Contact‐Electrification‐Regulated Amino Acid for Improving Materials' Triboelectric Properties

ABSTRACT

The dual sustainability of energy and materials, especially through the use of green molecular building blocks, is of critical importance for next‐generation self‐powered systems. Here, we propose a contact‐electrification (CE)‐regulated molecular‐engineering strategy to actively manipulate interfacial charge configurations in amino acid‐based triboelectric films. Amino acid molecules intrinsically contain both electron‐donating amino groups and electron‐accepting carboxyl groups; however, a dominant exposure of amino groups is more favorable for positive triboelectric properties. By exploiting CE‐driven interfacial electrostatic fields, preferential outward enrichment of amino groups and simultaneous interfacial anchoring of carboxyl groups and side chains to the underlying metal electrode are achieved, thereby maximizing interfacial charge transfer efficiency with a 1.51‐fold enhancement. These findings demonstrate that molecular‐level interfacial regulation, rather than material substitution alone, represents a fundamentally new paradigm for advancing triboelectric energy conversion. Notably, this CE‐regulated molecular orientation mechanism originates from the amino and carboxyl groups shared by amino acids. This finding provides a molecular perspective for understanding and regulating interfacial charge transfer in a conserved donor–acceptor framework.

A contact‐electrification (CE)‐regulated molecular engineering strategy is developed to manipulate charge configurations in amino acid‐based films without external bias. By inducing orientation of functional groups via a self‐generated electrostatic field, enhanced and stabilized charge transfer is achieved. Films exhibit improved electrical output, providing a general inspiration for high‐performance triboelectric materials.

1. Introduction

At the deepest level of matter, energy is “written” in the quiet migration of charges across interfaces. In triboelectric systems, this interfacial charge migration is governed not by mechanical contact itself, but by the intrinsic electron‐donating and electron‐withdrawing characteristics of surface functional groups [1, 2, 3]. Typical tribo‐negative materials, such as fluoropolymers, are dominated by strongly electron‐withdrawing F‐containing groups [4, 5], whereas tribo‐positive materials generally rely on electron‐donating functionalities, including amino [6], hydroxyl [7], and alkyl [8] groups. Therefore, the nature, density, and exposure orientation of surface functional groups play a decisive role in determining charge transfer efficiency in triboelectrification. Meanwhile, the urgent demand for dual sustainability in both energy and materials has driven intensive exploration of environmentally benign triboelectric materials [9, 10, 11, 12]. Unlike synthetic fluoropolymers, sustainable materials widely derived from nature intrinsically contain abundant polar functional groups, offering rich but underutilized electron‐donating and accepting sites [13, 14, 15]. However, how to precisely activate or regulate these molecular‐level charge‐determining groups for high‐performance triboelectric energy conversion remains a fundamental challenge.

Among the diverse sustainable molecular candidates, systems that inherently integrate both electron‐donating and electron‐accepting functionalities within a single molecular framework are particularly attractive for precise interfacial charge regulation [16, 17]. Amino acids represent a unique class of sustainable molecular building blocks for triboelectric applications because of their intrinsic molecular bipolarity [18, 19]. Each amino acid simultaneously contains electron‐donating amino groups and electron‐accepting carboxyl groups. Together with diverse side chains that provide abundant intermolecular interactions, anchoring sites, and dipole modulation capabilities [20]. This combination offers a rare opportunity to regulate interfacial electron transfer at the molecular scale using a single‐component system. In recent years, amino acid‐based and biomolecular triboelectric materials have been increasingly explored as environmentally benign alternatives to conventional polymers [21, 22]. However, in most reported systems, amino acid molecules are randomly distributed and isotropically oriented within the triboelectric layer. As a result, their intrinsic donor–acceptor asymmetry is only passively involved in triboelectrification, severely limiting the effective utilization of electron‐donating amino groups. Therefore, an external physical driving force capable of breaking this orientational randomness and selectively biasing molecular dipoles is highly desired.

Notably, contact‐electrification (CE) inherently generates intense transient interfacial electrostatic fields during repeated contact‐separation processes [23, 24, 25]. These localized fields serve as a universal physical origin for a variety of CE‐enabled enhancement strategies. As demonstrated by only a few highly impactful studies so far, including surface charge amplification in solid‐solid sliding systems [26] and polarization‐driven output boosting in liquid‐based TENGs [27]. Such transient fields can couple strongly to polar species at the interface, inducing local polarization and electron distribution; they also redistribute mobile charges and alter the interfacial potential landscape, thereby changing how polar moieties and ions interact with the surface. However, in these existing systems, the triboelectric field primarily serves to amplify or redistribute interfacial charges. Whether it can directly regulate molecular configuration at the interface remains largely unexplored. In particular, whether CE can be exploited as an active tool to break the intrinsic orientational randomness of molecular building blocks and selectively expose electron‐donating moieties at triboelectric interfaces is still an open question. Addressing this question would open a fundamentally new pathway for triboelectric regulation beyond conventional material substitution or structural engineering (Table S1).

Here, we propose a CE‐regulated molecular‐engineering strategy to activate and direct amino acid molecular orientation in triboelectric films. Using cysteine‐doped PEO films as a model system, CE‐driven interfacial electrostatic fields induce preferential outward enrichment of electron‐donating amino groups. Simultaneously, electron‐accepting carboxyl groups and side chains are anchored at the metal electrode interface. This molecular rearrangement maximizes interfacial charge transfer efficiency, resulting in a 1.51‐fold enhancement in transferred charges and more than a two‐fold increase in power density. Based on this mechanism, an optimized CE‐regulated amino acid‐TENG was developed and integrated into a smart insole system, achieving an average power density of 2.97 mW through steps during walking and successfully powering five serially connected temperature–humidity sensors. This CE‐regulated behavior is fundamentally dictated by the inherent coexistence of amino and carboxyl groups in amino acids, suggesting a potential mechanistic linkage between molecular structure and interfacial charge‐transfer behavior across amino acid systems.

