The mechanisms and applications of friction energy dissipation

About 30% of the world’s primary energy consumption is in friction. The economic losses caused by friction energy dissipation and wear account for about 2%–7% of its gross domestic product (GDP) for different countries every year. The key to reducing energy consumption is to control the way of energy dissipation in the friction process. However, due to many various factors affecting friction and the lack of efficient detection methods, the energy dissipation mechanism in friction is still a challenging problem. Here, we firstly introduce the classical microscopic mechanism of friction energy dissipation, including phonon dissipation, electron dissipation, and non-contact friction energy dissipation. Then, we attempt to summarize the ultrafast friction energy dissipation and introduce the high-resolution friction energy dissipation detection system, since the origin of friction energy dissipation is essentially related to the ultrafast dynamics of excited electrons and phonons. Finally, the application of friction energy dissipation in representative high-end equipment is discussed, and the potential economic saving is predicted.


Introduction
Tribology is a subject of the science and technology of friction, wear, and lubrication of interacting surfaces in relative motion [1,2]. It appears in high-end equipment such as microelectromechanical systems, integrated circuits, aerospace, and new energy vehicles. For example, the size of moving parts in the micro-nano devices reaches the order of micro-nano ( Fig. 1(a)), and the interface friction there replaces gravity as the dominant force. And the friction and wear interface effect will become the main bottleneck problem, affecting the device's performance. If car engines worldwide reduce friction by 18% ( Fig. 1(b)), 117,000 million in fuel can be saved, and 290 million tons of CO 2 emissions can be reduced [3]. The Kepler telescope launched by National Aeronautics and Space Administration (NASA), whose flywheel shaft system failed due to friction ( Fig. 1(c)), lost the function of the whole machine after some years of service [4]. If the wear of the flywheel can be reduced, its service quality can be significantly improved. According to statistics, around the world, friction causes ~30% of primary energy waste, and wear causes ~80% of machine parts to fail every year. The average economic loss caused by friction and wear in different countries accounts for about 2%-7% of their GDP [3,5,6]. As a major manufacturing country, China accounts for a higher proportion of losses caused by friction and wear. Calculated at only 5% [6], the losses caused by friction and wear may be as high as about 5 trillion RMB in 2020.
In recent years, the discovery of the superlubricity penomenon has provided a new and vital way to solve the problem of frictional energy consumption [7][8][9][10][11][12][13][14][15][16][17]. In the superlubricity state (Figs. 1(d)-1(f)), the friction coefficient is reduced by orders of magnitude compared with conventional oil lubrication [18,19]. And the wear rate is extremely low, close to zero. The realization and universal application of the superlubricity will significantly reduce the consumption of energy and resources, and significantly improve the service quality of essential mechanical moving parts. However, the mechanism of superlubricity is still unclear, and existing theories cannot explain some phenomena. The energy dissipation path and mechanism of the friction process are the key to predicting and controlling superlubricity.
In addition, the friction surfaces in mechanical products not only transfer motion and energy, but also play a decisive role in the efficiency, noise, accuracy, corrosion, reliability, and life of the equipment. New tribological theories and technologies are expected to provide practical solutions to the energy shortage, resource depletion, environmental pollution, and health problems faced by the development of human society. Therefore, we need to reveal the origin of friction and the nature of superlubricity from energy dissipation. Further, exploring the law of frictional energy dissipation is of great significance to high-end equipment manufacturing and the realization of energy conservation and emission reduction. At the same time, it also provides a strong guarantee for achieving the goal of carbon neutrality.

Classical microscopic mechanism of friction energy dissipation
The origin of friction has been a long-standing problem. In 1699, Amontons believed that the friction energy loss originated from the contact of the surface asperities when the contact surfaces moved relative to each other. And the elastoplastic behaviours such as meshing and collision between multiple asperities were the cause of friction. In 1785, Coulomb established the theory of unevenness caused by friction and generalized the classical friction law. In 1804, Leslie questioned the theory of friction bumps, thinking that the uphill and downhill would be roughly balanced when two asperities slide through each other [10]. Therefore, for the entire system, there was no change in energy. In the 1950s, Bowden and Tabor [21] proposed the theory of asperity adhesive friction, which assumed that friction originated from the shear and plastic deformation of the adhesive contact point. And they proposed that the friction force is proportional to the actual contact area and the normal load. Although the macro-scale solid friction theory system has become more and more complete, Fig. 1 (a) Micro-nano mechanical device, (b) automobile engine, (c) Kepler telescope, (d) schematic diagram of two-dimensional material solid superlubricity, (e) graphene self-recovery phenomenon, and (f) graphene-wrapped tip to achieve superlubricity. Reproduced with permission from Ref. [9] for (a, d), © Springer Nature Limited 2018; Ref. [20] for (e), © American Physical Society 2012; Ref. [7] for (f), © The Author(s) 2017.
www.Springer.com/journal/40544 | Friction the idea of friction control also stays on controlling the surface roughness and elastoplastic behaviour to influence the friction process. There is still a lack of in-depth research and analysis on the energy dissipation mechanism in the friction process.
With the development of modern instrument technology and simulation calculation methods, the theoretical system of tribology has gradually entered the micro-and nanoscale. In such scale friction experiments based on the atomic force microscopy (AFM), the researchers found that: (1) Relative sliding occurs between two ideally smooth surfaces, with neither asperities in contact nor sticking, but friction still exists; (2) an atom or atom-level tip is used for a slip experiment on an almost ideal smooth surface. The effect of asperities can be ignored, but there is still stick-slip friction, which consumes energy. At this time, theories such as rough peak contact and plastic deformation in macro-scale friction cannot explain the process of friction energy dissipation at the micro-nano scale, indicating that the mechanism of friction energy dissipation must involve more basic physical processes. Therefore, the origin of friction has once again raised people's interest. Before suitable detection equipment appears, computer simulation becomes a good research tool. And then, the atomiclevel friction model attracts people's attention, mainly phonon and electron dissipation models for microscopic friction energy dissipation.

