Triboelectric nanogenerator based on a water droplet spring with a concave spherical surface for harvesting wave energy and detecting pressure

1 Introduction

Micro and nano energy being harvested and utilized is an important research direction in recent years. The earth’s surface is 510 million km², of which the oceans cover about 71%, containing abundant wave energy [1]. The extraction of wave energy represents an environmentally sound and sustainable method for electricity generation, exhibiting a remarkable contrast in terms of pollutant emissions when compared to conventional fossil fuel-based power plants. This inherent characteristic not only underscores its potential as a clean energy source but also highlights its instrumental role in preserving the delicate balance and ecological integrity of marine ecosystems. At present, harvesting wave energy is mainly based on electromagnetic generators; however, they are not suitable for collecting low-frequency wave energy in distant seas and have high maintenance costs. Prof. Wang et al. have proposed a novel harvesting energy method based on electromagnetic induction, which is called a triboelectric nanogenerator (TENG) [2]. It is widely studied due to its simple design and low manufacturing cost. TENG can be used not only to harvest energy (wave energy, wind energy, and acoustic energy) [3,4,5,6] but also to detect/monitor human physiological activities [7,8,9], ship attitude [10,11,12], water level monitoring [3], chemical sensing [13,14], etc.

TENG is mainly divided into solid–solid [5,15,16,17,18,19,20,21,22,23] and solid–liquid [1,24,25,26,27,28] modes. On the one hand, the latter has less wear due to liquid lubricity and fluidity. On the other hand, the latter is less affected by the environment, such as pressure and air humidity [29]. Recently, Nguyen et al. propose a TENG based on water electrification for self-powered electronics that can output 57.6 µW [30]. Xue et al. propose a non-Hookean droplet spring in which the droplet expands and contracts nonlinearly between two plates as the applied pressure changes and develops a mechanical model for it [31]. Hu et al. design a highly sensitive and self-cleaning TENG that can be used for body fluid detection [32]. Lee et al. design a water droplet-driven TENG with superhydrophobic surfaces, which can generate up to 16 V and is used to harvest energy from raindrops [33]. Chen et al. design a TENG for harvesting environmental mechanical energy from water mist, which can produce a maximum open-circuit voltage of 9.5 V and a short-circuit current of 250 nA [34]. Sahu et al. design triple perovskite-based TENG, which is a facile method of energy harvesting and a self-powered information generator [35]. On the one hand, most of the existing solid–liquid TENGs are difficult to obtain regular outputs and cannot be correlated with the inputs; on the other hand, most of the solid–liquid TENGs for liquid droplets are open-ended and depend on the external environment, which makes them difficult to be applied to harsh environments. The above problems are solved in a dispersive-aggregative triboelectric nanogenerator (DA-TENG), which shows a method for pressure detection and wave energy harvesting.

In this work, we designed a TENG based on a water droplet spring with a concave spherical surface, which could harvest wave energy or detect pressure. The basic working principle was that external force disperses the water droplet, subsequently capitalizing on the concave spherical surface and gravitational forces to effectuate their retraction and aggregation. In the above process, the solid and liquid electrodes came into contact and separated, thus obtaining electrical output. It was therefore named a DA-TENG. In DA-TENG, we pasted a circular copper foil on the concave spherical surface and connected wires, then stuck polytetrafluoroethylene (PTFE) on the whole surface and placed a water droplet on PTFE, and connected wires. For one thing, as a harvesting energy device, the spherical surface replaced the elastic membrane to squeeze the water droplets, which can ensure DA-TENG longevity and increase its output performance. For another, as a self-powered pressure detector, the external force/acceleration input was positively correlated with the voltage, and an elastic membrane was used for encapsulation to ensure its output stability. Due to its cost-effectiveness, ease of processing, and adaptability to complex environments, this self-powered pressure detector held immense potential for widespread application across diverse contexts, promising a multifaceted range of practical uses. Since both core parts were the same, the basic DA-TENG was used as the subject to investigate how each variable affects its output performance, including the water droplet volume, the cross-sectional radius of the concave spherical surface, the force area of the elastic membrane, and the excitation frequency of the shaker.