2. Result

2.1. Fundamental Processes of CE and an Amino Acid Enhancement Strategy

Breaking the orientational randomness of biomolecular building blocks at triboelectric interfaces remains a long‐standing bottleneck for fully exploiting their intrinsic charge‐donating capability. Here, taking cysteine as a representative model, we demonstrate that the interfacial alignment of amino acid molecules can be actively regulated through charge engineering‐assisted molecular assembly. First, to establish the conceptual foundation of the CE‐regulated molecular‐engineering strategy, the physical principles governing CE are revisited, and the molecular constraints that limit its tunability are identified. As illustrated in Figure 1a, a typical contact‐separation device consists of two dielectric materials laminated with conductive electrodes on their back surfaces. Upon periodic contact and separation, charge transfer occurs between the two electrodes, driven by CE and the subsequent electrostatic induction, as schematically shown in Figure 1b. Electrostatic induction can modulate interfacial potential landscapes and redistribute internal charges within the dielectric‐electrode pair; however, such induction processes are inherently passive and lack effective routes for deliberate tuning. Current research, therefore, focuses mainly on enhancing the charge quantity generated per contact by tailoring intrinsic material properties, such as dielectric constant, surface charge density, and the type and density of surface functional groups (Figure 1c). As shown in Figure 1d, the fundamental mechanism of CE originates from the dynamic overlap of the electron clouds of atom A and atom B during approach and separation, which triggers interfacial electron redistribution and partial charge displacement across the two surfaces. Consequently, the electron‐exchange characteristics between two materials become the key determinant of the charge transferred in a single contact event. These material‐dependent characteristics lead to an empirical classification of surfaces into relatively “positive” and “negative” materials, as depicted in Figure 1e. A larger disparity in their intrinsic CE tendencies generally corresponds to a larger quantity of transferred charges. Such disparities originate from the molecular‐level structure of each material, including the chemical identity of its surface functional groups and their local electronic environments (Figure 1f). With sustainability becoming an increasingly important design principle, naturally abundant materials have gained attention as environmentally benign alternatives to synthetic polymers. Their rich chemical functionality and bio‐derived molecular structures make them attractive candidates for next‐generation CE applications. Among these natural building blocks, amino acids, which contain one amino group, one carboxyl group, and a unique side chain, offer a particularly intriguing framework for interfacial modulation (Figure 1g). The coexistence of amino and carboxyl groups within a single molecular scaffold inherently gives rise to a spatially asymmetric electronic structure. To characterize this feature, the molecular structure of a representative amino acid, cysteine, was optimized using molecular mechanics followed by density functional theory (DFT) refinement (optimized coordinates provided in Table S2). Based on the optimized geometry, the electrostatic potential (ESP) distribution was calculated (Figure S1). The results show that the carboxyl group corresponds to a region of low ESP, while the amino group is associated with a relatively high ESP. This spatial separation of ESP reveals an intrinsic asymmetric electrostatic distribution along the amino acid molecular backbone. Rather than being governed by a single functional group, the electrostatic landscape arises from the collective distribution of molecular functionalities. Such a distribution provides a natural directional axis, enabling the molecule to respond coherently to external or interfacial electric fields. This feature suggests a promising route for achieving controlled molecular alignment under CE‐induced electrostatic environments. DFT calculations further reveal that amino acids possess smaller HOMO–LUMO gaps compared to common polymeric dielectrics (Figure 1h), implying higher frontier‐orbital accessibility and potentially more active electron‐exchange behavior during contact [12, 19, 28, 29, 30, 31, 32]. Experimentally and theoretically, amino acids are often found to exhibit stronger positive tendencies than conventional materials. However, the presence of the electron‐accepting carboxyl group can partially counteract the electron‐donating character of the amino group, limiting its full expression during interfacial electron transfer. Therefore, a strategy that selectively suppresses the expression of carboxyl groups while amplifying the contribution of amino groups is highly desirable (Figure 1i). Such an approach would enable directional control over functional‐group exposure, providing a rational pathway toward programmable interfacial electron‐exchange behavior.

FIGURE 1.

Fundamental mechanism of CE and amino acid‐enabled molecular regulation strategy. (a) Schematic illustration of a typical contact‐separation mode triboelectric device. (b) Origin of the triboelectric output arising from the combined effects of CE and electrostatic induction. (c) Material‐property‐driven enhancement of triboelectric charge transfer through various factors. (d) Fundamental origin of CE at the atomic level. (e) Classification of triboelectric materials into tribo‐positive and tribo‐negative categories based on their intrinsic electron‐donating and electron‐withdrawing tendencies. (f) Molecular nature of triboelectric polarity with the types of surface functional groups. (g) Molecular configuration of amino acids. (h) Density functional theory calculated the electronic structure of representative amino acids compared with conventional polymers [12, 19, 28, 29, 30, 31, 32]. (i) Conceptual illustration of the CE‐regulated strategy to maximize the effective tribo‐positivity.

