Fabrication and characterization of graphite materials.
As depicted in Fig. 1a, a WG-HEG module was constructed by a P-type graphite foil and an N-type graphite foil, both of which were prepared using a straightforward and scalable impregnation method (see experimental section for details). Scanning electron microscopy (SEM) images indicate that the graphite foil, with a thickness of 50 μm, is composed of small graphite flakes (Supplementary Fig. 1, 2). The X-ray diffraction (XRD) pattern (Supplementary Fig.3) reveals two prominent peaks, located at approximately 26.5° and 54.6°. These peaks correspond to the (002) and (004) planes, respectively, of the hexagonal graphite crystal structure, with the (002) peak is sharper and more intense than the (004) peak. The Seebeck coefficient is a measure of the voltage difference generated when a temperature gradient is applied, and its sign changes from negative to positive when the majority carrier switches from electrons to holes46. The schematic of the Seebeck coefficient measurement setup is descripted in Supplementary Fig. 3a. Measurements were conducted using gold electrodes and in nitrogen atmospheres to prevent electrochemical reactions between the graphite foil and the electrode, and minimize the impact of moisture on Seebeck coefficient measurement. The Seebeck coefficient of P-type graphite is measured as positive at room temperature (290 K), approximately +7.8±0.3 μV/K, while pristine and N-type graphite exhibit negative values, around -2.9±0.6 μV/K and -5.9±0.4 μV/K, respectively (see Supplementary Figs. 4b and 5). This result indicate that adsorption-induced chemical doping is an effective approach for tailoring the electrical properties of graphite.
Electricity generation of a WG-HEG unit.
Figure. 1b illustrates the step-by-step working principle of the WG-HEG, which can be attributed to the difference in the EDLs formed when the P-type and N-type graphite are in contact with water. When a WG-HEG device is immersed into water, a stronger EDL is formed on the P-type graphite, causing the electrons in the N-type graphite to be driven towards the P-type graphite through an external load. During this process, the electron flow continues until the EDLs on both the P-type and N-type graphite reach an equilibrium state (Fig. 1(ii)). Subsequently, the WG-HEG device is removed from the water and heated at 100°C for 5 minutes to remove transferred charges and residual water molecules on the graphite surface. The electric characteristics of the WG-HEG are measured at 25°C in tap water using a graphite sheet with a 0.25 cm2 area and a 5 mm inter-electrode distance. The measured results of the open-circuit voltage and short-circuit current are displayed in Figure. 1c and d. It can be observed that a single WG-HEG unit can repeatedly generate a voltage of about 0.6 V and an impressive current of about 150 μA when immersed in and removed from water. Furthermore, the output voltage is measured with various load resistances, and the optimal output power of 23 μW (92 μWcm−2) is achieved at a resistance of 5.1 kΩ (Fig. 2a and Supplementary Fig. 6), which can be calculated using the formula P=V^2/RL, where P represents the power output, V represents the output voltage, and RL represents the load resistance. In addition, the WG-HEG exerts stable performance after multiple immersions into water and even after being exposed to the ambient environment for 30 days, implying its outstanding stability in both operational and natural environments (Supplementary Fig. 7, 8).
The impact of graphite area, distance between two graphite foils, temperature, and water source on the performance of the WG-HEG were further investigated. The induced voltage exhibited slight variations with changes in all the tested parameters, whereas the output current showed significant changes (Fig. 2b-e). The variation in output current can be attributed to the influence of the tested parameters on the internal resistance of the device. Specifically, increasing the graphite area, reducing the distance between the graphite electrodes, raising the solution temperature, or using a 0.6M NaCl solution can effectively decrease the internal resistance of the device, resulting in a significant increase in output current. Impressively, the output current showed a linear increase with an increase in the graphite area, indicating the excellent scaling properties of the WG-HEG (Fig. 2b). In addition, The WG-HEG exhibits excellent performance even in 0.6M NaCl solution, which is similar to the average ocean salinity, and can be applied across a broad temperature range from 0 to 50℃. This wide temperature range and choice of solutions make it possible to utilize WG-HEGs in most regions of the earth.
Verification for electricity-generating mechanism.
The above-mentioned working mechanism was investigated using both comparative experiments and calculations based on the plane wave density functional theory. Firstly, the interfacial potential between water and the graphite surface was measured at 25°C using an Ag/AgCl reference electrode (Fig. 3a). The results showed that P-type graphite displayed a higher potential of approximately 340 mV, while N-type graphite exhibited a lower potential of around 200 mV relative to the Ag/AgCl reference electrode (Fig. 3b). We further observed a similar difference in P-type and N-type graphite treated with lithium chloride and sodium carbonate, suggesting the universality of this phenomenon (Supplementary Fig. 9, 10). Moreover, the higher zeta potential and smaller contact angle observed on P-type graphite indicate a stronger electric double layer formation at the interface between P-type graphite and water (Fig. 3c and Supplementary Fig. 11). These results reveal the formation of an asymmetric electric double layer at the water-graphite interface when P-type and N-type graphite are immersed in water.
