Oxygen‐Synergized All‐Fiber Hydrovoltaics With Milliamp Output Toward Self‐Powered Breathable Electronics

ABSTRACT

Eruptive development of wearable technology has evoked great demand for decentralized energy supplies harvesting ubiquitous environmental stimuli, such as water evaporation. Conventional hydrovoltaic power generators (HPG) collect energy through directional ion migration originating from water gradient in functional materials, while the unsatisfactory electrical output severely hinders the practical applications. Herein, we develop a liquid‐induced high‐performance all‐fiber HPG for sustainable self‐powered electronics by constructing an ion‐enriched storage electrode and coincidently inducing an oxygen‐involved reaction in the solid‐liquid‐gas interface of the functional layer. Taking advantage of the hierarchical structural configuration of HPG, we verify the dual pathway synergistic electricity generation mechanism of hydrovoltaic effect and oxygen‐involved redox through in situ characterization and theoretical calculations. Significantly, the HPG exhibits an impressive power density of 164.5 µW cm⁻², an extraordinary current density of 1.25 mA cm⁻², a high air‐permeability of 428.4 mm s⁻¹, and an excellent sustainability with high output retention of 88% after 150 cycles of washing, outperforming most of the counterparts. As demonstration of applications, the as‐assembled HPG power supply packs can directly drive various wearable electronics without extra energy storage devices or rectification circuits, demonstrating the great expectation for the development of evaporation energy harvesters and self‐powered wearable electronics.

Oxygen‐synergized all‐fabric hydrovoltaic power generator (HPG) is developed by constructing a hierarchical structure with an ion‐enriched storage electrode and induced oxygen reduction reaction at solid‐liquid‐gas interface, which achieves an extraordinary milliamp‐scale short‐circuit current density (1.25 mAcm⁻²) and high peak power‐density (164.5 µWcm⁻²), maintains 88% output after 150 cycles of washing, directly driving electronics without energy storage devices or rectification circuits.

1. Introduction

The wearable technology market is experiencing significant growth with a compound annual growth rate (CAGR) of 14.60% [1, 2, 3]. The escalating demand for wearable electronics and Internet of Things (IoT) has spurred intense research into decentralized power sources harvesting potential energies from ubiquitous environmental stimuli, such as moisture [4, 5], heat [6, 7, 8], and mechanical motion [9, 10, 11]. Among these, evaporation‐enabled hydrovoltaic electricity generation technologies have emerged as promising candidates, leveraging water evaporation or moisture sorption to drive ion migration and electron transfer for power production [12, 13]. Water evaporation serves as a natural spontaneous process accompanied by substantial energy transfer (latent heat of ∼2260 kJ kg⁻¹). Despite that, this insensible energy conversion form unconsciously hindered the development and systematic study of hydrovoltaic devices. Since the pioneering work by Guo in 2017 [14], this fascinating research field of hydrovoltaic power generator (HPG) has been flourishing vigorously, with increasing electric output enabled by new materials and structural design, such as nanostructured carbon [15, 16, 17, 18], layered double hydroxide [19], metal‐organic frameworks [20, 21], transition metal dichalcogenides [22], and metal oxide [23, 24]. To date, the output voltage has surged by several orders of magnitude, from a few millivolts for a single HPG unit to several hundred volts for a large‐scale integrated HPG pack [25, 26], while progress in the output current and power density has considerably lagged behind. The majority of HPGs reported exhibit low current densities confined to the microampere level, with correspondingly limited power densities ranging from several to tens of µW cm⁻², making it difficult to scale up to a practical level for high‐power consumption devices like multi‐sensor data collection and wireless data transmission [27]. Therefore, it is crucially urgent but challenging to develop a direct‐current (DC) HPG with high power output capable of driving wearable electronics and IoTs devices.

Typical hydrovoltaic technology mainly relies on the directional ion migration driven by water/moisture gradients located on a single hydrovoltaic functional layer [28]. This structural configuration of HPG results in a limited area for electricity generation among the entire functional layer, thus leading to an inferior current/power density. The limitations of the output performance for practical applications are also related to the inefficient coupling between ionic current in functional layers and electronic current in external circuits [29]. To facilitate the ion‐electron conversion at the electrode interface, hybrid energy harvesting strategy is proposed to enhance the hydrovoltaic effect, such as electromagnetic energy [30], piezoelectric energy [31], solar energy [32], and triboelectric energy [33]. However, the proposed hybrid energy harvesting strategy still suffers from the specific scenarios and limited current output originating from the weak electron source. Accordingly, ubiquitous substance capable of supplying a steady sustained electron stream, such as the natural oxygen commonly existed in the air, come into our sight. As a typical oxidant, oxygen possesses a relatively high standard reduction potential of 1.23 V, showing favorable thermodynamic advantages for spontaneous reaction [34]. This oxygen‐involved reaction could be potentially induced to directly generate electron flow, which precisely fills the electron source gap in traditional HPG driven solely by ion gradients, and renders a dual power generation path for HPG. The existing directional ion migration pathway serves as the basis of hydrovoltaic power generation, while the oxygen‐induced electron replenishment simultaneously enhances electron supply. The oxygen‐synergized hydrovoltaic effect is capable of directly improving electron transfer efficiency, and simultaneously optimizing the coupling effect between ion current and electron current. Therefore, it is of great significance to fundamentally enhance the output performance of HPG through efficient and effective electron transfer by constructing a dual pathway of ion migration and electron supplementation at the electrode interface.

The liquid medium used in HPG has widely evolved from pure water to diverse ionic solutions [35, 36]. Inspiringly, human sweat serves as an ideal liquid medium for self‐powered hydrovoltaic devices to drive wearable electronics, beneficial from its natural substance of ∼99% water for evaporation and ∼1% inorganic salts as ion reservoir. Generally, the average evaporated sweat amount is 500–700 mL a day per person, which contains barely harnessed latent heat energy of 1130–1580 kJ (capable of continuously driving a smart watch for more than 400 h). The electrolyte ions carried in the sweat can serve as ionic fuel for energy storage/release through repeated adsorption/desorption. Furthermore, HPG enabled by sweat evaporation places an additional demand on the hydrovoltaic substrate for porosity and breathability. The previous reported HPG toward wearable applications mainly focus on the power output performance, while the used dense polymer substrates, like polyethylene terephthalate (PET) film [37], polyimide (PI) board [38], and Ecoflex film [39], sacrifice the air permeability causing the discomfort of human body. Fiber‐based fabrics present adorable air permeability and excellent sweat wicking properties, which are widely used in wearable scenarios [40]. Thus, it is essentially necessary to explore robust and sustainable all‐fiber HPG with wearing comfort to autonomously drive wearable electronics under dynamic conditions.