Phonon dissipation
In the research of micro-and nanoscale friction, the first to be developed and perfected is the phonon friction model, including the independent oscillator (IO) model [22] and the Frenkel-Kontorova-Tomlinson (FKT) model [23,24]. The IO model proposed by Tomlinson in 1929 was the most concerned. Since Prandtl first proposed this model in 1928, it is also called the Prandtl-Tomlinson (PT) model [25]. The model states that the frictional upper surface is regarded as non-interacting atoms fixed on rigid support by a spring. The lower surface is regarded as an atomic array with a fixed periodic potential energy field, as shown in Fig. 2(a). The relative sliding of the upper and lower surfaces will trigger the vibration of the atoms on the upper surface. This kind of atomic lattice vibration originating from the friction contact position or spreading to the entire plane is related to the phonon energy dissipation.
The PT model regards the interaction on the friction interface as the interplay between the atoms connected by the spring and the periodic potential field. In contrast, it does not consider the interaction between the atoms. In 1938, Frenkel and Kontorova [23] developed a composite vibrator model based on this model. It not only considered the interaction between atoms, but also replaced the periodic potential field with a series of interface atoms connected by a spring network ( Fig. 2(b)). This model is also known as the FKT model. Although the composite vibrator model is more complex in form, it does not change the core feature of the transformation of spring deformation energy to vibration energy.
In the 1990s, with the rise of molecular dynamics simulation (MDS) technology, researchers can use the above models to analyze the origin of friction at the atomic level [26][27][28][29][30]. Wang et al. [31] and Xu et al. [32] studied the dynamic instability and stick-slip friction behaviour of self-assembled films, graphene, and other systems ( Fig. 3(a)). The work showed that the surface structure commensurability and lateral stiffness affected stick-slip friction and energy dissipation. The dissipation form of sliding kinetic energy was mainly caused by the lattice vibration, so the energy is irreversibly dissipated in the form of phonons. Weiss and Elmer [33] studied the dynamic characteristics of  | https://mc03.manuscriptcentral.com/friction the two-atom plane sliding at a constant speed, and pointed out that the friction force would increase under the influence of the coherent motion of phonons. Chen et al. [34] showed the temperature-dependent friction in carbon nanotubes. They found that phonons would be excited by the mechanical vibration when the temperature reached a critical value, increasing the friction force ( Fig. 3(b)). Vink [35] investigated the influence of the phonon back-scattering effect of the substrate on the sliding friction by the MDS. They proved that during sliding friction, the energy of the upper friction pair is mainly transferred to the phonon mode propagating vertically on the substrate. At the same time, they proposed that adjusting the life span of vertically propagating phonons can control friction.
In addition, a series of micro-and nanoscale friction experiments have also verified the vibration-based friction phonon dissipation [36][37][38][39][40][41]. Filleter et al. [42] found that the friction of monolayer graphene on SiC was larger than that of the bilayer graphene ( Fig. 3(c)). They attributed the contrast to the dramatic difference in electron-phonon coupling. Duan et al. [43] found that nonequilibrium phonons could induce resonant vibration of the tip and lead to multiple friction force peaks with the increasing sliding velocity (Fig. 3(d)).

