# Fileset

[chemosensors-11-00564.pdf](https://mdr.nims.go.jp/filesets/9d0ef2e6-cc5f-4bf1-840c-6e406dfb9c24/download)

## Creator

Shinya Kano, [Jin Kawakita](https://orcid.org/0000-0002-4821-4150), Shohei Yamashita, Harutaka Mekaru

## Rights

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

## Other metadata

[Water Vapor Condensation in Nanoparticle Films: Physicochemical Analysis and Application to Rapid Vapor Sensing](https://mdr.nims.go.jp/datasets/fc7a0d1f-166b-4464-811b-5000d2f68ca3)

## Fulltext

Water Vapor Condensation in Nanoparticle Films: Physicochemical Analysis and Application to Rapid Vapor SensingCitation: Kano, S.; Kawakita, J.;Yamashita, S.; Mekaru, H. WaterVapor Condensation in NanoparticleFilms: Physicochemical Analysis andApplication to Rapid Vapor Sensing.Chemosensors 2023, 11, 564.https://doi.org/10.3390/chemosensors11110564Academic Editor: Pi-Guey SuReceived: 22 September 2023Revised: 31 October 2023Accepted: 9 November 2023Published: 14 November 2023Copyright: © 2023 by the authors.Licensee MDPI, Basel, Switzerland.This article is an open access articledistributed under the terms andconditions of the Creative CommonsAttribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).chemosensorsArticleWater Vapor Condensation in Nanoparticle Films:Physicochemical Analysis and Application to RapidVapor SensingShinya Kano 1,2,* , Jin Kawakita 3, Shohei Yamashita 4 and Harutaka Mekaru 1,21 Human Augmentation Research Center, National Institute of Advanced Industrial Science andTechnology (AIST), Kashiwa 270-0882, Japan; h-mekaru@aist.go.jp2 Sensing System Research Center, National Institute of Advanced Industrial Science and Technology (AIST),Tsukuba 305-8564, Japan3 National Institute for Materials Science, 1-1 Namiki, Tsukuba 305-0044, Japan; kawakita.jin@nims.go.jp4 High Energy Accelerator Research Organization, 1-1 Oho, Tsukuba 305-0801, Japan; yamasho@post.kek.jp* Correspondence: shinya-kano@aist.go.jpAbstract: Nanomaterial-based humidity sensors hold great promise for water vapor detection be-cause of their high sensitivity and fast response/recovery. However, the condensation of water innanomaterial films remains unclear from a physicochemical perspective. Herein, the condensation ofwater vapor in silica nanoparticle films was physicochemically analyzed to bridge the abovemen-tioned gap. The morphology of surface-adsorbed water molecules was characterized using infraredabsorption spectroscopy and soft X-ray absorption spectroscopy, and the effect of RH on the amountof adsorbed water was observed using a quartz crystal microbalance. The adsorbed water was foundto exist in liquid- and ice-like states, which contributed to high and low conductivity, respectively. Thelarge change in film impedance above 80% RH was ascribed to the condensation of water betweenthe nanoparticles. Moreover, RH alteration resulted in a colorimetric change in the film’s interferencefringe. The obtained insights were used to construct a portable device with response and recoverytimes suitable for the real-time monitoring of water vapor. Thus, this study clarifies the structure ofwater adsorbed on nanomaterial surfaces and, hence, the action mechanism of the correspondingnanoparticle-based sensors, inspiring further research on the application of various nanomaterials tovapor sensing.Keywords: humidity sensor; water vapor; capillary condensation; Kelvin equation; impedance;infrared absorption spectroscopy; X-ray absorption spectroscopy; quartz crystal microbalance; opticalsurface imaging; nanoparticle film1. IntroductionIn the field of improving industrial, agricultural, and human activities, much attentionhas been drawn to the monitoring of water evaporation dynamics, which typically involvesthe real-time detection of moisture [1]. The performance of humidity sensors, e.g., theirresponsiveness to water vapor, can be enhanced through the use of humidity-sensitivenanomaterials [2–5], particularly in the form of films with large surface-to-volume ratios.Among the several types of humidity sensors, impedance-type ones are easily incorporatedinto conventional electronic circuits and have therefore been most widely studied and usedin portable devices [6–12]. In our previous studies, insulator nanoparticles were used ashumidity-sensitive materials to fabricate sensors exhibiting rapid exponential impedancechanges in response to humidity variation [13,14]. This sensing mechanism was appropriateto detect high humid air because a high impedance ratio regarding high (95%) versus low(40%) humidity (Zhigh/Zlow = 102 using 1 Hz ac voltage [14]) was obtained. However, thephysicochemical origin of this dependence has not yet been elucidated experimentally.Chemosensors 2023, 11, 564. https://doi.org/10.3390/chemosensors11110564 https://www.mdpi.com/journal/chemosensorshttps://doi.org/10.3390/chemosensors11110564https://doi.org/10.3390/chemosensors11110564https://creativecommons.org/https://creativecommons.org/licenses/by/4.0/https://creativecommons.org/licenses/by/4.0/https://www.mdpi.com/journal/chemosensorshttps://www.mdpi.comhttps://orcid.org/0000-0001-8117-5838https://orcid.org/0000-0002-1120-6018https://doi.org/10.3390/chemosensors11110564https://www.mdpi.com/journal/chemosensorshttps://www.mdpi.com/article/10.3390/chemosensors11110564?type=check_update&version=1Chemosensors 2023, 11, 564 2 of 14Proton hopping through water layers on nanomaterials was suggested to become dominantat high humidity [7–9], although the detailed chemical morphology of water in such layershas not been studied.Capillary condensation is an important phenomenon affecting the behavior of nanomaterial-based humidity sensors. According to the theory of capillary effects (Kelvin equation), waterdroplets can form between particles with sizes of several nanometers in air with <100% relativehumidity (RH) [5,15–17]. For example, Baumgartner et al. demonstrated the presence ofcapillary-condensed water in mesoporous silica with pore diameters of 5–13 nm using infrared(IR) spectroscopy [18]. Cappelli et al. described the operation of semiconductor TiO2 nanoparticle-based humidity sensors by examining water adsorption on the nanoparticle surface with