To address the above issues, we herein present a sweat‐induced and oxygen‐synergized high‐performance HPG toward self‐powered breathable electronics. This HPG features all‐fiber component with hierarchical structure consisting of high‐capacitance carbon‐based hydrovoltaic fibrous layer, hydrophilic dry‐fit fabric separator and woven Zn counter electrode. Based on this architecture, we acknowledged dual pathway synergistic electricity generation mechanism of hydrovoltaic effect and oxygen‐involved redox through systematical experiments, in situ characterization and theoretical calculations. The electrons transferred between oxygen and hydrovoltaic electrode take part in the charge balance to match the high energy density of the counter electrode, realizing a remarkable electrical output. Coupling with evaporation‐enabled ionic capacity and electron‐involved redox, a single HPG unit can produce a high DC short‐circuit current density of 1.25 mA cm⁻², as well as an outstanding peak power density of 164.5 µW cm⁻², surpassing most of the counterparts. The breathable all‐fiber HPG endows with excellent air‐permeability with a high air flow rate of 428.4 mm s⁻¹, and presents favorable long‐term stability with a high output retention of 88% after 150 washing cycles, suggesting promising prospect for wearable technologies. Besides, the interfacial capacitance is significantly enlarged through the introduction of capacitive active carbon materials to increase ion adsorption sites. Ultimately, the HPG pack with enhanced peak discharge current can effectively drive a series of portable and wearable electronics. This work provides a conceptional “energy‐from‐the‐body‐for‐the‐body” solution for self‐powered wearable electronics, and offers meritorious insight into the design of evaporation energy harvester toward practical applications.

2. Results and Discussion

2.1. Functional Hydrovoltaic Layer

Polymer‐based (here, polypropylene, PP) non‐woven fiber (NWF), was chosen as the substrate considering the flexibility of its constituent fibers, porosity and chemical stability. Figure 1a shows the schematic diagram for hydrovoltaic functional fibers. In order to collect electricity generated by hydrovoltaic effect, carbon black (CB) was utilized for the preliminary modification of pristine insulated PP NWF through dip‐coating self‐assembly method. To demonstrate the positive effect of CB modification on electric conductivity, the changes in sheet resistance of CB@PP NWFs was explored under successive CB modifications (Figure S1). As the number of CB modification increases, the incremental mass of CB@PP NWF increases linearly, while the sheet resistance decreases sharply, reaching 2 orders of decrease in magnitude after the fourth modifications. While the resistance can be further reduced by CB modification, the resistance value cannot be reduced by another order of magnitude. Therefore, as a preference, four dip‐coating operations are selected for the modification of conductivity. The intermolecular interactions between CB and PP fiber were investigated through Fourier‐Transform Infrared Spectroscopy (FTIR). As shown in Figure S2, a new absorption peak at 892 cm⁻¹ appears on the spectrum of CB@PP NWF, which can be assigned to the stretching vibration of O─O of ─C─O─O─C‐ [41], suggesting a new chemical bond might be formed. Notably, slight blue shift from 1236 to 1242 cm⁻¹ in the CH₂ torsion motion is observed. This result indicates the presence of intermolecular CH‐π interaction between CB and PP fiber (Figure 1a) [42]. Compared with typical hydrogen bonds, CH‐π interaction are weak interactions with bond energies of 1–3 kcal mol⁻¹. However, the accumulation of multiple interaction sites in single CB molecule can also produce significant couple effects, which endows CB@PP NWF with superior durability and washability.

FIGURE 1.

Construction and characterization of hydrovoltaic functional layer. (a) Schematic illustration of fabrication of hydrovoltaic functional fiber. (b) Mechanism of electricity generation. The upper part of the bottom scheme shows the generating process of potential difference under open‐circuit conditions due to the evaporation of water at the three‐phase interface. The lower part of the bottom scheme reveals the movement of ions and electrons under short‐circuit conditions. (c) Assembled HPG for multiple wearable applications. (d) SEM images of PP NWF (d₁‐d₂) and CB@PP NWF (d₃‐d₄). (e) Wettability of PP NWF and CB@PP NWF. (f) Schematic illustration of water diffusion during the capillary attraction process of CB@PP NWF. (g) Water absorption heatmap and electric potential distribution heatmap on CB@PP NWF.

The design of all‐fiber HPG focuses on direct and sustainable hydrovoltaic‐enabled electricity generation with milliamperes level output for practical wearable applications. It is achieved by efficient ion adsorption/desorption owing to electron transfer originated from water evaporation, and further effectively enhanced by electron‐involved oxygen reduction reaction at the solid‐liquid‐gas three‐phase interface, as demonstrated in Figure 1b. The detailed mechanism of electricity generation is discussed in the subsequent section. The milliamperes level output of all‐fiber HPG endows the direct activation of numerous applications without any energy storage device or rectifier equipment (Figure 1c). The demonstration of wearable application can be found in the final section.

From the Scanning Electron Microscopy (SEM) images in Figure 1d, the bare PP fibers exhibit a smooth surface morphology, whereas for CB@PP NWF, the CB nanoparticles are uniformly wrapped on the PP fibers, and densely arranged into rough, porous structure (Figure S3). The surface structure of CB@PP fibers was estimated by roughness measurement, where the roughness average value (Ra) was determined to be 3.1 µm, which was much higher than that of bare PP NWF (0.7 µm) (Figure S4). X‐ray photoelectron spectroscopy (XPS) was performed to identify the functional groups formed on the CB surface (Figure S5). The peaks of hybridized carbon atoms were found at 284.3 (sp²) and 285.8 eV (sp³), while the peaks at 287.4, 289.3, and 290.9 eV corresponded to oxygen‐containing functional groups including C─O, C═O, and O─C═O, respectively [43, 44]. The microporous morphology of fibers and the oxygen‐containing functional group of CB are believed to facilitate the wettability of CB@PP NWF and improve the accessibility to water. The excellent wettability of CB@PP NWF was validated by the contact angle test and capillary rise experiment (Figure 1e; Figure S6). The bare PP NWF shows an obvious hydrophobic behavior with a contact angle of 120°. In sharp contrast, upon the droplet touching the CB@PP NWF, it was immediately and completely absorbed (Video S1). Notably, in the capillary attraction test, once the sample encounters DI water, driven by capillary force related to surface tension, the liquid spontaneously rises along the CB@PP NWF against gravity (Figure 1f; Figure S7). According to the Laplace‐young equation, the capillary force is inversely proportional to the equivalent pore radius of porous structure. In other words, constructing a rough and porous surface structure can effectively and efficiently expedite the diffusion of water in all directions on the CB@PP NWF. Also, this is beneficial for an enhanced hydrovoltaic effect through the enlargement of the three‐phase interface and the increase of the evaporation rate of water.

2.2. Hydrovoltaic Effect of HPG

Most reported hydrovoltaic devices adopt a typical direct‐connect‐electrode system (Figure S8), in which the current collector of the working electrode is directly connected to the top (evaporation side) of the functional hydrovoltaic layer, while the current collector of the counter electrode is connected to the bottom (immersion side). However, due to the relatively high conductivity of CB@PP NWF, the adoption of this direct‐connect‐electrode system might result in the potential short circuit risk (Figure S9), thus deteriorating the output performance and complicating the mechanism analysis (Figure S10). Therefore, to reveal the hydrovoltaic effect and demonstrate the voltage generation of CB@PP NWF, a “disconnect‐electrode (DE)” setup was constructed for the measurement of the electric performance, as illustrated in Figure S11. The DE setup is featured with significant separation between the functional hydrovoltaic layer and the counter electrode, ensuring that the entire process of water absorption, diffusion, and evaporation occurs on the working electrode side. To demonstrate the formation of diverse functional zones on CB@PP NWF, the variation of water content after capillary attraction was investigated by a subdivision method, as described in Experimental section. As presented in Figure 1g, the adsorbed water manifests an obvious gradient distribution, where the weight of the adsorbed water at the bottom of CB@PP NWF was almost 4 times of the weight of the fiber itself. The excessive water covering the bottom of CB@PP NWF maintains the inner bulk water for effective ions adsorption/desorption, while the moisture area provides sufficient three‐phase interface for spontaneous evaporation, which is beneficial for a desirable hydrovoltaic performance. The electric potential distribution heatmap was also carried out through a local measurement method, where the current collector of the working electrode and the counter electrode were inert Pt metal (Pt||CB@PP). From the heatmap, when the evaporation medium is pure water, the maximum potential different between CB@PP NWF and Pt counter electrode reaches up to 200 mV, and the high potential zone is located at the top of CB@PP NWF (dry area). The generation of the potential difference can be attributed to the formation of an electric double layer (EDL) at the electrode/water interface. The EDL is resulted from the absorption of H⁺/OH⁻ released by water ionization, as well as the polar water molecules. In addition, the potential difference conveys a characteristic trend of decreasing gradient from the top to the bottom (dry area to wet area). The lower potential at the bottom can be ascribed to the areal reduction of the functional hydrovoltaic layer. Without a doubt, this will result in a decrease in EDL, especially when there are less available ions (pure water) in the system.

