# Fileset

[Electrokinetic Power Generation, Feb26.pdf](https://mdr.nims.go.jp/filesets/9ec73470-349c-4518-b02a-978f3d4ac772/download)

## Creator

[Manpreet Kaur](https://orcid.org/0009-0008-5294-8465), [Avinash Alagumalai](https://orcid.org/0000-0002-6024-2760), [Omid Mahian](https://orcid.org/0009-0009-9117-0614), Sameh M. Osman, [Tadaaki Nagao](https://orcid.org/0000-0002-6746-2686), Zhonglin Wang

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Harvesting Energy Via Water Movement and Surface Ionics in Microfibrous Ceramic Wools](https://mdr.nims.go.jp/datasets/f1db6f3d-7b60-4c48-86d2-b53af6269057)

## Fulltext

1  Harvesting Energy via Water Movement and Surface Ionics in Microfibrous Ceramic 1 Wools 2 Manpreet Kaura,b*, Avinash Alagumalaib,  Omid Mahianc,d,e,i*, Sameh M. Osmanf, Tadaaki 3 Nagaoa, g*, Zhong Lin Wange,h 4 aInternational Center for Materials Nanoarchitectonics (WPI- MANA), National Institute for 5 Materials Science (NIMS), Tsukuba, Ibaraki, 305-0044, Japan 6 bDepartment of Chemical and Petroleum Engineering, University of Calgary, 2500 University 7 Dr. NW, Calgary, Alberta, T2N 1N4, Canada 8 cZhejiang Provincial Engineering Research Center for the Safety of Pressure Vessel and 9 Pipeline, Faculty of Mechanical Engineering and Mechanics, Ningbo University, Ningbo 10 315211, China 11 dDepartment of Chemical Engineering, Imperial College London, London SW7 2AZ, UK 12 e Beijing Institute of Nanoenergy and Nanosystems, Chinese Academy of Sciences, Beijing, 13 100083, People’s Republic of China 14 f Chemistry Department, College of Science, King Saud University, P.O. Box 2455, Riyadh 15 11451, Saudi Arabia 16 g Department of Condensed Matter Physics, Hokkaido University, Sapporo, Hokkaido, 060-17 0810, Japan 18 hSchool of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, GA, 19 USA 20 i Laboratory on Convective Heat and Mass Transfer, Tomsk State University, 634045, Tomsk, 21 Russia 22 Corresponding authors: Manpreet (manpreet.kaur7@ucalgary.ca);  Omid 23 (o.mahian@imperial.ac.uk; omid.mahian@gmail.com);  Tadaaki (nagao.tadaaki@nims.go.jp)    24 Abstract 25 Due to the push for carbon neutrality in various human activities, the development of 26 methods for producing electricity without relying on chemical processes or heat sources has 27 become highly significant. Also, the challenge lies in achieving microwatt-scale outputs due 28 to the inherent conductivity of the materials and diverting electric currents. To address this 29 challenge, our research has concentrated on identifying non-conductive mediums for water-30 2  based using low-cost microfibrous ceramic wools in conjunction with a NaCl aqueous 31 solution for power generation. The main source of electricity originates from the directed 32 movement of water molecules and surface ions through densely packed microfibrous ceramic 33 wools due to the effect of dynamic electric double layer. This occurrence bears resemblance 34 to the natural water transpiration in plants, thereby presenting a fresh and straightforward 35 approach for producing eco-friendly electricity. The generator module suggested in this study, 36 measuring 12×6 cm², demonstrated a noteworthy open-circuit voltage of 0.35 V, coupled 37 with a short-circuit current of 0.51 mA. Such low-cost ceramic wools are suitable for 38 ubiquitous, permanent energy sources and hold potential for use as self-powered sensors and 39 systems, eliminating the requirement for external energy sources such as sunlight or heat. 40 Keywords: Energy harvesting; ceramic microfibers; water evaporation; power generation; 41 self-powered systems; clean energy. 42 1. Introduction 43 Over the upcoming decades, easily accessible fossil energy resources are anticipated to 44 diminish significantly, exacerbated by their persistent adverse effects on the environmental 1-4. 45 Recently renewable energy research has advanced to the forefront, and zero-emission energy 46 sources and carbon neutrality have become top priorities in energy research 5-9. Several 47 devices have been successfully developed utilizing battery-free, self-powered systems, such 48 as piezoelectric or triboelectric nanogenerators 10-12. Indeed, the concept of self-powered 49 sensors and systems based on “Nanogenerators” was coined by Wang and Song in 2006 13, 14. 50 Most such energy-harvesting systems have limitations in regard to operational lifetimes, 51 useful geographic region, or fabrication costs 15, 16. Also, their nano-fabrication processes are 52 rather demanding and fall far short of meeting industrial needs 17-20. In recent times, there has 53 been a lot of interest in water-based power generation due to its capacity to harness clean 54 energy stored within the dynamic states of water 21-26. If the energy absorbed by water could 55 3  be harnessed at an efficiency of 1% using water-based power technology, the potential 56 contribution could amount to approximately one-third of the world's energy consumption, 57 which is comparable to the energy output of crude oil 27. However, this method typically 58 necessitates variations in elevation and is unsuitable for decentralized and widespread energy 59 extraction. In the recent past, innovative nanogenerators that capitalize on water flow have 60 emerged as a solution to overcome this limitation 28-31. In such devices, electricity generation 61 is based on streaming current, in which an ionic species near the electrical double layer 62 (EDL) at the solid-liquid 32-35 interface carries the electricity during water movement 36-38. 