2.2. Chemical and Structural Characteristics of CE‐Enhanced Amino Acid Films

To elucidate how CE can be exploited to steer the interfacial assembly of amino acid‐containing films, a classical CE phenomenon is first considered. When fur and a rubber rod undergo repeated contact and separation (Figure S2), part of the electrons migrate from the fur to the rubber surface. As a result, the fur becomes positively charged, whereas the rubber acquires an equal amount of negative charge. This basic charge‐transfer principle provides a physicochemical basis for tailoring surface potential on metal foils. Building on this principle, CE is applied between fluorinated ethylene propylene (FEP) and Al foil to create a positively biased Al surface (Figure 2a). Prior to establishing the interfacial electrification mechanism, the preparatory workflow is illustrated in Figure S3, including amino acid solution formulation, sliding CE treatment, and subsequent spin‐coating. This process establishes the experimental basis for the ensuing molecular alignment strategy. During FEP‐Al sliding friction, electrons preferentially accumulate on the highly electronegative FEP, while Al correspondingly develops a net deficit of electrons. Finite‐element simulations are further performed to evaluate the electrostatic environment established on the CE‐treated Al substrate. As shown in Figure S4a, CE induces a distinct surface potential on the Al foil. After introducing an H₂O layer above the substrate, the ESP is found to be maximized at the Al/H₂O interface and gradually decays into the liquid phase (Figure S4b), indicating that the CE‐induced potential can propagate across the solid–liquid boundary and establish an interfacial field gradient. Although dielectric screening attenuates the potential in the aqueous phase, the strongest electrostatic field remains localized near the interface, where molecular adsorption and initial organization occur. Given that the precursor is deposited from aqueous solution, this interfacial electrostatic environment provides a direct driving force for the preferential arrangement of polar functional groups during film formation. The positively enriched Al surface, therefore, becomes capable of electrostatically attracting amino acid molecules and imposing preferential adsorption. As a result, electron‐donating amino groups tend to orient toward the film–air interface, whereas electron‐accepting groups, such as carboxyl groups, preferentially anchor to the Al substrate. This establishes a mechanistically supported pathway linking macroscopic CE treatment with molecular‐level ordering, ensuring that the strategy is well grounded in established interfacial electrostatics. A representative photograph of the spin‐coated film on the CE‐treated Al substrate is shown in Figure 2b, illustrating the uniform macroscopic coverage obtained after the coating process. To further clarify the proposed mechanism, a schematic illustration is presented in Figure 2c, which conceptually delineates the evolution from an initially random molecular distribution to a CE‐assisted interfacial preorganization. As shown, the process involves: (i) a disordered distribution of amino acid molecules within the polymer matrix; (ii) the establishment of a local electrostatic environment during CE‐assisted interfacial formation; (iii) the emergence of a biased molecular orientation, leading to preferential exposure of specific polar groups at the interface; and (iv) the formation of a stabilized interfacial configuration that governs subsequent (v) dynamic triboelectric charge transfer. Once formed, this preorganized interface is considered structurally stable during cyclic operation. Accordingly, the dynamic behavior observed during contact–separation is primarily associated with charge redistribution at the interface, rather than continuous molecular reorientation. The evolution of Al's surface potential and working function during incremental CE cycles is shown in Figure 2d,e. Untreated Al exhibits an initial potential of 0.89 V (CE #0). After five CE cycles, the potential rises to 1.79 V, an increase by a factor of 2.01. A higher surface potential indicates enhanced positive charge enrichment and strengthened interfacial electric fields, which are conducive to directing molecular adsorption. Correspondingly, Kelvin‐probe measurement calibrated with a gold reference, shows that the surface work function of Al decreases from 4.31 to 3.41 eV, reflecting a facilitated electron‐emission tendency and consistent with the introduction of positive surface charges. Water contact angle (WCA) measurements (Figure 2f) provide further evidence for charge accumulation. In CE‐modified surfaces, higher surface charge density enhances electrostatic interactions with polar water droplets, resulting in reduced WCA. Among various material combinations tested (Kapton‐Al, PAN‐Al, paper‐Al, and FEP‐Al), FEP‐Al CE produces the lowest WCA on both FEP (53°) and Al (39°), confirming the strongest charge buildup in this pairing. Corresponding optical images of the WCA profiles are provided in Figure S5, offering direct visual validation of the wettability contrast induced by different CE treatments. The temporal evolution of the surface potential of the CE‐treated Al substrate was characterized under ambient conditions (Figure 2g). The surface potential decreases from 1.79 to 1.60 V within 1 h, retaining approximately 89.4% of its initial value, and further decreases gradually to 1.49 V after 3 h. The decay behavior shows a relatively faster initial decrease followed by a slower long‐term relaxation. This trend indicates a transition from the dissipation of loosely bound surface charges to a more stable, trap‐dominated regime. The relatively slow decay demonstrates that the CE‐induced surface potential remains sufficiently stable within the timescale of film fabrication. Considering that the spin‐coating process is completed within seconds to minutes after CE treatment, the interfacial electric field can be regarded as effectively constant during the critical stage of molecular organization. After film formation, the molecular configuration becomes fixed, while the surface charges gradually relax over time. Following CE conditioning, PEO is selected as the polymer matrix due to its favorable triboelectric performance among biodegradable polymers, good compatibility with amino acids, and its relevance to sustainable material design. A PEO/cysteine solution is deposited onto the activated Al via spin coating. As summarized in Table S3, spin coating is selected for its control over film uniformity and minimal introduction of additional directional effects, enabling clearer identification of the CE‐induced interfacial regulation. At this stage, the pre‐charged Al surface establishes an interfacial electrostatic field that can induce directional polarization at the solid–liquid interface, conceptually analogous to the triboiontronic electrical double layer (EDL) regulation mechanism (Figure S6). This field‐driven interfacial polarization promotes preferential organization of polar molecular groups during the initial stage of film formation. SEM images and sulfur mapping (Figure 2h‐i, ii) reveal a smooth surface with homogeneously distributed elements. Additionally, Figure S7a further confirms the excellent surface flatness of the deposited film. Cross‐sectional SEM and corresponding elemental mapping (Figure 2h‐iii, iv) show a uniform film thickness with no observable interfacial discontinuities, demonstrating the film's structural integrity and consistent deposition quality. As shown in Figure S7b, no interlayer bubbles are present, and the film exhibits a uniform color distribution with an overall thickness of approximately 1 µm. Optical profilometry of slow and fast spin‐coating conditions (Figure 2h‐v, vi) confirms that the film thickness remains stable across coating speeds, indicating robust wettability and spreading behavior on CE‐treated Al. To further evaluate the reliability of the observed elemental redistribution and exclude possible measurement uncertainty, we first examined the sample‐to‐sample reproducibility within a single fabrication batch (within‐batch, WB). Three independently prepared films were fabricated and characterized under identical conditions (25°C and 40% RH). As shown in Figure 2i, the untreated films exhibit averaged compositions of C = 59.79 ± 0.43 at%, N = 8.93 ± 0.16 at%, O = 22.68 ± 0.13 at%, and S = 8.47 ± 0.12 at%, while the CE‐treated films show C = 53.69 ± 0.46 at%, N = 13.60 ± 0.16 at%, O = 27.97 ± 0.09 at%, and S = 5.30 ± 0.34 at%. The small standard deviations (typically within ∼0.1–0.5 at%) indicate good statistical reproducibility, excluding random measurement fluctuations and local surface heterogeneity. For completeness, we further performed day‐to‐day (DD) and batch‐to‐batch (BB) measurements, as summarized in Figure S8a, b. The DD measurements, obtained by repeating the characterization on the following day under identical environmental conditions, show negligible variation, confirming temporal stability. In addition, independently prepared films from another batch exhibit consistent elemental compositions and the same compositional trend, demonstrating that the observed redistribution is not dependent on a specific batch.

FIGURE 2.

CE‐induced interfacial potential modulation and CE‐assisted amino acid film formation on Al foil. (a) Schematic of CE‐induced positively charged Al surface enabling directional adsorption and dipole‐biased alignment of amino acid molecules. (b) A representative photograph of the amino acid‐based spin‐coated film on the CE‐treated metal substrate. (c) Schematic illustration of CE‐assisted interfacial preorganization. (d) Surface potential evolution of Al foil after different CE‐treated cycles. (e) Corresponding reduction of the Al work function from 4.31 to 3.41 eV after CE treatment. (f) Water contact angle characterization of Al and FEP after mutual CE with different polymers. (g) Time‐dependent decay of surface potential of CE‐treated Al under ambient conditions. (h) (i) Planar SEM image and (ii) mapping of planar SEM image; (iii) cross‐sectional SEM images and (iv) mapping of cross‐sectional SEM image; Optical profilometry results under (v)slow and (vi)fast spin‐coating conditions. (i) EDS elemental analysis.