Electron dissipation
In addition to phonon excitation, ohmic heat is generated between the sliding surfaces due to the lattice scattering of excited electrons [44,45]. Friction electron dissipation essentially comes from the frictional Coulomb drag effect between two very close layers. The moving atoms of the top layer can drag the carriers of the bottom layer by Coulomb interaction [46][47][48][49]. Therefore, additional friction and ohmic energy loss are generated. Persson and Zhang [50] first used the Coulomb drag model to analyze the www.Springer.com/journal/40544 | Friction friction force between two atomically smooth metal surfaces separated by a vacuum slab of thickness d (Fig. 4). They found that when the distance is small enough (a few nanometers), even if there is no potential difference between the two surfaces (no electric field at the interface), due to the thermal or quantum fluctuations, an unsteady state of charge distribution will still be generated at certain positions. This unstable charge distribution gives rise to a temporal charge imbalance and a spatial electric field. If the two surfaces slide relative to each other, there is always an energy difference between them. The relatively moving surface will continue to transfer energy to the relatively stationary surface in the form of electron dissipation, triggering currents inside the friction surface or at the contact interface. Finally, the current is dissipated as the ohmic heating under the influence of material resistance. In the early years, the research on electronic frictional energy dissipation mainly focused on the changes in macroscopic physical and chemical properties, resulting from the electronic behaviour, such as the increase in interface resistance caused by the film adsorption effect [51][52][53], the appearance of infrared absorption peaks [54], and the broadening of the vibration peak [55]. Persson et al. [56] and Dou and Subotnik [57] found that the resistivity of the metal film increased when the adsorbent slid on the metal surface. The generated kinetic energy excited the conduction band electrons when the adsorbent moved on the metal surface, resulting in damping at the interface. Witte et al. [58] found a strong interaction between the octane molecules and the clean metal Ru substrate, which excited electron-hole pairs in the metal Ru substrate when they oscillated into and out of the surface. This implied that the electrons controlled the octane molecule's vertical vibration and damping motion, leading to the molecular's vibration peaks broadening. In addition, Highland and Krim [59] compared the friction changes between the lead (Pb) substrate and the nitrogen (N 2 ), helium (He), and water (H 2 O) molecules before and after the superconductivity, and found that the H 2 O molecules caused stronger electronic friction dissipation.
At present, there is no powerful tool to detect the electron dynamics in the friction process directly. Therefore, researchers apply an external voltage to affect the process of electronic friction dissipation, which can change the concentration and distribution of carriers on the surface of the friction pair. Kim et al. [60] found that VO2 exhibited metallic characteristics above a critical temperature, and its internal current and friction increased with the increasing temperature. The increasing current was attributed to the increase in the number of carriers in VO 2 . However, the increasing friction resulted from the strong Coulomb attractive interaction between the trapped charge and the tip. Park et al. [61,62] applied different voltages to the pn junction of silicon. They found a strong accumulation of carriers in the p-region with a large ohmic loss when a positive voltage was applied, leading to a significantly higher friction force than that in the n-region ( Fig. 5(a)). He et al. [63] found that in-plane carriers could suppress the high-energy dissipation process at the friction interface of MoS 2 , thereby reducing friction ( Fig. 5(b)). Fang et al. [64] used the electron beam injection method to change the carrier concentration in MoS 2 for controlling the friction force. In addition, an external conductive layer could modulate the electrical properties of the friction interface [65].
Some calculations have also shown that the interfacial electric field can affect the interlayer charge distribution of two-dimensional materials, affecting the interlayer interaction and surface energy barrier and leading to the change of friction behaviour. Wang et al. [66] and Wang et al. [67] studied the influence of electric field on the friction between graphene and MoS2. They found that the | https://mc03.manuscriptcentral.com/friction electric field could inject different electron densities, thus controlling the interlayer interaction. Wang et al. [68] showed that the atomic-scale friction was determined by the fluctuation of the sliding-induced interface charge density (Fig. 5(c)).
Although the existence of electronic friction dissipation has been fully verified [69][70][71][72][73][74][75][76][77], its systematic model has not been formed [78][79][80]. Because the electronic dissipation process is affected by a variety of factors, such as electric energy band structure, carrier distribution state, and charge relaxation time. Moreover, the origin of electron dissipation involves quantum theory, and their related experiments require strict experimental conditions and test accuracy. In addition, in the actual friction dissipation process, the phonon and electron dissipation process often occur simultaneously and couple with each other, which also brings some challenges to the study of electronic friction dissipation.

Non-contact friction energy dissipation
The non-contact friction issue has been widely studied because of its simplified model and ultrasensitive force detection. In 1997, Pendry [81] calculated the friction between two infinite, parallel (with very small spacing) and relatively moving with a constant speed. He found that friction still existed even in an environment of absolute zero temperature, resulting from electromagnetic fluctuations in the vacuum. Volokitin and Persson [82] further studied the friction between two metal conductors with tens of nanometers and sliding relatively parallel. Theoretically, when there www.Springer.com/journal/40544 | Friction were some adsorbents on the metal surface, they could cause resonance photon tunnelling or low-frequency surface plasma. Thus, the surrounding electromagnetic field had large fluctuations, giving rise to a large increase in the friction between the two metal planes. Although the influencing factors of quantum friction are different, such as distances [83], shapes [84], surface adsorbents [85], and materials [86], the essence is related to the electromagnetic field fluctuation, which in turn changes the friction.
There are four microcosmic mechanisms for non-contact friction energy dissipation in Fig. 6 [65, [87][88][89]. (1) Van der Waals friction: Due to quantum fluctuations with no external field, an electric dipole is induced in the above tip, further inducing a dipole in the bottom surface. When the dipole in the tip emits a photon, its wave is Doppler-shifted in the bottom body, resulting in a different reflection amplitude. The same is true for the second body. The exchange of Doppler-shifted photons is the origin of the van der Waals friction energy dissipation. (2) Electrostatic friction: When an electrical field is applied or an inhomogeneous tip-sample electric field is unavoidable, the induced charges follow the tip motion, leading to Joule energy losses. (3) Phononic friction: The local elastic deformation of the surface resulting from the van der Waals interaction or electrostatic interaction follows the tip motion and induces the energy dissipation in the form of phonons in the substrate. (4) Adsorbate drag friction: When there are adsorbates on the surface, a moving tip will induce a drag force, owing to the van der Waals interaction or electrostatic interaction between the tip and adsorbates.
Non-contact friction can be investigated by attaching an extremely sharp tip on an elastic cantilever, which allows to register the strength of the normal and lateral force components between the surface and the tip ( Fig. 7(a)). Kisiel et al. [39] studied the dissipation mechanism of phonon and electron in Nb film under non-contact friction conditions. They found that the non-contact friction coefficient Γ, closely related to the probe-sample distance d and the bias voltage V s , was reduced by a factor of three when the sample entered the superconducting state. Specifically, in the metallic state, Γ was proportional to d −1 and 2 s V ; while in the superconducting state, Γ was proportional to d −4 and 4 s V , as shown in Fig. 7(b). This indicated that electronic friction was the main dissipation channel in the metal state; in the superconducting state, electronic friction was suppressed, and phonon friction became the main dissipation channel.
The topological insulator has topologically protected surface electronic states from preventing the backscattering of electrons, in which Joule dissipation is suppressed. Yildiz et al. [88] combined the scanning tunnelling microscopy (STM) and pendulum atomic force microscopy (pAFM) to perform non-contact friction experiments on the surface of Bi 2 Te 3 in Fig. 7(c). They found that due to the topological protection of the surface state, the relationship between the friction coefficient and the bias deviated from the parabolic shape. And dissipation peaks appeared at some special bias positions, indicating that Joule dissipation was suppressed, as shown in Fig. 7(d). In addition, as the magnetic field intensity increased, which destroyed the topological protection surface state, it is observed  | https://mc03.manuscriptcentral.com/friction that the dissipation peak decreased and Joule dissipation reappeared. This work provides a new idea for regulating non-contact frictional energy dissipation channels.
Although the non-contact frictional energy dissipation mechanisms of van der Waals friction, electrostatic friction, phonon friction, and adsorbate drag friction have been confirmed in experiments, their internal mechanisms are still unclear. Taking phonon friction as an example, although we have found that the friction process can excite phonons, how the friction phonons dissipate energy is unknown. This requires further exploration of the types of phonons and their dynamic behaviour, which is the key to clarifying the mechanism of phonons. Therefore, the non-contact frictional energy dissipation mechanism needs to be studied at a more microscopic level.