63 Due to the ample availability of water resources and the increasing requirement for 64 environmentally friendly energy sources, a novel approach has been documented wherein 65 electricity generation is propelled by water. Carbon black, graphene oxide, and carbon 66 nanotubes have been studied together with their hybrid structures37, 39-45. However, their 67 achieved voltages have been constrained to millivolt ranges, while their wattage outputs 68 predominantly remain within the microwatt scale. This limitation is likely due to the inherent 69 conductive properties of these materials, which tend to divert the electric currents. To address 70 this challenge, our research has concentrated on identifying non-conductive mediums for 71 water-based power generation. Our focus has shifted towards water transpiration from 72 ceramic media, as we anticipate that these materials hold the potential to deliver higher 73 voltage and power outputs in comparison to systems reliant on carbon-based components. In 74 this work, an endeavor has been made to study the generation of electrical voltage through 75 the process of water transpiration across a ceramic microfiber wool (CW), which acts as an 76 "insulator". This specific ceramic microfiber wool possesses a notable combination of 77 attributes, including a substantial surface area and a multitude of narrow spaces that facilitate 78 capillary forces. Among the numerous options available in the realm of micro-fibrous 79 ceramic materials, CW stands out due to its cost-effectiveness and wide-scale industrial 80 4  production. Our previous research demonstrated that the fibrous structure of CW employs 81 capillary action for efficient water evaporation, leading to swift near-room-temperature 82 distillation through solar photothermal evaporation 46. Building on this, our current study 83 focuses on harnessing the natural movement of water through CW, driven by 84 evapotranspiration, to generate electrical power. 85 We have utilized commercially available ceramic wool (ISOWOOL Blankets 1260; Isolite 86 Insulating Products Co. Ltd) in our experiments. These ceramic wool products are widely 87 used in refractories for high-temperature equipment and are among the least expensive and 88 most mass-produced industrial materials in this category. The decision to use ceramic fibers 89 with the composition of microfibers comprising 54% SiO2 and 46% Al2O3 was based on their 90 suitable properties for filtration or capillary tasks in corrosive media, elevated temperatures, 91 or pressures. Furthermore, these environmentally friendly  and low cost ceramic fibers are 92 flexible, easy to handle, and can be woven into fabrics, allowing for the fabrication of 93 complex-shaped composites.  94 Through our experiments, we have demonstrated that surface-modified CW serves as a 95 proficient power generator fabricated using a device measuring 12×6 cm². Furthermore, we 96 explored the potential for voltage scaling by arranging such devices in series, leading to an 97 impressive voltage increase to 1.6 V. This level of voltage proves adequate for tasks such as 98 charging a supercapacitor to illuminate an LED or operate a motorized fan. 99 A remarkable advantage of this system is its fuel source – ionized water, which is naturally 100 evaporated, can serve as the generator's fuel. Consequently, this generator can operate 101 continuously, both during the day and night. This unique trait positions it as a valuable power 102 supply for self-sustaining systems, thereby complementing the operation of solar cells. 103  104 5  2. Methods 105 Initially, the CW underwent a cleaning process using a sodium hydroxide solution. To 106 transition the hydrophobic nature of CW to hydrophilic, hydroxyl (OH) groups were 107 introduced through the reaction with sodium hydroxide. The negative zeta potential indicates 108 the negative charges due to OH groups on the modified CW surface. In all measurements, 109 this OH modified hydrophilic CW was used, and the narrow spacing in between the fibers 110 induced a capillary force that resulted in efficient uptake and transpiration of water by this 111 material.  112 Figure 1 provides an overview of the entire apparatus and experimental arrangement 113 employed to measure the electric potential generated through evaporation. To fabricate the 114 ceramic wool electric generator (CWEG) device, two cotton sheets measuring 12×6 cm² each 115 were prepared. Subsequently, 1,000 mg of CW was reshaped into a thin layer measuring 116 12×6 cm² with a thickness of approximately 0.2 cm. This CW layer was then placed between 117 the two cotton sheets, and this assembly was further enclosed between two plastic grids as 118 depicted in Figure 1(a-i). The entire structure was secured using plastic clamps to fasten the 119 electrodes in place. The electrodes were copper plates or carbon-glass plates (Figure 1e and 120 1h). To simplify the handling of the CWEG device, copper (Cu) plates were utilized as 121 electrodes in all conducted experiments. This device, with the Cu electrodes, was positioned 122 within a beaker to facilitate the measurement of the potential variation across the CWEG 123 sample. 124 The employed CW comprised microfibers which exhibited an average diameter of around 3.2 125 µm (as depicted in Figure 1j). By utilizing the Brunauer–Emmett–Teller method, the specific 126 surface area of the CW was determined to be 70.99 m²g-1. Additionally, electrical 127 connections were established using two electrical clamps. Graphite paste was used to 128 safeguard the electrode. This paste-coated area was subsequently dried at a temperature of 129 6  50 °C for a duration of 3 hours. This measure was undertaken to prevent corrosion within a 2 130 mM saline environment. Figure S1 shows (a) Prior to and (b) Post graphite coating effect 131 pictures of the electrode.  132  133 Figure 1. Design of a ceramic wool electric generator (CWEG) (a) commercial grade cotton fiber 134 cloth (b) ceramic wool (CW) (c) plastic grid (d) plastic clamp, plastic screw, and Cu plates (e) 135 prepared CWEG (f) prepared CWEG using carbon glass plate (g) CWEG in a grid (h) complete 136 CWEG setup in NaCl (i) schematic diagram of the setup for current-voltage measurement (j) SEM of 137 CW. 138 The CWEG was introduced to a 2 mM NaCl aqueous solution. This solution covered a 4 cm 139 segment of the CWEG, while the rest remained exposed to the surrounding atmosphere 140 (depicted in Figure 1i). One of the electrodes positioned at the extremity of the CWEGs was 141 submerged in the water. The water infiltrated to a level several centimeters above the water 142 7  surface in the beaker due to capillarity effect of filler. This capillary effect emulates the water 143 transpiration observed in plants.  