High-voltage hydrovoltaic generator based on micro/nano multi-scale superhydrophilic SiO₂@activated carbon with enhanced capillary infiltration performance†

Luomin Wang^a, Weifeng Zhang*^bc and Yuan Deng*^ad
aSchool of Materials Science and Engineering, Beihang University, Beijing 100191, P. R. China
^bState Key Laboratory of Organic–Inorganic Composites, Beijing University of Chemical Technology, Beijing 100029, P. R. China
^cBeijing Engineering Research Center of Advanced Elastomers, Beijing University of Chemical Technology, Beijing 100029, P. R. China. E-mail: zhangweifeng@buct.edu.cn
^dKey Laboratory of Intelligent Sensing Materials and Chip Integration Technology of Zhejiang Province, Hangzhou Innovation Institute of Beihang University, Hangzhou 310051, P. R. China. E-mail: dengyuan@buaa.edu.cn

First published on 14th July 2025

---

---

New concepts Nowadays, the climate crisis has become a serious problem because of the unabated use of fossil fuels and the ascendance of greenhouse gas levels. Hence, it is of great significance to seek new clean energy sources and electricity generating methods. Among the promising renewable energy sources, water resources on Earth are distributed across diverse forms. With the rapid rise of materials science and innovative technology, hydrovoltaic intelligence has emerged and exhibited a novel concept of generating electricity compared with traditional methods. As a zero-carbon conversion mode with a negative heat energy emission property, it provides a promising prospect of upgrading the mode of water energy use, constructing a renewable energy industry and alleviating environmental issues. However, current research on hydrovoltaic generators predominantly focuses on single-component materials or monoscale architectures, limiting systemic efficiency. Thus, there still exist challenges, including insufficient liquid–solid interface contact and inadequate liquid transport. To address the aforementioned issues, we present a hierarchically structured, multi-scale hydrovoltaic composite material system fabricated via multistep ball milling processes for energy harvesting (an open-circuit voltage of >4.3 V) and environmental monitoring (response to variations in sunlight intensity and wind speed). This study provides insights into the future structural design of high-voltage and multifunctional hydrovoltaic generators.

Introduction

Among the promising renewable energy sources, water resources on Earth are distributed across diverse forms, such as rivers, creeks, lakes, oceans, and glaciers. In recent years, the hydrovoltaic effect, which makes use of the interaction with water for electricity generation via waving, dropping, absorption, flowing and evaporation, has garnered significant attention.^1–8 According to the moving forms of water energy mentioned above, hydrovoltaic effect mechanisms are classified into three fundamental categories: energy harvesting based on the drawing potential of liquids, energy harvesting based on moisture absorption and energy harvesting based on water evaporation.⁹ In contrast to the other two types, energy harvesting based on natural water evaporation, as an emerging field, has now been extensively studied owing to its distinctive advantages: (i) the process is a ubiquitous and spontaneous means of electricity generation without the need for an external power supply. (ii) Generators powered by the evaporation process can reliably convert air thermal energy into sustainable electric power without requiring environments characterized by high humidity. (iii) Evaporating potential and streaming potential synergistically facilitate the electrical output.^10–14 From a microscopic perspective, the wettability, capillary infiltration, porosity, and pore size of active materials greatly impact the solid/liquid interaction.^15,16 Thus, advancements in the fabrication technologies of materials and structural regulation strategies are imperative to design effective hydrovoltaic materials.

Over the past decade, advances in harvesting energy from water evaporation have been propelled by the development of functional materials, such as nanostructured carbon materials^17–21 and solid oxide nanomaterials.^22–25 As a typical example, Guo et al. found that an oxidized carbon black sheet could generate a sustained voltage of about 1 V by evaporation-driven water flowing in nanochannels.²⁶ Subsequent innovations by Qu's group developed scalable painting and blade coating approaches to expand the production of flexible hydrovoltaic films based on SiO₂,²⁷ which achieved a voltage output of 1.7 V and a current output of 450 nA. Most recently, Zhang and his team engineered a hydrovoltaic electricity generator via drop-casting an Al₂O₃/polyvinylidene fluoride solution,²⁸ attaining a peak short-circuit current density of 160 μA cm⁻² and an open-circuit voltage of 0.7 V. Despite these advancements, current research on hydrovoltaic energy harvesting predominantly focuses on single-component materials or monoscale architectures, limiting systemic efficiency. Thus, a critical challenge arises from the inverse relationship between nanochannel ion selectivity and flow resistance. Although reduced channel dimensions enhance ion selectivity, they concomitantly elevate ion flow resistance and diminish voltage output.^29,30 This trade-off underscores the need for hierarchical material design-integrating multiscale porosity, tunable surface chemistry, and optimized nanofluidic pathways to decouple selectivity from transport losses. Overall, solving these issues is essential for scaling hydrovoltaic generators toward practical applications.