thecapillary effect [19]. Asay et al. examined the effect of environmental humidity on the formationof water layers on flat silicon oxide surfaces using IR spectroscopy [20–22], revealing that ice-likestructures with approximately three molecular layers were produced below 30% RH, whereasa liquid-like structure began to form for RH above 60%. Seo et al. showed that the electricalconductivity of flat silicon oxide substrates was dependent on the phase of the water layers andused this dependence to rapidly detect humid breath air [23]. In our previous work, movablewater molecules began to adsorb on the flat surface of a galvanic action-based humidity sensorfor RH concentrations higher than 60% RH [12]. The abovementioned insights can help usbetter understand the behavior of nanomaterials exposed to humid air and improve moisturegas detection.Herein, to shed light on the detailed chemical morphology of water adsorbed oninsulator nanomaterials, we investigated the condensation of water vapor in nanoparticlefilms by physicochemical methods. The presence and structure of capillary-condensedwater in these films were evaluated using Fourier-transform infrared (FT-IR) spectroscopyand soft X-ray absorption spectroscopy (XAS) in ambient air and in dry He gas. A quartzcrystal microbalance (QCM) and impedance analysis were used to evaluate the effectsof humidity on the mass of adsorbed water and film impedance, respectively. Moreover,we examined the evolution of the film surface with the change in impedance to correlateimpedance variation with water condensation in the film, and simultaneously reveal thedynamics of the condensation in real time. Cappelli et al. investigated a physicochemicalsensing mechanism of semiconductor TiO2 nanoparticle-based humidity sensors usingQCM [19]. In this study, we used an insulator SiO2 nanoparticle-based humidity sensorto focus on the effect of capillary-condensed water. Proton hopping in physisorbed waterand condensed water was elucidated since the current through the SiO2 nanoparticles wasnegligible. The simultaneous observation of sensor output and optical surface imagingproved to be useful for determining the relationship between impedance change andmacroscopic water condensation [12]. Finally, we constructed a nanofilm-based systemfor detecting water vapor in real time, demonstrating that its response and recovery timesallowed for rapid vapor detection.2. Materials and Methods2.1. TheoryFigure 1a illustrates the response of a humidity sensor to an increase in RH and itssubsequent recovery upon the restoration of the initial conditions. The response time(t90) is defined as the time required to reach 90% of the total sensor output after the RHincreases [4], and the recovery time (t10) is defined as the time required for the output todrop by 90% upon the restoration of the original RH.Chemosensors 2023, 11, 564 3 of 14Figure 1. (a) Response and recovery of humidity sensor. (b) Theoretical relationship between poresize of nanoporous material and relative humidity (RH) corresponding to onset of water vaporcondensation at 25 ◦C. (c) Illustration of water vapor condensation in nanoparticle array betweentwo electrodes. At RH < threshold RH (RHt), voids between nanoparticles are partially filled, andcondensation occurs at points where nanoparticles contact each other. At RH > RHt, voids are fullyfilled with condensed water.In this study, we pay attention to the condensed water in the nanoparticle film andinvestigate the electrical conductivity of the water directly. In the field of humidity sensors,semiconductor nanoparticles such as TiO2 [19,24,25] and SnO2 [26] are widely used. Thissolution uses an electronic modulation of adsorbed molecular water on the semiconductors,which is used to detect low humidity. In this case, it is difficult to separate the conductionof the condensed water from that of semiconductor nanoparticles. Since the semiconductorgives sufficient conduction, the electrical conduction of the condensed water is unseparated.To see the effect of the capillary condensed water on conductivity, we chose a nonporousinsulator nanoparticle thin film as a humidity-sensitive material. This solution allows us tosee the electrical conductivity of the condensed water directly.According to the Kelvin equation, nanopore capillaries become spontaneously filledwith condensed water at RH < 100% [15,17,27] when the following condition is satisfied:ln(pp0)= −2γVmcos θrRT. (1)Here, p is the partial pressure of water vapor, p0 is the pressure of water vapor at satura-tion at temperature T, γ is the surface tension of liquid water at temperature T (72.0 mN/mat 25 ◦C), Vm is the molar volume of water (1.8 × 10−5 m3/mol), θ is the contact angle ofwater on the nanopore material, R is the universal gas constant (8.31 J/(K mol)), and r is theradius of the meniscus curvature. At T = 25 ◦C and θ = 0◦ (highly hydrophilic nanopores),the equation is reduced to:r = −1.05 × 10−9ln(pp0) . (2)If the pore size is assumed to be roughly two times r, the critical pore size providingcondensation equals −2.1 × 10−9/ln(p/p0). Figure 1b shows the dependence of the criticalpore size (2r) on relative humidity (p/p0) at 25 ◦C and presents the threshold humidity(RHt) at which capillary condensation occurs in nanoscale voids.Another key parameter is the nanoparticle diameter D. According to Hudson [28],simple trigonometric analysis suggests that vapor condensation occurs at a critical meniscusChemosensors 2023, 11, 564 4 of 14radius rc of 0.22475 (D/2) for the closest hexagonal packing model and at 0.4142 (D/2)for the cubic packing model. For a nanoparticle film with D = 50 nm, rc roughly rangesfrom 5.6 to 10.4 nm, i.e., the critical pore size equals 11.2–20.8 nm in the ideal case andis expected to be randomly distributed in this range in the case of a real nanoparticlefilm. According to Figure 1b, a critical pore size of 11.2 nm corresponds to RHt = 83%.At RH < 83%, the condensation of water vapor occurs only at points where nanoparticlescontact each other [29]. The nanoscale voids are partially filled with small liquid clusters. Inthis case, no large current is expected to flow between the electrodes upon the applicationof a bias