The presence of a charged surface is a necessary condition for the formation of an electric double layer at the solid-liquid interface47. The observation of similar induced voltages after removing possible pre-existing charges on the graphite surfaces by grounding or using an ionizing air blower rules out the possibility that these pre-existing charges were the source of the charged graphite surface (Supplementary Fig. 12). The dissolution of metal ions from the active metal electrode results in a negatively charged and low potential metal electrode (Supplementary Fig. 13). However, the mechanism underlying the formation of a charged graphite surface differs from that of the active metal electrode, owing to the stability of graphite in neutral water at ambient temperatures. Additionally, Similar voltages were generated in dark environment (Supplementary Fig.14), providing evidence that the photovoltaic effect was not contributing to the charged graphite surface. Remarkably, it has been reported that when the atoms of a substrate like graphite come in contact with a polar solvent like water, the molecules constituting the polar solvent exert a finite electric field on the atoms of the substrate. The influence of solvent polarity on the performance of WG-HEG was investigated by employing solvents with varying polarities. As depicted in Figure 3(d), the voltage outputs generated by solvents such as water, dimethyl sulfoxide, ethylene glycol, dichloromethane, and hexane were approximately 564, 387, 262, 5, and 0 mV, respectively, and exhibited a positive correlation with solvent polarity. These results confirm that solvent polarity is a crucial factor affecting the electricity output of WG-HEG. To provide a further understanding of the observed experimental phenomena, density functional theory (DFT) simulations were conducted to investigate the interaction between graphene and a polar water molecule (see the details of the first principles calculations in Supporting Information). The simulation results indicate that the oxygen atoms in water molecules were found to be located closer to the P-type graphite surface due to the cumulative effects of Coulomb interactions and van der Waals forces. Additionally, the orientation of water molecules on the N-type graphite surface differed from that on the P-type graphite surface. Specifically, the hydrogen atoms in water molecules were observed to be situated closer to the N-type graphite surface (Fig. 3e).
Based on the analysis of contrasting experiments and simulations presented above, we propose a possible mechanism for the formation of the asymmetric electric double layer, as illustrated in Figure 3f. The arrangement of polar water molecules on the surface of pristine graphite induces positive charges on the graphite surface and attracts negative ions in the solution due to electrostatic interactions, resulting in the formation of an electric double layer. The arrangement of water molecules is strengthened and weakened on the surfaces of P-type and N-type graphite, respectively, resulting in corresponding changes to the electric double layer strength on these surfaces. While we propose a possible mechanism for the formation of the asymmetric electric double layer based on initial findings, the underlying mechanism still needs to be further understood in future studies due to the complex interactions between imperfect experimental materials. Specifically, the rough graphite surface could adsorb airborne hydrocarbons spontaneously when contact with air48, and ions are often present in solution, such as H+ and OH- in water47. Moreover, the interaction between water molecules, ions, and the surface of graphite involves not only simple adsorption and desorption phenomena but also possible charge transfer between water molecules and graphite49,50.
Applications of WG-HEG.
The scalable integration of electric generators is crucial for increasing the energy output of devices. The utilization of inexpensive experimental materials and a simple fabrication process facilitates the large-scale connection of devices in either series or parallel, thereby increasing the output voltage or current. By connecting 20 units in series, the generated voltage can reach up to 7.3 V (Fig. 4a), and connecting 120 units in parallel results in a linearly boosted current output of about 58 mA (Fig. 4b), demonstrating the outstanding scaling properties of the WG-HEG. Notably, the integrated WG-HEGs have a favorable trade-off between output voltage and current, superior to previously reported integrated devices (Fig. 4c, Supplementary Table 1), indicating their potential as practical power sources. The electric power supplied by the WG-HEG can be stored in commercial capacitors without the need for extra rectifiers. Benefiting from the high current output of the integrated WG-HEG, capacitors with capacitances of 100, 470, and 1000 μF can be charged to 4, 3, and 2 V within 10 seconds, respectively, and maintain a fully charged state (Fig. 4d). To further demonstrate the practical applications of WG-HEGs, capacitors were charged by the WG-HEG integration to power commercial electronics such as an electric fan (Fig. 4e and Supplementary Video 1) and a full-color LCD screen (Fig. 4f and Supplementary Video 2).