144 The energy conversion performance of the CW device was determined using a VersaSTAT 145 potentiometer (VersaSTAT 4, Princeton Applied Research) under water evaporation. Before 146 the measurements, the fibers were readied with the NaCl solution for almost 1 h. All tests 147 were performed thrice to determine their repeatability. The following are the CWEG 148 dimensions; ∼ 12 × 6 cm2, thickness ∼0.2 cm, weight ∼1 g (optimized CWEG dimensions 149 and discussed below), followed by all other experiments in this work. The voltage and current 150 values with the mean and standard deviation of five devices is shown in Table S1. 151  152 3. Results and Discussion 153 The working principle of the evaporative nanogenerator is shown in Figure 2. Wang et al.'s 154 research underscores the primary involvement of electrons in the initial formation of the 155 charge layer during the phenomenon of liquid-solid contact electrification, commonly 156 referred to as triboelectrification 47-49. The nanogenerator, which relies on water evaporation, 157 utilizes this phenomenon for mechanical energy harvesting and investigating charge transfer 158 at these interfaces. These nanogenerators operate by combining contact electrification and 159 electrostatic induction, entailing surface electron/ion/material transfer and generating electric 160 potential differences upon separation in materials with contrasting triboelectric polarity 50, 51. 161 Ceramic fibers possess several desirable qualities over the previous used nanogenerators 162 materials, rendering them appealing for filtration or capillary applications in corrosive 163 environments, high temperatures, or under elevated pressures.  164 Measured zeta potential values (discussed later) are well aligned with the origin of the 165 evaporation induced streaming potential and chemistry of NaCl on the CW surface.  The zeta 166 potential values give an indication of the negative charge on the modified CW surface. The 167 8  noticeable hydrophilic characteristics of CW, attributable to the presence of surface -OH 168 groups (as confirmed by zeta potential values), facilitate the deep penetration of water 169 molecules into the interior of the CWEG. As a result, the water adsorbed by the hydrophilic 170 nature of CW promptly wets the surface, consequently triggering surface charging. Notably, 171 during the measurement, the bottom CWEG electrode functions as the negative electrode, 172 while the top electrode serves as the positive electrode.  173 In the aqueous environment, the presence of ions initiates the development of an interfacial 174 charge layer that closely resembles an EDL when these ions interact with the hydroxyl (OH) 175 groups situated on the ceramic fibers. Counterions, such as Na+, exhibit a distinct preference 176 for upward migration within the mobile layer of the EDL due to the capillary-induced 177 pressure gradient 52. This upward migration ultimately leads to the creation of charge 178 polarization, subsequently giving rise to the generation of an induced streaming potential. 179 Chemically, this is due to the dynamic electric double layer, which is produced by the change 180 in liquid covered surface area and/or ion migration at the vicinity of the surface due to charge 181 transfer between liquid and solid 49. The structure of the Capillary-Woven CWEG doesn't 182 inherently possess pores, but rather its porosity arises from the interstitial spaces between the 183 crisscrossed fibers. When an electrolyte solution flows through such a capillary or porous 184 medium with narrow spaces, comparable to the Debye length of the fluid, counterions tend to 185 occupy these spaces due to the overlapping EDLs. 186 In summary, upon inserting the CWEG into NaCl solution, contact electrification occurs, 187 resulting in the negatively charged CW walls and the spontaneous formation of an EDL at the 188 interface between the CW fibers and the surrounding water. This process is notably facilitated 189 by the combined influences of capillary forces and evaporation. The measured zeta potential 190 of CW is -49.09 mV. To maintain charge neutrality at the CW-NaCl interface, negatively 191 9  charged CW walls attract cations (Na+), while excluding anions (Cl−), creating the EDL. As 192 water flows, these ions move upward with it. In evaporation-driven flow, the EDL generates 193 a net positive charge transport, creating a streaming current and inducing a streaming 194 potential. This potential difference between CWEG electrodes drives a continuous 195 unidirectional electron flow, resulting in a conduction current. 196 Considering the complex fibrous CW architecture, numerous electrochemical reactions likely 197 occur on the electrodes during the passage of current. However, for clarity, a generalized 198 schematic illustrates electron movement from the induced cathode to the anode (Figure 2b). 199  200  201 Figure 2. (a) Schematic representation of the experimental setup for measuring induced potential and 202 (b) capillary transport of ions on individual fiber, showing the specific ion distribution on the fiber.  203 A modified hydrophilic CW with intricate fibers generates an electric potential difference 204 between integrated electrodes as liquid flows through it. This potential difference increases 205 10  until reaching its peak. When dry, no open-circuit voltage (Voc) is observed, but when 206 saturated with a 2 mM NaCl solution, a Voc of 0.35V arises after about an hour, as water 207 reaches certain height, indicating material and solution properties influence this voltage 208 generation (Figure 3a). A similar experiment was performed with a carbon-glass plate 209 electrode (Figure S2), which gives nearly the same result as Cu-plate electrodes, indicating 210 that the redox reaction of Cu electrodes is not the main source of voltage generation. A 1-µF 211 capacitor charges to 0.35 V in 10 min (Figure 3a, inset). Throughout all experiments, the 212 CWEG device maintains an angle of approximately 60° with respect to the water surface 213 within the beaker. This orientation permits the water to ascend to a height of roughly 4 cm 214 above the water level. Variations in the measured voltage can be attributed to the oscillations 215 in laboratory temperature and humidity during the testing period. Additionally, the suggested 216 model shows that the generated electricity can be significantly amplified by series connection 217 (Figure 3b). VOC increases linearly, with slight variations due to losses stemming from wiring 218 and contact resistance. Two, three, and six samples in series can produce up to 0.61 V, 0. 