voltage. On the contrary, the voids are completely filled with condensed waterat RH > 83%. At this point, conductance is expected to sharply increase because of ionconduction through the condensate. This phenomenon is similar to the percolation ofconductive filaments in conductive and insulator composites [30]. Figure 1c illustratescondensation in nanoparticle films. For a nanoparticle film with D = 10 nm, the criticalpore size is 2.4–4.2 nm. For a pore size of 2.4 nm, water vapor condensation is expected tostart at RHt = 42%.2.2. Experimental MethodsIn the present study, we adopted a nonporous silica nanoparticle as an insulatingstructure to condense water vapor. A colloidal solution of silica nanoparticles (50 nmdiameter particle: Sigma-Aldrich (St. Louis, MO, USA) and 10 nm diameter particle:Micromod Partikel Technology, Rostock, Germany) was diluted in ethanol (Fujifilm, Tokyo,Japan), with a proportion 10 times higher, and filtered through a syringeless filter with apore size of 0.45 µm (Whatman, Maidstone, UK) to remove agglomerated particles. For IRspectroscopy and XAS measurements, a dispersion of 10 nm nanoparticles was spin-coatedon a gold-coated silicon substrate to increase the amount of condensed water on the surface.Since the spectroscopic measurements were carried out under constant humidity, hysteresiseffect was not discussed in these results. For impedance measurements, it was important toconsider the hysteresis effect on humidity changes. In our previous work [14], we foundthat the hysteresis effect in the 50 nm nanoparticle device was smaller than that in the10 nm one. Thus, we used 50 nm nanoparticles coated over interdigitated electrodes for theimpedance measurement.FT-IR absorption spectra were recorded under ambient conditions (25.5 ◦C and 50%RH) using a FT-IR microscope (LUMOS, Bruker, Billerica, MA, USA) in reflection mode.The IR path was dehumidified using molecular sieves to prevent absorption loss due toambient vapor. The acquired spectra were deconvoluted using the Gaussian function fitimplemented in OriginPro 2023 software (Version 10.0.0.154, OriginLab).The presence of water molecules on the nanoparticle film was further confirmed bysoft XAS, which can detect surface adsorbates with high sensitivity, as the escape depth ofexcited electrons is of the order of several nanometers. XAS measurements were performedin vacuum (~10−6 Pa) and He gas (ambient pressure, G1 purity) using a BL19B beamlineat the Photon Factory of KEK (Tsukuba, Japan). Incident X-rays were generated usingsynchrotron radiation. An APPLE-II-type undulator (2.5 GeV maximum electron energy)with a variable included angle Monk-Gillieson monochromator and varied-line-spacinggratings was used. The beam size was approximately 200 µm × 50 µm. The beam intensitywas 2.4 × 1011 photons/s at 400 eV. Measurements were conducted in fluorescence modeusing a silicon drift detector (SDD with C2 window, Amptek, Bedford, MA, USA) placedat an angle of 45◦ relative to the incident beam. A SiN window was placed between thebeamline and the measurement chamber. Acquisition was performed within an energyrange of 520–560 eV using an energy step of 0.1 eV and a dwell time of 0.5 s.To correlate the film impedance with the amount of adsorbed water, we used QCMand impedance sensors (Figure 2a,b). The abovementioned solution was spin-coated on (i)a bare QCM sensor (SEN-20E-H-10, TAMADEVICE Co., Ltd.; resonant frequency = 20 MHz,diameter = 8 mm) and on (ii) interdigitated gold electrodes on a thermally oxidized siliconsubstrate. The interdigitated electrodes were designed and fabricated using standardChemosensors 2023, 11, 564 5 of 14photolithographic techniques. The impedance sensor size was 10 mm × 10 mm, and theinsulating oxide layer on the substrate had a thickness of 1 µm. The line spacing of theinterdigitated electrode was 10–20 µm. Figure 2c shows an optical microscopy image ofthe top surface of the impedance sensor. As the nanoparticle film had a thickness smallerthan 1 µm, it was almost transparent to visible light. Optical reflection at the interface(air/nanoparticle film/bulk SiO2) resulted in the emergence of interference fringe, whichwas an important feature for determining the presence of condensed water between thenanoparticles. Figure 2d shows a cross-sectional image of the silica film on the siliconsubstrate, revealing that the nanoparticles (D = 50 nm) were randomly packed.Figure 2. Schematic illustration of (a) quartz crystal microbalance (QCM) and (b) impedance sensors.(c) Optical microscopy image of impedance sensor surface. (d) Cross-sectional scanning electronmicroscopy image of thin silica nanoparticle film on silicon substrate.We carried out the evaluation of the sensor response to humidity changes in anenvironment-controlled chamber (SH-222, ESPEC, Osaka, Japan). The resonant frequencyshift of the QCM sensor was measured by a portable measurement system (THQ-100P-SW,TAMADEVICE Co., Ltd., Kawasaki, Japan). The sampling time of the frequency was 1 s.The impedance of the nanoparticle films was monitored by an LCR meter (ZM 2376, NFInstruments, Yokohama, Japan) controlled by LabVIEW software (National Instruments,Austin, TX, USA). The amplitude and frequency of the applied voltage were 1.0 V and0.2–10 kHz, respectively. A portable measurement system with an impedance-type sen-sor was developed for rapid vapor detection. The system with a portable data logger(MetaWearC, MbientLab, San Francisco, CA, USA) and a voltage divider comprising a resis-tor (10 or 100 MΩ) remotely monitored the changes in the film resistance via a custom-madeAndroid API.Simultaneous monitoring of the sensor surface and impedance was performed asfollows. The impedance sensor was mounted on an aluminum block cooled using a chiller(UC180, Solid-State Cooling Systems, Wappingers Falls, NY, USA), and placed under anoptical microscope (VHX-7000, Keyence, Osaka, Japan). The environmental humidity onthe surface was adjusted by controlling the temperature of the sensor using a thermo-controller. The temperature and RH of the ambient air were 23–24 ◦C and 30%, respectively,corresponding to a dew point of 3–4 ◦C. Once the temperature of the sensor decreased to3 ◦C, the effective environmental humidity around the sensor surface increased to 100%RH. During cooling, the sensor surface was observed by the optical microscope, and thevoltage was recorded by the portable measurement system as explained above.Chemosensors 2023, 11, 564 6 of 14For the demonstration of water vapor detection, we used water vapor naturallygenerated from liquid water in a beaker. The impedance sensor was intermittently exposedto the water vapor to obtain the transient sensor response in voltage. The sensor positionwas mechanically controlled by a solenoid-based linear actuator. The height of the sensorover the beaker was 2 cm from the edge. The water was mildly warmed at 60 ◦C to generatevapor. The temperature and relative humidity in the ambient air were 24 ◦C and 54%,respectively.3. Results and Discussion3.1. IR Spectroscopy and XAS MeasurementsThe critical pore size of the 10 nm silica nanoparticle film was 2.4 nm. Hence, at 50%RH, the voids of this film were expected to be filled with capillary condensed water, asshown in Figure 1b. To prove the applicability of the Kelvin equation to our nanoparticlefilm, we recorded its IR absorption spectrum. The IR absorption spectrum of the 10 nmsilica nanoparticle film on the gold surface (Figure 3a) featured water-related peaks at1645 (bending mode) and 3400 cm−1 (symmetric and asymmetric modes) [22,29]. Previousstudies have shown that the broad O–H stretching band at 3000–3700 cm−1 can be decon-voluted into the peaks of liquid-like (around 3400 cm−1) and ice-like (around 3200 cm−1)water [20,21,31–33]. The difference between these forms can be explained by the strengthof hydrogen bonding between their water molecules, which is larger for ice-like water [22].The broad peak around 3400 cm−1 in Figure 3b has a shape different from that of the bulkwater peak (dotted line). Moreover, compared to that of bulk water, the peak was shifted tolower wavenumbers, indicating that the condensate formed in the nanoparticle film mayhave a larger content of ice-like water and thus implies the presence of more ordered watermolecules on the film surface [22].Figure 3. (a) Infrared absorption spectrum of 10 nm silica nanoparticle film recorded in air at 25.5 ◦Cand 50% RH. (b) Results of water peak deconvolution. Black dots, dashed lines, red solid line, anddotted line represent experimental data, deconvoluted peaks, deconvoluted peak sum, and bulkwater peak (downscaled five-fold), respectively. (c) Illustration of liquid and ice-like water betweennanoparticles.The electrical conductivity of the nanoparticle film (σ) is determined by the number ofcharge carriers and their mobility in transfer channels. Figure 3c illustrates the proposedconformation of water molecules in the nanoparticle film. The presence of the liquid-likewater peak (3445 cm−1) suggested that water condensation occurred in the voids betweenChemosensors 2023, 11, 564 7 of 14the nanoparticles. Liquid-like water is well suited for the transfer of hydronium (H3O+)and hydroxide (OH−) ions, thus providing high conductivity [34,35]. The elevated numberof ions involved in charge transport can also account for high conductivity. When liquidclusters are macroscopically connected between the electrodes (Figure 1c), a sufficientcurrent is obtained in response to the applied voltage. Therefore, condensed water causes alarge change in sensor impedance. In contrast, water molecules physisorbed on the surface(up to three monolayers [20]) has an ice-like structure (3240 cm−1): the water molecules areweakly hydrogen-bonded each other. The first layer of the water molecules is also bondedto the silicon oxides. In this case, proton transfer through the water molecular network(Grotthuss mechanism) plays a major role in electrical conduction [34,36–38]. Given thelimitations of this transfer pathway and the small number of free protons, ice-like watercontributes to small σ.Figure 4a shows the oxygen K-edge spectrum of the 10 nm nanoparticle film recordedin He gas. The strong peak at 537–538 eV was ascribed to oxygen in silica nanoparticles,while the weak peak at 534 eV was attributed to oxygen in water molecules. Notably,the latter peak was located at a lower energy than that of bulk water (535 eV [39–41]),which indicates that several molecular water layers with an ice-like structure were stillpresent in an atmosphere of dry He gas. On the contrary, no peak of water moleculesphysisorbed on the surface was observed in vacuum (10−6 Pa), as these molecules wereforcibly removed by vacuum pumping. The thermogravimetric analysis of Zhuravlevshowed that water multilayers on a silica surface can be transformed into a single molecularlayer by vacuuming at 25 ◦C [42]. Thus, chemisorbed water molecules possibly remainedon the surface in vacuum. Figure 4b illustrates the abovementioned results.Figure 4. (a) Oxygen K-edge X-ray absorption spectra of 10 nm silica nanoparticle film recorded inHe gas (blue) and under vacuum (red). (b) Schematic illustration of water physisorbed in void spacebetween nanoparticles under vacuum and in He gas.3.2. QCM and Electrical Sensor MeasurementsFigure 5a shows the frequency shift (∆f ) of the QCM sensor in response to a change inRH from 35% to 95% and then back to 35% in steps of 5%. The mass of water adsorbed onquartz (∆m) was calculated using the Sauerbrey theory as∆m = − A2.26 × 10−6 f 20∆ f , (3)where A is the area of the sensor surface, and f 0 is the resonant frequency of quartz [11,43]. Inthis study, A equaled 0.196 cm2 (5 mm disk), and f 0 equaled 20 MHz (∆m/∆f = 0.22 ng/Hz).The mass of the sensing film was estimated to be 400 ng according to the change in the QCMfrequency before and after the film deposition. Figure 5b shows ∆m as a function of RH togetherwith the corresponding fit with a quadratic function, revealing that this function well modeledthe evolution of ∆m. The ratio of the adsorbed water to the sensing film was roughly 0.5 at80% RH and 0.8 at 95% RH.Chemosensors 2023, 11, 564 8 of 14Figure 5. (a) Dynamic frequency shift of QCM sensor upon RH change from 35% to 95% and back to35%. (b) Mass of adsorbed water as a function of RH for QCM sensors coated and not coated with50 nm nanoparticles. Red