Ultrafast friction energy dissipation
With the development of quantum tribology, people gradually realized that the origin of friction is mainly related to the dynamics processes of energy carriers such as phonons, electrons, and photons at the friction interface. These energy dissipation pathways involve the charge transfer, electron-electron interaction, Auger recombination of carriers, lattice vibration, as well as processes such as exciton recombination, diffusion, annihilation, etc. (Fig. 8). These processes usually occur in the femtosecond range to nanosecond [11,79,[90][91][92][93][94], which is difficult to detect. Thus, to reveal the friction energy dissipation mechanism, it is necessary to systematically study the energy dissipation issue at nanoscale and ultrafast. But the existing instruments can no longer satisfy the requirement. In addition, a real friction interface involves multiphysics coupling such as electricity, heat, force, and light. These complicated conditions have reached the detection limit of existing scientific instruments in the field of tribology and require extremely high space-time challenges for high-sensitivity in-situ detection.

Ultrafast phonon energy dissipation dynamics
The friction energy dissipation path is almost all related to the thermal energy carried by various vibrations at the atomic scale [80]. Phonon describes the lattice vibration with different frequencies [43,[95][96][97][98]. Then, the phonons dissipate their energy through ultrafast processes such as electron-phonon scattering [99], phonon diphase [100], and exciton-phonon coupling [101], which determine the phonon lifetime. Ding et al. [102] found that the energy accumulated in the adhesion stage and completely dissipated within 100 ps by the excitation of phonons at the Cu interface. Chen's research group [103] studied the energy dissipation between graphene layers through the MDS. They found that the sliding velocity affected the phonon lifetime, which proved that reducing phonon lifetime or increasing phonon scattering intensity could increase friction. Sakong et al. [104] calculated that the vibrational lifetime of Ge interface under saturated hydrogen was 1.56 ns. But under saturated both hydrogen and deuterium, the tensile vibrational life of Ge-D was six times longer than that of Ge-H.

Ultrafast electron energy dissipation dynamics
In the friction process, the kinetic energy of the friction pair can excite electrons, holes, and electron-hole pairs (excitons) at the material interface (Fig. 8). In the metal or semiconductor interface, electron dissipation becomes an important form of friction energy dissipation [105]. In metal interface materials, free electrons or holes are called carriers; while in semiconductor interface materials, the excited electron and hole are bound to form an exciton due to the bandgap and Coulomb interaction. Therefore, the concept of friction electron dissipation process includes carriers and excitons. When the excited state electrons or excitons return to the ground state, energy is released through radiative or non-radiative energy dissipation pathways. These dissipation channels include charge transfer [106], Auger recombination [107], electron-electron interaction [108,109], and electron-phonon coupling [110] of carriers, as well as processes such as energy transfer [111], radiative recombination [112], exciton-exciton annihilation [113,114], exciton-phonon interaction, and diffusion [115] of excitons. These processes usually occur in the range of femtosecond to nanosecond [116].