To address the aforementioned tradeoff of the hydrovoltaic generator, we present an innovative structural design for evaporation-driven hydrovoltaic materials. By compositing micron-scale activated carbon with nano-scale silicon dioxide via multistep ball milling processes (SiO₂@activated carbon, namely, S@AC), the as-prepared material can achieve rapid capillary pumping-transporting-evaporating coupling of water. Further thermal annealing and plasma treatment introduce oxygen functional groups on the material, thus enhancing the ion/proton mobility, surface energy and hydrophilicity.³¹ The resulting multi-scale hydrovoltaic generator (MS-HG) achieves a sustainable voltage of up to 4.3 V under ambient conditions, while it can also act as a responsive environmental sensor for real-time monitoring of various wind speeds and sunlight intensities. Notably, the output performance of the devices can be further scaled up easily through the series or parallel connections of multiple devices. Meanwhile, the electricity-generating performance under different conditions, such as relative humidity (RH), wind, water level, length and component, is further investigated. Furthermore, to deeply understand the underlying mechanism, we systematically study the capillary infiltration, hydrophilicity, effective zeta potential and other surface properties of the active layer during the spontaneous evaporation process of the water in the S@AC film. This work optimizes liquid transport dynamics, establishes a new framework for hydrovoltaic material design, and offers a pathway to overcome obstacles that further improve electrical performance.

Results and discussion

Design and fabrication of MS-HG

A schematic representation of the design of the MS-HG is demonstrated in Fig. 1. The fabricated MS-HG mainly consisted of a bottom electrode, hydrovoltaic active layer and top electrode, and an alumina strip was used as the substrate. First, two electrodes were printed on a cleaned Al₂O₃ substrate using a multiwalled carbon nanotube (MWCNT) ink. Then, a slurry composed of activated carbon (AC), silicon dioxide nanoparticles, polyvinylidene fluoride (PVDF), silane coupling agent KH-550 and N-methyl-2-pyrrolidone (NMP) was coated on the Al₂O₃ substrate crossing the two MWCNT electrodes. Given that the electrical output of materials is generally positively correlated with their zeta potential, which is an important and reliable indicator of surface charge, we selected silicon dioxide nanoparticles owing to their negatively charged surface, which is similar to that of activated carbon.²⁷ It is worth noting that the slurry was prepared through multistep ball milling processes, including sequential dry and wet ball milling (D + W). Initially, dry ball milling was employed to achieve a uniform particle size distribution of AC and to ensure homogeneous dispersion of SiO₂. This was followed by wet ball milling, which facilitated thorough mixing of the slurry components. Afterward, the device was thermally annealed at 370 °C for 2.5 h in an ambient atmosphere and cooled spontaneously to room temperature. Subsequently, the device was wired and precisely encapsulated by polydimethylsiloxane (PDMS) in the exposed electrode regions. Finally, the obtained device was treated with oxygen plasma for 1 min at 130 W to optimize its surface properties. For electrical characterization, the bottom end of the film was inserted into a container with deionized (DI) water, while a sustaining liquid flow was established in the film driven by capillary force and water evaporation. Based on the hydrovoltaic mechanisms, the dissociation of oxygen-containing functional groups on the film surface induced a negative surface potential, imparting the channels with ion selectivity to repel the hydroxide ions in the water.³² Thus, excess hydronium ions accumulated at the top of the channel form a high-potential region. In terms of hydrovoltaic capability, the MS-HG based on micro/nano superhydrophilic S@AC provides abundant overlapped electric double layers (EDLs) and enhances capillary infiltration performance, holding considerable significance for efficient energy harvesting from water as well as for environmental monitoring.

Structure and characterization of a hydrovoltaic generator

Fig. 2A depicts the structure of the MS-HG. Two terminals of the film were electrically contacted by patterned electrodes with silver wire for the measurement of the output voltage. To further evaluate the binding force and stability of the hydrovoltaic coating, a high-velocity water flow was used to impact the S@AC film. As illustrated in Fig. 2B, when water flow impinged on the film, a large amount of water splashed. Notably, after repeated impacts, the film still maintained its structural integrity without observable changes. Furthermore, we demonstrated that the S@AC-based MS-HG could generate electricity continuously with an open-circuit voltage (V_oc) of ∼4 V under ambient room conditions (25 °C and 35% RH), as illustrated in Fig. 2C. In contrast, the steady V_oc for AC(D + W)-based and AC(W)-based HGs are approximately 0.25 and 0.11 V, respectively (the inset in Fig. 2C). A reasonable explanation is that the microscopic structure of the material affects the output performance. When water is forced to flow in the narrow channels or hierarchical micro/nano channels, the special wettability and capillary effects can be amplified obviously to promote liquid transport and electrical output. Hence, a dry–wet ball milling strategy was employed. The scanning electron microscopy (SEM) image shown in Fig. 2D reveals that the film prepared via direct wet ball milling consists of interconnected activated carbon particles ranging from 5 to 15 μm in diameter (Fig. 2E). The SEM image (Fig. 2F) indicates that the film prepared using a combination of dry ball milling for particle size uniformity and wet ball milling for slurry synthesis is composed of interconnected activated carbon with diameters ranging from 0.5 to 1.5 μm (Fig. 2G). These results suggest that the AC particles subjected to dry ball milling exhibit distinct morphology compared to their pristine counterparts, as evidenced by a pronounced reduction in particle size (comprehensive data are shown in Fig. S1, ESI†). To further narrow the channel size, SiO₂ nanoparticles were incorporated and uniformly dispersed via dry ball milling (Fig. S2, ESI†). Subsequently, a multi-scale S@AC film with an average thickness of about 35.7 μm was prepared via wet ball milling (Fig. 2H and Fig. S3, ESI†). Here, dry ball milling was employed to reduce particle size and ensure the homogeneous dispersion of SiO₂, while wet ball milling was employed to mix the slurry components to form a hydrovoltaic film. Therefore, as depicted in Fig. 2I, the interconnected particle diameters in this hierarchical structure are predominantly less than 1 μm, with uniform SiO₂ distribution confirmed by elemental scanning and mapping of Si (the bottom diagram in Fig. 2J). Here, the tiny amounts of Si observed in the top diagram (Fig. 2J) arise from the addition of the silane coupling agent KH-550 to the slurry. X-ray photoelectron spectroscopy (XPS) characterizations not only verify the existence of SiO₂ in the S@AC film but also reveal the presence of various oxygen-containing functional groups within the hydrovoltaic film, such as C–OH, C–O–C, CO and OC–OH groups (Fig. 2K, L and Fig. S4, ESI†). Consequently, the intrinsic structure and surface properties of the film are crucial to output performance.