The structural and interfacial characteristics of the CE‐enhanced amino acid film (ACE‐film) were examined alongside the amino acid‐modified film (A‐film) and pristine PEO film. As demonstrated in Figure 3a, the FTIR spectra show a shared vibrational band near 2868 cm⁻¹ across all samples, while both ACE‐film and A‐film exhibit an additional absorption at 2539 cm⁻¹, indicating the preserved presence of cysteine groups and their effective incorporation into the polymer matrix. The XRD patterns in Figure 3b confirm that the crystalline phases of PEO and cysteine remain intact after blending. Notably, the ACE‐film exhibits stronger diffraction peaks at 8.54°, 25.1°, and 33.4°, a direct consequence of increased crystalline domain alignment and higher lattice coherence, which enhance the scattering intensity. The survey XPS spectra further reveal characteristic S‐ and N‐containing signals in amino acid‐based films, supporting the successful introduction of cysteine‐derived functionalities without altering the elemental composition (Figure 3c). In Figure 3d, comparison of the normalized C 1s‐referenced spectra highlights a ∼15% reduction in S 2p intensity and an ∼8% increase in N 1s intensity in the ACE‐film relative to the A‐film. The observed shift reflects a depth‐dependent redistribution of functional groups; thiol‐containing species reside closer to the substrate while amino‐containing species dominate the outer detection depth. To resolve the depth‐dependent distribution of chemical states, angle‐resolved XPS was employed, where larger take‐off angles provide more surface‐sensitive information, enabling differentiation of near‐surface versus subsurface chemical environments (Figure 3e). As shown in Figure 3f, analysis of the O 1s spectra across 0°, 30°, and 60° reveals distinct interfacial behaviors. For the PEO film, the O 1s component at 531.8 eV remains constant across angles, consistent with a uniform oxygen environment. In the A‐film, two oxygen components appear at 531.8 eV (C─O─C in PEO) and 530.2 eV (C═O from the cysteine carboxylate), with the lower‐binding‐energy component dominating at shallow angles but gradually approaching the intensity of the higher‐binding‐energy peak as the probing depth decreases. In the ACE‐film, this trend becomes more pronounced: the two O 1s components exhibit comparable intensities at 30°, and 60° the 531.8 eV contribution surpasses the 530.2 eV signal, indicating that carbonyl groups are preferentially buried while ether oxygens are enriched toward the surface after CE‐induced reorientation. Depth profiling via ToF‐SIMS further differentiates the molecular organization within the A‐film and ACE‐film. In the untreated A‐film as illustrated by Figure 3g, both ─NH₂ and ─SH fragments appear at similar surface intensities and gradually decay with sputtering, while ─OH shows surface enrichment and diminishes near the substrate. By contrast, ─COOH signals are relatively weak at the surface but strengthen closer to the Al substrate, consistent with partial interfacial coordination. In CE‐treated ACE‐film results in Figure 3h, the surface abundance of ─NH₂ notably exceeds that of ─SH, reflecting a clear outward orientation of amino groups. Near the film‐substrate interface, depleted ─SH and ─COOH signals indicate their involvement in interfacial binding. The depth‐resolved spectra support the formation of Al─S⁻ and Al─COO⁻ anchoring bonds at the bottom interface, consistent with the chemical states identified in high‐resolution spectra (Figure 3i). Compared with the depth profiling results of the A‐film shown in Figure S9, ─NH₂ fragments exhibit a relatively uniform distribution across the middle and bottom regions of the film, with their surface intensity lower than that in deeper layers. In contrast, ─SH groups show clear surface enrichment relative to the middle and bottom, while ─COOH signals remain largely uniform throughout the film thickness. The ─OH groups display weaker surface intensity compared to subsurface regions. Notably, neither ─SH nor ─COOH demonstrates significant anchoring interactions with Al, and both Al─S⁻ and Al─COO⁻ signals appear only at low intensity across the entire film, indicating the absence of strong interfacial coordination in the untreated A‐film. The surface potential and work function of each sample are measured (Figure 3j). The ACE‐film exhibits the highest surface potential, followed by the A‐film and PEO‐film. As demonstrated in Figure 3k, the corresponding work functions are 4.06, 3.80, and 3.59 eV for the PEO film, A‐film, and ACE‐film, respectively, with small standard deviations, indicating good reproducibility. The increase in surface potential is accompanied by a monotonic decrease in work function, indicating an enhanced electron‐donating tendency of the film surface. This evolution reflects the modulation of the surface electrostatic/electronic state induced by amino acid incorporation and CE‐assisted molecular organization.

FIGURE 3.

Structural and interfacial chemical characterization of PEO film, A‐film, and ACE‐film. (a) FTIR spectra (b), XRD patterns (c), and XPS survey spectra of PEO film, A‐film, and ACE‐film. (d) S 2p XPS intensity and N 1s intensity comparison of A‐film and ACE‐film. (e) Schematic illustration of angle‐resolved XPS probing depths at different take‐off angles. (f) Angle‐dependent O 1s spectra demonstrating CE‐induced redistribution. (g) ToF‐SIMS depth profiles of functional fragments in the untreated A‐film. (h) ToF‐SIMS depth profiles of functional fragments in the CE‐treated ACE‐film. (i) Chemical bond depth distribution in ACE‐film. (j) The surface potential of (i) P‐film, (ii) A‐film, and (iii) ACE‐film. (k) Related working function of P, A, ACE‐film.

2.3. Electrical Output Characteristics of ACE‐Based TENG