Ultrafast energy dissipation detection
To detect the energy dissipation of phonons and electrons in the friction process, there are still three major difficulties: (1) Both the phonon and electron dissipation processes occur in an ultrafast time scale (the order of fs-ps). Hence, the experimental detection of phonon and electronic dynamics is a challenge.
(2) Solids will not be ideal crystals like the Tomlinson model. The appearance of defects and impurities often accompanies the interface. Especially in the atmospheric environment, covalent bonds between layers and chemical pinning of macromolecule adsorption easily occur in friction interfaces. (3) The actual friction process, in addition to the phonon and electron dissipation, also emits X-rays, visible light, plasma, etc. Moreover, it involves material deformation, molecular orientation and chemical reactions, etc. [117,118].
The ultrafast spectroscopy is an effective tool for studying the dynamics of carriers or excitons in materials [119][120][121][122][123]. For example, time-resolved fluorescence technology is used to study the radiative dissipation process, and pump-probe technology is used to detect the non-radiative dissipation process. Liu et al. [124] used the fluorescence lifetime imaging system to study the relationship between the exciton radiative recombination lifetime and the layer number in transition-metal dichalcogenides (TMDCs) (Figs. 9(a)-9(c)). Exciton radiative lifetime decreased | https://mc03.manuscriptcentral.com/friction with the TMDCs' layer number reducing. The fastest exciton recombination rate was observed in monolayer TMDCs, attributed to their reduced dielectric screening. Liu et al. [125] studied the charge transfer and energy resonance transfer mechanism in quantum dots/MoS 2 heterojunction (Figs. 9(d) and 9(e)), whose structure was similar to the non-contact friction in Fig. 6. When quantum dots (QDs) were directly spun onto the surface of MoS 2 , the non-radiative energy dissipation pathway in the heterojunction was mainly charge transfer. As the layer number of MoS 2 increased, the non-radiative energy dissipation rate decreased. When h-BN was inserted between QDs and MoS 2 , the non-radiative energy dissipation pathway became the energy resonance transfer. As the thickness of h-BN increased, the non-radiative energy dissipation rate decreased. These results reveal the mechanism of the dielectric property, and layer number and distance on the energy dissipation rate, and realize the controlled energy transfer rate.
As mentioned above, the ultrafast electron energy dissipation process can be divided into radiative and non-radiative processes. Time-resolved fluorescence imaging technology is usually used to monitor radiative lifetime in the range of 100 picoseconds to nanoseconds. But non-radiative energy dissipation processes often require higher time resolution. For example, the exciton formation and the defect trapping exciton processes are all in the range of 100 femtoseconds to picoseconds, which requires the help of femtosecond transient absorption imaging technology detection that based on pump-probe [126][127][128][129][130][131], as shown in Fig. 10(a). This technology offers an effective tool to study the ultrafast electron energy dissipation at the nanoscale. Especially in two-dimensional semiconductors, the friction interface is always accompanied by a series of defects [9,[132][133][134][135][136][137], which often significantly impact excitons' ultrafast energy dissipation process. Therefore, studying the influence of defects on exciton dynamics provides insight into the friction mechanism [138,139]. Liu et al. [140,141] built a femtosecond transient absorption imaging system to study the effect of defects on the process of exciton radiative recombination and annihilation in WS 2 . The defect captured neutral exciton in 7.75-17.88 ps to form a long-lived defect-bound exciton (Figs. 10(b)-10(e)). Further, the process of defects hindering exciton diffusion was directly visualized (Fig. 10(f)), which realized the imaging detection of the ultrafast energy dissipation process. It is found that

High-resolution friction energy dissipation detection system
Although the above researchers have used the ultrafast spectroscopy to study the energy dissipation process of electrons and phonons at the interface of two-dimensional materials widely used in the field of friction, they have not yet detected ultrafast energy dissipation under friction. At present, some kinds of friction energy dissipation detection instruments play an important role, which includes the AFM, q-Plus AFM, and pAFM. The AFM is widely used in tribology, which detects frictional force and frictional energy dissipation by detecting the torsional deformation of the cantilever caused by the contact sliding of the probe with the surface. Due to the contact between the tip and the surface of the sample, the tip interacts with multiple atoms on the surface of the sample, resulting in complex tip-sample interactions. Therefore, it is difficult to achieve accurate analysis. Saitoh et al. [142] and Weymouth et al. [143] used a high-quality factor quartz tuning fork (q-Plus AFM) to detect frictional force and frictional energy dissipation. By placing the quartz tuning fork perpendicular to the sample and ensuring that the tip oscillates parallel to the sample, direct detection of frictional force and frictional energy dissipation can be achieved under non-contact conditions. In addition, this device integrates the STM function to spatially achieve atomic resolution. To further improve the resolution of frictional energy dissipation, Kisiel et al. [39] and Langer et al. [144] used a very soft and sensitive cantilever placed perpendicular to the sample surface, so that the  | https://mc03.manuscriptcentral.com/friction cantilever oscillated parallel to the surface, which likes a pendulum and is named the pAFM. The pAFM can achieve a friction coefficient resolution of 10 −12 kg·s −1 and a frictional energy dissipation resolution of 65 μW. Although these detection instruments have extremely high spatial and energy resolution, they are not time-resolved and cannot effectively detect the ultrafast dynamic process of phonons and electrons in frictional energy dissipation. However, these dynamic processes are the key to a deep comprehension of friction energy dissipation. There is an urgent need for an instrument with high temporal resolution, high spatial resolution, and high frictional energy dissipation resolution at the same time.
After long-term exploration, the research team of Luo [6] has recently developed a friction energy dissipation measurement system (Fig. 11), which has made breakthroughs in basic research such as ultra-high friction coefficient measurement resolution and ultrafast energy dissipation and the measurement of friction energy dissipation in-situ and real-time. The instrument architecture consists of six chambers with an H-shaped layout. In addition, it is composed of vacuum interconnection, multi-level vibration reduction, and a thousand-level clean atmosphere to ensure environmental consistency of scientific experiments under extreme conditions such as ultra-high vacuum and low temperature. The origin of friction energy dissipation and the nature of superlubricity can be explored from the perspectives of the ultra-low friction coefficient (0.0001), the ultrafast dynamics of friction phonon, the detection of the ultra-broad spectrum of tribo-emission in the order of eV, high-speed online measurement of molecular orientation, and structural evolution at friction interface. The successful development of this instrument will provide a powerful test tool for discovering new ways to reduce friction energy www.Springer.com/journal/40544 | Friction consumption, developing new superlubricity materials, exploring new characterization methods, and revealing the mechanism of material friction damage. It will be of great meaning to the development of new materials, advanced manufacturing, smart transportation, aerospace, and other fields.