Electricity generation mechanism of MS-HG

Pressure-driven ionic flow results in charge accumulation differences, yielding a streaming potential.^33–35 As illustrated in Fig. 3A, when liquid flows through a narrow channel with negative surface charges, an EDL is formed at the solid–liquid interface, and the counterions in the liquid are transported through channels owing to electrostatic interaction, leading to a streaming potential. The Debye screening length (λ_D) is an important parameter that indicates the influential range of EDL.^36,37 In the classical Gouy–Chapman model, the λ_D is inversely proportional to the ionic concentration of the solution, which decides the ion selectivity of the channels and affects the corresponding electrical output.^38,39 The λ_D can be calculated as follows:⁴⁰

	(1)

where ε_r is the dielectric constant of water, ε₀ is the permittivity of vacuum, k_B is the Boltzmann constant, T is the absolute temperature, e is the charge of an electron, Z_i is the valence of ion species i, and n^B_i is the bulk concentration of ion species i. Therefore, the theoretical calculation reveals λ_D of DI water is approximately 1 μm, exceeding the nanoscale dimensions of interstitial channels between S@AC particles, which suggests that the channels can efficiently transport counterions owing to an overlapped EDL with a high counterion permselectivity (Note S1 and Fig. S5, ESI†). Among them, nano-scale silicon dioxide enables more overlapped EDLs in hydrovoltaic materials to promote convective charge transport in a channel to boost streaming potential. Moreover, for wetting behavior, the incorporation of nano-scale silicon dioxide significantly augments the intrinsic wettability of micron-scale activated carbon, facilitating spontaneous water infiltration and directional flow alongside the film, thereby promoting the electrical output power (Fig. S6, ESI†). Aside from DI water, we further explored the electricity generation performance of MS-HG in the NaCl aqueous solution to confirm the dependence of voltage on ionic concentration (Fig. S7, ESI†). The potential output decreases as the ionic concentration increases from 10⁻⁷ to 10⁰ M because the number of charges in the solution exhibits a significant increase accompanied by a dramatic decrease in λ_D according to eqn (1).

Apart from the streaming potential, direct liquid evaporation in the precursor film ahead of the capillary front induces a more pronounced evaporating potential. As illustrated in Fig. 3B, when water molecules evaporate from the solid–liquid–gas three-phase interface, accumulated electrons on the upper surface of the S@AC film recombine with the departing water molecules, increasing the local hole concentration and thereby generating an evaporating potential.⁴¹ Specifically, for zeta potential, the ionization of intrinsic-functional-groups in aqueous solution negatively charges the surface of micron-scale activated carbon and nano-scale silicon dioxide, leading to an electron transfer from water molecules to the substrate at the wetted region of the film. Additionally, the higher zeta potential is useful for capturing ambient heat through natural evaporation, so micron-scale activated carbon and nano-scale silicon dioxide, which have the same negatively charged surface, are beneficial for enhancing the evaporating potential. Similarly, bulk water enters micro/nano-structured superhydrophilic materials, becomes confined/bound water through wetting and capillary effects, and eventually turns into vapor after the water desorption process, contributing further to the generation of evaporative potential. Consequently, we attribute the power generation contribution of the MS-HG to two main categories: streaming potential and evaporating potential. To fully clarify the contributions of the two hydrovoltaic potentials, we designed a specific three-electrode experiment to verify the coexistence of streaming potential and evaporating potential, as shown in Fig. 3C (left). In this setup, the bottom electrode is inserted into DI water, while the remaining electrodes are kept above the bulk water level, and the saturated voltages are measured between each pair of electrodes, V_I–II, V_I–III and V_II–III are depicted in Fig. 3C (right). Each voltage measurement was recorded for about 1000 s. The voltage between the two highest electrodes V_II–III is around 0.5 V primarily because the evaporation phenomenon causes carrier transfer in the precursor film, while streaming potential plays an ignorable role. However, the rest of the S@AC film shows a sustainable voltage (V_I–III) of up to 2.1 V, as both streaming and evaporating potentials contribute. The streaming potential emerges from the electrokinetic effect in the fully-wetted region, while the evaporating potential arises from the charge transfer in the partially wetted region. Notably, the summation of V_II–III and V_I–III is approximately equal to the measured V_I–II, confirming the additive nature of the two hydrovoltaic potentials. When the MS-HG is sealed in a beaker (Fig. S8, ESI†) or the S@AC film is fully immersed in DI water (Fig. S9, ESI†), the voltage response is negligible, indicating that possible electrochemical reaction at the interface of active materials and electrodes contributes little to the induced voltage; therefore, the potential arises exclusively from steady water streaming and evaporating in our experiments.