The electrical output characteristics of the prepared films were evaluated using a linear motor‐driven contact‐separation setup. In this configuration, P‐film, A‐film, and ACE‐film served as the positive dielectric layers, paired with FEP as the negative dielectric layer, resulting in P‐TENG, A‐TENG, and ACE‐TENG devices, respectively. The measurement platform enables a controlled, periodic approach‐separation motion to quantify charge‐transfer capability under identical mechanical conditions (Figure 4a). A schematic of the device structure is shown in Figure 4b. Notably, the interfacial molecular configuration in these films is pre‐established prior to device operation and remains structurally stable during the measurement process. When the two dielectric layers are brought into contact, triboelectric charge transfer generates opposite charges on their surfaces. Upon separation, the electrostatic field established between these oppositely charged interfaces drives electron flow between electrode 1 and electrode 2 until electrostatic equilibrium is reached. The alternating contact–separation sequence, therefore, induces periodic potential modulation and results in an alternating voltage output. Figure 4c summarizes the optimization pathway for enhancing output performance. In the absence of amino acid doping, the pristine P‐TENG (5 wt%) yields an output voltage of 62.88 V. Increasing the doping ratio from 5 to 20 wt% progressively improves the output, reaching 145.94 V, consistent with an increased density of polar functional groups contributing to charge transfer at the interface. Further enhancement is achieved by pre‐conditioning the aluminum substrate through CE. As the CE cycling number increases from #1 to #5, the output voltage rises from 200.59 to 289.93 V. Since the effect of CE‐induced molecular arrangement may be influenced by film thickness, the coating speed was adjusted to tune the thickness of the amino acid‐containing layer. Increasing the spin‐coating speed from 1000 to 2600 rpm results in an approximate 1.09‐fold increase in output voltage, suggesting that thinner films facilitate a higher effective surface density of active functional groups contributing to charge transfer. Furthermore, structural reproducibility should translate into consistent device performance. To verify this, electrical output measurements are conducted on all independently prepared samples under identical testing conditions (Figure S10). Notably, all output signals exhibit highly consistent voltage amplitudes, with negligible variation among different samples, measurement days, and fabrication batches. The nearly overlapping output curves indicate that the electrical performance is highly reproducible and not dominated by experimental uncertainty. This strong consistency demonstrates that the CE‐induced interfacial molecular orientation, once established during film formation, can stably govern the charge‐transfer behavior during contact–separation operation. To assess the influence of external actuation on device performance, the voltage and transferred charge of P‐TENG and ACE‐TENG are compared under identical operating conditions (Figure 4d). The output voltage increases from 62.88 V (P‐TENG) to 317.03 V (ACE‐TENG), and the transferred charge rises from 58.52 to 183.98 nC, corresponding to enhancements of 5.04‐fold and 3.14‐fold, respectively. Figure S11 further compares the voltage, current, and transferred charge outputs of P‐TENG, A‐TENG, and ACE‐TENG, revealing the progressive enhancement introduced by amino acid incorporation and CE‐treated molecular configuration. In addition, the dependence of ACE‐TENG on applied pressure (0–20 N) and operating frequency (0–2 Hz) is examined in Figures. S12 and S13, showing a monotonic rise in output with increasing mechanical stimulus, indicative of stable mechanical‐to‐electrical coupling across diverse loading conditions. In Figure 4e, based on the relation P = U²/(R A), where A denotes the active contact area, the power density peaks at 1.72, 4.81, and 10.36 W m⁻² for P‐TENG, A‐TENG, and ACE‐TENG at a load resistance of 10 MΩ. The load‐dependent characteristics of the ACE‐TENG are further presented in Figure S14. Voltage output and corresponding peak power density are summarized across a broad range of external resistances, providing a comprehensive assessment of the impedance–power matching behavior. It should be noted that the peak power density represents the maximum instantaneous output under controlled conditions and does not directly correspond to the average usable power available for practical electronics after power management. Therefore, this metric is primarily used here to evaluate the intrinsic charge‐transfer capability of the triboelectric interface under identical testing conditions. A broader benchmarking analysis places the ACE‐TENG in context with previously reported amino acid and protein‐based TENG devices (Figure 4f), demonstrating an improvement of approximately two orders of magnitude relative to several representative energy harvesting devices. The resulting comparison highlights the substantial improvement in output peak power density achieved through the CE‐enhancement. In addition, Table S4 provides a systematic comparison of processing strategies in amino acid and protein‐based TENG systems, further highlighting that the present CE‐induced approach represents a distinct self‐generated interfacial regulation pathway. Further evaluation of the energy‐storage capability is conducted by charging commercial capacitors using the rectified outputs of P‐TENG and ACE‐TENG (Figure 4g). The ACE‐TENG charges a 100 µF capacitor to 1.5 V within approximately 191.2 s, significantly faster than the P‐TENG under identical operating conditions. Extending this analysis across capacitors of different capacitances (Figure 4h) shows that each capacitor reaches a distinct voltage level within the same 191.2 s charging period, enabling quantitative comparison of charge‐delivery efficiency. As demonstrated in Figure 4i, using the relation E = C U²/2, the stored energies of these capacitors were calculated at 191.2 s (2 Hz operation), providing a direct metric of the ACE‐TENG's enhanced capability for powering low‐frequency energy‐storage components. In practical triboelectric systems, the stability of the dielectric layer under ambient perturbations is a decisive factor influencing long‐term device reliability. To assess environmental robustness, the output of the ACE‐TENG is measured across controlled humidity and temperature conditions. As relative humidity increased from 20% to 60% RH (Figure 4j), the current output decreased from 52.1 to 38.1 µA, a trend attributable to moisture‐assisted surface screening that weakens effective charge retention. In contrast, temperature variation produced a nonmonotonic response (Figure 4k): the current increased from 46.3 µA at 20°C to 51.2 µA at 40°C, likely due to enhanced molecular mobility facilitating charge separation, but subsequently declined to 45.8 µA at 60°C as thermally induced dissipation became dominant. These coupled environmental responses highlight both the sensitivity of interfacial charge processes and the resilience of the ACE‐film relative to its undirected counterpart. To further verify operational durability, a 20 000 s continuous cycling test at 2.5 Hz is performed (Figure 4l). The transferred charge remained nearly constant throughout the test. This result demonstrates that the CE‐guided amino acid arrangement forms a mechanically and electrostatically stable dielectric interface capable of maintaining performance under prolonged mechanical excitation.