Friction energy dissipation in high-end equipment
The widespread friction and wear phenomena in mechanical systems bring huge energy dissipation and lead to component failures, restricting mechanical system service life and reliability. Here, by analyzing the tribological problems encountered by some key parts of high-end equipment in the fields of navigation, high-speed rail, and precision machining, it illustrates the importance of friction energy dissipation for high-end equipment.

Friction energy consumption of ship propeller
With the increasingly severe global warming problem, energy conservation and emission reduction have become more and more important in the shipping industry. According to the statistics [145], the energy consumption of H 2 O transportation accounts for about 12% of the global transportation energy consumption, of which about 1/3 of the energy is used to drive the propeller ( Fig. 12(a)) and overcome the resistance of the ship during the travel. In actual working conditions, the maximum efficiency of the propeller is only about 70%. Even in the most efficient working conditions, the frictional resistance loss at the blade profile is as high as 20%. Therefore, reducing the frictional resistance of the propeller will significantly reduce frictional energy consumption and improve efficiency. The frictional energy consumption of the propeller comes from two aspects: One is the solid-liquid interface friction at the propeller blade profile, and the other is the biological attachment friction on the surface of the propeller [146]. Solid-liquid interface friction is related to the interaction between the interface. Wang et al. [147] prepared a superhydrophobic surface with a rough structure and a low surface energy coating. This structure could effectively reduce the solid-liquid interface contact area and interface bonding force, which caused slippage at the solid-liquid interface and reduced the friction near the wall. Choi et al. [148] studied the effect of superhydrophobic coating on the flow field around the propeller. They found that Never Wet superhydrophobic coating on the surface of a 42 mm diameter double-blade propeller reduced the turbulent kinetic energy (TKE) of the propeller's wake by an average of about 20%, as shown in Figs. 12(b) and 12(c). Compared with the actual ship propeller, the propeller used by Choi et al. is an order of magnitude smaller in size; and the Reynolds number in the experiment is small, which does not conform to the turbulent working condition of the large Reynolds number in the actual situation. Therefore, the effect of superhydrophobic coating on propeller performance under actual operating conditions needs further study.
The propeller is in long-term contact with seawater during service, and many marine organisms will adhere to its surface. Adhering organisms not only destroy the fluid linearity of the propeller and increase frictional energy consumption, but also produce corrosive extracellular secretions, destroy the surface of the propeller, and increase maintenance costs [149]. The use of marine antifouling coatings is an effective way to inhibit the adhesion of marine organisms. According to Ref. [150], the use of Intersleek900 fluorine-containing modified polymer antifouling coating can weaken the binding force between organisms and solid surfaces, thereby increasing the difficulty of biological attachment, reducing the resistance of the propeller, and saving 6% of fuel consumption.
By weakening the interaction between the solidliquid interface to reduce the interfacial frictional resistance, and by using an antifouling coating to solve the biological attachment, the frictional energy consumption of the propeller can be significantly reduced. However, the internal mechanism of superhydrophobic surface affecting the propeller's efficiency and how to improve the effective service life of the antifouling coating are still unclear. The precision of CNC is inseparable from the linear motion guideway (Fig. 13(a)), and the precision retention is highly related to the friction and wear of the guideway [151]. Li et al. [152] established a calculation formula for the wear depth based on the Archard model, solved it with a discretization method, and proposed a finite element analysis method of the wear of guideway. Zhang and Sun [153] studied the influencing factors of wear from the perspective of randomness in Fig. 13(b). They considered the main factors and micro-random influencing factors, established a wear prediction model, and carried out research on wear reliability prediction. However, these studies can only predict the regular law of wear, and cannot reduce wear from the source and improve precision retention.

Precision machine tool guide
To achieve smooth movement without friction and vibration, the linear guideway currently in ultraprecision photoetching machines is the air-floating guideway. The air-floating guideway is based on gas-dynamic and static pressure. It has the characteristics of high motion accuracy and low friction and wear, which can maintain manufacturing accuracy for a  www.Springer.com/journal/40544 | Friction long time. But, the air-floating guideway has a small bearing capacity, low rigidity, and high manufacturing and installation accuracy requirements, making it difficult to widely application [154]. The emergence of superlubricity technology may be an ideal candidate for traditional linear motion guideways to achieve near-zero friction and wear, in which there is little energy dissipation. By constructing a superlubricity interface on the surface of the guideway, while ensuring the bearing capacity and rigidity, it will greatly reduce the friction and wear between the friction pairs of the guideway, and improve the precision retention and processing reliability.