90 219 V, and 1.61 V, respectively.  220 Nonetheless, the induced voltage remains consistently close to 0.35 V during the entirety of a 221 6-hour test, while the short-circuit current (Isc) maintains a stable value of around 0.5 mA as 222 presented in Figure S3. This dependable performance occurs within a laboratory 223 environment characterized by temperature variations, along with relative humidity (RH) 224 levels spanning from 45% to 55%. The initial variations in the readings can be attributed to 225 the swift advancement of the capillary front. This front eventually stabilizes as a balance is 226 achieved between water adsorption and evaporation processes. The open-circuit voltage 227 (Voc) and short-circuit current (Isc) are 0.35 V and 0.5mA, respectively (Figure 3c). The 228 completely dried sample before adding water into beaker shows essentially zero current. 229 11  Submerging the CWEG in water leads to the manifestation of an IV curve with a discernible 230 slope and an offset.  231 Figure 3d shows that the zeta potentials of pristine and modified CW are 8.15 ± 0.035 mV, 232 and -49.09 ± 0.215 mV, respectively, where each sample was measured 3 times and presented 233 the mean and standard deviation values. The negative zeta potential indicates the negative 234 charges due to OH groups on the modified CW surface 53. Measured zeta potential values are 235 well aligned with the origin of the evaporation induced streaming potential and chemistry of 236 NaCl on the CW surface as presented in Figure 2.  The zeta potential values give an 237 indication of the negative charge on the modified CW surface. 238  239 Figure 3. Electrical attributes of the CWEG device: (a) Voltage generated by a CWEG over time (in 240 hours), while it is immersed to a depth of 4 cm in NaCl solution, (b) Voltage generated by 1 to 6 241 CWEG devices in series mode, (c) The Performance of CWEG devices in both dry and saturated 242 12  states within the ambient environment and (d) Zeta potential variations of pristine CW and modified 243 CW 244 Further insights into the generation of streaming potential come from additional experiments 245 using NaCl solution with different degrees of immersion of the CWEG in water (Figure 4a). 246 The bottom electrode of the nanogenerator was kept under the water, and the top electrode 247 was kept above it. For water depths of 1, 2, 3, 4 and 5 cm, voltages of the device were 0.27, 248 0.31, 0.33 0.35, and 0.34 V, respectively. The maximum voltage was observed, when the 249 CWEG was inserted 4 cm into the water, beyond which it decreased. With further increases 250 in the water level, the evaporation zone diminished, and the voltage decreased. Therefore, 251 placing the air-water interface near the middle of the CWEG is essential for voltage 252 generation, which effectively promotes evapotranspiration through the fibrous media.  253 On the other hand, to optimize the effect of device size on performance of the device, we 254 conducted an experiment with three different sample sizes. We first used a 6 x 6 cm2 device 255 which showed 0.24 V. To increase the device performance further, we increased the length of 256 the device to 12 x 6 cm2 and achieved 0.35 V. We then increased the width of the device to 257 12 x 12 cm2, which further increased the voltage to 0.38 V (Figure 4b). Hence, voltage 258 generation increases with increasing CWEG area. We chose 12 x 6 cm2 for the rest of our 259 measurements for ease of device setup in a laboratory beaker.  260 Next, we optimized the mass of CW sandwiched between the cotton tissues in Figure 4c and 261 Note S1. We tested voltage generation performance using cotton tissue without CW and the 262 voltage for cotton tissue only was considerably lower (0.012 V) than for the CW sandwich. 263 By sandwiching 500 mg of CW between cotton (~ 35 mg/cm3), we achieved 0.17 V, which 264 increases with higher amounts of CW, up to a maximum of 1000 mg (~ 69 mg/cm3), beyond 265 which the voltage actually decreased slightly. The higher CW fiber density (number of fibers 266 per unit length) increases the pore density, leading to higher device performance. However, 267 13  increasing the fiber density further increases the flow resistance, reducing the amount of 268 evaporation. Therefore, to avoid the excess spun fibers, which diminishes the spacing 269 efficiency of fibers, impeding ion migration, we have chosen 1000 mg CW weight (~ 69 270 mg/cm3) for the remaining measurements. 271 272 Figure 4. Voltage generation performance of CWEG with different (a) water levels (cm), (b) 273 sample sizes, (c) CW masses sandwiched between cotton tissues, and (d) voltage generated 274 continuously by a CWEG for 10 days, in which each cycle represents 8 hours of 275 measurement every day.  276 To assess the extended stability of the device, we conducted measurements continuously for 277 10 days, each day spanning 8 hours (Figure 4d). Notably, the recorded voltage exhibited 278 remarkable stability over this time frame. Throughout the course of the 10-day 279 experimentation period, the device's performance remained nearly constant, with only 280 marginal fluctuations in readings. These fluctuations can be attributed to alterations in 281 ambient conditions. Consequently, the device can be regarded as a "micro 'green' power 282 plant" capable of providing a consistent power supply with virtually negligible energy input.  283 14  Comparative studies on water evaporation were performed using CWEGs and NaCl solution. 