To assess water adsorption behavior, sequential optical images were captured during water contact angle testing, enabling a comparative analysis of the water absorption and transportation dynamics in the thermal annealing-treated and plasma processing-treated S@AC films. As displayed in Fig. 3D (top), compared to the original hydrophobic film with a water contact angle of approximately 141° (Fig. S10, ESI†), the thermal annealing-treated S@AC film took 2 s for DI water to reach a contact angle of 10° and 48 s to stabilize at 0° after contacting the surface, demonstrating its good hydrophilicity. In contrast, plasma processing-treated S@AC film swiftly and completely absorbed a water drop in about 3 s, exhibiting its superhydrophilicity (Fig. 3D, bottom). To further prove the water absorption capability under a natural environment, water sorption kinetics display that the S@AC film affords a water-adsorbing capacity of up to 30% under testing conditions of ∼80% RH and 25 °C (Fig. 3E). When the bottom end of the film is contacted with bulk water, a thin water film spontaneously climbs along the S@AC material through transpiration and capillary force. As shown in Fig. 3F, the real-time infrared thermal images show the dynamic rise of the capillary front in different films. The pristine film exhibited a slower water transport rate, with the capillary front climbing half the film after 3 min and achieving equilibrium (left). In contrast, both the annealing-treated (middle) and the plasma-treated (right) films exhibited rapid water transport, with the capillary front reaching the top of the film within the same time frame. Zeta potential measurements reveal that the S@AC film treated by thermal annealing and plasma processing displays negative charges, which are the fundamental property for power generation (Fig. 3G). In addition, the zeta potential of the original material is shown in Fig. S11 (ESI†). Once water flows through these negatively charged channels, the channels exhibit a high ion selectivity, leading to the development of a potential difference.⁴² As illustrated in Fig. 3H, an annealing-treated MS-HG can induce a V_oc of ∼0.4 V. Hence, all the above results provide support for the proposed electricity generation mechanism for MS-HG.

Performance and influencing factors of MS-HG

Since the intrinsic structure and geometric configuration critically influence output performance, the issue of SiO₂ nanoparticle content within the S@AC film is a key factor. The content of SiO₂ nanoparticle was regulated from 0 to 10 wt% in the S@AC film, and the corresponding V_oc was further tested. As shown in Fig. 4A, a small amount of SiO₂ nanoparticles exerts a negligible impact on nanostructure formation and hydrophilicity, thereby failing to enhance voltage output. However, as the SiO₂ nanoparticle content continuously increases to 4 wt%, the V_oc reaches a maximum of 3.4 V, suggesting a dominant role of SiO₂ nanoparticles in optimizing device performance. Beyond this threshold, excessive nanoparticle incorporation increases the flow resistance of the channel and introduces cracks, ultimately reducing the performance of MS-HG. In addition, the length of the S@AC film between the top and bottom electrodes was adjusted from 2 to 6 cm. As the length increases, the V_oc exhibits a continuous enhancement from 23 mV to 4.4 V likely attributable to the synergistic effect of accelerated water evaporation at higher capillary heights and accumulation of potential differences (Fig. 4B). Meanwhile, the short-circuit current density initially rises, reaching its peak at a film length of 3 cm, but then decreases as the length increases. The short-circuit current density (J_sc) can be defined as follows:

	(2)

where I_sc is the short-circuit current and S = w·h is the thickness area (w is the width and h is the thickness). This trend is ascribed to an extended hydrophilic region that contains a more efficient water flow for hydrovoltaic energy harvesting and introduces a greater internal resistance. Based on the comprehensive evaluation of the electricity generation performance, an optimal device dimension of 0.5 × 3 cm² was chosen in the following experiments. Owing to the poor interfacial adhesion between the thicker S@AC film and the underlying substrate, its influence was not explored further. The output performance of MS-HG is displayed in Fig. 4C. As the external load resistance increases from 1 to 10⁶ MΩ, the output voltage correspondingly rises from nearly zero to 3.5 V, while the current density gradually declines from 2.9 μA cm⁻² to nearly zero. The maximum output power can be expressed using the formula:⁴³

Pmax = Vmax × Jmax,	(3)

with a maximum power density of 1 μW cm⁻² when the load resistance is ∼5 × 10⁴ MΩ. The electrical output performance of MS-HG provides a foundation for its practical implementation in energy harvesting applications.