To elucidate the essence of ions migration on hydrovoltaic performance, we measured the open‐circuit voltage (V_oc) and short‐circuit current (I_sc) of the simplified HPG device with varying ion concentrations in the evaporation medium. As shown in Figure 2a, the V_oc and I_sc values present an evident monotonically increasing tendency with the increase of NaCl concentration. Specifically, in pure water, the V_oc and I_sc values are only 200 mV and 0.35 µA, respectively. However, once there are available electrolyte ions in the system, these electrical values immediately rise, reaching as high as 450 mV and 1.15 µA, respectively, in 2 m NaCl solution. Notably, NaCl in the solution dissociates into Na⁺ and Cl⁻, providing ionic fuel for the HPG. The higher the NaCl concentration, the more ionic fuel there is, and the more significant the electric signal (Figure S12). Figure 2b displayed the evolution of V_oc and I_sc over time at 0.1 m NaCl solution (close to the ion concentration in human sweat). The V_oc was recorded after short‐circuiting the HPG for 10 min, while the I_sc was measured with 10 min of resting. From the curves, due to the accumulation of ions at the solid‐liquid interface, the formation of EDL becomes prominent, and the V_oc value gradually increases. The accumulated capacity generated from the ion adsorption was further proved by the large current output at the initial stage of discharging. After releasing the accumulated ions, the I_sc value becomes stable and reaches to a current platform, indicating the presence of electron transfer in the system. The influence of environmental conditions on HPG performance was also systematically evaluated. As shown in Figure S13a, both V_oc and I_sc decrease with increasing relative humidity (RH), consistent with reduced evaporation rates at higher RH. Conversely, elevating the water temperature from 25°C to 60°C increases V_oc from 0.20 V to 0.33 V (Figure S13b), attributed to enhanced evaporation kinetics and ion mobility. These results confirm that the device output is sensitive to environmental parameters governing evaporation, and the reversible temperature response demonstrates excellent thermal stability. Figure 2c illustrates the simplified hydrovoltaic process and electricity generation. When in open circuit, the HPG was self‐charged through the formation of EDL originated from the hydrovoltaic effect. In the case of short‐circuit, the accumulated charges were released, while the hydrovoltaic layer still worked, thus producing the continuous electricity output. The detailed mechanism was carefully studied and discussed in the following sections.

FIGURE 2.

Electric output performance of HPG. (a) V_oc and I_sc of HPG at varying NaCl concentration. (b) Time evolution of V_oc and I_sc of HPG in 0.1 M NaCl solution. (c) Schematic illustration of the hydrovoltaic process and electricity generation. (d‐f) Electrical output curves (V_oc and I_sc) of HPG with different counter electrodes (Cu, Sn, Zn), respectively. (g) Areal capacitance of hybrid electrodes (HE) with different AC/CB weight ratios at various scan rates. (h) Self‐charging and galvanostatic discharging curves of HPG based on HE‐2 and Zn electrodes. (i) Self‐charging/discharging cycling performance of HPG under a discharge current density of 0.8 mA cm⁻². (j) Self‐charging curves of HPG under different conditions. (k) V_oc of HPG over time under different conditions. (l) I_sc of HPG under N₂‐filled and air‐refilled conditions.

2.3. Engineering of Enhanced HPG

To validate the hydrovoltaic effect and explore potential enhancements in electrical output, we modified the HPG with varying counter electrodes, including metallic Cu, Sn, and Zn. As presented in Figure 2d–f, the output curves for all configurations exhibit identical trends: an initial rapid rise to a stable voltage plateau for V_oc and a sharp decline to a stable current plateau for I_sc. The gradual increase in V_oc demonstrates the effectiveness of the hydrovoltaic process across different counter electrodes. Compared to the original case where Pt serves as the counter electrode, employing more electrochemically active metals significantly increased the stable I_sc value of the HPG.

As presented in Figure 2f, HPG with Zn||CB@PP configuration achieved a highest initial discharge current of several milliamperes, attributed to the rapid release of accumulated EDL capacity. To further improve the initial discharge current to match the large surge current required for the activation of high‐power electronics, high‐capacitance porous active carbon (AC) was incorporated into the functional hydrovoltaic fiber. This modification aims to boost the electrical output performance by increasing the specific surface area of fibers and providing considerable adsorption sites for ions. Hybrid electrodes (HE) were fabricated with varying AC/CB weight ratios (Table S1). The electrochemical performance of these electrodes was investigated in a typical three‐electrode system, where Pt, Ag/AgCl, and 1 m KCl served as the counter electrode, reference electrode, and electrolyte, respectively. As shown in Figure S14, the cyclic voltammetry (CV) curves of all electrodes displayed quasi‐rectangular shapes without distinct redox peaks, indicating a predominantly electrostatic energy storage mechanism: the absorption of electrolyte ions onto the polarized electrode surface under an applied electric field. Furthermore, all the doped electrodes demonstrated higher capacitance than the pure CB electrode, as reflected by the larger enclosed area of their CV curves. The specific areal capacitance was determined using the CV data at various scan rates. As depicted in Figure 2g, for each scan rate, the areal capacitance initially increases and then decreases with increasing AC/CB ratio. The HE‐2 electrode (AC/CB = 0.5) exhibited the highest areal capacitance and excellent rate performance. However, with further increases in AC content, the HE‐3 (AC/CB = 1) electrode showed degraded rate performance due to reduced CB content (which serves as conductive agent). Similarly, the HE‐4 (AC/CB = 2) electrode also showed reduced capacitance, as excessive AC occupied the CB sites and subsequently detached from the electrode during testing, due to its porous structure. As shown in Figure S15, the Nyquist plots were made for the analysis of kinetic processes of electrode reaction. In the high frequency, all electrodes exhibited the very same equivalent series resistance (R_s), suggesting the identical ohmic resistance of electrolyte and electrode materials. In the medium frequency, it can be seen that the HE‐2 electrode possesses the lowest charge transfer resistance (R_ct), indicating the highest migration rate of ions from the electrolyte to electrode. Based on the above discussion, HE‐2 (HE@PP) electrode was selected as the functional hydrovoltaic layer for subsequent electrical tests.