2 Materials and methods

2.2 Structure design and working principle of DA-TENG

DA-TENG mainly adopted a morphological bionic design. Water droplets accumulate on the lotus leaf center after a thunderstorm and travel upward along the lotus leaf edge when they are impacted by insects jumping in or other external effects (Figure 1a). Nevertheless, water droplets still fall back and gather at the bottom after reaching a certain height. The results show that the lotus leaf surface has a unique structure that causes them to exhibit hydrophobicity [24,36,37]. According to the lotus leaf biological configuration and surface structure, DA-TENG was designed by combining the morphological bionic design method with the superhydrophobic feature of PTFE.

Figure 1

The model of TENG and experimental layout: (a) Lotus leaf is subjected to external excitation. (b and c) Basic model and structural drawing of the DA-TENG. (d) The arrangement of the experimental equipment.

To investigate how each physical parameter affects its output performance, we designed a basic DA-TENG (Figure 1b). As shown in Figure 1c, the PTFE was pasted on the concave spherical surface made by PMMA plate engraving. A circular copper foil electrode was used as material to sense charge, which was pasted between PTFE and PMMA plate and concentric with a concave spherical surface; the load connected copper foil through wires. In addition, quantitative tap water was added to the bottom of the concave spherical surface with a syringe, and another wire was dipped into the water droplet; the wire connected the load. Finally, the elastic membrane covered the whole model. The water droplet made diffusion-aggregation with the loading–unloading external force on the elastic membrane. The contact and separation of the water droplets with the solid electrodes produced an induced charge, thus creating a potential difference between the two electrodes. We formed a loop by connecting it through an external circuit. Eventually electrical output was obtained.

3 Experiments and discussion

Figure 2a shows the motion of the water droplet that varies regularly with the excitation force period, and the red dashed line position is its edge. Its initial state is shown in Figure 2a Ⅰ. The excitation force caused the water droplet to spread after the shaker started. As illustrated in Figure 2a Ⅱ, the solid–liquid interface moved toward the solid electrode, instigating a transformation of kinetic energy into gravitational potential energy within the water droplet. When the excitation force was withdrawn, the gravitational potential energy stored in the water droplet was released, thus regrouping toward the bottom of the concave spherical surface (Figure 2a Ⅲ). Meanwhile, the solid–liquid interface gradually left the solid electrode. DA-TENG potential at the initial state is shown in Figure 2b Ⅰ. The water droplet, expanding incrementally from its base to cover the surrounding area while submerging the solid electrode, engendered an induced charge through the frictional interaction between the droplet and PTFE material. The electrons were transferred from the water droplet via the external circuit to copper foil (Figure 2b Ⅱ). During this process, the open-circuit voltage reached a positive peak (Figure 2c Ⅱ). A “jittering” edge appeared in Figure 2c Ⅰ as open-circuit voltage gradually increased to a positive peak. The phenomenon can be ascribed to the uneven force distribution within the water droplet upon the application of the excitation force. The force transmission from the center to the surrounding area in the water droplet causes different velocities at various points along the central ray direction; eventually, there will be an irregular velocity change at the water droplet edge. The open-circuit voltage reached a positive peak and then started to drop smoothly (Figure 2c Ⅳ). The water droplet merged from all around to the bottom and left the solid electrode. The electrons flow from the copper foil backed to the water droplet via the outer circuit (Figure 2b Ⅲ). Subsequently, further frictional interaction between two dielectric materials occurred, leading to a gradual decline in the open-circuit voltage until it reached a negative peak (Figure 2c Ⅲ). Eventually, the water droplet returned to the bottom of the concave spherical surface and was ready to enter the next cycle.

Figure 2

Kinematic state analysis and charge transfer process of DA-TENG. (a) The photo is in the initial state (Ⅰ), full-load (Ⅱ), or no-load (Ⅲ). (b) The initial state of DA-TENG (Ⅰ) and the power generation process of DA-TENG (Ⅱ) (Ⅲ). (c) Curve change of DA-TENG voltage.