284 Evaporation of water was tested using an electric balance (AUW220D, Shimadzu). The 285 amount of NaCl solution was fixed at 100 mL (1 ×105 mg), indicating the high evaporation 286 efficiency of the CWEG as compared with a beaker containing only saline (Figure 5a). After 287 3,600 s, the water mass loss with the CW reached 2250 mg. Therefore, due to high 288 evaporation efficiency of the CWEG, we have observed voltages as high as 0.35 V with high 289 stability from a 12 x 6 cm2 device.  290 In order to further confirm the directional nature of the voltage generation, we exchanged the 291 electrical leads, resulting in a change in polarity (Figure 5b). Upon reversing the connections, 292 the voltage's sign also reverses, yet it retains the same magnitude. This observation leads us 293 to the conclusion that the voltage's direction is contingent upon the direction of water flow 294 induced by evaporation (Figure 5b). 295  296 Figure 5. (a) Evaporation is significantly greater with a CWEG device than without. (b) 297 Induced potential for CWEGs connected in normal and reversed polarity (inset schematic 298 shows sample setup). 299 We conducted a comprehensive experiment to explore the relationship between induced 300 potential and pH value (electrolyte concentration). This involved systematically adjusting the 301 electrolyte concentration or pH value and measuring the corresponding voltage responses. 302 15  Three different electrolyte concentration solutions were prepared: a 0.2 mM NaCl aqueous 303 solution with a pH of ~7, a 2 mM NaCl aqueous solution with a pH of ~6 due to the increased 304 concentration of NaCl, and a 20 mM NaCl aqueous solution with a pH of ~5, reflecting a 305 further decrease in pH with increasing NaCl concentration. The induced voltage increased 306 initially as the NaCl solution molarity increased from 0.2 to 2 mM (Figure 6a). However, 307 when considering the 0.2 mM solution, the significantly low number of ions present might 308 not favor efficient surface ionization, leading to a diminished induced potential. In contrast, 309 the greater ion density in a 2 mM solution significantly enhances the voltage output, nearly 310 1.5 times greater compared to the 0.2 mM solution. Peculiarly, further elevating the ionic 311 concentration from 2 to 20 mM yields only minimal effects on the induced voltage. This 312 phenomenon can be comprehended through conventional electrokinetic theory. As the 313 concentration of the electrolyte rises, the thickness of the EDL diminishes. This outcome 314 leads to a reduction in the potential. 54, 55. 315 The electrolyte solutions of different composition such as NaOH solution, tap water, distilled 316 water, and ethanol were examined whether intrinsic properties of the liquid can affect the 317 performance of the CWEG device. Furthermore, Principal Component Analysis (PCA) was 318 applied to electrolyte solutions based on voltage and conductivity variables to gain valuable 319 insights (Figure S4). This analysis was conducted to clarify how these electrolytes interrelate 320 concerning their electrical and conductivity traits, revealing distinct behaviors for NaOH, 321 different water types, and ethanol. A positive correlation is observed in the case of the NaOH 322 electrolyte solution, indicating heightened electrical activity. Notably, tap water and distilled 323 water fall within the same quadrant, suggesting their similarity in terms of voltage and 324 conductivity attributes within this specific feature space. The juxtaposition of positive and 325 negative values in the first and second components respectively underscores the intricate 326 16  influence of NaCl on the voltage and conductivity data. In contrast, ethanol showcases 327 negative scores, signifying its distinct impact on the data's variability.  328 The induced potential (Figure 6) for different solutions varies in the following manner: NaCl 329 > NaOH > tap water > distilled water > ethanol, with voltages of 0.35, 0.21, 0.15, 0.11, and 330 0.06 V, respectively. The tap water was taken from the Tsukuba, Japan municipal water 331 supply. NaCl samples generate much higher voltage values due to the high ion concentration. 332 On the other hand, a low-polarity ethanol solution does not exhibit a strong EDL due to its 333 relatively low concentration of hydrogen protons. As a consequence, the resulting voltage is 334 lower. It's worth highlighting that the streaming potential is notably more pronounced in a 335 NaCl solution. A clear correlation is seen in the difference of measured conductivity, which 336 affects the ability to respond to the moving EDL boundary.  337 17   338 Figure 6. (a) Voltage generation with respect to electrolyte concentration, (b) mean values of voltage 339 generated from variety of electrolyte solutions and their conductivity data, (c) Effect of temperature 340 on device performance shown by Nyquist plot (d) Values for resistivity, conductivity, and voltage 341 with respect to temperature.   342 The charge carrier dynamics occurring at the interface between the CW and the electrolyte 343 can be characterized through the application of electrochemical impedance spectroscopy 344 (EIS). The electrochemical performance of the CWEG was studied using a NaCl electrolyte 345 in a two-electrode EIS system 56. In Figure 6 (c), Zim and Zre represent the imaginary and real 346 components, respectively, of the impedance (Z) as a function of frequency. Impedance is a 347 complex quantity that describes the opposition to the flow of an alternating current in a 348 circuit. Zim (Z imaginary) refers to the imaginary component of the impedance, which 349 18  represents the reactive part of the impedance. It is associated with the capacitive or inductive 350 elements in the system and is typically plotted along the vertical axis of a Nyquist plot. Zre (Z 351 real) refers to the real component of the impedance, which represents the resistive part of the 352 impedance. It is associated with the resistive elements in the system and is typically plotted 353 along the horizontal axis of a Nyquist plot. 354 In the context of Figure 6(c), the Nyquist plot displays the alteration of impedance (|Z|) as a 355 function of frequency, with Zim and Zre providing insights into the reactive and resistive 356 components of the impedance, respectively. The diameter of the semicircle within the 357 Nyquist plot serves as an indicator of the charge transfer resistance from the CWEG to the 358 electrolyte, particularly in relation to the temperature of the NaCl solution. 