The self‐charging process of the HPG based on HE@PP and Zn counter electrodes was demonstrated through the resting and galvanostatic discharging tests. As presented in Figure 2h, the HPG device can be spontaneously self‐charged to 0.9 V after 1200 s in an ambient environment, and the stored electricity can be released at varying current density for powering potential applications. The voltage curves at self‐charging stage experience a rapid initial increase, followed by a gradual stabilization, which is determined by the evaporation‐enabled energy generation process. The discharge time from 0.9 V to 0.2 V is about 2800 s at a current density of 0.3 mA cm⁻². As a result, the HPG delivered a high areal capacitance of 1327 mF cm⁻², which is several orders of magnitude greater than that of reported graphene‐based energy storage devices and also higher than that of hybrid ion capacitors (Figure S16). This excellent performance can be ascribed to the synergistic effect of electricity generation and energy release of stored ions. Figure 2i represents the self‐charging/discharging cycling performance under the discharge current density of 0.8 mA cm⁻². The HPG preserves durable self‐charging/discharging performance of more than 70 cycles, without significant degradation of discharge capacitance (Figure S17), indicating the great potential for practical applications.

The self‐charging performance of HPG was further evaluated at different evaporation conditions. First, the HPG was placed in a vacuum environment, where the air was pumped out while the water evaporation continued. As shown in Figure 2j, the HPG stored in vacuum self‐charged to a relatively lower V_oc of 0.31 V after 2000 s, compared to a maximum of 0.96 V under ambient conditions during the same period. The higher V_oc rise rate under ambient conditions highlights the positive role of air in enhancing hydrovoltaic energy generation. After standing for an ultra‐long time of 30000 s, the V_oc remained at approximately 0.836 V, which is still lower than that observed under ambient condition (Figure 2k). The self‐charging performance of HPG was also measured under a N₂‐filled atmosphere at ambient pressure. The shape of the self‐charging curves tested in N₂ atmosphere is quite similar to that obtained in vacuum testing, except that the V_oc is slightly higher after the same resting time. In vacuum, the increased distance between water molecules and the surface of hydrovoltaic layer reduces the total amount of charge transferred at the liquid‐solid interface compared to ambient pressure, thus resulting in a lower potential difference. Once the electrodes were hermetically sealed in a coin cell (preventing both air exchange and water evaporation), the V_oc value remained below 0.25 V even after a prolonged resting time of 30000 s (Figure 2k). These results confirmed that the electricity generation was initially induced by water evaporation, with additional enhancement from the air‐involved reactions. In order to quantitatively deconvolute theses contributions, measurement experiments were conducted to evaluate the I_sc under controlled evaporation conditions (Figure S18). As presented in Figure 2l, the I_sc output initially stabilized at 750 µA under ambient conditions. Upon N₂ starts to be filled, this value dramatically plummeted to 240 µA, remaining at this reduced level. In inert N₂ atmosphere, the current is attributed to the physical contribution from evaporation‐driven ion migration. Subsequently, when the system was refilled with air, this value gradually increased and returned to 750 µA. The current difference regarding to electrochemical contribution is determined to be 510 µA, which is 68% of the total current. These current measurements corroborate the V_oc analysis: evaporation initiates energy generation, while air‐involved reactions significantly amplify the electric output performance.

2.4. Mechanism Verification of Electricity Generation

First, an HPG with sandwich architecture was integrated, featuring vertically stacked hydrovoltaic fibers layer and the Zn counter electrode isolated by a hydrophilic dry‐fit fabric as the separator (Figure S19). As shown in Figure 3a, when the HPG was entirely underneath the 0.1 m NaCl solution (stage one, prevent both air exchange and water evaporation at solid‐liquid interface), the I_sc value was determined to be only ∼19 µA cm⁻², which probably resulted from the Zn corrosion in the salt solution. Notably, upon withdrawal from the solution, the I_sc value immediately surged to ∼200 µA cm⁻² (stage two) before gradually rising to a maximum of 1003 µA cm⁻² (stage three). As residual water on the HPG continues to evaporate, the I_sc value started to fluctuate, and decayed to zero upon complete dehydration (stage four). Figure 3b delineates the four distinct operational stages. In stage one, despite the evaporation of external water, the submerged HPG remains in environmental equilibrium. In the case of stage two, solution removal triggers immediate hydrovoltaic activation through the interfacial water evaporation, which results in the direct increase of I_sc. As for the stage three, continuous evaporation of interfacial water leads to the formation of the largest three‐phase interface, thus achieving the peak I_sc output of the HPG. Conversely, in stage four, the shrinking three‐phase interface inevitably induces a gradual attenuation of electricity generation.

FIGURE 3.

Mechanism investigation of electricity generation. (a) Time‐dependent I_sc of HPG under varying conditions. (b) Schematic illustration of the four stages corresponding to (a), showing the evolution of water distribution, three‐phase interface, and electron transfer during evaporation. (c) Electron transition model depicting energy levels (E_C : electron occupation energy levels of CB; E_B : electron occupation energy levels of bulk water) and potential barriers (E₁ : electron escape energy from the CB surface; E₂ : electron escape energy from the water surface) (d) Energy level diagrams of G showing shifts in Fermi level and vacuum level after contact with water. (e) Charge redistribution in three‐layer G upon adsorption of a water molecule. (f) Calculated work function values of G, G = O, G‐COOH, and G‐OH (G with different oxygen‐containing groups). (g) In situ Raman spectra of hydrovoltaic layer during evaporation. (h) Magnified Raman spectra highlighting the emergence of HO* peak (1016 cm⁻¹) during oxygen reduction reaction at the three‐phase interface. (i) Schematic illustration of voltage generation. (j) Schematic illustration of electricity generation.

The electron transfer mechanism occurring between water and the CB hydrovoltaic materials is elucidated through an electron transition model. Within this framework, atoms serve as potential wells, where the delocalized outer‐shell electron clouds occupy different energy levels. As illustrated in Figure 3c, E_C and E_B stand for the electron occupation energy levels in the CB materials and bulk water, respectively. E₁ and E₂ , on the other hand, correspond to the potential energies needed for electrons to escape from the surface of CB and water, respectively. Prior to an intimate contact of CB and water, electrons are confined on their own potential wells due to the high potential barrier. Nevertheless, once the two materials come into contact, their electron clouds overlap, merging into an asymmetric double‐well system. Significantly, the highest occupied energy level of the atom belonging to CB is higher than that of the atom belonging to water. Consistently, because of the elimination of the original energy barriers, electrons can transfer to any energy level which is below the highest occupied energy of atom CB [45]. When the two surfaces separate due to the water evaporation, the transferred electrons are dragged away from the CB surface, leaving behind holes in the hydrovoltaic layer, and forming a potential difference between working electrode and counter electrode. Notably, in the entire electricity generation process, all the energy generated originates from the latent heat absorbed when water evaporates, including the kinetic energy of water molecules leaving the solid‐liquid interface.

Quantum theory calculations have been employed to explain the interactions between water and atomically smooth solid surfaces by considering the coupling of charge fluctuations within the liquid to electronic excitations within the solid. Practically, the presence of oxygen‐containing functional groups on the solid surface might play a crucial role in the electron transfer process in the initial stage of water‐solid contact. First‐principles density functional theory (DFT) calculations were performed to explicate the effect of surface functional groups on the contact between water and CB in the initial stage. First, as a demonstration, monolayer graphene (G) was used to conceptually represent the CB without any functional groups. As shown in Figure 3d, the changes of Fermi level and vacuum level evidence the charge transfer behavior after contacting with water. The decrease of Fermi level and vacuum level suggest the declined electron concentration and the higher surface electrostatic potential, signifying the electron loss of CB and the accumulation of surface positive charges. In addition, the increased work function means CB plays a role in donating electrons during contact, indicating the tendency for electrons to be transferred from CB to water and trapped near the surface of water. When a water molecule was adsorbed onto the surface of G, the electron transfer from G to the molecule was determined to be 0.01 e (Figure 3e). Subsequently, the work functions of CB with varying oxygen‐containing chemical groups were calculated to evaluate the electron‐losing ability of the solid. As presented in Figure 3f, upon CB grafting with oxygen‐containing groups, the Fermi level and work function decrease accordingly. Remarkably, G─OH (CB grafted with OH groups) shows the lowest Fermi level and work function with 0.429 and 4.341 eV, suggesting the great electron transfer expectation between water and G‐OH.