FIGURE 4.

Electrical output performance and energy‐harvesting characteristics of P‐TENG, A‐TENG, and ACE‐TENG. (a) Contact–separation measurement platform. (b) Schematic illustration of the working mechanism of the TENG device. (c) Optimization pathway of output voltage through amino acid doping, CE treatment, and film thickness regulation. (d) Output voltage and transferred charge comparison between P‐TENG and ACE‐TENG under identical operating conditions. (e) Peak power density comparison between P‐TENG and ACE‐TENG under identical operating conditions. (f) Peak power density benchmarking of ACE‐TENG against previously reported amino acid‐ and protein‐based TENGs [12, 13, 15, 18, 22, 33]. (g) Charging curves of a series of capacitors powered by P‐TENG and ACE‐TENG. (h) Charging curves of a series of capacitors driven by ACE‐TENG in 191 s. (i) Charging behavior of a series of capacitors driven by ACE‐TENG in 191 s. Dependence of current output on ambient (j) humidity and (k) temperature. (l) Long‐term durability test of ACE‐TENG over 50 000 cycles of continuous operation.

2.4. Design and Characterization of the Stacked ACE‐TENG Structure

To enable higher output levels and broaden the applicability of the ACE‐TENG, a multilayer stacked architecture was engineered to amplify charge generation within a confined volume. Figure 5a presents the operating mechanism and structural composition of the stacked configuration. It depicts charge induction, interlayer coupling, and sequential contact‐separation across multiple dielectric interfaces. Figure S15 further presents the stacked design of the ACE‐TENG, in which a single ACE‐TENG unit is fabricated into five circular layers arranged in a Z‐type stacked architecture. According to the increasing diameters, the three individual stacked devices are defined as ACE‐TENG‐1, ACE‐TENG‐2, and ACE‐TENG‐3, and the entire integrated system is referred to as the S‐ACE‐TENG. In Figure 5b, the influence of the stacking layer number on the output current is investigated. Using the eight‐layer configuration as the normalized maximum, the current output gradually increases as the number of layers increases, but begins to deviate from linear superposition once the structure exceeds four layers. Such nonlinearity arises from partial mechanical coupling and incomplete contact synchronization among layers. Consequently, a five‐layer configuration is selected as the optimized design, balancing output performance and structural compliance. The circular geometry of the stacked device introduces a second key structural parameter: the chord length, which directly affects both the effective contact area and the natural separation distance. These factors are critical for integration into confined‐space self‐powered systems. Figure 5c examines how the chord length of ACE‐TENG‐1, ACE‐TENG‐2, and ACE‐TENG‐3 varies from 0 to half of the device radius. When Kapton is employed as the supporting layer, all chord lengths maintain an effective contact distance above 40 mm, meeting the practical requirement for reliable contact–separation motion in limited spaces. The choice of using one‐half of the radius as the representative chord length is due to its ability to preserve sufficient contact area while avoiding excessive mechanical strain that would otherwise accelerate structural fatigue. Based on this optimized geometric configuration, a 48 000 cycle durability test is conducted, as shown in Figure S16. The stacked device height decreases from approximately 150 to 68.1 mm, illustrating its progressive mechanical accommodation without catastrophic deformation. To further evaluate performance under confined displacement, Figure 5d measures the output current and voltage under different preset air‐gap sizes during the contact–separation motion. This behavior can be understood within the two regimes illustrated in Figure S17, namely the electrostatic equilibrium regime (d₁) and the field‐limited charge induction regime (d₂). The device reaches 99% of its maximum current output at an air gap below 2 mm, while the voltage increases nearly linearly until 7 mm and saturates at 100% beyond 8 mm. These results indicate that the device rapidly transitions from the d₂ regime to the d₁ regime with increasing separation distance. In the d₁ regime, the electrostatic field becomes fully developed, and further increases in spacing no longer contribute to additional charge transfer. Therefore, even under confined displacement conditions, the device operates near the effective saturation regime, ensuring minimal performance loss despite limited interlayer spacing. Consistent with this interpretation, COMSOL simulations (Figure S18) reveal the dynamic evolution of interlayer potential across the stacked structure during the contact–separation cycle. To further quantify the energy harvesting capability of the stacked architecture, the current generated by ACE‐TENG‐1, ACE‐TENG‐2, and ACE‐TENG‐3 reaches 23.6, 93.2, and 210.7 µA, respectively, demonstrating a clear enhancement with increasing device diameter (Figure 5e). In addition, Figure S19 reveals the multilayer phase‐shift behavior of single current pulses, where the stepping process induces a minor phase deviation of 8.72% relative to the maximum pulse, while the separation process recovers almost zero phase offset with the highest current output. The temporal characteristics of the device are further confirmed in Figure S20, showing a rapid 8 ms response time and an equally fast release time. Furthermore, Figure S21 provides a refined analysis of the low‐frequency pressure response of ACE‐TENG‐1, ACE‐TENG‐2, and ACE‐TENG‐3 within the 0–2 Hz regime (0.2 Hz step). The results show a clear linear growth in current output with increasing frequency, highlighting the stable charge‐transfer kinetics of the stacked architecture under low‐frequency mechanical excitation. The transferred charge of ACE‐TENG‐3, obtained by integrating the current signal, is 1.227 µC (Figure 5f), consistent with the experimentally measured value. The influence of interfacial contact quality on device performance was examined by introducing a soft foam spacer on one side of the stacked ACE‐TENG to enhance structural compliance. With this modification, the output current increases by approximately 1.21 times compared to the device without the foam layer (Figure S22). This improvement originates from the enhanced adaptability of the device to external pressure, which increases the effective contact area and promotes more uniform stress distribution across the multilayer interfaces. As a result, more complete and synchronized contact‐separation behavior is achieved among stacked layers, leading to more efficient charge transfer. The linearity of the pressure–response characteristics is presented in Figure 5g, where ACE‐TENG‐1, ACE‐TENG‐2, and ACE‐TENG‐3 all exhibit two distinct linear regions. The first region displays a higher force sensitivity, while the second exhibits a reduced slope at higher loads. Notably, all three devices show highly consistent normalized‐current behavior across both linear regimes, regardless of device size. This consistent scaling behavior highlights an advantageous structural invariance: the stacked ACE‐TENG maintains predictable and uniform mechano‐electrical coupling, an important feature for stable energy harvesting under varying mechanical pressures. When subjected to a single 30 N pressing force, ACE‐TENG1, 2, and 3 deliver 2.14, 27.39, and 114.02 mW of peak power, respectively, under a 10 MΩ load (Figure 5h). Integration of the load–current curves yields output energies of 34.9, 441.8, and 2436.1 µJ for ACE‐TENG‐1, ACE‐TENG‐2, and ACE‐TENG‐3 (Figure 5i), indicating the strong size‐dependent enhancement of mechanical‐to‐electrical conversion efficiency in the stacked architecture. Figure 5j compares the charging performance of ACE‐TENG‐1, ACE‐TENG‐2, and ACE‐TENG‐3, and the fully assembled S‐ACE‐TENG when charging a capacitor of 100 µF to 1 V. The S‐ACE‐TENG achieves the shortest charging time due to the cooperative contribution of multiple stacked layers, demonstrating the benefit of parallel geometric scaling for practical energy‐storage applications. Long‐term cycling stability is shown in Figure 5k, where 94.7% of the initial current is retained after 80 000 cycles, confirming the mechanical and electrical robustness of the stacked architecture. Moreover, as shown in Figure S23, ACE‐TENG‐3 maintains 88% of its original output after being left idle for ten days, further indicating the stability of its interfacial configuration under prolonged static conditions. Directional adaptability is assessed by applying downward forces at tilt angles of 5°, 10°, 15°, and 20° relative to the vertical direction. Even at a 20° inclination, the device maintains approximately 87% of its initial current output, demonstrating strong tolerance to off‐axis loading and supporting its suitability for irregular mechanical excitations in practical environments (Figure 5l).

FIGURE 5.

Structural design, electrical performance, and durability of the stacked ACE‐TENG (S‐ACE‐TENG). (a) Working mechanism and multilayer structural configuration of the stacked ACE‐TENG. (b) Dependence of output current on the number of stacked layers. (c) Influence of chord length on the effective contact distance for ACE‐TENG‐1, ACE‐TENG‐2, and ACE‐TENG‐3. (d) Output current and voltage of the stacked ACE‐TENG under different air‐gap distances. (e) Output current comparison of ACE‐TENG‐1, ACE‐TENG‐2, and ACE‐TENG‐3 with increasing device diameter. (f) Transferred charge of ACE‐TENG‐3 obtained from current integration. (g) Pressure–current response characteristics of ACE‐TENG‐1, ACE‐TENG‐2, and ACE‐TENG‐3. (h) Peak output power of ACE‐TENG‐1, ACE‐TENG‐2, and ACE‐TENG‐3 under a 10 MΩ load. (i) Output energy and (j) capacitor charging performance of ACE‐TENG‐1, ACE‐TENG‐2, and ACE‐TENG‐3 and the integrated S‐ACE‐TENG. (k) Long‐term cycling stability of the ACE‐TENG over 80 000 cycles. (l) Output current retention of the ACE‐TENG under different off‐axis loading angles.