359 The apparent trend indicates that higher temperatures substantially enhance charge transfer 360 efficiency. This is evidenced by the smaller arc observed at higher temperatures in 361 comparison to the arc displayed at lower temperatures (Figure 6c). This finding underscores 362 that the acceleration of charge transfer is markedly achieved through elevated temperatures. 363 As the temperature decreases, voltage also decreases due to a decrease in conductivity of the 364 NaCl solution (Figure 6d). The maximum VOC is recorded as ∼0.39 V at 50 °C, where the 365 effects of temperature and water evaporation raise the generated voltage significantly. The 366 performance of the device experiences a gradual decline from its optimal state as the 367 electrolyte temperature diminishes. This deterioration continues until it reaches a minimum 368 value of ∼0.31 V at a temperature of around 10 °C. Hence, the water temperature around the 369 CWEG influences the capacity of power generation. These results suggest that the CWEG 370 would be most effective at producing electric energy in warm, dry environments.  371 Additionally, to realize the influence of temperature and water evaporation on the device's 372 performance, we conducted an additional comparison under concentrated 100-mW cm⁻² 373 19  sunlight, utilizing a solar simulator (Figure 7a). The voltage increased and decreased as the 374 solar simulator was turned on and off, respectively. During illumination, CW absorbs light 375 and due to photothermal heat generation, evaporation is enhanced, leading to an increase in 376 voltage. The voltage returns to a low value after the light is turned off; thus the combined 377 effect of solar heating and evaporation is clearly beneficial.  378 Subsequently, in order to ascertain that evaporation indeed serves as the primary source of 379 the induced potential, the top surface of the CW was subjected to moving air at ~2 m/s (with 380 a temperature of around 22 °C). Air movement enhanced the evaporation rate from the CW, 381 leading to a slight increase in voltage up to ∼0.36 V (Figure 7b). Upon ceasing the airflow, 382 the induced potential decreased to ∼0.35 V. This clear correlation between the output voltage 383 and the rate of water evaporation underscores the significance of this process. Subsequently, 384 the behavior of the streaming voltage performance was further explored through the 385 controlled removal and injection of water (Figure 7c). The beaker was initially filled with 386 water so that the CWEG was immersed 4 cm. Then, after a few seconds, when all water was 387 removed from the beaker, voltage quickly dropped to ∼0.13 V. The voltage difference started 388 to increase again and reached the maximum and stabilized with further water addition. This 389 behavior was highly reproducible. 390  391 Figures 7. Device performance in different ambient conditions. Effect of induced potential 392 on periodic application of (a) simulated solar light (b) air flow (c) ejecting and injecting water. 393 20  In this work, we also studied the effect of surface modification with an ultrathin TiO2 coating 394 on the CW fibers (CW-TiO2). A uniform 20-nm TiO2 coating was deposited on CW fiber 395 surfaces by atomic layer deposition (ALD) for a comparative study. The TiO2 coating was 396 annealed at 300 °C for 3 h after deposition. Microscopic morphology of the CW-TiO2 was 397 inspected with scanning electron microscopy (Hitachi FE-SEM SU8230) and EDX (Figure 398 8). SEM data show that CW-TiO2 comprises fibers with a diameter of about 5 µm (Figure 8a). 399 The EDX spectrum (Figure 8b), indicates peaks of O, Al, Si, and Ti and EDX mapping 400 confirms the presence of TiO2 on CW fibers. A carbon peak appears due to the supporting 401 carbon tape holding the sample. The overall morphology of CW-TiO2 reveals successful 402 coating of TiO2 on the CW. Crystallinity of the samples was characterized with X-ray 403 diffraction (XRD, RINT 2000, Rigaku). All peaks are indexed to the anatase phase of TiO2 404 (JCPDS card No. 21-1272). The diffraction pattern of the CW exhibited a broad feature at 2 405 theta - 23°, corresponding to amorphous SiO2 and Al2O3 structures57 (Figure S5a). However, 406 after the CW surface modification with TiO2, the XRD pattern of the CW-TiO2 exhibited 407 sharp peaks corresponding to TiO2 NPs. From the Nyquist plot (Figure S5b), it is clear that 408 TiO2 coating exhibits a notable increase of charge transfer efficiency, resulting in a smaller 409 arc than that of bare CW, which means that charge transfer is drastically enhanced by TiO2 410 NPs. The maximum short-circuit current density (Jmax) for CW and CW-TiO2 was found to be 411 0.51 mA and 0.82 mA, respectively, while the corresponding open-circuit voltages (Vmax) 412 were 0.35 V and 0.30 V, respectively (Figure S5c-S5d). Utilizing the formula Pmax = Jmax × 413 Vmax, where Pmax represents the maximum output power density, we calculated the power 414 densities for CW and CW-TiO2 to be 0.18 mW and 0.25 mW, respectively, based on a sample 415 size of 12×6 cm2. Although the voltage decreased with the TiO2 coating, the higher current 416 density resulted in a significant increase in the overall power density. This observation 417 indicates that while the TiO2 coating may lead to a reduction in voltage, it concurrently 418 21  enhances the current output, thereby improving the overall power generation performance of 419 the CWEG. 420 Hence, the substantial increase in power output with the TiO2 coating demonstrates the 421 potential for strategically improving CWEG performance through surface modifications with 422 appropriate materials.  423  424 Figure 8. EDX mapping of CW-TiO2, (a) SEM image, (b) EDX spectrum of CW and CW-425 TiO2, (c), (d), (e), (f), (g) and (h) elemental mapping of Al, Si, O, Ti and C (scale bar 426 represents 2 µm).  