The variables in this experiment were the water droplet volume, the cross-sectional radius of the concave spherical surface, the force area of the elastic membrane, and the excitation frequency of the shaker. The main objective was to determine how the four elements mentioned above affect DA-TENG output performance.

First, our investigation focused on examining the impact of water droplet volume on voltage generation. This analysis was conducted under an excitation frequency of 5 Hz and a force area radius of 10 mm while considering three distinct concave spherical surface cross-sectional radii. With the cross-sectional radius of 30 mm on the concave spherical surface, the open-circuit voltage raised from 0.73 to 1.99 V when water droplet volume increased from 3.0 to 4.5 ml (Figure 3a). Then, the open-circuit voltage dropped rapidly to 1.27 V if the volume kept increasing. When the volume increased to 5.5 ml, the open-circuit voltage was irregular. This is due to the volume of the concave spherical surface being limited, which leads to the water droplet being chaotic. In other words, the dispersion of the water droplet in all directions is limited. The contact-separation of solid–liquid electrodes becomes irregular, which leads to voltage instability. Furthermore, the waveform and crest of the voltage are analyzed by the movement of the water droplet. Before reaching a positive peak, the contact-separation area between the water droplet and solid electrode increases during the movement as the water droplet volume increases, so the open-circuit voltage tends to rise. The open-circuit voltage shows a sharp drop after reaching a positive peak and continuing to increase water droplet volume. This phenomenon can be attributed to the incomplete detachment of the water droplet from the solid electrode, resulting in a diminished effective contact and separation area between them. If analyzed from the point of electrostatic induction, the increase in the effective area of contact and separation will generate more induced charges in a certain range, which leads to an increase in the potential difference between two electrodes. The cross-sectional radii of 35 (Figure 3b) and 40 mm (Figure 3c) show similar results. However, the voltage remained stable at 5.5 ml due to the increased the motion of range for the water droplet. The former had a 2.55 V open-circuit voltage at 5 ml water droplet, and the latter had a 3.45 V open-circuit voltage at 4.5 ml water droplet. The cross-sectional radius of the concave spherical surface also has an impact on voltage. The curvature for cross-sectional radii of 30, 35, and 40 mm were 0.0166, 0.0124, and 0.0096 mm⁻¹ in that order. Consequently, as the curvature decreases, the rate of droplet spreading accelerates, leading to the voltage increase. The root-mean-square (RMS) voltage for the three models is shown in Figure 3d with peaks of 1.24, 1.41, and 1.80 V, respectively. Second, under the experimental conditions with a cross-sectional radius of 40 mm, a water droplet of 4.5 ml, and an excitation frequency of 5 Hz, the force area of the elastic membrane affected the open-circuit voltage as shown in Figure 3e. For force radii below 10 mm, the elastic film deformation experienced a decrease as the force area expanded; however, the force area of the water droplet intensified. Consequently, the voltage exhibited an upward trajectory, peaking at 4.17 V. Nevertheless, with the ongoing increase in radius, the voltage curve depicted a rapid decline. This can be attributed to the excessive force area impeding the effective deformation of the water droplet, ultimately leading to a reduction in the submerged area of the solid electrode. Other things being equal, the open-circuit voltage increases as the force area expands at a particular range. The RMS voltage variation is shown in Figure 3f, with a maximum output of 2.39 V.

Figure 3

The effect of model size, droplet volume, and force area on voltage. (a–c) The variation of the open-circuit voltage with the water droplet volume for different model sizes. (d) The variation of the RMS voltage with the water droplet volume for different model sizes. (e) The open-circuit voltage at different force areas. (f) The RMS voltage at different force areas.