2.5. Performance Validation of the S‐ACE‐TENG in Real‐World Applications

To demonstrate the practical utility and system‐level advantages of the S‐ACE‐TENG, real‐world application tests are conducted to validate its biomechanical energy harvesting capability and ability to drive low‐power wearable electronics. The overall technical roadmap of the S‐ACE‐TENG system is illustrated in Figure 6a, outlining the progression from biomechanical energy harvesting to stacked‐architecture enhancement, energy management optimization, and final deployment in self‐sustained wearable electronics. With the global population approaching 8 billion, continuous harvesting of ubiquitous, low‐grade mechanical energy is of critical importance for the large‐scale deployment of distributed wearable electronics and self‐powered systems. Among various biomechanical energy sources, plantar mechanical energy is particularly attractive due to its high availability, repetitive nature, and relatively large force amplitude generated during daily walking and locomotion. To efficiently capture this energy, S‐ACE‐TENG are embedded into an insole at three biomechanically dominant pressure regions, the hallux, the first metatarsal head, and the heel, which collectively account for the majority of vertical plantar load during gait. The insole is then integrated into the shoe sole, forming a fully enclosed and wearable energy harvesting system, as shown in Figure 6b. The electrical outputs generated at the hallux, first metatarsal head, and heel during walking are presented in Figure 6c. The three harvesting units operate synchronously and complementarily, enabling comprehensive collection of plantar mechanical energy across the entire gait cycle. This distributed configuration effectively converts the spatially heterogeneous plantar forces into a unified electrical output, realizing a closed‐loop and full‐coverage biomechanical energy‐harvesting process. As a result, the smart insole delivers a power output of 67.04 mW Hz⁻¹ per step during walking, as summarized in Figure S24. Figure 6d presents the time‐resolved voltage waveform under a ∼1 Hz stepping condition together with the corresponding instantaneous power. The average output power is obtained by integrating the instantaneous power over one complete cycle. The calculated average output power reaches approximately 2.97 mW under the corresponding load condition. This result provides a direct evaluation of the practical energy output beyond peak power metrics. The corresponding average power density is on the order of ∼0.7 W m⁻² based on the projected device area. The performance of the present smart insole is further evaluated in the context of previously reported human biomechanical energy harvesters, as summarized in Table S5 [34, 35, 36, 37]. The comparison includes device location, output characteristics, and demonstrated functionalities. Despite the advantages of high voltage output, TENGs inherently exhibit high internal resistance and low current characteristics, which necessitate an efficient energy management circuit for practical energy utilization. The integrated energy management circuit shown in Figure 6e consists of five key functional components: a bridge rectifier for AC to DC conversion, a gas discharge tube for voltage‐triggered energy release, an inductor for buffering the transient discharge, an energy storage capacitor for charge accumulation, and a diode for suppressing reverse current. Together, these components enable regulated energy transfer from the TENG to downstream electronic loads. Benefiting from this circuit architecture, the charging performance of capacitors with different capacitances is markedly enhanced. As shown in Figure 6f, the charging rates for 47, 100, and 220 µF capacitors to 1.5 V are increased by factors of 10.3, 11.5, and 12.3, respectively. The acceleration in charging becomes more pronounced for larger capacitors. This improvement arises because the regulated energy release mitigates the cumulative losses associated with slow charging and leakage during prolonged energy accumulation. To further demonstrate continuous energy utilization under real walking conditions, a 220 µF capacitor is employed as the primary energy storage unit and connected to five capacitors in parallel. During human stepping, the capacitor voltage periodically rises to the device turn‐on threshold (charging stage), upon which the load is activated, and electrical energy is consumed (discharging stage). With continuous stepping, the system automatically alternates between charging and discharging states, forming a stable self‐sustained power supply cycle, as illustrated in Figure 6g. The real‐time operation of this charging‐discharging process is provided in Movie S1, and the practical configuration of the five parallel connection commercial temperature‐humidity sensors is shown in Figure 6h‐i–iii. These results collectively demonstrate that the stacked ACE‐TENG, when integrated with an optimized energy management circuit, is capable of efficiently harvesting plantar biomechanical energy and converting it into stable, usable electrical power for real‐world wearable applications. The smart insole system enables full‐coverage collection of plantar forces, supports rapid energy storage, and achieves autonomous charging–discharging operation without any external power supply. This work, therefore, establishes a viable pathway for scalable, self‐sustained wearable energy systems driven entirely by human motion.

FIGURE 6.

System‐level application and practical performance of the S‐ACE‐TENG for self‐sustained wearable energy harvesting. (a) Overall technical roadmap of the S‐ACE‐TENG system from biomechanical energy harvesting to wearable self‐powered electronics. (b) Photograph of the (i) smart insole with S‐ACE‐TENG unit integrated into the (ii) shoe. (c)Real‐time open‐circuit voltage signals of ACE‐TENG units at the hallux, first metatarsal head, and heel during walking (time (s) vs. voltage (V)) under realistic gait conditions. (d) Average output power derived from the time‐resolved voltage waveform under ∼1 Hz stepping condition. (e) Schematic diagram of the integrated energy management circuit. (f) Charging curves of 47, 100, and 220 µF capacitors with and without the energy management circuit. (g) Self‐sustained charging‐discharging cycles of the energy storage system during continuous human stepping. (h) Practical energy‐harvesting demonstration of the S‐ACE‐TENG for powering five commercial temperature‐humidity sensors connected in parallel: (i) arrangement of the five parallel‐connected sensors; (ii) sensors in the off state (switch off); (iii) sensors in the on state (switch on).

Beyond energy harvesting, the molecular‐level tunability of this strategy also provides opportunities for multifunctional self‐powered systems integrating sensing, energy management, and wearable electronics, which will require further optimization of molecular design, device architecture, and system integration. While the wearable demonstration validates the practical feasibility of the proposed strategy, several challenges remain for its broader implementation. Maintaining CE‐induced molecular orientation under long‐term mechanical deformation and environmental fluctuations will be critical for ensuring stable interfacial electronic regulation.

3. Discussion

In summary, we established a CE‐regulated molecular engineering strategy to actively control amino acid orientation within triboelectric films. By selectively enriching electron‐donating amino groups toward the triboelectric interface while anchoring electro‐accepting carboxyl groups and side chains at the electrode surface, the interfacial charge transfer is enhanced by 1.51‐fold, leading to a power density exceeding 10 W m⁻². These results demonstrate that molecular‐level interfacial regulation offers an effective and generalizable pathway for boosting triboelectric energy conversion efficiency. This molecular‐oriented strategy complements conventional interfacial engineering while preserving its distinct design philosophy. The practical feasibility of this strategy is further validated using a smart insole system, which, time‐resolved waveform under stepping conditions, yields an average output power of approximately 2.97 mW (peak value 67.04 mW Hz⁻¹) based on cycle integration, corresponding to an average power density on the order of ∼0.7 W m⁻² (the projected geometric area). Successfully powers five parallel connection commercial temperature‐humidity sensors, highlighting the potential of CE‐regulated amino acid films for sustainable, high‐performance wearable energy harvesting.

Importantly, the CE‐enabled molecular orientation originates from the intrinsic donor‐acceptor framework of amino acids, defined by the conserved amino and carboxyl groups. This backbone‐dominated mechanism ensures that the fundamental charge‐transfer behavior remains consistent across different amino acids, while variations in side chains primarily introduce localized modulation of interfacial charge density and orientation degree, rather than altering the underlying mechanism. These findings establish a molecular engineering principle in which conserved donor–acceptor motifs determine preferential orientation, whereas structural variations provide fine regulation of interfacial charge transfer. Accordingly, this strategy provides a mechanistic basis for extension to diverse amino acid systems, as supported by the systematic classification of natural amino acids with varied side‐chain characteristics (Figure S25). Future efforts should focus on quantitatively correlating side‐chain‐dependent electronic modulation with macroscopic triboelectric output. In parallel, improving film uniformity and interfacial stability will be important for achieving reliable device performance. Extending this CE‐regulated molecular engineering framework to other amino acids or bio‐derived molecular systems may further expand its applicability and open new avenues for next‐generation triboelectric energy harvesting technologies.

4. Methods

4.1. Materials

Polyethylene oxide (PEO, Mw 100 000) and cysteine (≥ 99% purity) were obtained from Macklin. Formic acid (85%) used in the preparation was also sourced from Macklin. FEP and polyurethane (PU) films employed in this study were purchased from Daikin Industries and Shanghai Hongsheng Medical Technology, respectively.

4.2. Preparation of P‐Film

2 g of PEO was dissolved in 10 g of formic acid and stirred magnetically at 30°C for 2 h to obtain a homogeneous solution. Aluminum substrates were then placed on glass slides and coated with the solution via a two‐step spin‐coating process: initially at 100 rpm for 20 s, followed by a high‐speed spin at 1000 rpm for 40 s. The coated films were subsequently dried in an oven at 80°C for 1 h.