427 The evaporative nanogenerator is capable of generating direct current (DC) electricity, which 428 can be harnessed and stored in flexible supercapacitors. These supercapacitors can 429 subsequently be interconnected to meet the energy demands of diverse electronic devices, 430 effectively providing power for various applications. Figure 9 depicts the capillarity-coupled, 431 evaporation-induced voltage generated by a series of 6 CWEGs, yielding a maximum 432 potential of around 1.6 V. The configuration of these 6 CWEGs, along with the measurement 433 setup, is detailed in Figure 9. The power derived from these devices can be utilized to 434 illuminate a red LED, as demonstrated in Figure S6. Notably, the generated electricity is 435 scalable by employing devices in series, enabling it to charge a 1-µF capacitor to 1.5 V 436 22  within a 4-hour timeframe using 6 devices in series (Figure S7). The energy stored through 437 this process proves sufficient to maintain the illumination of a red LED at a substantial 438 intensity for over 1 hour and to power a small DC motor for a duration of 50 seconds (Video 439 S1). The comparison of CWEG’s electrical performance with latest water evaporation-based 440 power devices is presented in Table S2. 441 The outcomes of these experiments indicate that the proposed CWEG has the capacity to 442 generate sufficient power to operate small-scale electronic devices like LEDs, capacitors, and 443 micro motors. A single CWEG, measuring 12 × 6 cm², can produce approximately 0.35 V.  444 Furthermore, the overall output performance of the system can be maximized through proper 445 selection and matching  of load impedance as the real world application involves variable or 446 unpredictable load applied to the energy harvesters58, 59.  447 As the CWEG with size of 12 × 6 cm2 can generate ∼ 0.35 mV which is slightly more than 448 the standard redox potential of copper 60. Thus, the redox reactions did not occur on the 449 copper electrodes actively. However, the redox reaction can adversely influence long-term 450 practical application of the series-connected CWEG system 61. It's worth considering that 451 such redox reactions could potentially hinder the long-term practical application of series-452 connected CWEG devices. 453 The implications of these findings are significant, suggesting the potential application of micro-454 structured insulating materials for electricity generation. Moreover, the comparison with reported 455 phenomena, such as a single droplet lighting up LEDs or powering a light bulb 62, highlights the 456 remarkable efficiency of water evaporation-based electricity generation. 457 23   458 Figure 9. Six CWEG devices (a) as prepared (b) in scaffolds (c) connected in series (d) 459 connected in electrolyte solutions. (e) Scaling of CWEG performance and its use to power an 460 LED by charging of a commercial 1-µF supercapacitor. The inset shows a glowing LED 461 driven through a charged supercapacitor. 462  463  464 24  4. Conclusion 465 This research demonstrates a successful utilization of water's evaporative energy to create 466 straightforward, economical devices for producing clean and sustainable power. The 467 investigation demonstrates that the evaporation of ionic water from ceramic fibers generates 468 electric voltages, and remarkably, the Voc remains stable as long as the device is partially 469 immersed in water. Notably, the voltage resulting from evaporation effects can be extended 470 to around 1.6 V by connecting 12×6 cm² CWEGs in series 471 These CWEGs offer an environmentally friendly solution by utilizing water for power 472 generation. The unidirectional water movement, akin to transpiration, propels efficient charge 473 transfer. This dynamic yields a voltage drop between the two electrodes that are 474 asymmetrically positioned in water and air environments. This difference in environments 475 can also result in additional chemical potential difference. 476 Our results imply that a bundle of proposed CWEG can generate sufficient power for 477 operating commercial electronic devices. The results reported here demonstrates the great 478 potential for adopting micro-structured insulating material for generating electricity for night-479 time self-powered LEDs, capacitors, micro motors and household minigadgets, 480 environmental sensors, and so forth. This approach can improve the contact electrification 481 effect and its energy conversion efficiency in practical applications. Such low-cost ceramic 482 wools are suitable for self-powered sensors and systems that do not rely on input sources, 483 such as sunlight and heat. 484 In forthcoming research, our primary focus will be on addressing the redox reaction issue. 485 We plan to develop redox reaction-free electrodes to ensure the practical viability of the 486 CWEG system over extended operational periods. One future direction is to consider the use 487 25  of polymer meshes coated with conductive carbon materials to create electrodes that can 488 mitigate or eliminate these redox reactions. 489  490 ASSOCIATED CONTENT 491 Supporting Information. Figures S1-S7, Table S1-S2, Note S1 and Video S1.  492 Funding Sources 493 This work was funded by JSPS KAKENHI (16H06364) and CREST “Phase Interface 494 Science for Highly Efficient Energy Utilization” (JPMJCR13C3) from the Japan Science and 495 Technology Agency. 496 Acknowlegements: We would like to thank Dr. Takemura Taro, Miss Li Xianglan from the 497 Molecules and Materials Synthesis Platform, NIMS for Zeta Potenial measurment support. 498 The authors extend their appreciation to the Deputyship for Research and Innovation, 499 "Ministry of Education" in Saudi Arabia for funding this research (IFKSUOR3-615-1). O.M. 500 would like to thank the support of Tomsk State University Development Programme 501 (priority-2030) for this work. 502 References  503 1. H. J. Hovel, NASA STI/Recon Technical Report A, 1975, 76. 504 2. C. J. Brabec, N. S. Sariciftci and J. C. Hummelen, Adv. Funct. Mater., 2001, 11, 15-26. 505 3. I. Katsouras, K. Asadi, M. Li, T. B. Van Driel, K. S. Kjaer, D. Zhao, T. Lenz, Y. Gu, P. W. Blom and 506 D. Damjanovic, Nature materials, 2016, 15, 78. 