The oxygen‐synergized effect of electricity generation is tracked by in situ Raman spectroscopy. As shown in Figure 3g,h, the intensity of Raman characteristic peaks of carbon (e.g., the D band at 1350 cm⁻¹ and the G band at 1610 cm⁻¹) changes along with the evaporation process, which is associated with the doping state of carbon materials [46]. Specifically, the intensity of D band and G band increased first and then decreased over reaction time. The highest intensity of the Raman characteristic peaks of carbon was attributed to the positive doping of carbon in hydrovoltaic functional layer and anions adsorption on the carbon surface. Subsequently, the physically adsorbed ions were released due to the evaporation of the near surface water, resulting in the decrease of the intensity of the characteristic peaks. Consistently, while the high intensity of D band and G band appeared, a featured peak located at 1016 cm⁻¹ emerged, which corresponds to the HO* species (intermediates adsorbed on the surface of the hydrovoltaic layer) related to the oxygen reduction reaction (ORR) [47]. The electroreduction of O₂ to H₂O involves four coupled electron and proton transfers:

$${\mathrm{O}}_2+\ast +2\left({\mathrm{H}}^{+}+{\mathrm{e}}^{-}\right)\to \mathrm{HO}\ast$$

$$\mathrm{HO}\ast +2\left({\mathrm{H}}^{+}+{\mathrm{e}}^{-}\right)\to {\mathrm{H}}_2\mathrm{O}$$

where * represents an unoccupied active site. The direct electrochemical evidence of ORR has been provided by supplementing rotating ring‐disk electrode (RRDE) measurements and CV tests to characterize ORR activity, selectivity, and reaction pathway of the hydrovoltaic functional materials. As shown in Figure S20a, in a O₂‐saturated electrolyte, a distinct cathodic reduction peak appears at approximately 0.6 V (vs. RHE) in the CV curve. In contrast, in the case of N₂‐saturated electrolyte, this reduction peak completely disappears, and the CV curve exhibits only the featureless rectangular shape typical of EDL capacitance, with no faradaic current. To further quantify the ORR process, RRDE measurements are performed to determine the electron transfer number and H₂O₂ yield. As presented in Figure S20b, the disk electrode delivers a distinct cathodic reduction current, confirming the excellent ORR activity. Notably, the ring electrode shows negligible current, indicating a highly selective 4‐electron ORR pathway with minimal generation of H₂O₂ intermediates. Figure S21 quantitatively verifies that the CB exhibits a stable electron transfer number close to 4.0 and an H₂O₂ selectivity below 5% during ORR. These results demonstrate that ORR on CB proceeds primarily via the 4‐electron pathway, which is consistent with the in situ Raman analysis and the high current output requirement of the HPG device.

To summarize, the dual pathway synergistic electricity generation mechanism of hydrovoltaic effect and oxygen‐involved redox was depicted in Figure 3i,j. In the open‐circuit case, when water and hydrovoltaic layer come into contact, there is an overlap of the electron clouds of the two materials. The difference in electron potential energy inevitably leads to electron deviation. Once the evaporation of water occurs at the solid‐liquid interface, the transferred electrons cannot return to their original positions because of the elimination of energy barrier, thus leaving holes in the hydrovoltaic layer. To balance the formed holes, anions (Cl⁻) are attracted to the hydrovoltaic electrode, simultaneously, cations (Na⁺) move to the counter electrode, which results in the formation of the potential difference. Additionally, once the HPG is short connected, the accumulated positive and negative charges are neutralized, and the formed EDL is released, leading to the generation of current. Significantly, the oxygen involved reaction at the hydrovoltaic layer side and the oxidation of active Zn at the counter electrode side cooperatively expand the electron transfer process, therefore producing a significantly enhanced electricity output.

2.5. Output Performance of One All‐Fiber HPG Unit

To integrate hydrovoltaic functionality into wearables, all‐fiber based HPG was made by sewing method, in which HE@PP NWF, dry‐fit fabric, and Zn woven fabric served as the working electrode, separator, and counter electrode, respectively (Figure 4a). The Zn woven fabric was fabricated through cross‐weaving Zn wire and nylon conductive wire. The as‐made Zn fabric possesses an excellent flexibility, showing great prospect for wearable applications (Figure S22). As presented in Figure S23, the HE@PP NWF, dry‐fit fabric separator, and Zn woven fabric exhibit excellent air permeability with air flow rates of 1439.7, 770.9, and 864.9 mm s⁻¹, respectively. Even after integrating all three layers into the final HPG unit, the assembled structure retains a high air permeability with an air flow rate of 428.4 mm s⁻¹, which is notably higher than a commercial cotton T‐shirt (324.8 mm s⁻¹) and more than double that of a surgical mask (199.3 mm s⁻¹) (Figure 4a). Given the uneven secretion of human sweat, artificial sweat was utilized as the liquid medium for performance evaluation of all‐fiber HPG. Figure 4b presents the time evolution of I_sc of HPG with the amount of artificial sweat. Once the artificial sweat was added to the electrode interface, the PHG unit immediately responded with I_sc. In addition, all I_sc curves exhibited the same tendency, which is to rise first and then gradually fall over time. Notably, with the increase in artificial sweat amount, the current density platform rose from 316 µA cm⁻² at 5 µL cm⁻² to 1190 µA cm⁻² at 28 µL cm⁻² (generally equivalent to chest sweating during 10 min jogging), indicating a stable electricity generation. With further increase artificial sweat amount to 40 µL cm⁻², the current density curve exhibited slight decrease, which may be attributed to the changes in three‐phase interface of hydrovoltaic layer due to the excessive liquid cover. To achieve superior output performance, a certain sweat amount of 28 µL cm⁻² was used in the following tests.

FIGURE 4.

Output performance of all‐fiber HPG. (a) Comparison of air permeability of HPG and other textiles. Insert: schematic illustration of the HPG structure, consisting of HE@PP NWF as the working electrode, dry‐fit fabric as the separator, and Zn woven fabric as the counter electrode. (b) Time evolution of I_sc of HPG with different amount of artificial sweat. (c) Schematic diagrams of HPG samples with different electrode widths. (d) V_oc, I_sc and current density of HPG samples. (e) Electric output of HPG with external resistance varied from 3 Ω to 620 KΩ. (f) Dependence of power density on external resistance. (g) Performance comparison between HPG and previously reported moisture‐electric generators and hydrovoltaic devices [17, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66]. (h) V_oc and I_sc profiles of HPG under different bending states from 0 to 180°. (i) V_oc and I_sc profiles of HPG under twisting conditions. (j) V_oc and I_sc profiles of HPG overtime under dynamic sweating conditions. (k) The MMT results of HPG. (l) Durability and washing cyclability of HPG.