dashed line presents quadratic fit for nanoparticle-coated sensor.According to other literature measuring nanoparticle-based sensors by QCM, the QCMsensor response as a function of humidity typically showed two regimes: linear responsein lower humidity and exponential increase in higher humidity [19,33]. In Figure 5b, thethreshold relative humidity in the transition is around 60% RH. From the Kelvin equationin Figure 1b, the void between the nanoparticles was filled with the condensed water at83% RH. This indicates that the large increase in mass above 75% RH can be attributed to thecapillary condensation in the void and physisorbed water molecular networks are formed.It should be noted that a chemisorption of water molecules on SiO2 nanoparticles occurredin lower humidity than 60% RH because a detectable current due to proton conductionthrough the molecules was not observed as shown later. The mass increase observed forthe QCM sensor without nanoparticles (control) at any given RH was roughly nine timeslower than that observed for the sensor coated with the 50 nm nanoparticle film.Figure 6a shows the dependence of the sensor impedance on frequency at differentRH. The impedance in the lower-frequency region decreased with increasing RH. Thisindicates that the adsorption of water molecules reduced the impedance. The impedanceat a lower frequency than 10 Hz in 60% RH is attributed to the proton transfer, whichcorresponds to the resistive component of the impedance. As increasing the humidity, theresistive component of the impedance was reduced, and the plateau plots were extendedto a higher frequency region. Considering the impedance as a function of frequency, weused impedance at 5 Hz to detect water vapor in the nanoparticle film. In this frequency, aresistive component was dominant on the electrical conduction. We noted that a detectablecurrent due to proton conduction was not observed in lower than 60% RH: a physiosorbedmolecular network over the nanoparticles was too thin to transfer mobile protons.Figure 6. (a) Impedance as a function of frequency at different RH. Dotted line represents inverserelationship between impedance and frequency. (b) Effects of RH on mass of water adsorbed onnanoparticle film and impedance at 5 Hz.Chemosensors 2023, 11, 564 9 of 14According to the work of Cappelli [19], the electrical model for nanoparticle films inhumid environments ideally includes three capacitance components: non-ideal capacitancein electrode/film interface (CE), nanoparticle-film capacitance (CF), and fixed parasiticcapacitance (Cp). In this work, the effects of CE and Cp appear as follows.The impedance dominated by CE is observed in the plot of 95% RH around 1 Hz. Thevalue of the capacitance is 800 nF using the equation C = 1/2πfZ where the impedance Z is200 kΩ at f = 1 Hz at 95% RH (noted by the square in Figure 6a). The electrode surface underthe nanoparticles S is approximately 0.10 cm2. If we assume that the relative permittivity εof water is 80 and the vacuum permittivity ε0 is 8.85 × 10−14 F/cm, we roughly estimatethe thickness of the electric double layer d = εε0S/C = 8.9 × 10−7 cm = 8.9 nm. The effectof the capacitance is dependent on humidity and the capacitance gradually decreases (i.e.,impedance increases) as the humidity is reduced. Thus, it is reasonable that the origin ofCE is the electrical double-layer capacitance formed by a thin film of condensed water.The impedance at a higher frequency than 10 Hz in 60% RH is due to Cp. The originof the parasitic capacitance is the silicon substrate. The top electrode area including padelectrodes S is approximately 0.12 cm2. The thickness of the thermally oxidized siliconin the substrate d is 10−4 cm (=1 µm). A silicon layer used in the base of the substrateworks as the other side of the electrode. The relative permittivity of SiO2 ε is 3.9 and thevacuum one ε0 is 8.85 × 10−14 F/cm. According to the calculation of a capacitance in aparallel plate (C = εε0S/d), we obtain C = 0.4 nF. When we assume the ac frequency is10 Hz, the impedance due to C is 40 MΩ. This value corresponds to the triangle in Figure 6aon the curve. Since the capacitance is not dependent on humidity, we conclude that thiscapacitance is parasitic, not an electric double layer one.Figure 6b shows the effects of RH on impedance at 5 Hz (filled circles) and the massof adsorbed water (open circles), revealing that impedance exponentially decreased from28 MΩ at 70% RH to 0.06 MΩ at 95% RH. As in the case of the QCM sensor, condensedwater contributed to an increase in electrical conductivity at over 80% RH. The change inimpedance at smaller than 80% RH was mainly ascribed to the molecular layers of ice-likewater formed on the nanoparticles. Notably, the electrical conductivity of liquid water isthree orders of magnitude larger than that of ice-like water because of the higher mobileion concentration of the former [44]. It should be noted that the impedance at lower than60% RH is much larger than 100 MΩ because the water molecular layers are too thin forprotons to percolate in. In this study, we focus on the formation of the water molecularnetwork.3.3. Water Vapor Detection by Portable SystemFigure 7a shows the circuit of the portable system used for rapid water vapor detection.Here, a 100 MΩ resistor was serially connected to the sensor, and the partial voltage atthe midpoint was monitored using the GPIO pin of the portable data logger. Figure 7bshows the effects of RH on the GPIO output, revealing that the developed system was wellsuited for the detection of highly humid air. When the sensor resistance is much larger than100 MΩ (RH < 60% in this device), the circuit shows a constant value of approximately zeroof the GPIO output.Subsequently, we tested the ability of our portable system to monitor water conden-sation in ambient air. Figure 8a shows the effects of RH alteration on the GPIO output.RH was varied in the range (70–95%) where the sensor had sufficient sensitivity. Withincreasing RH, the resistance of the sensor decreased, and the sensor output thereforeincreased. Figure 8b shows the optical images of the sensor surface captured at differentRH. According to the Kelvin equation (Figure 1b), water molecules were physisorbed onthe nanoparticle surface at 70% RH, and liquid water formed by capillary condensationinside the voids between the nanoparticles at above 80% RH. Notably, no macroscopicwater droplets were detected across the electrodes (10–20 µm). After dehumidification to25% RH, the condensed water re-evaporated, and the sensor output recovered to its initialstate. Figure 9a–c show the optical images of the sensor surface captured at 70, 95 and 100%Chemosensors 2023, 11, 564 10 of 14RH. At 100% RH (saturation), the condensed water could be macroscopically formed asdew droplets on the nanoparticle film. The line and spacing of the electrodes were 20 µmin this case. When the water droplets grew over the size, the resistance largely decreased to105 Ω. The size of the water droplets in Figure 9c is approximately 10 to 20 µm, which canbe observable in the optical microscope. It should be noted that the photo in Figure 9c is thesnapshot of the growing water droplets. If we wait for a long time at 100% RH, the dropletsare gradually coalesced, and the surface will be covered with a large amount of water.Figure 7. (a) Circuit design of portable system used to monitor water vapor. (b) Effects of RH onGPIO output.Figure 8. Simultaneous monitoring of sensor output and surface. (a) Change in sensor output inresponse to humidity changes over 50 nm nanoparticle surface. (b) Optical microscopy images ofsensor surface captured during humidification and dehumidification. The scale bar is 100 µm.Owing to the emergence of interference fringe, the visible color in Figure 8b changedwith increasing humidity, which was qualitatively explained by considering the change inthe refractive index. The refractive index of a composite of multiple materials is defined bythe composition ratio γ and component refractive indices n. In a nanoparticle film, the totalrefractive index depends on the composition of voids between nanoparticles. When thevoids are filled with liquid water, the refractive index increases because of the exchangeof air (n = 1) for water (n = 1.33). This mechanism forms the basis of a method previouslyproposed for the colorimetric detection of humidity changes [4,45,46]. The abovementionedresult indicates that our system allows for humidity sensing based on both optical andelectrical measurements.Chemosensors 2023, 11, 564 11 of 14Figure 9. Monitoring of macroscopic water condensation. Optical microscopy images of nanoparticlesurface at (a) 70, (b) 95, and (c) 100% RH and illustration of condensation on film surface. The scalebar is 100 µm.Finally, we carried out a feasible study to evaluate the sensor response to water vapor.Water vapor (i.e., 100% of relative humidity) was naturally generated from liquid wateras well as wet materials. Figure 10a shows a photograph of the portable system used fordetecting water vapor. The sensor with the portable logger was intermittently exposedto water vapor generated from water in a beaker as shown in Figure 10b. Water vaporwas cyclically applied to the sensor for ~1 s. In this experiment, the sensor detected watervapor (nearly 100% RH) and relatively dry ambient air (54% RH) alternately. A 10 MΩresistor was used for monitoring the sensor output. In Figure 10c, the sensor outputvoltage rapidly changed during response and recovery. The response (t90) and recovery(t10) times were determined as described in Section 2.1 and equaled 0.17 s on average, withthe corresponding histogram shown in Figure 10d. In Figures 5 and 8, the sensor responseincludes an environmental humidity change over the sensor surface (extrinsic responsetime [47,48]) as well as an intrinsic sensor response (resistance change by the environmentalhumidity change). The response time corresponds to the intrinsic response because humidair is directly applied to the sensor surface. The extrinsic response time is ideally negligible.The limitation of this demonstration is that we could not control humidity of air. Evaluationof response time depending on humidity of air is in our future study. We only did thecalculation of the Kelvin equation and the porosity in the nanoparticle film has not beeninvestigated experimentally. The evaluation of the porosity is important to support ourdiscussion in further study.Figure 10. Demonstration of water vapor detection. (a) Portable water vapor monitor. (b) Schematicof the measurement. (c) Cyclic response of sensor to water vapor. (d) Histogram of response andrecovery times determined by analyzing responses to water vapor.Chemosensors 2023, 11, 564 12 of 144. ConclusionsThe condensation of water vapor in silica nanoparticle films was physicochemicallyanalyzed to shed light on the chemical morphology of water adsorbed on nanomaterials innanoparticle-based impedance-type sensors. By using IR and XAS, we found that watervapor in the air was capillary-condensed in the voids between 10 nm nanoparticles at 50%RH, and residual ice-like water was detected on the nanoparticles in an atmosphere of He,whereas no ice-like water layers remained present after exposure to vacuum. The valueof RH was correlated with the mass of water adsorbed on the nanoparticle film, and alarge change in film impedance was observed above 80% RH, which was ascribed to vaporcondensation. The continuous optical observation of the nanoparticle film surface revealeda colorimetric change in the interference fringe with increasing humidity in the near-surfaceregion. A portable sensor featuring a silica nanoparticle film was designed to detect watervapor. The response/recovery times (0.17 s each) were sufficiently small for the real-timemonitoring of water vapor in automated systems. Thus, this study bridges an importantknowledge gap, revealing the importance of the capillary condensation of water vapor forthe operation of nanoparticle-based sensors, and is expected to inspire further research onthe application of various nanomaterials to vapor sensing.Author Contributions: Conceptualization, S.K., J.K., S.Y. and H.M.; investigation, S.K., J.K. and S.Y.;resources, S.K., J.K. and S.Y.; writing—original draft preparation, S.K.; writing—review and editing,J.K., S.Y. and H.M. All authors have read and