High-power gear transmission system in highspeed railway
High-speed rail is a "golden business card" of China's equipment manufacturing. The high-power gear transmission system ( Fig. 14(a)), as the core component of high-speed train energy transmission, is not only the key to achieving higher speed, but its reliability and stability directly affect the safe operation of trains.
In the gear transmission, friction energy consumption mainly comes from two aspects: One is the contact friction of the surface roughness peak, and the other is the shear friction of the liquid lubricating film. As the gear speed increases, the vibration frequency increases. This leads to the increase of the surface impact strength, the contact frequency of the surface roughness peak, and the solid-solid contact friction. According to traditional lubrication theory, increasing lubricating oil's viscosity is necessary to increase the lubricating film's thickness, thereby reducing the rough peak contact. Still, high-viscosity lubricating oil will increase the shear friction of the lubricating film. Therefore, we hope to reduce the contact of rough peaks while using low-viscosity lubricating oil and the solid-solid contact friction in the gearbox. Solid superlubricity is an effective way to prevent solid-solid contact friction. But under the extreme working conditions of high-speed rail gearboxes, the superlubricity state is difficult to maintain stably. The solid-liquid coupling can improve the stability of the superlubricity state [6], as shown in Fig. 14(b). By introducing the additive of two-dimensional material into the lubricating fluid, it is adsorbed on the surface of the friction pair to form a low-shear interface, reducing the rough peak contact friction under extreme conditions. Meanwhile, the additives are dispersed in the lubricating liquid molecules, which can reduce the interaction between the molecules, thereby forming a shearing lubricating oil film to improve the stability of the superlubricity state. The application of superlubricity in high-speed rail gearboxes will greatly reduce the frictional energy consumption of gear transmission and ensure the safe operation of trains.

Economic analysis of frictional energy dissipation-A case study of high-speed railway
As mentioned above, shipping, railway, machinery manufacturing, and other industries are the important parts of a country's industry and play essential roles in the national economy, environmental protection, and sustainable development. However, the problem of frictional energy dissipation is becoming more and more prominent in these industries. If frictional energy dissipation can be effectively reduced, it will generate huge economic benefits and promote the green development of human society. This section will explore the impact of tribological applications in the machinery industry and predict potential economic savings accordingly.
To better approach the sense of tribology applications in the machinery industry, we must have an idea first of what our Mother Earth is contemporarily facing. The environmental crisis that the world is currently dealing with has been more serious than ever. On the Climate Ambition Summit 2020, the UN secretarygeneral Antonio Guterres urged all countries to declare climate emergencies. He said, "can anybody still deny that we are facing a dramatic emergency? I urge all others to follow." [155] Hence, the term carbonneutral has been brought up as a priority. Multiple countries have updated their longer-term objectives towards zero emissions. China's action on achieving carbon-neutral objectives is now known as the Fourteenth Five-Year Plan from 2021 to 2025. It specifically and precisely set two goals: decreasing energy consumption per unit of GDP by 13.5% and reducing carbon dioxide emissions by 18% [156]. The machinery industry, in this case, holds indisputable responsibility for achieving these carbon-neutral objectives [157], and so do the tribology applications to the industry. Therefore, the reduction of friction energy dissipation in the machinery industry is necessary for completing the objective of international carbon neutralization.
Next, we take the high-speed rail industry as an example to deeply explore the economic value brought by the application of tribology from the perspective of economics.

Methodology of economic savings forecasting for China Railway High-speed (CRH)
CRH is one of China's most influential transportation infrastructure and machinery industry sectors. As of the end of 2019, the total mileage of domestic highspeed rail reached 35,000 km, which was ranked in the first place worldwide [158]. In addition to the convenience of travelling for citizens, the achievement of carbon neutralization objectives must be supported by the rapid development of high-speed railways. Compared to the traditional internal combustion engines, the high-speed railway is driven solely by electricity, making it much more environmentally friendly. In the following, we will focus on the electrical energy consumed to overcome friction dissipation and potential economic savings by applying tribology theories and technologies. Some researchers have established models to analyze the energy consumption in cars and trucks [3,[159][160][161]. According to these methodologies of economic forecasting, Fig. 15 shows the flowchart for CRH. There are two general expected outputs from this evaluation, namely: 1) Estimating the energy dissipation of CRH caused by friction; calculation output with an energy unit of TJ.
2) Estimating the potential economic saving by tribology applications; calculation output with a currency unit of U.S. dollar $.
For the evaluation of energy dissipation caused by friction, the process will contain the following steps: 1) Due to the differences in the applied technology, quality of products, working conditions, etc., an "average" equipment of CRH must be defined. Specifically, the equipment is under theoretical conditions and setups based on national data to gather representative conditions. 2) The overall energy consumption process and status are analyzed based on the "average" equipment conditions according to investigations and research. Then determine the proportion of friction loss, hence calculating the total friction-related energy dissipation.
3) Evaluating the energy dissipation savings by tribology applications. The time span of saving includes the highest level of technologies that can be applied on a large scale contemporarily; the potential level of technologies that may be realized in 5 years.
With the previous steps completed, the methodology should develop from the output of the first objective of forecasting to the second, which is the potential economic savings. The forecast method for the second objective is shown as follows: 1) Contemporary economic savings: converting the energy dissipation saved by tribological applications into the saved amount of electricity used by the equipment with the most leading lab technology. Then calculate it using the national average price of industrial electricity.
2) 5-year economic savings: converting the energy dissipation saved by tribological applications into the saved amount of electricity used by the equipment. Then calculate it using the estimated future national average price of industrial electricity.