To verify the origination of the remarkable output, structure‐performance relations of HPG were explored. A typical HPG can be divided into the HE@PP working electrode and Zn fabric counter electrode. Specified HPG samples with different electrode widths (Table S2) were fabricated to investigate the diversified outputs (Figure 4c). As shown in Figure 4d, all HPG samples exhibit almost the same V_oc around 0.93 V. Interestingly, despite the less influence on V_oc, the varying electrode area significantly affects the generated I_sc of HPG. Specifically, for samples S1, S4 and S5, the output current is passively correlated with the area of the hydrovoltaic electrode. Given the substantial energy density of the counter electrode, the larger area of hydrovoltaic electrode leads to more reaction sites and increased three‐phase interface, thus reaching a higher output current. Despite this, the current density of all HPG samples (based on the area of hydrovoltaic electrode) remains stable at around 1200 µA cm⁻².

The electrical output performance of HPG was further reflected by the power density. Connecting with external resistance from 3 Ω to 620 KΩ, the output voltage increases while the current density gradually decreases (Figure 4e). As presented in Figure 4f, a maximum peak power density value of 164.5 µW cm⁻² is achieved at an optimal resistance of 330 Ω for one single HPG unit. Notably, the as‐fabricated HPG exhibits an excellent current density of 1250 µA cm⁻², surpassing previously reported moisture‐electric generators and hydrovoltaic generators by orders of magnitude (Figure 4g; Table S3). The exceeding current density and power density are expected to directly boost the practical application of wearable electronics. Based on the latent heat of water evaporation as the input energy, the energy conversion efficiency is defined as: $\eta =\frac{P_{output}}{P_{input}}=\frac{{\int }_0^{t}UIdt}{mL}$, where U and I are the real‐time output voltage and current of the HPG unit at the optimal load resistance (330 Ω), m is the mass of water evaporated by the HPG functional layer, L is the latent heat of water evaporation (2260 kJ kg⁻¹). The evaporation to electricity energy conversion efficiency is calculated to be ∼0.098%, demonstrating that a significant amount of energy can still be efficiently and effectively converted.

To investigate the favorable mechanical flexibility of the HPG, the electric output performance under varying bending states has been evaluated. As presented in Figure 4h, the HPG remained almost unchanged in voltage profile at different bending angles from 0 to 180°, demonstrating outstanding flexibility and electrical stability. Notably, when bent, the HPG responded immediately, with a sudden increase in current density, and then a gradual decrease to a platform. This phenomenon can be ascribed to the closer contact of the liquid‐solid interface, which increases the active area of the functional layer and effectively reduces the interfacial resistance of the HPG. The mechanical robustness of the HPG was further evaluated under twisting deformation and repeated bending cycles. As shown in Figure 4i, the device remained almost unchanged voltage and current profiles under various twisting state. Furthermore, after 2000 continuous bending cycles, the V_oc retention remained above 95% (Figure S24), demonstrating exceptional durability. These results confirm that the all‐fiber HPG can reliably withstand the complex mechanical deformations encountered in wearable applications. To assess the viability of HPG device for practical wearable applications, the electrical performance under dynamic sweating condition was evaluated in a long‐term aerobic exercise of 90 min. As shown in Figure 4j, both V_oc and I_sc exhibited a gradual and sustained increase throughout the exercise period. The enhancement of electrical signals can be ascribed to the gradual increase in electrolytic ion concentration in sweat as water evaporates. This result demonstrates that our HPG not only maintains functionality under continuous sweating conditions, but benefits from the natural concentration process, with performance improving over time. Further, we investigated the influence of sweat pH, which can vary from 5.0 to 7.0 in real scenarios. Across this physiologically relevant pH range, the values of V_oc and I_sc were slightly increased with the increase of H⁺ ion concentration, as presented in Figure S25. The mild enhancement under weakly acidic conditions can be explained by accelerated Zn oxidation kinetics and the proton involvement in the ORR. Critically, the device maintained robust milliampere‐level output across the entire pH range, demonstrating excellent tolerance to sweat composition fluctuations.

To assess the feasibility of the HPG for practical wearable applications, we systematically evaluated its scalability to larger areas and its performance in terms of coating uniformity and sweat distribution. A large‐area HE@PP NWF with dimensions of 20 cm × 35 cm was successfully fabricated using the same dip‐coating self‐assembly method (Figure S26a). To evaluate the uniformity of the carbon coating across this large area, sheet resistance was measured at ten different zones distributed across the entire fabric. As shown in Figure S26b, the sheet resistance values across all ten zones are nearly identical, which confirms that our dip‐coating self‐assembly method is highly scalable and produces a uniform, reproducible coating over large areas. This uniformity is essential for ensuring consistent electrical performance across the entire wearable textile. Efficient sweat transport and uniform distribution across the active area are critical for consistent power generation in a large‐scale wearable device. To evaluate this, we performed moisture management testing (MMT) using a standardized moisture management tester (SDL ATLAS, Ltd., China) in compliance with the AATCC 195–2009 standard. As shown in Figure 4k, the blue dotted line corresponded to the water content in top layer (skin side), while the solid line symbolized the water content in bottom layer (air side). The overall water content in the bottom layer is higher than that in top layer, which evidences that the fabricated HPG exhibits great water transport performance and excellent sweat penetration behavior. In other words, the fabric stack effectively wicks sweat from the skin side (top) to the evaporation side (bottom), ensuring a continuous liquid supply for the hydrovoltaic effect. Figure S27 visually describes the sweat distribution of each surface after the test. Both the top and bottom layers present superior sweat diffusion capacity with high spreading speed of ∼2 mm s⁻¹. The rapid spreading speed ensures that sweat is quickly and evenly distributed, preventing localized saturation or drying that could lead to performance hotspots or dead zones.

The practical reliability of the HPG were further evaluated through multiple washing/testing experiments. As shown in Figure 4l, despite the fluctuation in maximum current density, the HPG maintained exceptional current output of milliampere level with a remarkable performance retention of 88%, after a long‐term washing cycles of 150. Specifically, the time evolution curves of current density exhibited similar profiles in the initial, 50^th, 100^th, and 150^th washing cycles, demonstrating the excellent electrical stability and cyclability for potential wearable applications. Since Zn counter electrode is susceptible to corrosion and potential passivation in sweat, which could impact device performance over extended operation, we have conducted a comprehensive long‐term stability study lasting 200 h, combining continuous voltage monitoring with periodic electrochemical impedance spectroscopy (EIS) analysis to probe the evolution of the electrode interface. As shown in Figure S28, the HPG device exhibits an excellent hydrovoltaic stability and durability, as the V_oc signal is consistently above 0.9 V even after an ultra‐long time of 200 h. EIS measurement has been utilized to investigate the dynamics of the HPG electrode interface after different hydrovoltaic period. As depicted in the insert of Figure S28, all EIS plots exhibit the very similar shapes, demonstrating the consistent and robust electrode interface. Table S4 lists the specific impedance parameters of HPG during long‐term hydrovoltaic process. The internal equivalent series resistance (R_s) remains at a low value of around 8 Ω, indicating that the bulk conductivity of the device is unaffected by prolonged operation and that no significant degradation of the carbon‐based functional layer or electrolyte pathways occurs. The liquid‐solid interfacial charge transfer resistance (R_ct) increased slightly from the initial 90.67 Ω to 112.4 Ω after 200 h of hydrovoltaic process. The increased R_ct values indicate that the Zn counter electrode indeed occurs the inevitable corrosion and passivation, and mildly affect the interfacial charge transfer behavior. Notably, the steady V_oc values during the long‐term testing demonstrate that the corrosion and passivation of Zn electrode would not significantly deteriorate the electrical output performance. In addition, in a real‐world wearable scenario, where sweat production is intermittent, the device would likely be used in cycles rather than continuously for 200 h, further mitigating any long‐term degradation effects. For future iterations, it is noted that alternative counter electrode materials (e.g., coated Zn) could be explored to further enhance longevity.