The frequency range of the shaker is 1–10 Hz to simulate the frequency of a human keyboard (4–6 Hz). The open-circuit voltages at various excitation frequencies are shown in Figure 4a with a cross-sectional radius of 40 mm, a water droplet of 4.5 ml, and a force area radius of 10 mm. At 1–7 Hz, the open-circuit voltage rose rapidly with increasing excitation frequency, reaching 4.95 V. At 8–10 Hz, the voltage was reduced and unstable because the motion of the water droplet becomes chaotic. The reason on that while the water droplet is doing the previous aggregation, the next diffusion is already underway. Therefore, two movements occur at overlapping times, which causes the water droplet not to be completely separated from the solid electrode and the induced charge generated is reduced. The ratios between the RMS voltage and shaker peak acceleration at 4–7 Hz were 0.4078, 0.5170, 0.6205, and 0.7287 in that order (Figure 4b). From the data, the mathematical relationship they have is that the difference between the two adjacent data lies at around 0.1. Analogous to the influence of the force area on voltage, an optimal parameter point emerges as the independent variable progressively increases. Next, we conducted the DA-TENG output stability test. The voltage remained stable for any 600 s when DA-TENG was operating at 1 Hz excitation frequency (Figure 4d). Furthermore, it worked for 30 min at an excitation frequency of 5 Hz. Due to a large amount of data, the peak voltage was studied as a group every 2 min and used a boxplot to show (Figure 4c). The peak voltage reached 4.17 V after a short rise in its median value during the first 8 min. The voltage began to gradually drop after 10 min, and then it dropped by approximately 2.87% in 30 min. At this time, the DA-TENG completed 9,000 cycles. The reason for the gradual voltage drop is a defect in processing. To address this issue, we can change the adhesive method or choose a better-performance glue.

Figure 4

The frequency effect on the voltage and the output stability test. (a) The open-circuit voltage at different excitation frequencies. (b) The RMS voltage and peak acceleration of the shaker at different excitation frequencies. (c and d) Stability tests of the voltage at excitation frequencies of 5 and 1 Hz.

Next, we tested DA-TENG output performance. The experimental model and equipment are shown in Figure 5a. DA-TENG could easily light up 18 green light-emitting diodes (Figure 5b). In addition to the four influencing factors studied above, the peak voltage was related to force magnitude and loading–unloading force acceleration. The volunteer held the acceleration sensor to press DA-TENG, which obtained voltage and acceleration waveforms simultaneously. The voltage peaks (Figure 5c) were positively correlated with the acceleration (Figure 5d). Based on the aforementioned research, DA-TENG demonstrates potential as a discernment tool for specific human movement behaviors. Thus, three volunteers were recruited to perform a set of pressure detector experiments. Under a cross-sectional radius of 40 mm, a force area radius of 10 mm, and a water droplet of 4.5 ml, volunteers pressed DA-TENG according to their usual pressing habits with the keyboard. Then, we selected the voltage data with the acceleration positive peak at 25–30 m/s². Figure 5e shows the positive/negative peaks that are generated by pressing/releasing DA-TENG. Without loss of generality, the factors that affect the positive peak are the depth, acceleration, and area (finger size) of pressing; the factors that affect the negative peak are the release acceleration and finger position. Due to the differences in individual physiological conditions and habits, there were large differences in peak voltage wave output by pressing DA-TENG in different people. However, for the same volunteer, the wave crest was almost the same for a certain range of accelerations. Averaging the positive peak, negative peak, and positive-to-negative peak ratios, and then finding the absolute values for three parameters, which also revealed dissimilarities in each tested parameter between different volunteers (Figure 5f). As a result, it is possible to identify a user by this data. Apart from that, if the DA-TENG is fixed to the mechanical equipment and a weight is placed on the elastic membrane surface, it will be able to detect mechanical vibrations. The weight produces displacement in a vibrating environment causing a change in the water droplet state, thus achieving a contact and separation of the two electrodes. However, the relationship between voltage and frequency/amplitude is to be investigated.