To further optimize the performance of the MS-HG, the possible environmental factors influencing its operation are systematically explored. As shown in Fig. 4D, the corresponding voltage of MS-HG was tested at different water levels, with the bottom electrode consistently submerged. With an increase in the water level, the effective evaporation area becomes shorter, so the voltage of the device decreases. Fig. 4E illustrates the strong correlation between the evaporation dynamics of water absorbed on the S@AC film and electricity generation. The V_oc immediately decreases from 2.8 V to the background voltage upon cessation of airflow. In contrast, the V_oc gradually returns to its original level when airflow is reinstated and is repeatable when the airflow is periodically switched on/off. Additionally, the relative humidity inside the glove box is monitored using a commercial hygrometer (Fig. S12, ESI†). As illustrated in Fig. 4F, the measured voltage output remains stable at around 4.5 V at 20% RH and 26 °C, while a drastic drop in voltage output is observed with increasing ambient humidity. This phenomenon may be attributed to the suppression of evaporation rates under higher humidity conditions. These findings suggest that improving the water evaporation effect can enhance performance output.

Electrical output, scalable integration and application of MS-HG

As shown in Fig. 5A, the MS-HG demonstrates prolonged operational stability under ambient conditions, with consistent performance observed over periods spanning several weeks to one year. A slight decline is detected beyond the seventh week, which probably originates from seasonal temperature variations (corresponding environmental test conditions shown in Table S1, ESI†). Notably, the maximum recorded voltage for a single device reaches approximately 4.3 V, remarkably outperforming previously reported hydrovoltaic devices made of different materials (Fig. 5B).^{17,26,27,29,44–49} The specific V_oc and dimension parameters of the hydrovoltaic devices in previous reports are depicted in Table S2 (ESI†). In addition to analyzing the impact of environmental factors, the correlation between sunlight intensity, wind speed, and the output voltage of MS-HG provides critical insights into its potential applications. As illustrated in Fig. 5C, when the sunlight intensity increases from 600 to 1100 W m⁻², the V_oc of MS-HG increases from 1.25 to 3.15 V correspondingly. The real-time voltage variation curve increases dramatically at first and then slowly with an increase in sunlight intensity. This is mainly because of the limited water transport capacity under photothermal conversion.^50,51 Furthermore, MS-HG displays a rapid response to variations in airflow velocity (Fig. 5D), with an increase in V_oc from 1.1 V at a wind speed of 1.5 m s⁻¹ to 2.5 V at 3.5 m s⁻¹ under ambient conditions. Therefore, these findings highlight the dual functionality of MS-HG, not only as an energy-harvesting device but also as a self-powered sensor for sensitively monitoring environmental fluctuations, including light intensity and airflow.^52–54 The scalability and continuous power generation capability of the MS-HG arrays are summarized in Fig. 5E and F. The insets show the circuit configuration. As anticipated, when 2 and 3 similar devices are connected in a series, the voltages reach ∼7.7 and 11.5 V, respectively. When 2 and 3 similar devices are connected in parallel, the current densities reach ∼5, and 6.4 μA cm⁻², respectively. Meanwhile, the V_oc of three devices in parallel is about 3.8 V, as presented in Fig. S13 (ESI†), which is approximately equal to the voltage of a single device. The distance between series- and parallel-connected devices was 4.5 cm, a spacing sufficient to avoid water evaporation interference because each device was individually isolated within a single beaker. The electrical output of MS-HGs enlarges linearly with the number of units connected in arrays, with negligible fluctuations attributed to device heterogeneity. These results ascertain the feasibility of MS-HG for scalable energy applications.

Conclusions

In summary, micro/nano multi-scale superhydrophilic hydrovoltaic SiO₂@activated carbon (S@AC) material with enhanced capillary infiltration performance is fabricated using multistep ball milling processes. Comprehensive experimental results confirm that power output arises from both streaming potential and evaporating potential. Correspondingly, the influencing principles of capillary infiltration dynamics, ion transport, hydrophilicity, zeta potential and other surface properties on the hydrovoltaic effect are discussed systematically. Notably, a single S@AC-based multi-scale hydrovoltaic generator (MS-HG) with dimensions of 0.5 × 3 cm² can generate a sustainable voltage exceeding 4.3 V under ambient conditions. The electrical performance of MS-HG can be modulated by environmental factors that affect water evaporation. MS-HG possesses higher voltage output with low water level, high airflow, low RH, and high sunlight intensity. Therefore, the demonstrated versatility of the MS-HG, which encompasses both energy harvesting and environmental monitoring, supports its potential as a scalable power source and an environmental sensor for meeting diverse requirements. Consequently, this multi-scale S@AC material system breaks through the limitations of liquid transport, thereby providing new opportunities for the advancement of hydrovoltaic intelligence.

Experimental section

Materials

Alumina (Al₂O₃) strips were purchased from New Materials Technology Co., Ltd. Multiwalled carbon nanotube (MWCNT) ink was purchased from Shenzhen Hongdachang Evolutionary Technology Co., Ltd. Activated carbon (AC) was purchased from Foshan Porous Carbon Tech Co., Ltd. Silicon dioxide (SiO₂) nanoparticles, silane coupling agent KH-550, N-methyl-2-pyrrolidone (NMP) and ethanol were purchased from Sinopharm Chemical Regent Co., Ltd. Polyvinylidene fluoride (PVDF) was purchased from Arkema. Polydimethylsiloxane (PDMS), Sylgard 184 base, and curing agents were purchased from Dow Corning, America. Silver (Ag) wire was purchased from Shanghai Xinxi Alloy Material Co., Ltd. Conductive carbon tape (no. 7321) was purchased from Nisshin EM Co., Ltd.