2.6. Application Potential in Wearable Electronics

The as‐made HPG units can be combined into a power supply pack through series and parallel connection to adapt to the operating voltage and current of varying electronics (Figure 5a). The V_oc and I_sc curves of the power supply pack were measured. As shown in Figure 5b, the V_oc value increases linearly with the number of HPG units connected in series, demonstrative of the successful integration. Moreover, upon the HPG units being parallelly connected, the I_sc value correspondingly raised, indicating the exceptional uniformity of the as‐prepared HPG units. As a demonstration, three serially connected HPG units exhibited an output voltage of 2.56 V (Figure 5c). Thanks to the milliampere level output of single HPG unit, the power supply pack assembled with multiple units is capable of delivering expected electrical output to activate various portable electronics.

FIGURE 5.

Demonstration of HPG driving functional applications and wearable electronics. (a) Schematic illustration of HPG power supply packs via series/parallel connections for scalable voltage/current outputs. (b) Electrical performance of HPG pack with different HPG units connected. (c) Photograph demonstrating the open‐circuit voltage of the HPG pack (3 units in series/3) measured and photographs showing direct driving of portable electronics, such as calculator, digital timer, and hygro‐thermometer. (d) A wireless locator driven by an HPG pack. (e) An LED light group and a digital watch driven by HPG packs. (f) A chest‐worn heart rate monitoring system. Insert: real‐time heart rate data curve and corresponding health metrics. (g) An enlarged HPG pack driving a motor and fan. (h) Left: electric heating fabric driven by an enhanced HPG pack (3 units in series/8 groups in parallel); right: infrared thermal images showing rapid temperature rise. (i) Core temperature evolution of the heating fabric. (j) Scalable demonstration of HPG packs: 24‐series units delivering a high voltage of 21.1 V, HPG pack with 3 units in series/8 groups in parallel configuration achieving a large current output of 22.9 mA.

The potential of HPG in driving portable electronic was first investigated. For example, a calculator can be directly driven by two HPG units connected in series without rectifier equipment, while the clear display in complex computation processes suggests the robust output of HPG (Video S2). In addition, powered by the power supply pack, a digital timer was readily and steadily activated, as demonstrated in Figure 5c and Video S3. The outstanding performance is benefit from the stable DC output originated from the hydrovoltaic effect, indicating the prospect of HPG as a DC power source to directly drive diverse electronics.

Moreover, the potential of HPG in driving portable multi‐functional devices was also investigated. Compared with the timer and calculator with simple functions, digital multi‐sensor devices usually require much more energy and power input. Therefore, a hygro‐thermometer with high power consumption was utilized as an example to show the stable energy supply of the HPG pack (Figure 5c). For a higher energy output, an enhanced HPG pack was assembled by connecting 3 units in series to drive the hygro‐thermometer. As shown in Video S4, the hygro‐thermometer can continuously detect and clearly display changes in temperature and humidity in the environment. During the sensor data collection process, the screen of the hygro‐thermometer is always clear without any flickering, indicating the excellent stability of HPG pack against variable power dissipation of electrical appliances. Furthermore, smart electronics functioning with wireless signal transmission are another high‐power consumption device that requires higher energy input. Herein, a wireless locator was employed for the demonstration, while an HPG pack with 5 serially connected units served as the enhanced power supply. Based on intermittent high‐frequency signal transmission with the cloud server, the wireless locator can be precisely tracked in real time by a smartphone, as presented in Figure 5d and Video S5.

The prospect of HPG in wearable applications was explored. As shown in Figure 5e, a light group consisting of 18 LEDs can be steadily illuminated by an HPG assembly (5 units in series). Significantly, the HPG assembly with sweat amount of 28 µL cm⁻² can drive this light group for a long‐lasting duration over 50 min (Video S6), which exhibits great potential in making wearable luminous products (e.g., the luminous shoes, Video S7). Consistently, a digital watch was easily driven by the assembled HPG power supply pack with only 3 units in series (Figure 5e). To further explore the potential of the HPG pack in powering wearable functional appliances, the integrated device with real‐time heart rate measurements was constructed for health monitoring and wireless data transmission (Figure 5f). As demonstrated in Video S8, once powered by the PHG pack (5 units in series), the heart rate monitor was immediately activated and wirelessly connected with the smartphone. Afterward, the heart rate data can be captured and recorded as visible fluctuated curves over time. Notably, the heart rate monitoring application is perfectly compatible with sweating conditions, since the human sweat produced during exercise serves as the continuous liquid medium triggering the hydrovoltaic effect for electricity generation. A series of activities including aerobic activities and recovery exercise were performed for long‐term heart rate monitoring. As demonstrated in Figure 5f, taking advantages of continuous production and evaporation of sweat, the as‐fabricated HPG pack can drive the assembled heart rate monitoring system throughout the entire training process.

Further, considering the body comfort during training process, the prospect of HPG in driving electronic devices for body thermal management was explored. As a demonstration, cooling fan and heating fabric were assembled and driven by HPG pack. To offer a matched electrical output for the high‐power cooling devices (motor and fan), 5 HPG units were serially connected and elaborately stitched within wearable (Figure 5g). As presented in Video S9, despite the high‐power consumption of the motor, the as‐assembled cooling fan can be activated and stably operate for up to 8 min when driven by the enhanced HPG pack. It is worth mentioning that when the electric motor was first energized, a maximal instantaneous input current (inrush current, several times the normal full‐load current) is required for rotor picking up speed. Encouragingly, this large switch‐on surge current can be sufficiently provided by the HPG pack through the accumulated capacity stored in porous carbon materials during the hydrovoltaic process. Additionally, an electric heating mesh was fabricated by integrating carbon fiber resistance wire into a fabric substrate (Figure 5h). For a higher energy output, an enhanced HPG pack (3 units in series/8 groups in parallel) was engaged for driving the heating device using Joule heating. As presented in Figure 5h, infrared thermal images show that the fabric has a significant temperature difference after being powered by the HPG pack. In specific, under natural convection conditions, the core temperature of the fabric rapidly increased by ∼1°C in several seconds and remained stable when the heating mesh was working (Figure 5i). Consistently, once powered off, heat accumulated in the fabric was gradually dissipated into the environment due to heat exchange. Moreover, the superior scalability of HPG is also presented, in which the HPG pack can generate enlarged output with dozens of volts and milliamperes through series and parallel connections (Figure 5j). As a demonstration, an HPG pack with 24 HPG units in series connection can provide an output voltage of 21.1 V, while the I_sc of the HPG pack (3 units in series/8 groups in parallel) can reach a maximum current output of 22.9 mA. Such an outstanding output is sufficient and necessary for most commercial‐available electronics.