507 4. Y. Zi, J. Wang, S. Wang, S. Li, Z. Wen, H. Guo and Z. L. Wang, Nat Commun., 2016, 7, 10987. 508 5. C. F. Maduko and U. B. Akuru. 509 6. R. Newell, D. Raimi and G. Aldana, Resources for the Future, 2019, 8-19. 510 7. A. M. Omer, Renewable and sustainable energy reviews, 2008, 12, 1789-1821. 511 8. C. Okkerse and H. Van Bekkum, Green Chemistry, 1999, 1, 107-114. 512 9. O. Ellabban, H. Abu-Rub and F. Blaabjerg, Renewable and Sustainable Energy Reviews, 2014, 513 39, 748-764. 514 10. V. Slabov, S. Kopyl, M. P. Soares dos Santos and A. L. Kholkin, Nano-Micro Letters, 2020, 12, 515 1-18. 516 11. K. Zhao, Y. Wang, L. Han, Y. Wang, X. Luo, Z. Zhang and Y. Yang, Nano-Micro Letters, 2019, 11, 517 1-19. 518 26  12. J. Yi, K. Dong, S. Shen, Y. Jiang, X. Peng, C. Ye and Z. L. Wang, Nano-micro letters, 2021, 13, 1-519 13. 520 13. Z. L. Wang and J. Song, Science, 2006, 312, 242-246. 521 14. F.-R. Fan, Z.-Q. Tian and Z. L. Wang, Nano energy, 2012, 1, 328-334. 522 15. A. J. Williams, M. F. Torquato, I. M. Cameron, A. A. Fahmy and J. Sienz, IEEE Access, 2021, 9, 523 77493-77510. 524 16. J. Singh, R. Kaur and D. Singh, International Journal of Energy Research, 2021, 45, 118-140. 525 17. R. Yang, Y. Qin, L. Dai and Z. L. Wang, Nature nanotechnology, 2009, 4, 34. 526 18. M. Grätzel, nature, 2001, 414, 338-344. 527 19. X. Wang, J. Song, J. Liu and Z. L. Wang, Science, 2007, 316, 102-105. 528 20. Y. Jia, Y. Pan, C. Wang, C. Liu, C. Shen, C. Pan, Z. Guo and X. Liu, Nano-Micro Letters, 2021, 13, 529 1-12. 530 21. G. J. Snyder and E. S. Toberer, in materials for sustainable energy: a collection of peer-531 reviewed research and review articles from Nature Publishing Group, World Scientific, 2011, 532 pp. 101-110. 533 22. Z.-H. Lin, Y. Yang, J. M. Wu, Y. Liu, F. Zhang and Z. L. Wang, The journal of physical chemistry 534 letters, 2012, 3, 3599-3604. 535 23. Y. Yang, G. Zhu, H. Zhang, J. Chen, X. Zhong, Z.-H. Lin, Y. Su, P. Bai, X. Wen and Z. L. Wang, 536 ACS nano, 2013, 7, 9461-9468. 537 24. Z. Liu, C. Liu, Z. Chen, H. Huang, Y. Liu, L. Xue, J. Sun, X. Wang, P. Xiong and J. Zhu, 2023. 538 25. L. Li, S. Feng, Y. Bai, X. Yang, M. Liu, M. Hao, S. Wang, Y. Wu, F. Sun and Z. Liu, Nature 539 Communications, 2022, 13, 1043. 540 26. R. Zhang, S. Wang, M. H. Yeh, C. Pan, L. Lin, R. Yu, Y. Zhang, L. Zheng, Z. Jiao and Z. L. Wang, 541 Advanced Materials, 2015, 27, 6482-6487. 542 27. J. Yin, J. Zhou, S. Fang and W. Guo, Joule, 2020, 4, 1852-1855. 543 28. J. Feng, M. Graf, K. Liu, D. Ovchinnikov, D. Dumcenco, M. Heiranian, V. Nandigana, N. R. 544 Aluru, A. Kis and A. Radenovic, Nature, 2016, 536, 197. 545 29. J. Osterle, Applied Scientific Research, Section A, 1964, 12, 425-434. 546 30. D. Burgreen and F. Nakache, J. Appl. Mech., 1965, 32, 675-679. 547 31. J. Yang, F. Lu, L. W. Kostiuk and D. Y. Kwok, Journal of Micromechanics and Microengineering, 548 2003, 13, 963. 549 32. P. Král and M. Shapiro, Physical review letters, 2001, 86, 131. 550 33. S. Ghosh, A. Sood and N. Kumar, Science, 2003, 299, 1042-1044. 551 34. J. Liu, L. Dai and J. W. Baur, Journal of applied physics, 2007, 101, 064312. 552 35. P. Dhiman, F. Yavari, X. Mi, H. Gullapalli, Y. Shi, P. M. Ajayan and N. Koratkar, Nano letters, 553 2011, 11, 3123-3127. 554 36. W. Guo, C. Cheng, Y. Wu, Y. Jiang, J. Gao, D. Li and L. Jiang, Adv.  Mater., 2013, 25, 6064-555 6068. 556 37. G. Xue, Y. Xu, T. Ding, J. Li, J. Yin, W. Fei, Y. Cao, J. Yu, L. Yuan and L. Gong, Nature 557 nanotechnology, 2017, 12, 317-321. 558 38. T. Ding, K. Liu, J. Li, G. Xue, Q. Chen, L. Huang, B. Hu and J. Zhou, Advanced Functional 559 Materials, 2017, 27, 1700551. 560 39. F. Zhao, H. Cheng, Z. Zhang, L. Jiang and L. Qu, Advanced Materials, 2015, 27, 4351-4357. 561 40. F. Zhao, Y. Liang, H. Cheng, L. Jiang and L. Qu, Energy & Environmental Science, 2016, 9, 912-562 916. 563 41. Y. Lu, H. Wu, Q. Yang, J. Ping, J. Wu and J. Liu, Adv. Sustainable Sys,, 2019, 1900012. 564 42. X. Gao, T. Xu, C. Shao, Y. Han, B. Lu, Z. Zhang and L. Qu, J. Mater. Chem. A, 2019, 7, 20574-565 20578. 566 43. J. Yin, X. Li, J. Yu, Z. Zhang, J. Zhou and W. Guo, Nature nanotechnology, 2014, 9, 378-383. 567 44. J. Yin, Z. Zhang, X. Li, J. Yu, J. Zhou, Y. Chen and W. Guo, Nature communications, 2014, 5, 1-6. 568 27  45. B. Ji, N. Chen, C. Shao, Q. Liu, J. Gao, T. Xu, H. Cheng and L. Qu, Journal of Materials 569 Chemistry A, 2019, 7, 6766-6772. 570 46. M. Kaur, S. Ishii, S. L. Shinde and T. Nagao, ACS Sustainable Chemistry & Engineering, 2017, 5, 571 8523-8528. 572 47. Z. H. Lin, G. Cheng, S. Lee, K. C. Pradel and Z. L. Wang, Advanced Materials, 2014, 26, 4690-573 4696. 574 48. Z. H. Lin, G. Cheng, L. Lin, S. Lee and Z. L. Wang, Angewandte Chemie International Edition, 575 2013, 52, 12545-12549. 576 49. S. Lin, X. Chen and Z. L. Wang, Chemical Reviews, 2021, 122, 5209-5232. 577 50. M. Kaur, S. Ishii, R. Nozaki and T. Nagao, Scientific reports, 2021, 11, 1-10. 578 51. M. Kaur and T. Nagao, Energy & Fuels, 2022, 36, 11443-11456. 579 52. Z. L. Wang, Journal, 2012. 580 53. M. R. Das, J. M. Borah, W. Kunz, B. W. Ninham and S. Mahiuddin, Journal of Colloid and 581 Interface Science, 2010, 344, 482-491. 582 54. F. H. van der Heyden, D. Stein and C. Dekker, Physical review letters, 2005, 95, 116104. 583 55. F. H. van der Heyden, D. J. Bonthuis, D. Stein, C. Meyer and C. Dekker, Nano letters, 2007, 7, 584 1022-1025. 585 56. H.-C. Ho, K. Chen, T. Nagao and C.-H. Hsueh, The Journal of Physical Chemistry C, 2019, 123, 586 21103-21113. 587 57. O. Saber and H. M. Gobara, Egyptian Journal of Petroleum, 2014, 23, 445-454. 588 58. L.-B. Huang, W. Xu, C. Zhao, Y.-L. Zhang, K.-L. Yung, D. Diao, K. H. Fung and J. Hao, ACS 589 applied materials & interfaces, 2020, 12, 24030-24038. 590 59. M. Mariello, E. Scarpa, L. Algieri, F. Guido, V. M. Mastronardi, A. Qualtieri and M. De Vittorio, 591 Energies, 2020, 13, 1625. 592 60. A. Bard, Standard potentials in aqueous solution, Routledge, 2017. 593 61. Z. Lin and Z. Yang, Droplet, e97. 594 62. W. Xu, H. Zheng, Y. Liu, X. Zhou, C. Zhang, Y. Song, X. Deng, M. Leung, Z. Yang and R. X. Xu, 595 Nature, 2020, 578, 392-396. 596  597