Synthesis of AC(W) slurry

Typically, 0.14 g of pristine AC was mixed with PVDF, KH-550 and NMP with a weight ratio of 1:3:2:4, followed by wet ball milling for 160 min to prepare a slurry. The revolution speed of ball milling is 560 rpm, and the size of the agate milling media is 12 mm, 10 mm and 6 mm.

Synthesis of AC(D + W) slurry

Initially, pristine AC underwent dry ball milling using ZrO₂ milling media (15 mm, 12 mm, 10 mm, 8 mm, 5 mm and 2 mm) at 560 rpm for 4 h to achieve uniform particle size distribution. Then, the treated AC was mixed with PVDF, KH-550 and NMP according to the above ratio. This mixture was subsequently subjected to wet ball milling for 160 min to produce a homogeneous slurry.

Synthesis of S@AC(D + W) slurry

Typically, the treated AC was blended with 2–10 wt% SiO₂ nanoparticles via dry ball milling for 128 min to ensure uniform dispersion, followed by wet ball milling with PVDF, KH-550 and NMP using agate milling media in a 100 mL vial for 160 min to achieve thorough mixing.

Fabrication of S@AC-based MS-HG

Al₂O₃ strips were sequentially cleaned ultrasonically in ethanol and DI water, followed by drying with a dry gas flow. Then, the MWCNT ink was screen-printed onto the Al₂O₃ strips, which served as the two electrodes. Subsequently, the S@AC(D + W) slurry was deposited into the desired patterns, overlapping the two MWCNT electrodes with a floating knife coater. Afterwards, the Al₂O₃ strip coated with the S@AC(D + W) slurry was annealed at 370 °C for 2.5 h in a muffle furnace and cooled naturally to room temperature. The MS-HG was then wired and encapsulated with the exposed electrode areas meticulously covered by PDMS. Finally, to further enhance the hydrophilicity of the S@AC film, the MS-HG was treated with oxygen plasma (pressure 100 Pa, RF power 130 W) for 1 min by a plasma cleaner system.

Characterization and measurement

The morphology and cross-sectional image of the film were characterized through scanning electron microscopy (SEM, Sigma 300, Zeiss, Germany) at 10 kV. The element distribution was analyzed using EDS (X-act SDD, Oxford Instruments). X-ray photoelectron spectroscopy (XPS) analyses were carried out using a Thermo Scientific K-Alpha (Thermo Fisher Scientific, USA), and the spectra were analyzed using Avantage software. The water contact angle was characterized by applying optical contact angle measurement equipment (SL250, KINO, USA). Dynamic vapor sorption (DVS) isotherm analysis was performed using a DVS Resolution (Surface Measurements Systems, UK). Infrared photos were taken using an infrared camera (T650sc, Teledyne FLIR). The zeta potential of the film was obtained with a zeta potential analyzer (Malvern Nano ZS90). The voltage and current signals were recorded in real-time using a Keithley 2400 source meter. The ambient temperature and humidity were controlled using a digital thermo-hygrometer manufactured by Shandong Renke Control Technology Co., Ltd. Sunlight was provided using a Tri-Sol Solar Simulator System (TSS-100, OAI, USA). Solar density was measured using a radiometer (SM206-SOLAR, Shenzhen Sanpo Instrument Co., Ltd). The wind at different speeds was stimulated by an axial flow fan and standardized with an anemometer (UT362, UNI-TREND Technology Co., Ltd, China).

Author contributions

Luomin Wang: conceptualization, methodology, data curation, visualization, investigation, formal analysis, writing – original draft; Weifeng Zhang: conceptualization, methodology, formal analysis, funding acquisition, project administration, writing – review & editing, supervision; Yuan Deng: conceptualization, resources, writing – review & editing, supervision, validation.

Conflicts of interest

The authors declare that they have no conflicts of interest.

Data availability

The data that support the findings of this study are available in the ESI,† of this article.

Acknowledgements

The authors gratefully acknowledge the financial support for this work from the Zhejiang Provincial Natural Science Foundation of China (grant no. LZ23E020004), the Fundamental Research Funds for the Central Universities (grant no. buctrc202506, JD2516) and the National Natural Science Foundation of China (grant no. 52003015).