Figure 5

The testing of DA-TENG output performance and the study of DA-TENG as a pressure detector. (a) Light up the LED experimental equipment. (b) The photographs of lighting 18 light-emitting diodes. (c and d) The voltage and acceleration are generated by finger pressing. (e) Pressing DA-TENG by three volunteers generates voltage signals. (f) Absolute values of the average values of the following parameters, including positive peaks, negative peaks, and the ratio of positive to negative peaks.

Figure 6 shows the schematic diagram and working principle. Multiple harvesting energy devices are stacked in the vertical direction and mounted on top of the floats (Figure 6a). When they are subjected to sea surface fluctuation or wave impact (Figure 6c), each harvesting energy device squeezes the other, thus achieving contact/separation of solid and liquid electrodes (Figure 6b). Due to the complex wave motion, the harvesting energy device floats and shakes. On the one hand, shaking also allows the solid and liquid electrodes to come into contact and separate, thus generating electrical energy. On the other hand, due to the harvesting energy device being connected in a large area, the whole device mainly moves up and down. Meanwhile, when the device is used in a small area, the shaking can be reduced or eliminated by changing the distance between the center of gravity and the metacenter or installing a bilge keel or fin stabilizer.

Figure 6

The schematic diagram and working principle: (a) The schematic of the harvesting energy device on the sea surface. (b and c) How the interior and exterior of the harvesting energy device work.

Based on the structure, the working principle, and the movement of the harvesting energy device, we simplified it to study how the load resistance, water droplet volume, and the speed of external force application affect the voltage. Given that the harvesting energy device predominantly operated in a vertical floating motion, the output performance was investigated utilizing a waterless experimental setup (Figure 7a), subjecting the device to periodic shocks at 3-s intervals. Since the sea surface flow velocity was about 0.4 ft/s (0.206 m/s), we connected different load resistors to the harvesting energy device under the condition of the peak velocity of external impact 0.192 m/s and the water droplet volume of 2.0 ml, so we got the RMS voltage at the terminals of different load resistors. Therefore, the internal resistance could be obtained according to the following equation:

where U _TENG is the total voltage generated by the harvesting energy device; U _i and U _j are the RMS voltages of the load; R _TENG is the internal resistance of the device. By calculation, we could get that R _TENG is about 5.6–6.1 MΩ. The effective power was calculated from the RMS voltage and the load resistance:

where P ₀ is the actual power; U _RMS is the RMS voltage of the load; R is the load resistance. The RMS voltages are shown in Figure 7b, and the effective power could reach 8.82 μW. Without loss of generality, the power is maximum when the internal and external resistance is equal, so we can know that the internal resistance of the device is about 6 MΩ, which is consistent with the theoretical calculations. To better measure the output performance of the harvesting energy device, we obtained curves for currents up to 1.7 μA (Figure 7c). Compared to previous studies [8,38], the device has a low internal resistance, high power, and high current. It is worth emphasizing that DA-TENG has almost no electrode wear problems. Under the experimental conditions where the peak velocity of external impact is 0.192 m/s and the load resistance is 6 MΩ, the relationship between water droplet volume and voltage is shown in Figure 7d. The volume is too large or too small to obtain the optimal output, which is similar to our study above. Finally, we investigated the relationship between peak velocity and voltage for the external impact. At 0.032–0.256 m/s, elevated velocity resulted in a corresponding surge in the peak voltage attained (Figure 7e), which might be related to the surface charge density [39]. When the water droplet comes in contact with the solid electrode, electron transfer causes the solid electrode to be negatively charged, as a result, the positive ions in the water droplet are attracted and come close to the solid electrode [40]. When the separation velocity of the solid–liquid electrode is low, the positive ions are adsorbed on the solid electrode surface, which affects the charge transfer.