3. Discussion

In summary, we have developed a sweat‐induced and oxygen‐synergized all‐fiber HPG system achieving milliampere‐level output, tailored for sustainable wearable electronics. The accumulation of ion adsorption and introduction of oxygen induced redox at the interfaces facilitate evaporation‐driven electricity generation and increase electrode capacitance. The dual pathway synergistic electricity generation mechanism of hydrovoltaic effect and oxygen‐involved redox is determined through an electron transition model and theoretical calculations, as well as in situ Raman examination. As a result, one single HPG unit is capable of generating a remarkable power density of 164.5 µW cm⁻² and current density of 1.25 mA cm⁻², exceeding previous reported devices by orders of magnitude. Furthermore, all‐fiber HPG exhibits excellent air‐permeability with a high air flow rate of 428.4 mm s⁻¹, robust mechanical flexibility against diverse deformations, and durable washability with a high output retention of 88% after 150 cycles of washing, demonstrating promising prospect for wearable applications. As expected, diverse multi‐functional appliances are directly and effectively driven by enhanced HPG power pack without extra energy storage devices or rectification circuits. This work presents a significant step toward the design and implementation of high‐performance wearable evaporation‐driven energy harvesters and points out a valuable direction for the development of miniaturized self‐powered wearable electronics.

4. Experimental

4.1. Materials

Non‐woven fabric (NWF) and dry‐fit fabric were purchased from local supplier. Silver‐nylon yarns were bought from Qingdao Hengtong XSilver Specialty Textile Co., Ltd. Ketjen carbon black (CB, EC‐300J) was offered by Canrd Technology Co., Ltd. Ultrahigh capacitance porous active carbon (AC, C196580), sodium dodecyl sulfate (SDS, ≥99.0%), and sodium chloride (NaCl, ≥99.5%) were bought from Shanghai Aladdin Bio‐Chem Technology Co., LTD. Deionized (DI) water was obtained from the ultra‐pure water system (Thermo Scientific, GenPure Standard). Pt sheet (99.999%), Cu foil (99.99%), Sn foil (99.999%), Zn wire (300 µm) and Zn foil (99.999%) were supplied by High‐Purity Metal Materials Research Institute. Simulated sweat (pH = 6.5 ± 0.1) was provided by Shenzhen Zhong Wei Equipment Co., Ltd.

4.2. Preparation of HPG Unit

The commercial‐available NWF was first modified by CB through a typical dip‐coating self‐assembly method. Specifically, the impregnation solution was prepared by dispersing CB (0.5 g) in DI water (100 mL), in which SDS (0.5 g) was added serving as dispersant. The original NWF (2 mm thickness) was elaborately tailored and washed by DI water and anhydrous ethanol. After purification, the NWF was submerged into the above‐mixed solution, followed by ultrasonic treatment for 5 min. The wetted NWF then was dried at 60°C to evaporate the residual DI water. For sufficient coating, the above‐mentioned dip‐coating and drying processes were repeated several times. In order to achieve optimized performance output, ultrahigh capacitance porous AC was added to the impregnation solution in different weight ratios. Afterward, breathable Zn fabric was manufactured by crisscross weaving Zn wire and silver‐nylon yarns into customized size. Finally, all‐fabric HPG was assembled by sewing NWF, dry‐fit fabric and Zn fabric together into a sandwich‐like hierarchical structure with sewing.

4.3. Characterization and Measurement

The morphologies, structures, roughness, and chemical analysis of CB@PP NWF was characterized by scanning electron microscopy (SEM, Tescan VEGA3, Czech), 3D laser scanning microscopy (VK‐X200, KEYENCE), Fourier transform infrared spectrum (FTIR, Spectrum 100, PerkinElmer), and X‐ray photoelectron spectroscopy (XPS, Thermo Fisher Scientific Nexsa). Water contact angles were measured with 5 µL water droplet by an optical contact angle measuring instrument (SDC‐350, Guangdong Dynetech Intelligent Technology Co., LTD.). Water absorption heatmap was produced based on the water content of each piece of CB@PP NWF. Specifically, the bottom of a piece of CB@PP NWF (1.6 cm × 5.4 cm) was immersed in the water. After the water level stabilized, the CB@PP NWF was taken out and cut into 16 pieces (0.8 cm × 0.8 cm). The water content of each piece was weighed and calculated. Electric potential distribution heatmap was made by recording the electric potential value of each location of CB@PP NWF in a “disconnect‐electrode (DE)” setup. The DE setup features significant separation between the functional hydrovoltaic layer and the Pt counter electrode. The infrared photographs were obtained by a thermal camera (FLIR E95, Teledyne FLIR). Sheet resistance measurements were performed on a digital multifunctional for‐probe tester (ST‐2258A, Suzhou Lattice Electronics Co., Ltd.). Open‐circuit voltages, short‐circuit currents, and voltage/current with external loads were tested by a digital multimeter (Fluke 179, True‐RMS Digital Multimeter). Electrochemical workstation (CHI600E) was employed to collect the cyclic voltammetry (CV), galvanostatic charge/discharge (GCD), and electrochemical impedance spectroscopy (EIS). ORR activity and selectivity were investigated using a rotating ring‐disk electrode (RRDE, Pine AFMSRXE 1523) in oxygen‐saturated 0.1 M Na₂SO₄ at a scan rate of 5 mV s⁻¹. Air‐permeability was tested by an automatic air‐permeability tester (KFS‐AP1, KATO Tech Co., Ltd.). Moisture management testing (MMT) was measured by standardized moisture management tester (SDL ATLAS, Ltd., China) in compliance with the AATCC 195–2009 standard. The heart rate monitoring test was assessed by wearing a heart rate monitor powered by an HPG pack to the chest of a volunteer (one of the authors of this article), and informed written consent was also obtained from the volunteer.

4.4. Computational Methods

Density functional theory (DFT) for periodic structures was implemented in the CASTEP module package. The exchange‐correlation energy was represented through the generalized gradient approximation (GGA) with the exchange‐correlation functional PBE. The cut‐off energy was set to 500 eV, and all structures were relaxed to an energy convergence of 10–5 eV atom⁻¹, a force convergence of 0.03 eV Å⁻¹, and a stress convergence of 0.05 GPa, respectively. The vacuum spacing in a direction perpendicular to the plane of the structure was set to 15 Å for the surface. To avoid the interactions between the periodic structures, a vacuum layer of 15 Å was introduced in the z‐direction of the structure. In addition, the charge density differences of the geometric conformations were investigated to study the electronic structures.

Author Contributions

Y.W. contributed to the conceptualization, investigation, data curation and writing – original draft. B.X. leaded the conceptualization, leaded the methodology, supervised the research work, held the funding acquisition, leaded the review & editing. S.W. and X.L. contributed to material characterizations and simulations. Z.G., and X.W. performed the electrochemical experiments. K.C. and Y.Y. analyzed the data and gave suggestions on applications. B.F. and W.Y contributed to review and editing of the manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1: adma73057‐sup‐0001‐SuppMat.doc.

Supporting File 2: adma73057‐sup‐0002‐Video S1.mp4.

Supporting File 3: adma73057‐sup‐0003‐Video S2.mp4.

Supporting File 4: adma73057‐sup‐0004‐Video S3.mp4.

Supporting File 5: adma73057‐sup‐0005‐Video S4.mp4.

Supporting File 6: adma73057‐sup‐0006‐Video S5.mp4.

Supporting File 7: adma73057‐sup‐0007‐Video S6.mp4.

Supporting File 8: adma73057‐sup‐0008‐Video S7.mp4.

Supporting File 9: adma73057‐sup‐0009‐Video S8.mp4.

Supporting File 10. adma73057‐sup‐0010‐Video S9.mp4.

Data Availability Statement

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