References

1. J. Yin, Z. Zhang, X. Li, J. Yu, J. Zhou, Y. Chen and W. Guo, Nat. Commun., 2014, 5, 3582 CrossRef CAS PubMed .
2. L. Li, X. Li, W. Deng, C. Shen, X. Chen, H. Sheng, X. Wang, J. Zhou, J. Li, Y. Zhu, Z. Zhang, J. Yin and W. Guo, Sci. Adv., 2023, 9, eadi2993 CrossRef CAS PubMed .
3. Z. Sun, L. Feng, X. Wen, L. Wang, X. Qin and J. Yu, Mater. Horiz., 2021, 8, 2303–2309 RSC .
4. J. Chen, X. Zhang, M. Cheng, Q. Li, S. Zhao, M. Zhang, Q. Fu and H. Deng, Mater. Horiz., 2025, 12, 2309–2318 RSC .
5. Y. Zhang, Z. Yu, H. Qu, S. Guo, J. Yang, S. Zhang, L. Yang, S. Cheng, J. Wang and S. C. Tan, Adv. Mater., 2024, 36, 2208081 CrossRef CAS PubMed .
6. B. Fan, A. Bhattacharya and P. R. Bandaru, Nat. Commun., 2018, 9, 4050 CrossRef CAS PubMed .
7. S. Ghosh, A. K. Sood and N. Kumar, Science, 2003, 299, 1042–1044 CrossRef CAS PubMed .
8. N. He, H. Wang, F. Li, B. Jiang, D. Tang and L. Li, Energy Environ. Sci., 2023, 16, 2494–2504 RSC .
9. L. Wang, W. Zhang and Y. Deng, ACS Nano, 2023, 17, 14229–14252 CrossRef CAS PubMed .
10. Z. Zhang, X. Li, J. Yin, Y. Xu, W. Fei, M. Xue, Q. Wang, J. Zhou and W. Guo, Nat. Nanotechnol., 2018, 13, 1109–1119 CrossRef CAS PubMed .
11. Y. Yang, L. Yang, F. Yang, W. Bai, X. Zhang, H. Li, G. Duan, Y. Xu and Y. Li, Mater. Horiz., 2023, 10, 268–276 RSC .
12. J. Yang, J. Yu, K. Zhang, F. Qiao and G. Yu, Adv. Mater. Technol., 2024, 9, 2301602 CrossRef CAS .
13. L. Wang, H. Wang, C. Wu, J. Bai, T. He, Y. Li, H. Cheng and L. Qu, Nat. Commun., 2024, 15, 4929 CrossRef CAS PubMed .
14. R. Yuan, H. Li, Z. Zhao, A. Li, L. Xue, K. Li, X. Deng, X. Yu, R. Li, Q. Liu and Y. Song, Nat. Commun., 2024, 15, 9796 CrossRef CAS PubMed .
15. J. Tan, X. Wang, W. Chu, S. Fang, C. Zheng, M. Xue, X. Wang, T. Hu and W. Guo, Adv. Mater., 2024, 36, 2211165 CrossRef CAS PubMed .
16. J. Lin, Z. Zhang, X. Lin, X. Cai, S. Fu, X. Fang, Y. Ding, X. Wang, G. Sèbe and G. Zhou, Adv. Funct. Mater., 2024, 34, 2314231 CrossRef CAS .
17. T. Ding, K. Liu, J. Li, G. Xue, Q. Chen, L. Huang, B. Hu and J. Zhou, Adv. Funct. Mater., 2017, 27, 1700551 CrossRef .
18. T. G. Yun, J. Bae, A. Rothschild and I. D. Kim, ACS Nano, 2019, 13, 12703–12709 CrossRef CAS PubMed .
19. J. Tan, S. Fang, Z. Zhang, J. Yin, L. Li, X. Wang and W. Guo, Nat. Commun., 2022, 13, 3643 CrossRef CAS PubMed .
20. S. Fang, W. Chu, J. Tan and W. Guo, Nano Energy, 2022, 101, 107605 CrossRef CAS .
21. L. Li, M. Hao, X. Yang, F. Sun, Y. Bai, H. Ding, S. Wang and T. Zhang, Nano Energy, 2020, 72, 104663 CrossRef CAS .
22. L. Li, S. Feng, Y. Bai, X. Yang, M. Liu, M. Hao, S. Wang, Y. Wu, F. Sun, Z. Liu and T. Zhang, Nat. Commun., 2022, 13, 1043 CrossRef CAS PubMed .
23. S. G. Yoon, Y. Yang, J. Yoo, H. Jin, W. H. Lee, J. Park and Y. S. Kim, ACS Appl. Electron. Mater., 2019, 1, 1746–1751 CrossRef CAS .
24. H. Xiao, J. Zhu, L. Ding, J. Zheng, C. Liu, B. Du, S. Chen and Y. Wang, Adv. Funct. Mater., 2024, 34, 2407669 CrossRef CAS .
25. W. Deng, G. Feng, L. Li, X. Wang, H. Lu, X. Li, J. Li, W. Guo and J. Yin, Energy Environ. Sci., 2023, 16, 4442–4452 RSC .
26. G. Xue, Y. Xu, T. Ding, J. Li, J. Yin, W. Fei, Y. Cao, J. Yu, L. Yuan, L. Gong, J. Chen, S. Deng, J. Zhou and W. Guo, Nat. Nanotechnol., 2017, 12, 317–321 CrossRef CAS PubMed .
27. C. Shao, B. Ji, T. Xu, J. Gao, X. Gao, Y. Xiao, Y. Zhao, N. Chen, L. Jiang and L. Qu, ACS Appl. Mater. Interfaces, 2019, 11, 30927–30935 CrossRef CAS PubMed .
