contact pressure, Nano Energy, 2022, 99, 107370.
166. W. Qiao, L. Zhou and Z. Zhao, et al., MXene Lubricated Tribovoltaic Nanogenerator with High Current Output and Long Lifetime, Nano-Micro Lett., 2023, 15(1), 218.
167. L. Zhang, H. Cai and L. Xu, et al., Macro-superlubric triboelectric nanogenerator based on tribovoltaic effect, Matter, 2022, 5(5), 1532–1546.
168. M. Benner, R. Yang and L. Lin, et al., Mechanism of In-Plane and Out-of-Plane Tribovoltaic Direct-Current Transport with a Metal/Oxide/Metal Dynamic Heterojunction, ACS Appl. Mater. Interfaces, 2022, 14(2), 2968–2978.
169. X. Xu, J. Li and X. Tao, et al., Study of interfacial design for direct-current tribovoltaic generators, Nano Energy, 2022, 94, 106957.
170. R. Yang, Z. He and S. Lin, et al., Tunable Tribovoltaic Effect via Metal-Insulator Transition, Nano Lett., 2022, 22(22), 9084–9091.
171. Z. Zhao, J. Zhang and W. Qiao, et al., Contact efficiency optimization for tribovoltaic nanogenerators, Mater. Horiz., 2023, 10(12), 5962–5968.
172. Y. Lu, Y. Yan and X. Yu, et al., Polarized Water Driven Dynamic PN Junction-Based Direct-Current Generator, Research, 2021, 2021, 7505638.
173. S. Ferrie, A. P. Le Brun and G. Krishnan, et al., Sliding silicon-based Schottky diodes: Maximizing triboelectricity with surface chemistry, Nano Energy, 2022, 93, 106861.
174. Z. Zhang, T. He and J. Zhao, et al., Tribo-thermoelectric and tribovoltaic coupling effect at metal-semiconductor interface, Mater. Today Phys., 2021, 16, 100295.
175. T. Lv, R. Cheng and C. Wei, et al., All-Fabric Direct-Current Triboelectric Nanogenerators Based on the Tribovoltaic Effect as Power Textiles, Adv. Energy Mater., 2023, 13(29), 2301178.
176. S. H. Sohn, G. J. Choi, B. D. On and I. K. Park, Synergistic Coupling of Tribovoltaic and Moisture-Enabled Electricity Generation in Layered-Double Hydroxides, Adv. Energy Mater., 2024, 14(10), 2304206.
177. M. Li, L. Yang and H. Wang, et al., Coupling mechanism between photogenerated carriers and triboelectric charges and photoinduced reinforcement of a triboelectric nanogenerator, Appl. Phys. Lett., 2021, 119(6), 063903.
178. J. Meng, C. Pan and L. Li, et al., Durable flexible direct current generation through the tribovoltaic effect in contact-separation mode, Energy Environ. Sci., 2022, 15(12), 5159–5167.
179. Q. Zhang, R. Xu and W. Cai, Pumping electrons from chemical potential difference, Nano Energy, 2018, 51, 698–703.
180. S. Lin and Z. Lin Wang, The tribovoltaic effect, Mater. Today, 2023, 62, 111–128.
181. X. Liang, T. Jiang and G. Liu, et al., Spherical triboelectric nanogenerator integrated with power management module for harvesting multidirectional water wave energy, Energy Environ. Sci., 2020, 13(1), 277–285.
182. L. Zhao, X. Guo and Y. Pan, et al., Triboelectric gait sensing analysis system for self-powered IoT-based human motion monitoring, InfoMat, 2024, 6(5), e12520.
183. Z. Wang, A. Berbille and Y. Feng, et al., Contact-electro-catalysis for the degradation of organic pollutants using pristine dielectric powders, Nat. Commun., 2022, 13(1), 130.

---

Footnote

† These authors contributed equally to this work.

---