agreed to the published version of the manuscript.Funding: This research was supported by the TIA Collaborative Research Program and KAKENHI(Grant No. 18KK0141) from the Japan Society for the Promotion of Science (JSPS).Institutional Review Board Statement: Not applicable.Informed Consent Statement: Not applicable.Data Availability Statement: The data presented in this study are available upon request from thecorresponding author.Acknowledgments: The authors thank K. Hirayama for assistance with the real-time imaging of thesensor surface during the humidification process. This study was approved by the Photon Factory(Approval Nos. 2022PF-G001 and 2023PF-G005).Conflicts of Interest: The authors declare no conflict of interest.References1. Bi, H.; Yin, K.; Xie, X.; Ji, J.; Wan, S.; Sun, L.; Terrones, M.; Dresselhaus, M.S. Ultrahigh humidity sensitivity of graphene oxide. Sci.Rep. 2013, 3, 2714. [CrossRef]2. Borini, S.; White, R.; Wei, D.; Astley, M.; Haque, S.; Spigone, E.; Harris, N.; Kivioja, J.; Ryhänen, T. Ultrafast Graphene OxideHumidity Sensors. ACS Nano 2013, 7, 11166–11173. [CrossRef] [PubMed]3. Zhang, D.; Zong, X.; Wu, Z.; Zhang, Y. Hierarchical Self-Assembled SnS2 Nanoflower/Zn2SnO4 Hollow Sphere Nanohybrid forHumidity-Sensing Applications. ACS Appl. Mater. Interfaces 2018, 10, 32631–32639. [CrossRef] [PubMed]4. Kano, S.; Jarulertwathana, N.; Mohd-Noor, S.; Hyun, J.; Asahara, R.; Mekaru, H. Respiratory Monitoring by Ultrafast HumiditySensors with Nanomaterials: A Review. Sensors 2022, 22, 1251. [CrossRef] [PubMed]5. Farahani, H.; Wagiran, R.; Hamidon, M.N. Humidity sensors principle, mechanism, and fabrication technologies: A comprehen-sive review. Sensors 2014, 14, 7881–7939. [CrossRef]6. Cai, J.; Lv, C.; Aoyagi, E.; Ogawa, S.; Watanabe, A. Laser Direct Writing of a High-Performance All-Graphene Humidity SensorWorking in a Novel Sensing Mode for Portable Electronics. ACS Appl. Mater. Interfaces 2018, 10, 23987–23996. [CrossRef]7. Choi, S.; Yu, H.; Jang, J.; Kim, M.-H.; Kim, S.; Jeong, H.S.; Kim, I.-D. Nitrogen-Doped Single Graphene Fiber with Platinum WaterDissociation Catalyst for Wearable Humidity Sensor. Small 2018, 14, 1703934. [CrossRef]8. An, H.; Habib, T.; Shah, S.; Gao, H.; Patel, A.; Echols, I.; Zhao, X.; Radovic, M.; Green, M.J.; Lutkenhaus, J.L. Water Sorption inMXene/Polyelectrolyte Multilayers for Ultrafast Humidity Sensing. ACS Appl. Nano Mater. 2019, 2, 948–955. [CrossRef]9. Li, B.; Tian, Q.; Su, H.; Wang, X.; Wang, T.; Zhang, D. High sensitivity portable capacitive humidity sensor based on In2O3nanocubes-decorated GO nanosheets and its wearable application in respiration detection. Sens. Actuators B Chem. 2019, 299,126973. [CrossRef]10. Zhou, C.; Zhang, X.; Tang, N.; Fang, Y.; Zhang, H.; Duan, X. Rapid response flexible humidity sensor for respiration monitoringusing nano-confined strategy. Nanotechnology 2020, 31, 125302. [CrossRef]https://doi.org/10.1038/srep02714https://doi.org/10.1021/nn404889bhttps://www.ncbi.nlm.nih.gov/pubmed/24206232https://doi.org/10.1021/acsami.8b08493https://www.ncbi.nlm.nih.gov/pubmed/30176720https://doi.org/10.3390/s22031251https://www.ncbi.nlm.nih.gov/pubmed/35161997https://doi.org/10.3390/s140507881https://doi.org/10.1021/acsami.8b07373https://doi.org/10.1002/smll.201703934https://doi.org/10.1021/acsanm.8b02265https://doi.org/10.1016/j.snb.2019.126973https://doi.org/10.1088/1361-6528/ab5cdaChemosensors 2023, 11, 564 13 of 1411. Mekawy, M.; Noguchi, H.; Kawakita, J. Quantitative and qualitative studies for real monitoring of interfacial molecular water. J.Colloid Interface Sci. 2022, 613, 311–319. [CrossRef] [PubMed]12. Kubota, Y.; Mishra, V.L.; Ando, T.; Sakamoto, Y.; Kawakita, J. Micro/nano galvanic-coupled arrays for early and initial detectionand prediction of dew condensation. Sens. Actuators A Phys. 2020, 303, 111838. [CrossRef]13. Kano, S.; Mekaru, H. Nonporous Inorganic Nanoparticle-Based Humidity Sensor: Evaluation of Humidity Hysteresis andResponse Time. Sensors 2020, 20, 3858. [CrossRef] [PubMed]14. Kano, S.; Mekaru, H. Capillary-condensed water in nonporous nanoparticle films evaluated by impedance analysis for nanoparti-cle devices. Nanotechnology 2020, 31, 455701. [CrossRef]15. Fisher, L.R.; Gamble, R.A.; Middlehurst, J. The Kelvin equation and the capillary condensation of water. Nature 1981, 290, 575–576.[CrossRef]16. Fisher, L.R.; Israelachvili, J.N. Direct experimental verification of the Kelvin equation for capillary condensation. Nature 1979, 277,548–549. [CrossRef]17. Yang, Q.; Sun, P.Z.; Fumagalli, L.; Stebunov, Y.V.; Haigh, S.J.; Zhou, Z.W.; Grigorieva, I.V.; Wang, F.C.; Geim, A.K. Capillarycondensation under atomic-scale confinement. Nature 2020, 588, 250–253. [CrossRef]18. Baumgartner, B.; Hayden, J.; Loizillon, J.; Steinbacher, S.; Grosso, D.; Lendl, B. Pore Size-Dependent Structure of Confined Waterin Mesoporous Silica Films from Water Adsorption/Desorption Using ATR-FTIR Spectroscopy. Langmuir 2019, 35, 11986–11994.[CrossRef]19. Cappelli, I.; Fort, A.; Lo Grasso, A.; Panzardi, E.; Mugnaini, M.; Vignoli, V. RH sensing by means of TiO2 nanoparticles: Acomparison among different sensing techniques based on modeling and chemical/physical interpretation. Chemosensors 2020, 8,89. [CrossRef]20. Asay, D.B.; Kim, S.H. Evolution of the Adsorbed Water Layer Structure on Silicon Oxide at Room Temperature. J. Phys. Chem. B2005, 109, 16760–16763. [CrossRef]21. Asay, D.B.; Barnette, A.L.; Kim, S.H. Effects of surface chemistry on structure and thermodynamics of water layers at solid-vaporinterfaces. J. Phys. Chem. C 2009, 113, 2128–2133. [CrossRef]22. Chen, L.; He, X.; Liu, H.; Qian, L.; Kim, S.H. Water Adsorption on Hydrophilic and Hydrophobic Surfaces of Silicon. J. Phys.Chem. C 2018, 122, 11385–11391. [CrossRef]23. Seo, M.-H.; Yang, H.-H.; Choi, K.-W.; Lee, J.-S.; Yoon, J.