28. K. Ni, B. Xu, Z. Wang, Q. Ren, W. Gu, B. Sun, R. Liu and X. Zhang, Adv. Mater., 2023, 35, 2305438 CrossRef CAS PubMed .
29. L. Li, Z. Zheng, C. Ge, Y. Wang, H. Dai, L. Li, S. Wang, Q. Gao, M. Liu, F. Sun and T. Zhang, Adv. Mater., 2023, 35, 2304099 CrossRef CAS PubMed .
30. C. Ge, Y. Wang, M. Wang, Z. Zheng, S. Wang, Y. Kong, Q. Gao, M. Liu, F. Sun, L. Li and T. Zhang, Adv. Mater., 2024, 36, 2310260 CrossRef CAS PubMed .
31. J. Li, K. Liu, G. Xue, T. Ding, P. Yang, Q. Chen, Y. Shen, S. Li, G. Feng, A. Shen, M. Xu and J. Zhou, Nano Energy, 2018, 48, 211–216 CrossRef CAS .
32. M. A. Brown, A. Goel and Z. Abbas, Angew. Chem., Int. Ed., 2016, 55, 3790 CrossRef CAS PubMed .
33. W. L. Hsu, H. Daiguji, D. E. Dunstan, M. R. Davidson and D. J. E. Harvie, Adv. Colloid Interface Sci., 2016, 234, 108–131 CrossRef CAS PubMed .
34. F. H. J. van der Heyden, D. Stein and C. Dekker, Phys. Rev. Lett., 2005, 95, 116104 CrossRef PubMed .
35. F. H. J. van der Heyden, D. J. Bonthuis, D. Stein, C. Meyer and C. Dekker, Nano Lett., 2007, 7, 1022–1025 CrossRef CAS PubMed .
36. M. M. Kohonen, M. E. Karaman and R. M. Pashley, Langmuir, 2000, 16, 5749–5753 CrossRef CAS .
37. J. Wu, Chem. Rev., 2022, 122, 10821–10859 CrossRef CAS PubMed .
38. R. Parsons, Chem. Rev., 1990, 90, 813–826 CrossRef CAS .
39. R. H. Tunuguntla, R. Y. Henley, Y.-C. Yao, T. A. Pham, M. Wanunu and A. Noy, Science, 2017, 357, 792–796 CrossRef CAS PubMed .
40. A. M. Smith, A. A. Lee and S. Perkin, J. Phys. Chem. Lett., 2016, 7, 2157–2163 CrossRef CAS PubMed .
41. S. Fang, J. Li, Y. Xu, C. Shen and W. Guo, Joule, 2022, 6, 690–701 CrossRef .
42. Q. Liu, P. Guo, X. Zhang, H. Ren, J. Liang, Q. Li, W. Tang and W. Wu, Adv. Sci., 2025, 12, 2413779 CrossRef CAS PubMed .
43. F. Zhao, Y. Liang, H. Cheng, L. Jiang and L. Qu, Energy Environ. Sci., 2016, 9, 912–916 RSC .
44. Y. Chen, J. He, C. Ye and S. Tang, Adv. Energy Mater., 2024, 14, 2400529 CrossRef CAS .
45. M. Wu, Z. Liang, M. Peng, B. Zhao, D. Li, J. Zhang, Y. Sun and L. Jiang, Nano Energy, 2023, 116, 108771 CrossRef CAS .
46. B. Shao, C. Xing, Y. Song, C. Jiang, G. Bai, T. Song, R. Liu, Y. Wang and B. Sun, Nano Energy, 2023, 106, 108081 CrossRef CAS .
47. Y. T. Wu, H. Q. Huang, W. M. Zhou, C. T. You, H. L. Ye, J. Chen, S. Zang, J. H. Yun, X. Q. Chen, L. W. Wang and Z. H. Yuan, ACS Appl. Mater. Interfaces, 2022, 14, 29099–29110 CrossRef CAS PubMed .
48. Q. Ma, Q. He, P. Yin, H. Cheng, X. Cui, Q. Yun and H. Zhang, Adv. Mater., 2020, 32, 2003720 CrossRef CAS PubMed .
49. H. X. He, T. M. Zhao, H. Y. Guan, T. Y. Zhong, H. Zeng, L. L. Xing, Y. Zhang and X. Y. Xue, Sci. Bull., 2019, 64, 1409–1417 CrossRef CAS PubMed .
50. Z. Sun, C. Han, S. Gao, Z. Li, M. Jing, H. Yu and Z. Wang, Nat. Commun., 2022, 13, 5077 CrossRef CAS PubMed .
51. L. Zhu, M. Gao, C. K. N. Peh and G. W. Ho, Mater. Horiz., 2018, 5, 323–343 RSC .
52. T. Zhong, H. Guan, Y. Dai, H. He, L. Xing, Y. Zhang and X. Xue, Nano Energy, 2019, 60, 52–60 CrossRef CAS .
53. J. Yang, S. Wu, J. Yu, Y. Deng, F. Qiao and K. Zhang, Diamond Relat. Mater., 2024, 148, 111401 CrossRef CAS .
54. X. Zhang, J. Guo, Z. Zou, B. Luo, N. Li, Z. Yang, S. Xiong, X. Wang, Y. Xu and Y. Li, Mater. Horiz., 2025, 12, 5211–5224 RSC .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5mh01101a

This journal is © The Royal Society of Chemistry 2025