Forecasting potential economic savings on CRH by tribological applications
The first step is to build up average equipment, in this case, a certain type of CRH represented in the entire field. When defining "average", our priority criterion is to pick an existing type of train instead of building up an ideal but not real equipment. Because the purpose of this economic forecast is to give people with interests to understand the potential realistic effect of applying high-end technology of tribology, we decided that the most representative type of CRH should be the one that is mostly employed and provides the services. According to The World Encyclopedia of High-speed Trains [162], all types of CRH are classified by their speed levels, including 300-350, 200-250, and 160 km/h. A part of the data extracted from the encyclopedia is shown by the following Table 1. Even though the count for CRH2A is the highest, we decided to use CRH380B as the average equipment for two reasons. First, the total count for 300-350 km/h speed level CRH is 1,809, holding 59.18% of the total CRH in service. On the other hand, the future development of China's high-speed railways will replace the slower types with higher speed and hence develop better efficiencies. Only by sticking with that direction, we may achieve the carbon-neutral objectives by 2025. The following presents the tech parameters of CRH380B in Table 2. The next step is to build up a starting point of its energy consumption rate, hence estimating the annual energy consumption of the CRH system. By conducting research and summarizing multiple reports [164-167], the energy consumption efficiency of CRH380B is estimated to be 3.8 kWh per 100 passenger-km. With energy consumption efficiency, the annual passenger volume with total distance travelled is also required. According to the reports [168, 169] and statistics [170], the national annual passenger transport quantity and passenger turnover quantity are concluded in Table 3. Based on the two parameters, the average transportation distance can then be calculated.
Since the total energy consumption is obtained, the proportion of energy used to overcome friction can be estimated. For the equipment like CRH train, the assumption is that the major running resistance, where most energies are used to overcome aerodynamic and frictional resistances. For CRH, when the train runs at 200 km/h, the aerodynamic resistance accounts for about 70% of the total resistance. When it runs at 486.1 km/h, the aerodynamic resistance exceeds 92% of the total resistance. If it runs at more than 500 km, the proportion becomes more than 95% of the total [171]. Moreover, the proportion of total energy consumption used to overcome aerodynamic resistance is estimated to be 80%, hence leaving 20% to frictional resistance [172]. Therefore, the energy converted to overcome friction is determined [173]: The last step is to convert the energy savings into economic ones by adapting the current China's industrial electricity price of 0.1549 $/kWh, which is also considered stable for the following five years. Noticeably, the potential reduction in the energy used to overcome friction in advanced tribological applications today and future five years are 25% and 55%, respectively [161,174] Saving proportion compared to CRH380B 2020: 55% Saving in electricity energy: 7,300.26 (10 3 TJ) Saving in economic value: 314.11 (10 6 $) The proportion of saving above is relatively conservative already. Optimistically, superlubricity applications will reduce the friction resistances ultimately close to zero. Figure 16 represents the ideal case of long-term saving that may be reached by the development of tribology considering approximate inflations of 7% annually. The immense amount of economic savings with the potential tribological application, dedicated research, and innovations are required for its realization. The world achieving carbon-neutral objectives is not just saving the contemporary human being, and it is our legacy for the following generations, our commitment to the future.

Conclusions
In summary, we have reviewed the classical microscopic mechanism of friction energy dissipation, including the phonon dissipation model (PT and FKT) www.Springer.com/journal/40544 | Friction and the electron dissipation model (Coulomb drag). Research about these atomic models mainly focuses on simulation calculations, and some experiments have proved the significance of phonon and electron dissipation in friction. However, due to the complex interaction of the friction interface, these experiments usually study the friction energy dissipation through changes in macroscopic properties. The emergence of non-contact friction avoids various influencing factors compared with direct contact, in which vacuum and low temperature provide a pure experimental environment for studying the electron and phonon dissipation. Furthermore, since the electron and phonon dissipation usually occur in an ultrafast time range (femtoseconds to nanoseconds) at the nanoscale, this has reached the detection limit of existing scientific instruments in the field of tribology. Therefore, the study of friction energy dissipation should be transformed from mechanics to the perspective of ultrafast phonon and electron energy dissipation. The ultrafast spectroscopy is an effective tool to detect ultrafast energy dynamics. If the ultrafast spectroscopy and friction detection platform are combined, it is expected to realize in-situ real-time detection of ultrafast friction phonon and electron energy dissipation, which provides a powerful means to explore the nature of superlubricity and the origin of friction.
One of our research purposes on friction energy dissipation is to clarify the energy dissipation channels of friction interfaces. By regulating the energy dissipation channels, friction can be reduced or inhibited from the source to realize the active design of superlubricity interfaces. Superlubricity is a disruptive technology with huge application prospects in high-end equipment fields such as precision manufacturing, aviation, aerospace, and new energy vehicles. It provides an important way to solve common problems such as high energy consumption and poor stability of the equipment. Here, by analyzing the tribology problems encountered by some key parts of high-end equipment in the fields of navigation, high-speed rail, and precision machining, it illustrates the importance of friction energy dissipation for high-end equipment. Finally, we analyze the impact of friction consumption reduction on the high-speed rail from an economic point of view. We predict that huge energy losses will be saved if superlubricity is applied to the highspeed railway. Therefore, the study of friction energy dissipation is of great significance to high-end equipment, green energy, and economic development, and is a strong support for the world to achieve carbon-neutral.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.