-B. A simple breathing rate-sensing method exploiting a temporarilycondensed water layer formed on an oxidized surface. Appl. Phys. Lett. 2015, 106, 053701. [CrossRef]24. Su, P.G.; Huang, L.N. Humidity sensors based on TiO2 nanoparticles/polypyrrole composite thin films. Sens. Actuators B Chem.2007, 123, 501–507. [CrossRef]25. Garcia-Belmonte, G.; Kytin, V.; Dittrlch, T.; Bisquert, J. Effect of humidity on the ac conductivity of nanoporous TiO2. J. Appl. Phys.2003, 94, 5261–5264. [CrossRef]26. Kuang, Q.; Lao, C.; Wang, Z.L.; Xie, Z.; Zheng, L. High-Sensitivity Humidity Sensor Based on a Single SnO2 Nanowire. J. Am.Chem. Soc. 2007, 129, 6070–6071. [CrossRef]27. Seiyama, T.; Yamazoe, N.; Arai, H. Ceramic humidity sensors. Sens. Actuators 1983, 4, 85–96. [CrossRef]28. Hudson, D.R. Density and packing in an aggregate of mixed spheres. J. Appl. Phys. 1949, 20, 154–162. [CrossRef]29. Delenne, J.; Richefeu, V.; Radjai, F. Liquid clustering and capillary pressure in granular media. J. Fluid Mech. 2015, 762, R5.[CrossRef]30. Last, B.J.; Thouless, D.J. Percolation theory and electrical conductivity. Phys. Rev. Lett. 1971, 27, 1719–1721. [CrossRef]31. Ewing, G.E. Ambient thin film water on insulator surfaces. Chem. Rev. 2006, 106, 1511–1526. [CrossRef] [PubMed]32. Brubach, J.B.; Mermet, A.; Filabozzi, A.; Gerschel, A.; Lairez, D.; Krafft, M.P.; Roy, P. Dependence of water dynamics uponconfinement size. J. Phys. Chem. B 2001, 105, 430–435. [CrossRef]33. Torun, B.; Kunze, C.; Zhang, C.; Kühne, T.D.; Grundmeier, G. Study of water adsorption and capillary bridge formation for SiO2nanoparticle layers by means of a combined in situ FT-IR reflection spectroscopy and QCM-D set-up. Phys. Chem. Chem. Phys.2014, 16, 7377–7384. [CrossRef] [PubMed]34. Kreuer, K.D. Proton conductivity: Materials and applications. Chem. Mater. 1996, 8, 610–641. [CrossRef]35. Hassanali, A.; Giberti, F.; Cuny, J.; Kühne, T.D.; Parrinello, M. Proton transfer through the water gossamer. Proc. Natl. Acad. Sci.USA 2013, 110, 13723–13728. [CrossRef]36. Agmon, N. The Grotthuss mechanism. Chem. Phys. Lett. 1995, 244, 456–462. [CrossRef]37. Chen, Z.; Lu, C. Humidity Sensors: A Review of Materials and Mechanisms. Sens. Lett. 2005, 3, 274–295. [CrossRef]38. Kano, S.; Mekaru, H. Proton transport over nanoparticle surface in insulating nanoparticle film-based humidity sensor. Jpn. J.Appl. Phys. 2022, 61, SE1011. [CrossRef]39. Fransson, T.; Harada, Y.; Kosugi, N.; Besley, N.A.; Winter, B.; Rehr, J.J.; Pettersson, L.G.M.; Nilsson, A. X-ray and ElectronSpectroscopy of Water. Chem. Rev. 2016, 116, 7551–7569. [CrossRef]40. Sellberg, J.A.; Kaya, S.; Segtnan, V.H.; Chen, C.; Tyliszczak, T.; Ogasawara, H.; Nordlund, D.; Pettersson, L.G.M.; Nilsson, A.Comparison of X-ray absorption spectra between water and ice: New ice data with low pre-edge absorption cross-section. J.Chem. Phys. 2014, 141, 034507. [CrossRef]41. Petit, T.; Yuzawa, H.; Nagasaka, M.; Yamanoi, R.; Osawa, E.; Kosugi, N.; Aziz, E.F. Probing Interfacial Water on Nanodiamonds inColloidal Dispersion. J. Phys. Chem. Lett. 2015, 6, 2909–2912. [CrossRef] [PubMed]https://doi.org/10.1016/j.jcis.2022.01.025https://www.ncbi.nlm.nih.gov/pubmed/35042031https://doi.org/10.1016/j.sna.2020.111838https://doi.org/10.3390/s20143858https://www.ncbi.nlm.nih.gov/pubmed/32664356https://doi.org/10.1088/1361-6528/aba7e0https://doi.org/10.1038/290575a0https://doi.org/10.1038/277548a0https://doi.org/10.1038/s41586-020-2978-1https://doi.org/10.1021/acs.langmuir.9b01435https://doi.org/10.3390/chemosensors8040089https://doi.org/10.1021/jp053042ohttps://doi.org/10.1021/jp806815phttps://doi.org/10.1021/acs.jpcc.8b01821https://doi.org/10.1063/1.4906815https://doi.org/10.1016/j.snb.2006.09.052https://doi.org/10.1063/1.1610805https://doi.org/10.1021/ja070788mhttps://doi.org/10.1016/0250-6874(83)85012-4https://doi.org/10.1063/1.1698327https://doi.org/10.1017/jfm.2014.676https://doi.org/10.1103/PhysRevLett.27.1719https://doi.org/10.1021/cr040369xhttps://www.ncbi.nlm.nih.gov/pubmed/16608189https://doi.org/10.1021/jp002983shttps://doi.org/10.1039/C3CP54912Ghttps://www.ncbi.nlm.nih.gov/pubmed/24623070https://doi.org/10.1021/cm950192ahttps://doi.org/10.1073/pnas.1306642110https://doi.org/10.1016/0009-2614(95)00905-Jhttps://doi.org/10.1166/sl.2005.045https://doi.org/10.35848/1347-4065/ac4b0ehttps://doi.org/10.1021/acs.chemrev.5b00672https://doi.org/10.1063/1.4890035https://doi.org/10.1021/acs.jpclett.5b00820https://www.ncbi.nlm.nih.gov/pubmed/26267179Chemosensors 2023, 11, 564 14 of 1442. Zhuravlev, L.T. The surface chemistry of amorphous silica. Colloids Surf. A Physicochem. Eng. Asp. 2000, 173, 1–38. [CrossRef]43. Yao, Y.; Chen, X.; Guo, H.; Wu, Z. Graphene oxide thin film coated quartz crystal microbalance for humidity detection. Appl. Surf.Sci. 2011, 257, 7778–7782. [CrossRef]44. Hakimian, A.; Mohebinia, M.; Nazari, M.; Davoodabadi, A.; Nazifi, S.; Huang, Z.; Bao, J.; Ghasemi, H. Freezing of few nanometerswater droplets. Nat. Commun. 2021, 12, 6973. [CrossRef]45. Mohd-Noor, S.; Jang, H.; Baek, K.; Pei, Y.-R.; Alam, A.-M.; Kim, Y.H.; Kim, I.S.; Choy, J.-H.; Hyun, J.K. Ultrafast humidity-responsive structural colors from disordered nanoporous titania microspheres. J. Mater. Chem. A 2019, 7, 10561–10571. [CrossRef]46. Kim, S.; Han, S.G.; Koh, Y.G.; Lee, H.; Lee, W. Colorimetric humidity sensor using inverse opal photonic gel in hydrophilic ionicliquid. Sensors 2018, 18, 1357. [CrossRef]47. Ménil, F.; Susbielles, M.; Debéda, H.; Lucat, C.; Tardy, P. Evidence of a correlation between the non-linearity of chemical sensorsand the asymmetry of their response and recovery curves. Sens. Actuators B Chem. 2005, 106, 407–423. [CrossRef]48. Steele, J.J.; Taschuk, M.T.; Brett, M.J. Response time of nanostructured relative humidity sensors. Sens. Actuators B Chem. 2009,140, 610–615. [CrossRef]Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individualauthor(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury topeople or property resulting from any ideas, methods, instructions or products referred to in the content.https://doi.org/10.1016/S0927-7757(00)00556-2https://doi.org/10.1016/j.apsusc.2011.04.028https://doi.org/10.1038/s41467-021-27346-whttps://doi.org/10.1039/C9TA01394Fhttps://doi.org/10.3390/s18051357https://doi.org/10.1016/j.snb.2004.08.027https://doi.org/10.1016/j.snb.2009.05.016 Introduction  Materials and Methods  Theory  Experimental Methods  Results and Discussion  IR Spectroscopy and XAS Measurements  QCM and Electrical Sensor Measurements  Water Vapor Detection by Portable System  Conclusions  References