4.3. Preparation of A‐Film

The preparation procedure for the A‐film followed the same protocol as that for the P‐film, except that cysteine was additionally introduced into the PEO solution with a mass fraction ranging from 5 to 20 wt% prior to spin coating. The mixed solution was stirred until fully homogeneous, and the subsequent spin‐coating and drying conditions were identical to those used for the P‐film.

4.4. Preparation of ACE‐Film

The ACE‐film was fabricated based on the A‐film with an additional CE pretreatment applied to the Al substrate. Specifically, before solution deposition, the Al foil was subjected to one–five cycles of CE using FEP under a normal force of 10 N. The precursor solution was then drop‐cast onto the CE‐treated Al surface and spin‐coated using a two‐step program at 100 rpm for 20 s, followed by 1000–2600 rpm for 40 s. The resulting film was subsequently dried in an oven at 80°C for 1 h.

4.5. Preparation of P‐TENG, A‐TENG, and ACE‐TENG

The P‐TENG, A‐TENG, and ACE‐TENG devices were fabricated using P‐film, A‐film, and ACE‐film, respectively, all with identical dimensions. In each device, a 3 × 3 cm P‐film, A‐film, or ACE‐film laminated with aluminum foil was employed as the tribo‐positive layer as well as its corresponding electrode. Meanwhile, a 3 × 3 cm FEP film served as the tribo‐negative layer, and copper tape was attached to the back of the FEP film as the electrode. All TENG devices were assembled in the same configuration to ensure a fair performance comparison.

4.6. Preparation of ACE‐TENG‐1, ACE‐TENG‐2, and ACE‐TENG‐3

The ACE‐TENG‐1, ACE‐TENG‐2, and ACE‐TENG‐3 devices were fabricated using the same material system and assembly protocol, while their effective diameters were deliberately varied to investigate size‐dependent output characteristics. The diameters of ACE‐TENG‐1, ACE‐TENG‐2, and ACE‐TENG‐3 were designed as 10, 40, and 60 mm, respectively. Each circular device was further cut with a chord length equal to its own radius to define a controlled contact‐separation geometry. The ACE‐film and the FEP‐film were prepared in advance. For each device, the ACE‐film and the FEP‐film were assembled in a face‐to‐face configuration to form the triboelectric contact pair. Copper tape was attached to the back side of the FEP‐film as the corresponding back electrode, while aluminum foil was used as the back electrode for the ACE‐film side. To ensure stable and repeatable contact–separation motion, a soft foam spacer was placed behind the aluminum foil electrode as an elastic buffer layer. The entire device was finally encapsulated using a flexible Kapton frame to maintain structural integrity and electrical insulation. All three devices share identical material compositions, surface modification chemistry, electrode configuration, spacer materials, and mechanical layouts, with only the lateral size and chord‐cut geometry being systematically varied. This structural consistency enables a direct assessment of the size‐dependent scaling behavior of electrical output, transferred charge, and output power.

4.7. Detailed Information of Energy Management Circuit

The energy management circuit utilizes a DB107 bridge rectifier to convert the alternating output of the TENG into a direct current signal. An SX51‐230 ceramic gas discharge tube is employed as a fast‐switching component, allowing the accumulated charge to be rapidly discharged once the threshold voltage is reached, thereby facilitating efficient energy transfer to the storage unit. A 10 mH H‐type inductor is introduced to temporarily buffer the released energy and regulate the transfer process. In addition, a storage capacitor is used to collect the electrical output, while a 1N4007 diode is incorporated to suppress reverse current and protect the circuit from backflow‐induced damage.

4.8. Characterizations and Measurements

The surface morphologies of the as‐prepared films were examined using a field‐emission scanning electron microscope (FE‐SEM, ZEISS GeminiSEM 300, Germany) and a three‐dimensional optical profilometer (Bruker Contour GT‐K 3D, Germany). The electrical surface potential and nanoscale topography were further characterized by atomic force microscopy (AFM, Bruker Dimension Icon, Germany). The chemical structures and surface compositions of the films were identified by Fourier‐transform infrared spectroscopy (FTIR, Thermo Fisher Scientific Nicolet iS20, USA), x‐ray diffraction (XRD, Rigaku Ultima IV, Japan), and x‐ray photoelectron spectroscopy (XPS, Thermo Scientific K‐Alpha, USA). Depth profiling analysis of the films was carried out using time‐of‐flight secondary ion mass spectrometry (TOF‐SIMS and ION‐TOF TOF.SIMS 5). The electrical output performance, including voltage, current, and transferred charge, was recorded using a programmable electrometer (Keithley 6514) and an oscilloscope (InfiniiVision 2000 X). The mechanical excitations during performance evaluation were provided by a commercial linear motor (J‐Best RP‐200). A magnetic stirrer used in this work was obtained from Lichen Technology.

4.9. Detailed Information of Energy Management Circuit

The energy management circuit integrates a rectification unit, a voltage‐triggered switching module, an energy buffering element, an energy storage unit, and a unidirectional protection component to enable efficient conditioning of the TENG output. The alternating electrical signal produced by the TENG is first converted into a unipolar form through the rectification stage. Once the accumulated voltage reaches the predefined threshold, the switching unit is activated to release the stored energy in a controlled manner. The released energy is then temporarily accommodated by an inductive buffer to moderate the transient discharge process and stabilize the energy transfer. Subsequently, the regulated electrical energy is delivered to the storage capacitor for accumulation. Meanwhile, a unidirectional conduction element is introduced to isolate the load from reverse current and suppress undesirable backflow, thereby ensuring the operational safety and reliability of the energy management system.

4.10. Statistical Analysis

All electrical performance data reported in this work were obtained from three independent measurements under identical experimental conditions, and all error bars represent the corresponding standard deviation. For compositional analysis, the atomic percentage obtained from EDS measurements (Figures 2i and S8) was statistically evaluated based on repeated measurements, including both DD and BB variations, to ensure data reliability and reproducibility. The optimization pathway of output voltage through amino acid doping, CE treatment, and film thickness regulation (Figure 4c) was evaluated using the averaged values. The current output under different ambient humidity and temperature conditions (Figure 4j,k) was analyzed based on the same statistical criterion. The dependence of output current on the number of stacked layers (Figure 5b) and the output retention under different off‐axis loading angles (Figure 5l) were also quantified using the same method. The long‐term operational stability of the multi‐layer device (Figure S16) was assessed through continuous measurements under identical working conditions, and the output signals were analyzed using the same statistical framework. All statistical analysis and data visualization were performed using Origin and Microsoft Excel.

Author Contributions

Y.H. was responsible for conceptualization, methodology, validation, and investigation. Y.H., J.S. and X.Z. contributed to the original draft writing, resource acquisition, and visualization. Y.Q. and D.Y. were involved in formal analysis and data curation. Y. Q., H.Z., Z.L.W., and X.L. provided review & editing, project administration, supervision, and funding acquisition.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1: adma74012‐sup‐0001‐SuppMat.docx.

Supporting File 2: adma74012‐sup‐0002‐MovieS1.mp4.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.