Figure 7

Experiments for testing the output performance of energy harvesting devices and the study of their influencing factors. (a) Simplified model of the harvesting energy device and the experimental setup without water. (b) The effective power and voltage of the harvesting energy device at different load resistances. (c) Current generated by the harvesting energy device at a load resistance of 6 MΩ. (d) The effect of water droplet volume on voltage. (e) How the speed of the applied force affects the voltage. (f) Change in volume fraction of water droplets when squeezed. (g) Single (Ⅰ), parallel (Ⅱ), and series (Ⅲ) charging and discharging circuits. (h) Experimental arrangement of three harvesting energy devices. (i) Charge–discharge curves of the wave harvesting energy device in single/parallel/series after full-wave rectification. (j) The current is generated by three harvesting energy devices connected in parallel when the load resistance is 6 MΩ.

To visually depict the diffusion-aggregation motion of the water droplet, we conducted simulations to observe the dynamic changes in volume fraction. Figure 7f shows the change in the volume fraction occupied by air as the sphere (upper arc) squeezes the water droplet (blue part). Subsequently, a charge/discharge experiment was conducted employing the circuit illustrated in Figure 7g; the capacitor was 10 μF and the load resistance was 2 MΩ. The experimental setup is shown in Figure 7a and h. After working at 0.256 m/s on the device with a 3-s interval impact for 500 s, the charged curve of a single harvesting energy device (Figure 7g Ⅰ) is shown in Figure 7i (S1 closed and S2 opened). The voltage could reach 2.1 V. The full-wave rectifier bridge was connected in each of the three harvesting energy units, and then, the direct-current poles of the rectifier bridges were connected in parallel (Figure 7g Ⅱ)/series (Figure 7g Ⅲ) to obtain the charging curves shown in(Figure 7i). Upon establishing a parallel connection, the circuit yielded a maximum voltage of 3 V, whereas a series configuration enabled the attainment of a higher voltage, peaking at 6 V. Under the above experimental conditions, three energy harvesting devices were interconnected in parallel/series configuration and subjected to full-wave rectification, which accelerates the charging rate and increases the voltage. The analysis shows that the parallel circuit does not increase the input voltage, but the device processing error and the input frequency error make it difficult to synchronize the motion between the three devices, which makes the voltage peaks intensive. Therefore, the charging speed is accelerated. Figure 7j shows the current curves with good output synchronization when the three devices are connected in parallel. The maximum current could reach 5.3 μA. The series circuit increases the input voltage, which directly leads to an increase in capacitor voltage. The selection of circuit connections is based on the synchronization among the units within the integrated device, resulting in enhanced efficiency for energy harvesting. In summary, the harvesting energy device with DA-TENG as the core has a good superposition effect.

4 Conclusion

In summary, a TENG based on a water droplet spring with a concave spherical surface can be used to harvest wave energy and detect pressure. According to electrostatic induction, water droplet in DA-TENG is dispersed/aggregated to contact/separate from solid electrodes to generate voltage. When an external force is applied, the water droplet disperses in all directions, which produces a positive peak. When the external force is withdrawn, it regroups at the bottom of the surface, which produces a negative peak. To facilitate the study, a basic DA-TENG is fabricated to explore its output performance. By optimizing the model parameters, the maximum open-circuit voltage is generated at a water droplet of 4.5 ml, a cross-sectional radius of 40 mm for the concave spherical surface, a force area radius of 10 mm for the elastic membrane, and an excitation frequency of 7 Hz. DA-TENG has stable output, high wear resistance, and strong environmental adaptability. With the 5 Hz for 30 min, the voltage fluctuated between 4.01 and 4.18 V, which indicates that DA-TENG can be used in industrial production. It also verified that DA-TENG is more suitable for harvesting energy and monitoring conditions in higher-frequency environments than previous studies. Next, a new type of pressure detector is proposed by collecting and analyzing the relationship between acceleration and voltage. The detector can sensitively detect changes in acceleration/force, which can be applied for pressure measurement and the initial determination of personal identification. Finally, we design the DA-TENG as a harvesting wave energy device. Under a 6 MΩ load, a peak voltage of 10.2 V, a peak current of 1.7 μA, and an effective power of 8.82 μW can be obtained by operating one unit; a peak current of 5.3 μA can be obtained by operating three units. Therefore, the device has a broad application prospect in harvesting wave energy.